Knowledge Management 9780765616371

KNOWLEDGE
MANAGEMENT
AN EVOLUTIONARY VIEW
Becerra-FernandezTitleHalf.qxd 5/5/2008 10:01 AM Page 2
Advances in Management Information Systems
Advisory Board
Eric K. Clemons
University of Pennsylvania
Thomas H. Davenport
Accenture Institute for Strategic Change
and
Babson College
Varun Grover
Clemson University
Robert J. Kauffman
University of Minnesota
Jay F. Nunamaker, Jr.
University of Arizona
Andrew B. Whinston
University of Texas
ADVANCES IN MANAGEMENT
I NFORMATI ON SYSTEMS
VLADIMIR ZWASS SERIES EDITOR
AMS
M.E.Sharpe
Armonk, New York
London, England
IRMA BECERRA-FERNANDEZ
DOROTHY LEIDNER
EDITORS
KNOWLEDGE
MANAGEMENT
Copyright 2008 by M.E. Sharpe, Inc.
All rights reserved. No part of this book may be reproduced in any form
without written permission from the publisher, M.E. Sharpe, Inc.,
80 Business Park Drive, Armonk, New York 10504.
Library of Congress Cataloging-in-Publication Data
References to the AMIS papers should be as follows:
Maryam Alavi and Gerald C. Kane. Social networks and information technology: Evolution and new frontiers.
Irma Becerra-Fernandez and Dorothy Leidner, eds., Knowledge Management: An Evolutionary View. Advances
in Management Information System. Vol. 12 (Armonk, NY: M.E. Sharpe, 2008), 6385.
ISBN 978-0-7656-1637-1
ISSN 1554-6152
Printed in the United States of America
The paper in this publication meets the minimum requirements of
American National Standards for Information Sciences
Permanence of Paper for Printed Library Materials,
ANSI Z 39.48-1984.
~
IBT (c) 10 9 8 7 6 5 4 3 2 1
ADVANCES IN MANAGEMENT INFORMATION SYSTEMS
AMIS Vol. 1: Richard Y. Wang, Elizabeth M. Pierce,
Stuart E. Madnick, and Craig W. Fisher
Information Quality
ISBN 9780-765611338
AMIS Vol. 2: Sergio deCesare, Mark Lycett, and
Robert D. Macredie
Development of Component-Based Information Systems
ISBN 9780-765612489
AMIS Vol. 3: Jerry Fjermestad and Nicholas
C. Romano, Jr.
Electronic Customer Relationship Management
ISBN 9780-76561327-1
AMIS Vol. 4: Michael J. Shaw
E-Commerce and the Digital Economy
ISBN 9780-76561150-5
AMIS Vol. 5: Ping Zhang and Dennis Galletta
Human-Computer Interaction and Management
Information Systems: Foundations
ISBN 9780-765614865
AMIS Vol. 6: Dennis Galletta and Ping Zhang
Human-Computer Interaction and Management
Information Systems: Applications
ISBN 9780-765614872
AMIS Vol. 7: Murugan Anandarajan, Thompson S.H.
Teo, and Claire A. Simmers
The Internet and Workplace Transformation
ISBN 9780-765614452
AMIS Vol. 8: Suzanne Rivard and Benoit Aubert
Information Systems Sourcing
ISBN 9780-765616852
AMIS Vol. 9: Varun Grover and M. Lynne Markus
Business Process Transformation
ISBN 9780-765611918
AMIS Vol. 10: Panos E. Kourouthanassis and George
M. Giaglis
Pervasive Information Systems
ISBN 9780-765616890
AMIS Vol. 11: Detmar W. Straub, Seymour Goodman,
and Richard Baskerville
Information Security Policy and Practices
ISBN 9780-765617187
AMIS Vol. 12: Irma Becerra-Fernandez and Dorothy
Leidner
Knowledge Management: An Evolutionary View
ISBN 9780-765616371
Forthcoming volumes of this series can be found on the series homepage.
www.mesharpe.com/amis.htm
Editor in Chief, Vladimir Zwass (zwass@fdu.edu)
v
CONTENTS
Series Editors Introduction
Vladimir Zwass vii
Preface xiii
1. On Knowledge, Knowledge Management, and Knowledge Management Systems:
An Introduction
Irma Becerra-Fernandez and Dorothy Leidner 3
Part I. A Conceptual Lens for Knowledge Management 11
2. Individual, Group, and Organizational Learning: A Knowledge
Management Perspective
Irma Becerra-Fernandez and Rajiv Sabherwal 13
3. Knowledge Management and Organizational Culture
Dorothy E. Leidner and Timothy R. Kayworth 40
Part II. The Role of Information Technology in Knowledge Management 61
4. Social Networks and Information Technology: Evolution and New Frontiers
Maryam Alavi and Gerald C. Kane 63
5. The Evolution of Knowledge Management Technology: From Explicit
Rules to Implicit Profles
Ulrike Schultze 86
6. A Four-Layer Model for Information Technology Support of Knowledge Management
Matteo Bonifacio, Thomas Franz, and Steffen Staab 104
Part III. The Role of Knowledge Management Systems in the Organization 125
7. Mobilizing Knowledge in a Yu-Gi-Oh! World
Youngjin Yoo 127
8. The Impact of Computer-Mediated Communication on Knowledge Transfer
and Organizational Form
David G. Schwartz and Dov Teeni 145
vi CONTENTS
9. Moving Toward a Knowledge Management Maturity Model (K3M) for
Developing Knowledge Management Strategy and Implementation Plans
Jay Liebowitz and Tom Beckman 163
10. Building Knowledge Management Systems to Improve Profts and Create
Loyal Users: Lessons from the Pharmaceutical Industry
Alan R. Dennis, Dong-Gil Ko, and Paul F. Clay 180
11. Can We Learn from Our Past? Managing Knowledge Within and Across Projects
Patrick S. W. Fong 204
Part IV. Knowledge Management and Teams Within and Across Organizations 227
12. Managing Knowledge in Virtual Communities Within Organizations
Naren B. Peddibhotla and Mani R. Subramani 229
13. Dynamic Team Memory Systems: Enabling Knowledge Sharing Effectiveness
in Structurally Diverse Distributed Teams
Ann Majchrzak, Arvind Malhotra, and Richard John 248
14. Electronic Knowledge Networks: Processes and Structure
Samer Faraj, Molly McLure Wasko, and Steven L. Johnson 270
15. Ad Hoc Interorganizational Collaboration: Safeguards for Balancing Sharing
and Protection of Knowledge
Yongsuk Kim, Sirkka L. Jarvenpaa, and Ann Majchrzak 292
Editors and Contributors 309
Series Editor 317
Index 319
vii
SERIES EDITORS INTRODUCTION
Vladimir Zwass, Editor-in-ChiEf
We live in a knowledge economy. National prosperity, employment and employability, and corporate
and personal success all depend on the ability of societal units and individuals to assimilate the
existing knowledge, generate new knowledge, and bring knowledge into action. These capabilities
determine a countrys progress in the globalizing world, a frms success in the heightening global
competition, a teams effectiveness in an enterprise, and an individuals chance to become a leader
within his or her domain of action. It is no longer suffcient to stand on the shoulders of giants: we
need to run very fast and fnd ever new giants to shoulder the innovation. The exponential growth of
scientifc and practical knowledge has been taking place for over two centuries, since the Industrial
Revolution (Mokyr, 2002). Over the past half-century, this process has accelerated owing to the
information and communication technologies, acting as tools, media, and the complements of the
human mind. The accessibility of the existing knowledge, and that of the individuals who create
it and who know, has increased vastly. This has, in its turn, amplifed the ability to learn and to
create new knowledge. Since the early 1990s, the InternetWeb compound has further increased
the velocity of knowledge creation and application and, together with the political change, has
truly globalized the power of knowledge to change human pursuits and circumstances.
The pioneer who recognized and conceptualized this shift was Fritz Machlup (1962). Without
identifying at that time the distinction between knowledge and information, he computed that in
the advanced industrial economy of the United States half a century ago, its sectors related to these
entities accounted for 29% of the Gross National Product and engaged 31% of labor force. We
have since come to realize that knowledge-related activities can no longer be compartmentalized.
They are not limited to certain sectors of the economy, organizational units (such as research and
development), or professions. In our society and in our organizations, knowledge is everybodys
business. A mathematician will prove a theorem that defed proof for a century and that will
leadafter a period that cannot be determineda software engineer to design a more semanti-
cally rich Web-search algorithm, resulting in increased revenue in B2C e-commerce. With her
experiential knowledge, a fork-lift operator will come up with the superior method of stacking
pallets, resulting in the better utilization of warehouse space, faster operation of container ports,
and access to new markets.
In consequence, knowledge management (KM) has been recognized as a source of vital or-
ganizational capabilities, leading to a successful competitive positioning. Pursuing the general
objectives of the Advances in Management Information Systems, the present AMIS volume shows
both the advances in our knowledge about KM and the research methods deployed to generate this
knowledge. The volume has been edited by two leading experts in KM, with contributions from
the top scholars in the feld. It remains to me here to provide additional context.
What is knowledge? How does this seemingly abstract concept relate to the pragmatics of
viii SERIES EDITORS INTRODUCTION
organizations? What is KM? It is, of course, well to say that knowledge cannot be managed,
butas the volume shows and as I argue herethe ability of an organization to know what it
knows and to be able to apply it is a necessary and generalized capability.
Since Plato, a common defnition of knowledge has crystallized as that of justifed true belief.
The important part of this, rather unsatisfying, defnition is the need for evidence as the measure
of truth. More recent epistemological work indeed centers on what external evidence is needed
for the belief to be justifed. The evidence that warrants scientifc knowledge differs from that
underwriting the more pragmatic organizational knowledge types, yet in all cases warrants for
truth are necessary.
In our approach to KM, it is important to distinguish knowledge from information. Knowledge
is an organized structure of facts, relationships, experience, skills, and insights that produces a
capacity for action. It follows that some knowledge inheres in individuals, while some has been
codifed to transcend the limitationsand advantagesof individual minds. The degrees of or-
ganization, the validity of knowledge, and its infuence on actions differ among individuals. (The
distinction between knowledge and an accumulation of facts has received its unrivaled depiction
in the last, unfnished, novel of Gustave Flaubert, Bouvard and Pcuchet.) With advances in in-
formation technology (IT) and telecommunications, the capacity and degree of organization and
accessibility of external knowledge stores, combined with the means of communication among
knowers, have made the effective knowledge a potent economic resource (Foray, 2004). Informa-
tion is an increment to knowledge, meaningful to its individual recipient. This increment may
become another node in a conceptual schema, affect the contents of a node, or, indeed, cause a
structural change in the overall knowledge framework. The distinctions between knowledge and
information are important, as they determine the difference in our approach to KM compared with
information management.
Three distinctions appear most salient. Knowledge is always provisional as an aggregate
whether in a human mind, in a technical handbook, as an organizational knowledge base, or in a
science. This goes to the very core of human understanding of the world. The gravity in the Newto-
nian physics was a force drawing us to the planets surface; in the Einsteinian physics, it is a warp,
induced by Earths mass. Biology leavened by the discoveries of genetics has become a largely
different discipline. Obviously, in the aggregate, we do not know what we do not know.
By far, not all knowledge is explicit or codifed in the great variety of repositories. An important
part of knowledge is personal (Polanyi, 1962). It inheres in the knowing individual, is tacit, and is
diffcult to access by others. A superb diagnostician, an experienced geologist, a recognized pianist,
a successful commodities trader, or a fully skilled chocolate maker with 40 years on the job is not
replaceable by a knowledge system. Techniques of sharing and articulating tacit knowledge have
been offered, along with a stepwise progression in the creation of new knowledge in organizations
(Nonaka and Takeuchi, 1995). However, much of knowledge is gained with experience and remains
tacit. This is the more sticky knowledge that is at the source of the competitive advantage of the
entities to which it sticks. Beyond that, all knowledge is accessible to organization through the
mediation of individuals with knowledge. An individuals assessment of the context, interpretation
of the new information, recognition of an analogy or a mental template, and capability to use the
knowledge to actualize a new state of affairs make all the difference in performance.
Further, knowledge is always in a certain sense collective. The validation of knowledge compo-
nents, including mathematical proofs, is a social process. The deployment of individual knowledge
in a frm occurs within the context of a team or another social group that can amplify, attenuate,
complement, enhance, elaborate, distort, or articulate what an individual can accomplish. Knowl-
edge sharing and collective reinterpretation, particularly across the organizational boundaries, are
SERIES EDITORS INTRODUCTION ix
key to innovation (Lester and Piore, 2004). Informal exchanges across disciplinary domains are at
the source of a great variety of product and process innovations. Communities of practice have been
recognized as the bearers and generators of collective knowledge (Wenger and Snyder, 2000).
Knowledge management has to be conducted in full recognition of the above distinctions. Only
a limited part of the overall task of KM can be automated; we cannot meaningfully talk about
KM systems in the same sense as we speak about information-processing systems. We may, in
a sense, speak about symbiosis of people and IT, as Licklider (1960) did in an early visionary
article. Beyond that, collective processes need to be supported to share, elaborate, converse about,
and debate. It is about connecting people across space, time, and disciplinary or organizational
borders. Yet the people can be supported by the systems that make available the externally stored
knowledge, help circulate electronic boundary objects, and store the activity traces as desired.
Vigilant IT-supported processes have to keep the current knowledge up to date, even as the histori-
cal components are archived. The apportioning of cognitive tasks between people and IT changes
as we learn more about KM.
The distinctive properties of knowledge shape the general contours of KM as the general hu-
man pursuit in organizations, with the ever more comprehensive assistance from IT. KM is the
organizational policy and set of practices aimed at recognizing, creating, categorizing, maintaining,
sharing, and applying the collective knowledge of people assisted by IT. The corpora of explicit
knowledge that can have impact on organizational performance are expanding rapidly, with much
of it stored outside of the frm. Extensive deployment of IT is necessary to cope with the exist-
ing knowledge and to generate new knowledge. Organizational knowledge includes not only the
technologies needed for its production and product innovation. It also encompasses the process and
administrative knowledge, the relationship knowledge about its customers, suppliers, and partners,
fnancial knowledge specialized to its businessand so much more. Organizations products can
be seen to containor even becongealed knowledge (Arthur, 1996). This is particularly so with
more advanced software, containing as it does knowledge about a specifc domain. New business
models and organizational forms congeal around knowledge. As one example, Cobalt International
Company of Houstonwith its 35 employeesis in the business of deepwater oil exploration, a
traditional bailiwick of large multinationals (Mouawad, 2007). The frm buys immense volumes
of seismic data to be analyzed by IT-assisted experts. Cobalt outsources everything other than its
core, which consists of the knowledge of the highly experienced geologists and geophysicists,
minimizing the number of dry wells that would be drilled at a great expense and to no avail. A
large virtual company has been created around a small knowledge-based core.
Organizational KM can be seen sparsely within a framework of four process sets, aiming at
creation, storage/retrieval, transfer, and application of knowledge (Alavi and Leidner, 2001).
Another way to perceive KM is as the management of the organizational memory, assisted by
an organizational memory information system (OMIS) that supports the fundamental activities
leading to organizational effectiveness (Stein and Zwass, 1995). Organizational memory is de-
fned in this conceptualization as the means by which knowledge of the past is brought to bear
on present activities (Stein and Zwass, 1995, p. 89). Electronic traces left today by most orga-
nizational performances can be categorized, structured, and organized to assist knowledge work
in the future. This and all other knowledge are organized around the frms experience. Thus, the
presence of an electronic product prototype will certainly assist in the development of the future
product; but this assistance will be so much more fruitful in innovation if the design rationale had
been provided by storing the narratives and partial designs as they evolved into the fnal design
solution. As contexts and constraints change in the future, so will the design rationale, and so will
the product. By consistent management of the experiential knowledge of people and in electronic
x SERIES EDITORS INTRODUCTION
repositories, the mechanisms for assimilating and renewing corporate knowledge are enacted.
Such conceptualization evokes different cultural phenomena and different habits of performance
(Ricoeur, 2004). An OMIS may also be incorporated within the larger scheme of KM.
The developments in IT, particularly since the assimilation of the InternetWeb compound,
have enabled a far more comprehensive support for KM. Search engines, social networking, in-
tranets, wikis, and myriad other tools have joined expert systems, case bases, and document- and
content-management systems on the widespread organizational and interorganizational platforms.
Knowledge discovery and elaboration through data mining, elicitation from experts, case-based
reasoning, collaborative work with groupware systems, and systems for the aggregation of collec-
tive wisdom are supported by a variety of electronic repositories, as often as not distributed around
the world. However, as constantly stressed in this volume, it is the human thought and action, the
organizational culture and processes, and incentive alignment that are at the core of KM.
As you will conclude in reading this volume, the research domain of KM in the overall manage-
ment information systems research ranges widely, refecting the importance of KM for organiza-
tional well-being and competitive posture. The work published here represents the great variety
of research streams contributed to the overall KM domain. Our understanding of KM strategies
is being systematized (Earl, 1991). Motivation to share ones knowledge within an organization
generally does not come naturally and motivational schemes are being studied (Quigley et al.,
2007). Deep interdependencies exist between the enactment of KM practices and organizational
culture (Alavi, Kayworth, and Leidner, 20052006). Knowledge-based approaches are being ap-
plied to the coordination of globally distributed project teams (Espinosa et al., 2007). The methods
of disciplined expertise location over the Web are being explored (Becerra-Fernandez, 2006). A
design theory for systems aimed to support knowledge processes has been proposed (Markus,
Majchrzak, and Gasser, 2002). The measures of success of organizational KM are being studied
(Kulkarni, Ravindran, and Freeze, 2007). The chapters of the volume are a worthy contribution
to and an illustration of this research tradition in formation.
Looking forward, we will study the changing conception of knowledge production and produc-
tive assimilation as the realm of the possible is expanding and as boundaries blur. The warrant for
scientifc knowledge now includes computer-based experimentation. Corporations are a vital source
of scientifc knowledge and innovation. Universities produce extensive intellectual property. The
management of knowledge will be an ever more weighty undertaking, undergirded by IT.
REFERENCES
Alavi, M.; Kayworth, T.R.; and Leidner D.E. 20052006. An empirical examination of the infuence of or-
ganizational culture on knowledge management practices. Journal of Management Information Systems,
22, 3 (Winter), 191224.
Alavi, M., and Leidner, D.E. 2001. Knowledge management and knowledge management systems: Conceptual
foundations and research issues. MIS Quarterly 25, 1 (March), 107136.
Arthur, W.B. 1996. Increasing returns and the new world of business. Harvard Business Review, July-August,
101109.
Becerra-Fernandez, I. 2006. Searching for experts on the Web: A review of contemporary expertise locator
systems. ACM Transactions on Internet Technology 6, 4 (November), 333355.
Earl, M. 2001. Knowledge management strategies: Toward a taxonomy. Journal of Management Information
Systems 18, 1 (Summer), 215233.
Espinosa, J.A.; Slaughter, S.A.; Kraut, R.E.; and Herbsleb, J.D. 2007 Team knowledge and coordination
in geographically distributed software development. Journal of Management Information Systems 24, 1
(Summer), 135169.
Foray, D. 2004. Economics of Knowledge. Cambridge, MA: MIT Press.
Kulkarni, U.; Ravindran, S.; and Freeze, R. 2007. A knowledge management success model: Theoretical
SERIES EDITORS INTRODUCTION xi
development and empirical validation. Journal of Management Information Systems 23, 3 (Winter),
309347.
Lester, R.K., and Piore, M.J. 2004. Innovation: The Missing Dimension. Cambridge, MA: Harvard Univer-
sity Press.
Licklider, J.C.R. 1960. Man-computer symbiosis. IRE Transactions on Human Factors in Electronics, HFE-
1, March, 411.
Machlup, F. 1962. The Production and Distribution of Knowledge in the United States. Princeton, NJ:
Princeton University Press.
Markus, M.L.; Majchrzak, A.; and Gasser, L. 2002. A design theory for systems that support emergent
knowledge processes. MIS Quarterly, 26, 3, 179212.
Mokyr, J. 2002. The Gifts of Athena: Historical Origins of the Knowledge Economy. Princeton, NJ: Princeton
University Press.
Mouawad, J. 2007. A wildcatter pounces. The New York Times, Sunday Business Section, May 20, BU-1,
BU7.
Nonaka, I., and Takeuchi, H. 1995. The Knowledge-Creating Company: How Japanese Companies Create
the Dynamics of Innovation. New York: Oxford University Press.
Polanyi, M. 1962. Personal Knowledge: Towards a Post-Critical Philosophy (corr. ed.). Chicago: University
of Chicago Press.
Quigley, N.R.; Tesluk, P.E.; Locke, E.A.; and Bartol, K.M. 2007. A multilevel investigation of the motiva-
tional mechanisms underlying knowledge sharing and performance. Organization Science, 18, 1 (Janu-
ary-February), 7188.
Ricoeur, P. 2004. Memory, History, Forgetting. Chicago: University of Chicago Press.
Stein, E.W., and Zwass, V. 1995. Actualizing organizational memory with information systems. Information
Systems Research 6, 2, 85117.
Wenger, E., and Snyder, W.M. 2000. Communities of practice: The organizational frontier. Harvard Business
Review (January-February), 139145.
xiii
PREFACE
This research volume offers a collection of refereed research chapters in the feld of knowledge
management (KM). Written by researchers who have made signifcant contributions to the feld
of KM and KM systems (KMS) in the past several years, each chapter seeks to synthesize how
research perspectives on a particular topic in KM have progressed over the years, thereby illustrating
the dynamic and evolutionary nature of KM theories and documenting the intellectual milestones
of KM. In presenting how thinking about KM has evolved over a period of time, the chapters
elucidate how ideas have moved from being novel to being commonly accepted, how seemingly
obvious truths have turned out to be more nuanced and multifaceted over time, how new lines of
KM research have emerged from the shortcomings of other lines, and how the presumed purpose
of KM (and thus, KMS) has changed over time.
In short, we hope that this volume will become a key resource by presenting the journey of KM
over time, which is signifcantly different than the time-bound perspectives that peer-reviewed
articles typically present.
We would frst like to thank all the chapter authors, who contributed their knowledge and time
to the preparation of the chapters in this volume, and who also assisted in the review of their
peers contributions. Much like in KMS, the authors of this volume are the mind and spirit of
this knowledge base. We would also like to thank our editor-in-chief of the Advances in Manage-
ment Information Systems (AMIS) series, Vladimir Zwass, for all his advice and support during
the development of this monograph. We are also thankful to Harry Briggs, executive editor, and
Elizabeth Granda, associate editor, both of M.E. Sharpe. Finally, we want to thank our families,
for their love, encouragement, and understanding of the importance of us devoting the necessary
time to the completion of this public good.
KNOWLEDGE
MANAGEMENT
3
ChaptEr 1
ON KNOWLEDGE, KNOWLEDGE
MANAGEMENT, AND KNOWLEDGE
MANAGEMENT SYSTEMS
An Introduction
irma BECErra-fErnandEZ and dorothy lEidnEr
Abstract: The discipline of knowledge management (KM) has evolved over time. This intro-
duction presents the motivation for organizations to continue to pursue the goal of effectively
managing their intellectual resources, even in light of prior failures. This research monograph
offers a collection of ffteen refereed research chapters in the feld of KM. Written by researchers
who have made signifcant contributions to the feld of KM and KM systems in the past several
years, each chapter seeks to synthesize how research perspectives on a particular topic in KM
have progressed over the years, thereby illustrating the dynamic and evolutionary nature of KM
theories and documenting the intellectual milestones of KM. In presenting how thinking about
KM has evolved over a period of time, the chapters elucidate how ideas have moved from being
novel to being commonly accepted, how seemingly obvious truths have turned out to have more
nuances and be more multifaceted over time, how new lines of KM research have emerged from
the shortcomings of other lines, and how the presumed purpose of KM, and thus of KM systems,
has changed over time.
Keywords: Knowledge Management, Knowledge Management Systems
How do we want to leave the world for the next generation? What are good initial condi-
tions for them? One desideratum would be a world offering as many alternatives as pos-
sible to future decision makers, avoiding irreversible commitments that they cannot undo.
. . . A second desideratum is to leave the next generation of decision makers with a better
body of knowledge and a greater capacity for experience. The aim here is to enable them
not just to evaluate alternatives better but especially to experience the world in more and
richer ways.
Herbert A. Simon (1996, p. 163)
INTRODUCTION
In the past few years, KM has been viewed as an increasingly important feld of study that promotes
the creation, capture, sharing, and application of an organizations knowledge. It has been argued
that the most vital resource of todays enterprise is the collective knowledge residing in the minds
4 BECERRA-FERNANDEZ AND LEIDNER
of an organizations employees, customers, and vendors. Learning how to manage organizational
knowledge therefore may produce many benefts, including leveraging core business competencies,
accelerating innovation and time-to-market, improving cycle times and decision-making, strength-
ening organizational commitment, and building sustainable competitive advantage. There is little
doubt that the fundamental challenge of organizing and making available important knowledge has
probably been present in all types of organizations for thousands of years. Yet, this need became
particularly acute in the latter half of the twentieth century, as competitive pressures increasingly
pushed employees in large and geographically distributed organizations to specialize in cognitively
demanding tasks and to integrate their knowledge across location and specialty.
Much of the emphasis in KM has been on knowledge that has been validated and articulated in
some form, including knowledge about processes, procedures, intellectual property, documented
best practices, forecasts, lessons learned, and solutions to recurring problems. A greater challenge
rests in the development of ways to manage the expertise of employees that resides solely in their
minds, and to enhance the returns of such knowledge.
A wide range of scholars has made important contributions toward understanding the man-
agement of knowledge and learning in organizations, and by now it is clear that managers see
information technologies as very relevant in solving some kinds of KM problems. The tremen-
dous technological advances that made it possible for organizations to manage large amounts of
raw data have culminated to a point where information technology (IT) solutions also have been
developed to accomplish a wide range of KM-related tasks. Early innovators capitalized on the
superior speed of transmission of IT to enhance knowledge sharing, as well as accelerating the
growth of knowledge. Later developments included technology to locate and map the distribution
of expertise in organizations, to guide users effciently to fnd the solutions that uniquely meet
their needs from a repository of thousands of best practices, and to create virtual communities.
IT thus has provided a major impetus for the development of new approaches toward managing
knowledge in organizations. The promise of effective IT-based KMS was grand, and in the 1990s
many organizations experimented with the implementation of such systems, often prior to under-
standing their implications for the frm.
Despite these developments, it seems clear that KM research is still in the early stages. KM
has facilitated the integration of knowledge from a variety of different perspectives. The works
of fve philosophersLeibniz, Locke, Kant, Hegel, and Singerhave signifcantly infuenced
the development of KMS, as has the integration of organizational theory into the literature of
management information systems. Likewise, information systems researchers have made signif-
cant contributions to KM that also extend into related disciplines, from psychology to artifcial
intelligence. KM research has infuenced our understanding of other relevant research streams,
including virtual teams, social networks, and organizational learning (OL).
ABOUT DATA, INFORMATION, AND KNOWLEDGE
Early papers in KM described the differences among data, information, and knowledge in terms
of a richness hierarchy, which considers knowledge as the richest and deepest of the three: data
consist of facts, observations, or perceptions. Data may be devoid of context, meaning, or intent but
can easily be captured, stored, and communicated via electronic or other media. For example, the
wind component (u and v) coordinates for a particular hurricanes trajectory, at specifc instances
of time, are considered to be data (Becerra-Fernandez, Gonzalez, and Sabherwal, 2004).
Information has been described as data that possess context, relevance, and purpose. Information
involves the manipulation of raw data to obtain a more meaningful indication of trends or pat-
ON KNOWLEDGE, KM, AND KM SYSTEMS 5
terns in the data. Continuing with the above example, based on the u and v components, hurricane
software models may be used to create a forecast of the hurricane trajectory, which is considered
information (Becerra-Fernandez, Gonzalez, and Sabherwal, 2004).
Finally, knowledge is distinguished from data and information in that knowledge refers to infor-
mation that enables action and decisions, or information with a direction. Knowledge is then seen
at the highest level of a hierarchy, with information at the middle level, and data at the lowest level.
Knowledge is then the most valuable of the three. Western philosophers have in general accepted
Platos defnition of knowledge as a justifed true belief. More recent perspectives treat knowledge
as a transformation process, enabling the conversion of data into information or information into
decisions. Continuing with the same example used above, the knowledge of a hurricane researcher
is used to analyze the u and v wind components, as well as the hurricane forecast produced by the
different software models, to determine the probability that the hurricane will follow a specifc
trajectory. Thus, knowledge helps produce information from data, or more valuable information
from less valuable information. Knowledge helps to use information to make decisions. Knowledge
is typically described as being either tacit or explicit (Polanyi, 1966). Explicit knowledge refers
to knowledge that has been expressed in words and numbers, and it can be easily codifed and
shared as data, specifcations, computer programs, videos, etc. Tacit knowledge, on the other hand,
includes knowledge that is diffcult to express and share, such as intuitions and insights. Nonaka
(1994) describes how knowledge can be converted from these explicit and tacit forms through
four different conversion modes, which are (1) from tacit knowledge to tacit knowledge through
socialization, (2) from tacit knowledge to explicit knowledge through externalization, (3) from
explicit knowledge to explicit knowledge through combination, and (4) from explicit knowledge
to tacit knowledge through internalization.
THE STRUCTURE OF THIS VOLUME
This research monograph offers a collection of ffteen refereed research chapters in the feld of
KM. Each chapter seeks to synthesize how research perspectives on a particular topic in KM have
progressed over the years, thereby illustrating the dynamic and evolutionary nature of KM theories
and documenting the intellectual milestones of KM. The chapters in this volume are grounded in
multiple disciplines, including computer science, engineering, management information systems,
psychology, sociology, economics, strategy, and organizational behavior. In addition, the research
methodologies represented in the chapters in this volume include conceptual analysis, empirical
research, and qualitative methods. All the chapters synthesize prior work and highlight the key
contributions of different streams of research on the topic of analysis.
This research monograph is organized as follows: the Introduction consists of this chapter,
which helps to introduce the topic of KM and KM systems and the topics discussed in the volume.
Part I presents a conceptual lens for the treatise of organizational topics related to the concept
of KM, such as organizational learning and organizational culture. Part II presents a collection
of papers that describe the role of information technology in KM, including how KM systems
and other forms of organizational memories can support many organizational processes, from
supporting social networks to solving knowledge problems. In addition this chapter presents the
role of intelligent technologies in KM. Part III presents a collection of papers that describe the
changing roles of KM and KM systems in organizations. As practitioners have experimented and
found ways to leverage knowledge in their organizations, academics have been developing new
theories to help explain the success or failure of KM implementation efforts. Finally, Part IV
presents a collection of chapters that describe how KM connects with related research streams
that have their theoretical bases in other disciplines, including social psychology, organizational
behavior, strategy, and organization theory. Topics include virtual teams, expert teams and teams
of experts, and communities of practice.
ABOUT KNOWLEDGE MANAGEMENT
Knowledge Management (KM) refers to identifying and leveraging the collective knowledge in an
organization to help the organization compete: the companys overall performance depends on the
extent to which managers can mobilize all the knowledge resources held by individuals and teams
and turn these resources into value-creating activities (von Krogh, 1998, p.133). KM can also
be defned as performing the activities involved in discovering, capturing, sharing, and applying
knowledge so as to enhance, in a cost-effective fashion, the impact of knowledge on the units goal
achievement (Becerra-Fernandez, Gonzalez, and Sabherwal, 2004, p. 31). The chapters in this part
of the monograph trace the origins of KM along theories of OL and organizational culture.
In Chapter 2 Individual, Group, and Organizational Learning: a Knowledge Management
Perspective, Irma Becerra-Fernandez and Rajiv Sabherwal trace the foundations for the current
KM literature to lie in the early discussion of OL, which is said to take place through individu-
als (Simon, 1991). Without a doubt, KMS have had an impact on learning, at the individual,
group, and organizational levels, and these are seen as evolutionary shifts in KM. For some OL
researchers, KM is seen as a subset of OL, while KM researchers claim that KM lies beyond OL
boundaries. Regardless, KMS and OL tie together via the development of the Internet and other
collaboration technologies that provide opportunities for organization-wide and interorganizational
socialization.
Equally important as the impact of KM in OL is the understanding about the impacts that KMS
may have on the frms ability to discover, capture, and share intellectual resources. Dorothy E.
Leidner and Timothy R. Kayworth address those questions in Chapter 3 Knowledge Management
and Organizational Culture by examining the work related to organizational culture and KM. An
understanding of organizational culture and its relationship to KM is key to understanding how
frms can effectively deploy organization-wide KMS.
ABOUT THE ROLE OF INFORMATION TECHNOLOGY IN
KNOWLEDGE MANAGEMENT
KMS have been defned as an emerging line of systems [which] target professional and manage-
rial activities by focusing on creating, gathering, organizing, and disseminating an organizations
knowledge as opposed to information or data (Alavi and Leidner, 2001). A framework for
classifcation of KMS (Becerra-Fernandez, Gonzalez, and Sabherwal, 2004) suggests the follow-
ing types of KMS:
1. Knowledge discovery systems create new knowledge through the implementation of intel-
ligent algorithms such as data mining, and through the inference of data relationships.
2. Knowledge capture systems preserve and formalize the knowledge of experts so it can
be shared with others. Knowledge capture systems formalize knowledge in models such
as concept maps, which allow others to learn the domain.
3. Knowledge sharing systems organize and distribute knowledge. Knowledge repositories
comprise the majority of the KMS currently in place. Several types of knowledge reposi-
tories are described to support the capture and reuse of experience in different contexts.
In addition to corporate memories, there are lessons-learned systems, incident report
databases, alert systems, best practices databases, and expertise locator systems that are
also described as knowledge sharing systems. The differences among these systems are
based on their origin (does the content originate from experience?), application (do they
describe a complete process, or perhaps a task or decision?), results (do they describe
failures or successes?), and orientation (do they support an organization or a whole indus-
try?) (Weber, Aha, and Becerra-Fernandez, 2001). Expertise-locator systems (ELS, also
called knowledge yellow pages or people-fnder systems) are a special type of knowledge
sharing systems that point to experts, those who have the knowledge (Becerra-Fernandez,
2006).
4. Knowledge application systems assist in solving problems. Organizations with signifcant
intellectual capital require eliciting and capturing knowledge for reuse in solving new
problems as well as recurring old problems.
In Chapter 4 Social Networks and Information Technology: Evolution and New Frontiers,
Maryam Alavi and Gerald C. Kane trace the evolution of KM from the purview of two different
schools: IT and social networks. This chapter traces the development of knowledge sharing systems,
recognizing that IT has played an essential role in the development of knowledge sharing networks.
This chapter suggests that the continued evolution of social network research on knowledge shar-
ing should incorporate IT artifacts in the examination of these activities.
Ulrike Schultze in Chapter 5 The Evolution of Knowledge Management Technology: From
Explicit Rules to Implicit Problems examines the evolution of IT in the support of KM and the
progression of its assumptions about knowledge and knowledge work, by studying a specifc as-
pect of the KM movement, namely the technologies designed to solve knowledge problems. The
systems described in this chapter clearly support a KM process (e.g., creating, sharing, capturing,
and applying knowledge)a slice of the evolution of KM. This chapter highlights how early KMS
were more focused on supporting decision-making tasks using narrow, domain-specifc knowledge,
whereas more recent solutions focus on the sharing of common knowledge.
Chapter 6 entitled A Four-Layer Model for IT Support of Knowledge Management by Matteo
Bonifacio, Thomas Franz, and Steffen Staab, presents a review of existing knowledge sharing systems
and argues that there is not one paradigm that suits all of the needs of all organizations. The authors
consider KMS as enablers of organizational communication: frst, through a vocabulary used to
capture and organize knowledge using means such as the Semantic Web and, second, through a way
of organizing access and sharing of knowledge in a distributed manner involving technologies such
as distributed databases, agent-based systems, or peer-to-peerbased systems. This topic promises
to be forward looking in that it presents a new approach for the design of KMS.
ABOUT THE ROLE OF KNOWLEDGE MANAGEMENT SYSTEMS IN
THE ORGANIZATION
The successful implementation of KMS has presented numerous challenges to KM practitioners,
who initially followed a build-and-they-will-come approach to the deployment of these sys-
tems, only to quickly realize that success factors in the implementation of KMS are differentiated
with those associated with traditional IS. While traditional IS research has concentrated much
of its efforts in understanding what factors are leading users to accepting, and thereby using, IT
(perceived usefulness and perceived ease of use) (Davis, 1989), the successful implementation of
KMS requires that its users not only effectively use such systems as in traditional IS but also
contribute to the knowledge base of such systems, thereby playing an active role in building the
content of such systems. The chapters in this part describe the changing roles of KM and KM
systems in organizations, and help explain the theoretical underpinnings for successful KM sys-
tems implementation efforts.
In Chapter 7 Mobilizing Knowledge in a Yu-Gi-Oh! World, Youngjin Yoo argues that early
conceptualizations for KM failed to provide new ideas and inspiration for management in the
knowledge (postindustrial) economy, which he names the Yu-Gi-Oh! (after the popular childrens
trading card game) economy. As a result, KM initiatives have not lived up to the promise of
creating organizations where knowledge fows freely to support the creation of new and novel
solutions. Successful KM implementations must effectively deal with key KM challenges based
on the principle of interaction design.
David G. Schwartz and Dov Teeni in Chapter 8, The Impact of Computer-Mediated Com-
munication on Knowledge Transfer and Organizational Form, trace the evolution and cyclical
interrelationships of computer-mediated communication, knowledge transfer, and organizational
form. This chapter suggests the applicability of Churchills assertion that we shape our environ-
ments, then our environments shape us as it applies to how the infuence of KMS into our orga-
nizational environments has shifted the evolutionary course of knowledge transfer and continues
to bring about a change in the form of those systems.
Chapter 9, Moving Toward a Knowledge Management Maturity Model (K3M) for Develop-
ing Knowledge Management Strategy and Implementation Plans, by Jay Liebowitz and Tom
Beckman, describes the importance of alignment between a KM strategy and the organizations
business strategy based on the formulation of a Knowledge Management Maturity Model (K3M)
as the basis for successful implementation of KMS. Much like the successful capability maturity
model, the K3M blends diverse schools of thought to better structure the assessment and formula-
tion of business and KM strategies and to provide a clear implementation path based on mapping
stakeholder needs, competencies, and capabilities.
In Chapter 10 entitled Building Knowledge Management Systems to Improve Profts and Cre-
ate Loyal Users: Lessons from the Pharmaceutical Industry, Alan Dennis, Dong-Gil Ko, and Paul
F. Clay describe the many attempts that a large multinational pharmaceutical frm in the United
States underwent in developing and deploying KMS in support of their feld sales representa-
tives. The chapter traces the early two unsuccessful attempts, leading to the successful design
and implementation of the third KMS implementation. This chapter also describes measures of
performance that described how this system was able to have a positive impact on the bottom
line for this frm.
Finally, Patrick S.W. Fong in Chapter 11Can We Learn from Our Pastdescribes the in-
fuence of KM in project management. The chapter describes characteristics of projects, features
of project-based organizations, and different perspectives of projects. The chapter also observes
the development of KMS in support of project management, including aspects related to project-
based learning.
ABOUT KNOWLEDGE MANAGEMENT AND TEAMS WITHIN AND
ACROSS ORGANIZATIONS
Perhaps one of the most signifcant events of the late 1900s was the advent of globalization, a
term used to describe the political, economic, social, and technological changes that have resulted
in fundamental changes in the notions of space and time. The proliferation of the Internet has
been one of the forces contributing to globalization, diminishing the signifcance of geographical
distance as a measure of time and space. The chapters in the fourth part describe the intersection
of KM related disciplines, such as social psychology, organizational behavior, strategy, and or-
ganization theory, in explaining topics such as virtual teams, expert teams and teams of experts,
and communities of practice.
In Chapter 12 Managing Knowledge in Virtual Communities Within Organizations, Naren B.
Peddibhotla and Mani R. Subramani describe the establishment of virtual communities, among the
most successful approaches toward KM adopted within frms. Virtual communities are described
as contexts where individuals within frms can come together and interact, bound by shared roles,
expertise, or passion for specifc topics.
Ann Majchrzak, Arvind Malhotra, and Richard John observe in Chapter 13, Dynamic Team
Memory Systems: Enabling Knowledge Sharing Effectiveness in Structurally Diverse Distributed
Teams, how frms are increasingly leveraging their knowledge resources through the deployment
of structurally diverse distributed teams, which relate to virtual teams that are diverse in terms of
geographical location, functional assignment, reporting managers, and business units. Typically
co-locating such expertise may not be economically feasible; therefore, these teams must conduct
their work electronically, integrating their work products through the use of virtual workspaces.
This chapter describes how to effectively use virtual workspaces, proposing that teams must
use dynamic team memory systems that provide the ability to (1) intelligently, continuously,
and easily acquire and encode knowledge from information producers, (2) maintain integrity of
knowledge over time, and (3) match patterns during search and retrieval.
In Chapter 14 Electronic Knowledge Networks: Processes and Structure, Samer Faraj,
Molly McLure Wasko, and Steven L. Johnson deal with one specifc type of virtual community
electronic knowledge networks (EKNs), which are described as online forums that cross organi-
zational boundaries. EKNs are open activity systems characterized by their virtual presence and
self-organizing structure, and they focus on a shared interest or practice. Examples of EKNs include
USENET newsgroups and organizational discussion groups, and their members congregate and
collaborate via asynchronous technology to exchange, for example, technical advice and specifc
products primarily using conversations rather than documentation.
Finally, Yongsuk Kim, Sirkka Jarvenpaa, and Ann Majchrzak in Chapter 15, Ad Hoc Inter-
Organizational Collaboration: Safeguards for Balancing Sharing and Protection of Knowledge,
describe the importance of managing the tension between knowledge sharing and protection in
ad hoc interorganizational collaboration, which is critical for interorganizational joint action and
long-term learning. Because virtual interorganizational collaborations are particularly volatile
and diffcult to monitor, safeguards must be established a priori to manage the tension between
effective knowledge sharing and the possible leakage of corporate secrets. The authors argue that
safeguards must be established based on assumptions of stable and hierarchical interorganizational
relations.
CONCLUSION
This research monograph covers a broad range of perspectives associated with the feld of KM
via the contribution of authors insights into (1) a conceptual lens for knowledge management,
(2) the role of information technology in knowledge management, (3) the role of knowledge
management systems in the organization, and (4) knowledge management and teams in and across
organizations.
We hope that this research monograph will serve a variety of audiences. One audience is those
who teach and conduct research on KM and KM systems. Researchers in the feld of KM represent
a variety of disciplines: management information systems, engineering, computer science, orga-
nizational behavior, strategy, psychology, sociology, and education among others. Regardless of
their background disciplines, research in the feld of KM has increasingly gained momentum and it
impacts a variety of other research areas as can be seen in this monograph, including organizational
learning, organizational culture, project management, social networks, virtual organizations, and
others. The second audience is information systems researchers. The 1990s witnessed an explo-
sion in the proliferation of information systems in support of the organization. In many instances,
KM systems served the parochial needs of a subunit, and were not properly integrated with other
business processes represented in the organizations enterprise systems. Furthermore, KM systems
are differentiated with traditional information systems in that KM systems users play an active
role in building the content of such systems. While traditional information systems research has
concentrated much of its efforts in understanding what factors are leading users to accepting, and
thereby using IT (perceived usefulness and perceived ease of use), the successful implementation
of KM systems requires that its users not only effectively use such systems as in traditional
information systems but also contribute to the knowledge base of such systems. Therefore, seeking
to understand the factors that lead to the successful implementation of KM systems is an important
area of research that is still in its infancy. Finally, a third audience for this book is students, from
undergraduate to graduate levels. For them, they will fnd useful the comprehensive literature
reviews that the authors who have contributed the chapters of this monograph have painstakingly
constructed as they developed the topics treated here, including those related to social networks,
organizational culture, organizational learning, virtual teams, and many other themes.
Together these ffteen chapters portray a varied and thought-provoking set of views, and how
these views have evolved over time. In short, we hope that this monograph will become a key
resource by presenting the journey of KM over time, which is signifcantly different than the
time-bound perspectives that peer-reviewed articles typically present. The extent to which these
views contribute to stimulating future research in these topics is the extent to which this endeavor
to provide an evolutionary perspective of knowledge management was successful.
REFERENCES
Alavi, M., and Leidner, D. 2001. Knowledge management and knowledge management systems: Conceptual
foundations and research issues. MIS Quarterly, 25, 1, 107136.
Becerra-Fernandez, I. 2006. Searching for experts on the Web: A review of contemporary expertise locator
systems. ACM Transactions on Internet Technology, 6, 4, 333355.
Becerra-Fernandez, I.; Gonzalez, A.; and Sabherwal, R. 2004. Knowledge Management: Challenges, Solu-
tions, and Technologies. New York: Prentice-Hall.
Davis, F.D. 1989. Perceived usefulness, perceived ease of use, and user acceptance of information technol-
ogy. MIS Quarterly, 13, 3, 319340.
Nonaka, I. 1994. A dynamic theory of organizational knowledge-integration. Organization Science, 5, 1,
1437.
Nonaka, I., and Takeuchi, H. 1995. The Knowledge Creating Company. Oxford: Oxford University Press.
Polanyi, M. 1966. The Tacit Dimension. London: Routledge and Keoan.
Simon, H.A. 1991. Bounded rationality and organizational learning. Organization Science, 2, 1, 125134.
. 1996. The Sciences of the Artifcial. Cambridge, MA: MIT Press.
Weber, R.; Aha, D.W.; and Becerra-Fernandez, I. 2001. Intelligent lessons learned systems. Expert Systems
with Applications, 20, 1, 1734.
PART I
A CONCEPTUAL LENS FOR
13
ChaptEr 2
INDIVIDUAL, GROUP, AND
ORGANIZATIONAL LEARNING
A Knowledge Management Perspective
irma BECErra-fErnandEZ and rajiV saBhErwal
Abstract: This chapter traces the feld of knowledge management (KM) along the discipline
of organizational learning (OL). Organizational learning theories recognize the importance
of cognitive development and the value of learning as an organizational-level phenomenon,
which is synergistic and not simply the cumulative result of each members learning. However,
OL takes place through individuals and the positive and negative outcomes that their members
encounter from their behaviors. In this chapter, we discuss some of the changes that have oc-
curred over the past three decades in the feld of KM, highlighting how the feld has evolved to
the current state and some of the ongoing changes. In doing so, we examine how the interplay
between information and communication technologies and KM processes (including various
social and structural mechanisms) contributes to learning activities at the individual, group,
and organizational levels. In short, the goals of KM and OL are intertwined: KM systems sup-
port the goals of learning at the individual, group, and organizational levels. Also, KM systems
and OL tie together via the development of the Internet and other collaboration technologies
that provide opportunities for organization-wide socialization. Finally, KM systems support
the processes of perspective making and perspective taking at the individual, group, and orga-
nizational levels.
Keywords: Knowledge Management (KM), Knowledge Management Systems (KMS), Organiza-
tional Learning (OL), KM Processes
INTRODUCTION
Knowledge management systems (KMS) support the discovery, capture, sharing, and application
of organizational knowledge (Becerra-Fernandez, Gonzalez, and Sabherwal, 2004). Learning how
to manage an organizations knowledge provides numerous benefts, including leveraging core
business competencies, accelerating innovation and time to market, improving cycle times and
decision making, strengthening organizational commitment, and building sustainable competi-
tive advantage (Davenport and Prusak, 1998; Sabherwal and Sabherwal, 2005). Furthermore, the
effects of knowledge management (KM), which is defned as performing the activities involved
in discovering, capturing, sharing, and applying knowledge so as to enhance, in a cost-effective
fashion, the impact of knowledge on the units goal achievement (Becerra-Fernandez, Gonzalez,
14 BECERRA-FERNANDEZ AND SABHERWAL
and Sabherwal, 2004, p. 31), progress from individuals to groups and then to organizational and
interorganizational levels.
Knowledge is said to reside in people in all organizations. This is most clearly evident in profes-
sional service frms, such as consulting or law frms, where considerable knowledge resides in the
minds of individual members of the frm (Argote and Ingram, 2000). KM can enhance employees
learning and exposure to the latest knowledge in their felds. For example, in preparing a report on
lessons learned from a project, consultants document the tacit knowledge they acquire during the
project. Individuals embarking on later projects could then acquire the knowledge gained by the
earlier team by reading this report and thereby reexperiencing what others have gone through.
Considerable knowledge resides in groups because of the relationships among the members of
the group. When individuals have worked together for a long time, they instinctively know each
others strengths, weaknesses, expertise, and preferences and recognize aspects that need to be
communicated as well as those that can be taken for granted (Skyrme, 2000). Consequently, groups
form beliefs about what works well and what does not, and this knowledge is over and above the
knowledge residing in each individual member. Thus, the collective knowledge is synergistic,
and greater than the sum of each individuals knowledge. Communities of practice illustrate such
embedding of knowledge within groups.
The impact of KM at the organizational level has been clearly highlighted in some recent stud-
ies (e.g., Hult, 1995; Slater and Narver, 1995; Hult et al., 2000). Organizational learning (OL) has
been found to enhance frms innovativeness and capacity for adaptation (Hurley and Hult, 1998).
Firms with a greater learning also possess a higher capacity to innovate, which in the presence
of adequate resources, results in increased competitive advantage and performance for the frm
(Hurley and Hult, 1998). Finally, OL can translate into business agility, by positively infuenc-
ing business processes, such as the cycle time of the purchasing process (Hult et al., 2000), and
eventually into superior stock market returns (Sabherwal and Sabherwal, 2005).
Rapid changes in the feld of KM have largely resulted from the dramatic progress in the feld
of information and communication technologies (ICT) (Becerra-Fernandez et al., 2004). Also,
experience with KM has enabled organizations to develop better social and structural mechanisms
for managing knowledge and to improve the deployment of existing mechanisms. These develop-
ments have enabled organizations to develop KMS that best leverage KM mechanisms by deploying
sophisticated technologies. Consequently, rather than using ICT and social mechanisms indepen-
dently for different KM tasks, organizations utilize ICT and social mechanisms in an increasingly
synergistic fashion. KMS combine a variety of KM mechanisms and technologies to support the
KM processes of knowledge discovery, capture, sharing, and application.
In this chapter, we discuss some of the changes that have occurred over the last three decades
in the feld of KM, highlighting how the feld has evolved to the current state and some of the
ongoing changes. In doing so, we examine how the interplay between ICT and KM processes
(including various social and structural mechanisms) contributes to learning activities at the indi-
vidual, group, and organizational levels.
EVOLUTION OF KNOWLEDGE MANAGEMENT
In this section we describe how the feld of KM has evolved to the current state. In particular, we
focus on how the interplay between increased development in ICT and social mechanisms supports
the activities of learning organizations. We also present perspectives that relate KM to OL. Finally,
we discuss the shift from perspective taking or perspective making to perspective taking and per-
spective making; and the increased embeddedness of KM across organizational processes.
INDIVIDUAL, GROUP, AND ORGANIZATIONAL LEARNING 15
From KM Mechanisms OR ICT to KM Mechanisms AND ICT
The ICT have provided a major impetus to KM by enabling the implementation of KMS. The
ICT that support KM include artifcial intelligence technologies encompassing those used for
knowledge acquisition and case-based reasoning systems, discussion groups, computer-based
simulations, databases, decision support systems, enterprise resource planning systems, expert
systems, management information systems, expertise locator systems, videoconferencing and
information repositories encompassing best practices databases, and lessons learned systems.
In addition, social mechanisms can also promote and enable KM. They may or may not utilize
technology, but they do involve some kind of organizational arrangement or social or structural
means of facilitating KM. Examples of KM mechanisms include on-the-job training, learning
by observation, face-to-face meetings, cooperative projects across departments, mentoring for
knowledge sharing, and employee rotation across departments. We call the applications resulting
from such synergy between the latest technologies and social mechanisms knowledge management
systems. KMS and the role of ICT on its evolution over time, including how this development has
been synergistic with KM mechanisms, is discussed further this chapter.
From Organizational Learning to Knowledge Management
The foundations for the current KM literature lie in the early discussion of OL (e.g., Argyris, 1977;
Argyris and Schon, 1978; Duncan and Weiss, 1979; Hedberg, 1981). Organizational learning
takes place through individuals (Simon, 1991), but it is not simply the cumulative result of each
members learning (Fiol and Lyles, 1985). According to Simon (1991):
All learning takes place inside our heads; an organization learns in only two ways: (a) by
the learning of its members, or (b) by ingesting new members who have knowledge the
organization didnt already have. (p. 125)
Furthermore, what members of an organization know may be related, and what individuals
learn depends on what they already know (Simon, 1991). Organizational learning requires the
transmission of information from one member to another; therefore, learning is a social phenom-
enon (Simon, 1991). According to March (1991):
Two distinctive features of the social context are considered. The frst is the mutual learning
of an organization and the individuals in it. Organizations store knowledge in their proce-
dures, norms, rules, and forms. They accumulate such knowledge over time, learning from
their members. At the same time, individuals in an organization are socials to organizational
beliefs. (p. 73)
In essence, learning is viewed as an organizational-level phenomenon (Argyris and Schon,
1978; Huber, 1991). According to Hedberg (1981):
Organizations do not have brains, but they have cognitive systems and memories. As in-
dividuals develop their personalities, personal habits, and beliefs over time, organizations
develop worldviews and ideologies. Members come and go, and leadership changes, but
organizations memories preserve certain behaviors, mental maps, norms, and values over
time. (p. 6)
Organizations learn through the positive and negative outcomes that their members encounter
from their behaviors. The amount of such experiential learning depends on the extent to which
each specifc individual learns and the extent to which individual learning gets embedded in orga-
nizational memory (Argyris and Schon, 1978). The eventual impact of OL is on the organization,
especially when behavioral changes produced from learning are incorporated (Slater and Narver,
1995; Hurley and Hult, 1998).
Organization learning may be viewed from two broad perspectives (Weick, 1991). Accord-
ing to one perspective, OL is said to occur when new knowledge is generated, even if this new
knowledge does not produce any change in behavior (Duncan and Weiss, 1979; Huber, 1991).
For example, Huber states: An organization learns if any of its units acquires knowledge that it
recognizes as potentially useful to the organization (p. 89). According to the second perspective,
cognitive development is necessary but not suffcient for OL; instead, according to this perspective,
OL requires behavioral development as well (Argyris, 1977; Stata, 1989). For example, Argyris
states: An organization may be said to learn to the extent that it identifes and corrects errors
(p. 113). However, both these perspectives on OL theories recognize the importance of cognitive
development.
A review of citation patterns of published KM works within the 10-year period of 1991 to 2001
reveals that the fve most central literature domains identifed in KM are knowledge creation and
epistemology, OL, evolutionary economics, intellectual capital, and organizational capabilities
and competencies (Ponzi, 2002, 2004). The literature for KM and OL started to converge around
1996, when the literature began to distinguish between the two notions and to include them both
within the same paper (Ponzi, 2002). Table 2.1 presents a summary of the top citations in the KM
literature with a focus on OL. Although this table does not represent a comprehensive review of
the literature for the intersection of these two domains, it serves to compare and contrast their
emphasis. Organizational learning and KM are both processes through which the valuable resource
of knowledge is changed (Vera and Crossan, 2003). The relation between OL and KM is implicit
in the discussion of organizations as distributed knowledge systems (Tsoukas, 1996), streams
of knowledge (e.g., Van Krogh, Roos, and Slocum, 1994), and systems of distributed cognition
(Boland and Tenkasi, 1995; Weick and Roberts, 1993), wherein individuals act autonomously
while understanding their interdependence with others. For some OL researchers, KM is seen as
a subset of OL (Fulmer, Gibbs, and Keysl, 1998; Ponzi, 2002), while KM researchers claim that
KM lies beyond OL boundaries (Nevis, DiBella, and Gould, 1995; Ponzi, 2002).
However, a focus on KM implies a more prescriptive view than a focus on OL. In addition, the
literature on OL and KM agrees that these processes may lead to cognitive development, which is
usually followed by behavioral change, and subsequently produces organizational impacts. Fur-
thermore, these theories agree that KM can produce impacts at various levels, including impacts
on the overall organization, with learning originating at the individual level, and then moving
up through groups, and then to the overall organization. Nonaka (1994; Nonaka and Takeuchi,
1995) makes a connection between OL and KM, when he describes organizational knowledge
creation as a spiral process, which starts at the individual level, expanding to the group, and the
organizational levels. He contends that only individuals can create knowledge, and the interactions
among individuals are essential to develop organizational knowledge. In his view, organizational
knowledge creation is a process that amplifes the individuals knowledge, as original ideas
emanate from autonomous individuals, diffuse within the team, and then become organizational
ideals (Nonaka and Takeuchi, 1995, p. 76).
Despite the similarities between the literature on OL and the literature on KM, the shift from
OL to KM has been accompanied by three signifcant developments. First, the OL literature fo-
17
T
a
b
l
e

2
.
1
T
o
p

C
i
t
a
t
i
o
n
s

i
n

t
h
e

K
n
o
w
l
e
d
g
e

M
a
n
a
g
e
m
e
n
t

L
i
t
e
r
a
t
u
r
e

w
i
t
h

a

F
o
c
u
s

o
n

O
r
g
a
n
i
z
a
t
i
o
n
a
l

L
e
a
r
n
i
n
g
A
u
t
h
o
r
/
R
a
n
k
i
n
g
*
C
i
t
a
t
i
o
n
C
o
n
t
r
i
b
u
t
i
o
n
N
o
n
a
k
a

a
n
d

T
a
k
e
u
c
h
i

(
1
9
9
5
)
/
1
T
h
e

K
n
o
w
l
e
d
g
e

C
r
e
a
t
i
n
g

C
o
m
p
a
n
y
:

H
o
w

J
a
p
a
n
e
s
e

C
o
m
p
a
n
i
e
s

C
r
e
a
t
e

t
h
e

D
y
n
a
m
i
c
s

o
f

I
n
n
o
v
a
t
i
o
n
.

N
e
w

Y
o
r
k
:

O
x
f
o
r
d

U
n
i
v
e
r
s
i
t
y

P
r
e
s
s
F
o
u
r

m
o
d
e
s

o
f

k
n
o
w
l
e
d
g
e

c
o
n
v
e
r
s
i
o
n

(
s
o
c
i
a
l
i
z
a
t
i
o
n
,

e
x
t
e
r
n
a
l
i
z
a
t
i
o
n
,

i
n
t
e
r
n
a
l
i
z
a
t
i
o
n
,

a
n
d

c
o
m
b
i
n
a
t
i
o
n
)
.

S
p
i
r
a
l

o
f

k
n
o
w
l
e
d
g
e

c
r
e
a
t
i
o
n
:

f
r
o
m

t
h
e

i
n
d
i
v
i
d
u
a
l
,

t
o

t
h
e

g
r
o
u
p
,

t
o

t
h
e

o
r
g
a
n
i
z
a
t
i
o
n
,

t
o

i
n
t
e
r
o
r
g
a
n
i
z
a
t
i
o
n
.
S
e
n
g
e

(
1
9
9
0
)
/
4
T
h
e

F
i
f
t
h

D
i
s
c
i
p
l
i
n
e
:

T
h
e

A
r
t

a
n
d

P
r
a
c
t
i
c
e

o
f

t
h
e

L
e
a
r
n
i
n
g

O
r
g
a
n
i
z
a
t
i
o
n
.

N
e
w

Y
o
r
k
:

D
o
u
b
l
e
d
a
y
A
n

o
r
g
a
n
i
z
a
t
i
o
n

m
u
s
t

p
r
a
c
t
i
c
e

f
v
e

d
i
s
c
i
p
l
i
n
e
s

t
o

r
i
d

i
t
s
e
l
f

o
f

l
e
a
r
n
i
n
g

d
i
s
a
b
i
l
i
t
i
e
s
:

a

s
h
a
r
e

v
i
s
i
o
n
,

p
e
r
s
o
n
a
l

m
a
s
t
e
r
y
,

s
t
r
o
n
g

m
e
n
t
a
l

m
o
d
e
l
s
,

g
r
o
u
p

(
t
e
a
m
)

l
e
a
r
n
i
n
g
,

a
n
d

s
y
s
t
e
m

t
h
i
n
k
i
n
g
.
L
e
o
n
a
r
d
-
B
a
r
t
o
n

(
1
9
9
5
)
/
5
W
e
l
l
s
p
r
i
n
g
s

o
f

K
n
o
w
l
e
d
g
e
.

B
o
s
t
o
n
:

H
a
r
v
a
r
d

B
u
s
i
n
e
s
s

S
c
h
o
o
l

P
r
e
s
s
M
a
n
a
g
e
r
s

m
u
s
t

t
h
i
n
k

c
o
n
s
t
a
n
t
l
y

a
b
o
u
t

t
h
e

p
o
t
e
n
t
i
a
l

k
n
o
w
l
e
d
g
e
-
b
u
i
l
d
i
n
g

i
m
p
a
c
t

o
f

e
v
e
r
y

t
e
c
h
n
o
l
o
g
y
-
r
e
l
a
t
e
d

d
e
c
i
s
i
o
n

t
h
e
y

m
a
d
e
.

T
h
o
s
e

d
e
c
i
s
i
o
n
s

m
u
s
t

o
r
i
g
i
n
a
t
e

i
n

a
n

u
n
d
e
r
s
t
a
n
d
i
n
g

o
f

t
h
e

l
o
n
g
-
t
e
r
m
,

s
y
s
t
e
m
i
c
,

a
n
d

p
e
o
p
l
e
-
b
a
s
e
d

n
a
t
u
r
e

o
f

t
e
c
h
n
o
l
o
g
i
c
a
l

a
d
v
a
n
t
a
g
e
.
A
r
g
y
r
i
s

a
n
d

S
c
h
o
n

(
1
9
7
8
)
/
7
O
r
g
a
n
i
z
a
t
i
o
n
a
l

L
e
a
r
n
i
n
g
:

A

T
h
e
o
r
y

o
f

A
c
t
i
o
n

A
p
p
r
o
a
c
h
.

R
e
a
d
i
n
g
:

A
d
d
i
s
o
n
-
W
e
s
l
e
y
S
i
n
g
l
e
-

a
n
d

d
o
u
b
l
e
-
l
o
o
p

t
h
e
o
r
y
.
H
u
b
e
r

(
1
9
9
1
)
/
1
0
O
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g
:

T
h
e

c
o
n
t
r
i
b
u
t
i
n
g

p
r
o
c
e
s
s
e
s

a
n
d

t
h
e

l
i
t
e
r
a
t
u
r
e
s
.

O
r
g
a
n
i
z
a
t
i
o
n

S
c
i
e
n
c
e
,

2

(
1
)
:

8
8
1
1
5
E
v
a
l
u
a
t
e
s

o
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g

l
i
t
e
r
a
t
u
r
e

a
n
d

d
e
t
a
i
l
s

f
o
u
r

c
o
n
s
t
r
u
c
t
s

t
h
a
t

a
r
e

m
o
s
t

o
f
t
e
n

l
i
n
k
e
d

w
i
t
h

o
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g
:

k
n
o
w
l
e
d
g
e

a
c
q
u
i
s
i
t
i
o
n
,

i
n
f
o
r
m
a
t
i
o
n

d
i
s
t
r
i
b
u
t
i
o
n
,

i
n
f
o
r
m
a
t
i
o
n

i
n
t
e
r
p
r
e
t
a
t
i
o
n
,

a
n
d

o
r
g
a
n
i
z
a
t
i
o
n
a
l

m
e
m
o
r
y
.
W
e
i
c
k

(
1
9
9
5
)
/
1
7
S
e
n
s
e
m
a
k
i
n
g

i
n

O
r
g
a
n
i
z
a
t
i
o
n
s
.

T
h
o
u
s
a
n
d

O
a
k
s
:

S
a
g
e

P
u
b
l
i
c
a
t
i
o
n
T
h
e

n
a
t
u
r
e

a
n
d

p
r
o
p
e
r
t
i
e
s

o
f

s
e
n
s
e
m
a
k
i
n
g

i
n

o
r
g
a
n
i
z
a
t
i
o
n
s
:

a
n

o
n
g
o
i
n
g

p
r
o
c
e
s
s

o
f

c
r
e
a
t
i
o
n

o
f

r
e
a
l
i
t
y

w
h
e
n

p
e
o
p
l
e

m
a
k
e

r
e
t
r
o
s
p
e
c
t
i
v
e

s
e
n
s
e

o
f

t
h
e

s
i
t
u
a
t
i
o
n
s

i
n

w
h
i
c
h

t
h
e
y

f
n
d

t
h
e
m
s
e
l
v
e
s
.
S
i
m
o
n

(
1
9
7
6
)
/
2
2
A
d
m
i
n
i
s
t
r
a
t
i
v
e

B
e
h
a
v
i
o
r
.

N
e
w

Y
o
r
k
:

T
h
e

F
r
e
e

P
r
e
s
s
L
e
a
r
n
i
n
g

t
a
k
e
s

p
l
a
c
e

i
n

i
n
d
i
v
i
d
u
a
l
s
.

O
r
g
a
n
i
z
a
t
i
o
n
s

l
e
a
r
n

i
n

t
w
o

w
a
y
s
:

(
a
)

b
y

t
h
e

l
e
a
r
n
i
n
g

o
f

i
t
s

m
e
m
b
e
r
s

o
r

(
b
)

b
y

i
n
g
e
s
t
i
n
g

n
e
w

m
e
m
b
e
r
s

w
h
o

h
a
v
e

k
n
o
w
l
e
d
g
e

t
h
e

o
r
g
a
n
i
z
a
t
i
o
n

d
i
d

n
o
t

h
a
v
e

p
r
e
v
i
o
u
s
l
y
.
(
c
o
n
t
i
n
u
e
d
)
18
B
r
o
w
n

(
1
9
9
1
)
/
2
3
O
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g

a
n
d

c
o
m
m
u
n
i
t
i
e
s

o
f

p
r
a
c
t
i
c
e
:

T
o
w
a
r
d

a

u
n
i
f
e
d

v
i
e
w

o
f

w
o
r
k
i
n
g
,

l
e
a
r
n
i
n
g
,

a
n
d

i
n
n
o
v
a
t
i
o
n
.

O
r
g
a
n
i
z
a
t
i
o
n

S
c
i
e
n
c
e
,

2

(
1
)
:

4
0
5
7
K
n
o
w
l
e
d
g
e

s
h
a
r
i
n
g

b
e
h
a
v
i
o
r

a
n
d

c
u
l
t
u
r
e

i
s

a

f
o
r
m

o
f

p
r
o
f
e
s
s
i
o
n
a
l

c
o
m
p
e
t
e
n
c
e
.

C
o
m
m
u
n
i
t
i
e
s

o
f

p
r
a
c
t
i
c
e

b
u
i
l
d

t
r
u
s
t

a
n
d

t
r
a
n
s
f
e
r

k
n
o
w
-
h
o
w
,

w
h
i
c
h

i
n

t
u
r
n

c
r
e
a
t
e

o
r
g
a
n
i
z
a
t
i
o
n
a
l

v
a
l
u
e
.
S
p
e
n
d
e
r

(
1
9
9
6
)
/
2
5
C
o
m
p
e
t
i
t
i
v
e

a
d
v
a
n
t
a
g
e

f
r
o
m

t
a
c
i
t

k
n
o
w
l
e
d
g
e
?

U
n
p
a
c
k
i
n
g

t
h
e

c
o
n
c
e
p
t

a
n
d

i
t
s

s
t
r
a
t
e
g
i
c

i
m
p
l
i
c
a
t
i
o
n
s
.

I
n

B
.

M
o
s
i
n
g
e
o
n

a
n
d

A
.

E
d
m
o
n
s
o
n

(
e
d
s
.
)
,

O
r
g
a
n
i
z
a
t
i
o
n
a
l

L
e
a
r
n
i
n
g

a
n
d

C
o
m
p
e
t
i
t
i
v
e

A
d
v
a
n
t
a
g
e

(
p
p
.

5
6
7
3
)
.

L
o
n
d
o
n
:

S
a
g
e
.

S
t
r
a
t
e
g
i
c

i
m
p
l
i
c
a
t
i
o
n
s

o
f

t
h
e

p
r
o
p
o
s
i
t
i
o
n

t
h
a
t

t
a
c
i
t

k
n
o
w
l
e
d
g
e

i
s

a
t

t
h
e

r
o
o
t

o
f

c
o
m
p
e
t
i
t
i
v
e

a
d
v
a
n
t
a
g
e
.

A

f
o
u
r
-
f
o
l
d

t
y
p
o
l
o
g
y

i
s

s
u
g
g
e
s
t
e
d

t
h
a
t

d
i
s
t
i
n
g
u
i
s
h
e
s

t
h
r
e
e

t
y
p
e
s

o
f

t
a
c
i
t

k
n
o
w
l
e
d
g
e

f
r
o
m

o
b
j
e
c
t
i
v
e

k
n
o
w
l
e
d
g
e
:

c
o
n
s
c
i
o
u
s

p
r
a
c
t
i
c
a
l

k
n
o
w
l
e
d
g
e
,

a
u
t
o
m
a
t
i
c

p
r
a
c
t
i
c
a
l

k
n
o
w
l
e
d
g
e
,

a
n
d

c
o
l
l
e
c
t
i
v
e

p
r
a
c
t
i
c
a
l

k
n
o
w
l
e
d
g
e
.
M
a
r
c
h

(
1
9
9
1
)
E
x
p
l
o
r
a
t
i
o
n

a
n
d

e
x
p
l
o
i
t
a
t
i
o
n

i
n

o
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g
.

O
r
g
a
n
i
z
a
t
i
o
n

S
c
i
e
n
c
e
,

2

(
1
)
:

7
1
8
7
E
x
p
l
o
r
a
t
i
o
n

a
n
d

e
x
p
l
o
i
t
a
t
i
o
n

o
f

o
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g

a
r
e

e
s
s
e
n
t
i
a
l

f
o
r

o
r
g
a
n
i
z
a
t
i
o
n
s
.

A
d
a
p
t
i
v
e

p
r
o
c
e
s
s
e
s
,

b
y

r
e
f
n
i
n
g

e
x
p
l
o
i
t
a
t
i
o
n

m
o
r
e

r
a
p
i
d
l
y

t
h
a
n

e
x
p
l
o
r
a
t
i
o
n
,

a
r
e

l
i
k
e
l
y

t
o

b
e
c
o
m
e

e
f
f
e
c
t
i
v
e

i
n

t
h
e

s
h
o
r
t

r
u
n

b
u
t

s
e
l
f
-
d
e
s
t
r
u
c
t
i
v
e

i
n

t
h
e

l
o
n
g

r
u
n
.
G
a
r
v
i
n

(
1
9
9
3
)
B
u
i
l
d
i
n
g

a

l
e
a
r
n
i
n
g

o
r
g
a
n
i
z
a
t
i
o
n
.

H
a
r
v
a
r
d

B
u
s
i
n
e
s
s

R
e
v
i
e
w
,

J
u
l
y
A
u
g
u
s
t
:

7
8
9
1
D
i
s
c
u
s
s
e
s

s
k
i
l
l
s

t
h
a
t

a
r
e

c
h
a
r
a
c
t
e
r
i
s
t
i
c

o
f

l
e
a
r
n
i
n
g

o
r
g
a
n
i
z
a
t
i
o
n
s
:

s
y
s
t
e
m
a
t
i
c

p
r
o
b
l
e
m

s
o
l
v
i
n
g
,

e
x
p
e
r
i
m
e
n
t
a
t
i
o
n
,

l
e
a
r
n
i
n
g

f
r
o
m

t
h
e
i
r

o
w
n

e
x
p
e
r
i
e
n
c
e
,

b
e
s
t

p
r
a
c
t
i
c
e
s
,

a
n
d

t
r
a
n
s
f
e
r
r
i
n
g

k
n
o
w
l
e
d
g
e
.
L
e
v
i
t
t

a
n
d

M
a
r
c
h

(
1
9
8
8
)
O
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g
.

A
n
n
u
a
l

R
e
v
i
e
w

o
f

S
o
c
i
o
l
o
g
y
,

1
4
:

3
1
9
3
4
0
O
r
g
a
n
i
z
a
t
i
o
n
a
l

d
e
c
i
s
i
o
n
s

d
e
p
e
n
d

o
n

r
u
l
e
s

a
n
d

o
r
g
a
n
i
z
a
t
i
o
n
a
l

r
o
u
t
i
n
e
s

t
h
a
t

e
n
c
o
d
e

o
r
g
a
n
i
z
a
t
i
o
n
a
l

e
x
p
e
r
i
e
n
c
e
s

a
n
d

s
o
l
u
t
i
o
n
s
.
S
c
h
e
i
n

(
1
9
8
5
)
O
r
g
a
n
i
z
a
t
i
o
n
a
l

C
u
l
t
u
r
e

a
n
d

L
e
a
d
e
r
s
h
i
p
.

S
a
n

F
r
a
n
c
i
s
c
o
:

J
o
s
s
e
y
-
B
a
s
s
D
e
s
c
r
i
b
e
d

o
r
g
a
n
i
z
a
t
i
o
n
a
l

c
u
l
t
u
r
e

a
s

s
h
a
r
e
d

a
s
s
u
m
p
t
i
o
n
s

b
y

a

g
r
o
u
p
,

w
h
i
c
h

w
e
r
e

l
e
a
r
n
e
d

a
s

t
h
e

o
r
g
a
n
i
z
a
t
i
o
n

s
o
l
v
e
s

p
r
o
b
l
e
m
s
,

w
h
i
c
h

l
e
a
d

t
o

e
x
t
e
r
n
a
l

a
d
a
p
t
a
t
i
o
n

a
n
d

i
n
t
e
r
n
a
l

i
n
t
e
g
r
a
t
i
o
n
.

A
l
s
o

l
i
n
k
s

o
r
g
a
n
i
z
a
t
i
o
n
a
l

c
u
l
t
u
r
e

t
o

t
h
e

l
e
a
r
n
i
n
g

o
r
g
a
n
i
z
a
t
i
o
n
.
S
c
h
o
n

(
1
9
8
3
)
T
h
e

R
e
f
e
c
t
i
v
e

P
r
a
c
t
i
t
i
o
n
e
r
:

H
o
w

P
r
o
f
e
s
s
i
o
n
a
l
s

T
h
i
n
k

i
n

A
c
t
i
o
n
.

L
o
n
d
o
n
:

M
a
r
a
c
a

T
e
m
p
l
e

S
m
i
t
h
T
h
e

m
o
s
t

c
r
u
c
i
a
l

c
o
m
p
e
t
e
n
c
e

f
o
r

a
l
l

p
r
o
f
e
s
s
i
o
n
a
l
s

i
s

r
e
f
e
c
t
i
o
n
,
i
n
-
a
c
t
i
o
n
,

a
n
d

a
b
o
u
t
-
a
c
t
i
o
n
.

T
h
i
s

i
s

t
h
e

k
n
o
w
i
n
g
-
i
n
-
a
c
t
i
o
n

p
r
o
c
e
s
s
,

i
m
p
o
r
t
a
n
t

f
o
r

c
o
n
t
i
n
u
o
u
s

i
m
p
r
o
v
e
m
e
n
t
.
T
a
b
l
e

2
.
1

(
c
o
n
t
i
n
u
e
d
)
A
u
t
h
o
r
/
R
a
n
k
i
n
g
*
C
i
t
a
t
i
o
n
C
o
n
t
r
i
b
u
t
i
o
n
19
K
i
m

(
1
9
9
3
)
T
h
e

l
i
n
k

b
e
t
w
e
e
n

i
n
d
i
v
i
d
u
a
l

a
n
d

o
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g
.

S
l
o
a
n

M
a
n
a
g
e
m
e
n
t

R
e
v
i
e
w
,

F
a
l
l
:

3
7
5
0
T
h
e
o
r
y

a
b
o
u
t

t
h
e

p
r
o
c
e
s
s

t
h
r
o
u
g
h

w
h
i
c
h

i
n
d
i
v
i
d
u
a
l

l
e
a
r
n
i
n
g

a
d
v
a
n
c
e
s

o
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g
.
D
e

G
e
u
s

(
1
9
9
7
)
T
h
e

L
i
v
i
n
g

C
o
m
p
a
n
y
:

H
a
b
i
t
s

f
o
r

S
u
r
v
i
v
a
l

i
n

a

T
u
r
b
u
l
e
n
t

B
u
s
i
n
e
s
s

E
n
v
i
r
o
n
m
e
n
t
.

B
o
s
t
o
n
:

H
a
r
v
a
r
d

B
u
s
i
n
e
s
s

S
c
h
o
o
l

P
r
e
s
s
R
e
s
e
a
r
c
h

o
n

o
r
g
a
n
i
z
a
t
i
o
n
a
l

l
o
n
g
e
v
i
t
y

p
l
a
c
e
s

o
r
g
a
n
i
z
a
t
i
o
n
a
l

c
u
l
t
u
r
e

a
s

o
n
e

o
f

t
h
e

k
e
y

c
o
m
p
o
n
e
n
t
s

t
o

c
o
m
p
e
t
i
t
i
v
e

s
t
r
e
n
g
t
h
.
N
e
v
i
s

e
t

a
l
.

(
1
9
9
5
)
U
n
d
e
r
s
t
a
n
d
i
n
g

o
r
g
a
n
i
z
a
t
i
o
n
s

a
s

l
e
a
r
n
i
n
g

s
y
s
t
e
m
s
.

S
l
o
a
n

M
a
n
a
g
e
m
e
n
t

R
e
v
i
e
w
,

W
i
n
t
e
r
:

7
3
8
4
F
r
a
m
e
w
o
r
k

f
o
r

e
x
a
m
i
n
i
n
g

a
n

o
r
g
a
n
i
z
a
t
i
o
n
s

a
b
i
l
i
t
y

t
o

l
e
a
r
n
.

B
a
s
e
d

o
n

7

l
e
a
r
n
i
n
g

o
r
i
e
n
t
a
t
i
o
n
s

a
n
d

1
0

f
a
c
i
l
i
t
a
t
i
n
g

f
a
c
t
o
r
s

o
r

p
r
o
c
e
s
s
e
s
.
C
y
e
r
t

a
n
d

M
a
r
c
h

(
1
9
6
3
)
A

B
e
h
a
v
i
o
r
a
l

T
h
e
o
r
y

o
f

t
h
e

F
i
r
m
.

E
n
g
l
e
w
o
o
d

C
l
i
f
f
s
,

N
J
:

P
r
e
n
t
i
c
e
-
H
a
l
l
.
O
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g

i
s

a
n

a
d
a
p
t
i
v
e

p
r
o
c
e
s
s

i
n

w
h
i
c
h

g
o
a
l
s

a
n
d

p
r
o
c
e
d
u
r
e
s

d
i
r
e
c
t

t
h
e

o
r
g
a
n
i
z
a
t
i
o
n

t
o

f
n
d

s
o
l
u
t
i
o
n
s

t
o

p
r
o
b
l
e
m

t
h
a
t

a
d
a
p
t
e
d

t
o

t
h
e
i
r

o
w
n

e
x
p
e
r
i
e
n
c
e
s
.
F
i
o
l

a
n
d

L
y
l
e
s

(
1
9
8
5
)
O
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g
.

A
c
a
d
e
m
y

o
f

M
a
n
a
g
e
m
e
n
t

R
e
v
i
e
w
,

1
0

(
4
)
:

8
0
3
8
1
3
L
o
w
e
r
-
l
e
v
e
l

v
e
r
s
u
s

h
i
g
h
e
r
-
l
e
v
e
l

l
e
a
r
n
i
n
g
.
D
o
d
g
s
o
n
,

1
9
9
1
T
e
c
h
n
o
l
o
g
y
,

l
e
a
r
n
i
n
g
,

t
e
c
h
n
o
l
o
g
y

s
t
r
a
t
e
g
y

a
n
d

c
o
m
p
e
t
i
t
i
v
e

p
r
e
s
s
u
r
e
s
.

B
r
i
t
i
s
h

J
o
u
r
n
a
l

o
f

M
a
n
a
g
e
m
e
n
t
,

2
/
3
:

1
3
2
1
4
9
T
a
c
t
i
c
a
l

v
e
r
s
u
s

s
t
r
a
t
e
g
i
c

l
e
a
r
n
i
n
g
.
L
e
v
i
n
t
h
a
l

a
n
d

M
a
r
c
h

(
1
9
9
3
)
T
h
e

m
y
o
p
i
a

o
f

l
e
a
r
n
i
n
g
.

S
t
r
a
t
e
g
i
c

M
a
n
a
g
e
m
e
n
t

J
o
u
r
n
a
l
,

1
4
:

9
5
1
1
2
T
w
o

m
e
c
h
a
n
i
s
m
s

o
f

o
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g
:

s
i
m
p
l
i
f
c
a
t
i
o
n

(
s
i
m
p
l
i
f
y

e
x
p
e
r
i
e
n
c
e

a
n
d

m
i
n
i
m
i
z
e

i
n
t
e
r
a
c
t
i
o
n
s
)

a
n
d

s
p
e
c
i
a
l
i
z
a
t
i
o
n

(
l
e
a
r
n
i
n
g

t
e
n
d
s

t
o

f
o
c
u
s

a
t
t
e
n
t
i
o
n

a
n
d

n
a
r
r
o
w

e
x
p
e
r
i
e
n
c
e
)
.

O
r
g
a
n
i
z
a
t
i
o
n
s

c
o
d
e

r
e
s
u
l
t
s

i
n
t
o

s
u
c
c
e
s
s
e
s

a
n
d

f
a
i
l
u
r
e
s

a
n
d

d
e
v
e
l
o
p

i
d
e
a
s

a
b
o
u
t

t
h
e
i
r

c
a
u
s
e
s
.
S
o
u
r
c
e
:

A
d
a
p
t
e
d

f
r
o
m

P
o
n
z
i
,

2
0
0
2
.
*
R
a
n
k
i
n
g
s

s
p
e
c
i
f
e
d

o
n
l
y

f
o
r

t
h
e

o
r
g
a
n
i
z
a
t
i
o
n
a
l

l
e
a
r
n
i
n
g

a
u
t
h
o
r
s

i
n

t
h
e

t
o
p

2
5

m
o
s
t

c
i
t
e
d

a
u
t
h
o
r
s

i
n

t
h
e

k
n
o
w
l
e
d
g
e

m
a
n
a
g
e
m
e
n
t

l
i
t
e
r
a
t
u
r
e

a
c
c
o
r
d
i
n
g

t
o

P
o
n
z
i
,

2
0
0
2
.
cused more on the creation of new knowledge and its transfer within the organization, whereas the
KM literature explicitly recognizes the importance of other processes, namely knowledge capture
(through externalization and internalization) and knowledge application (through direction and
routines) (Becerra-Fernandez, Gonzalez, and Sabherwal, 2004). When the OL literature considers
the use of knowledge acquired through learning, the knowledge is utilized by the individual or
group that acquires the knowledge. In contrast, according to the KM literature, knowledge could
be utilized by individuals who do not themselves have that knowledge (Grant, 1996b). This is
refected in the notion of knowledge substitution, which is implemented through directions and
routines, and is a key characteristic of the knowledge-based theory of the frm (Conner and Pra-
halad, 1996; Grant, 1996a, 1996b).
Second, the OL literature focused more on the social aspects of KM, focusing on individuals
within an organization maximizing their individual potential through a process of self-develop-
ment, assisted by the organization. The KM literature increasingly recognizes the importance of
individuals as well as their networks within and across organizations, seeking overall gains for
the organization, through exploiting the individuals ability to learn from past experiences or
lessons.
Finally, whereas the OL literature focused on social and structural aspects, the KM literature
explicitly recognizes the importance of ICT in knowledge creation, sharing, and application pro-
cesses (Sabherwal and Sabherwal, 2006). The effects of ICT are discussed later in this chapter.
From Perspective Taking OR Perspective Making to Perspective Taking AND
Perspective Making
Perspective making is the process whereby a community of knowing develops and strengthens its
own knowledge domain and practices (Boland and Tenkasi, 1995, p. 356). It is a social practice
of creating and communicating via narratives to make sense of observations. Perspective making
is a learning activity at the individual level. For instance, developing the technical paper associ-
ated with an experiment or documenting the lessons learned from a project is each an example
of perspective making.
Perspective taking is the process through which individuals are able to appreciate and syn-
ergistically utilize their distinct knowledge . . . it requires a process of mutual perspective taking
where distinctive individual knowledge is exchanged, evaluated, and integrated with that of others
in the organization (Boland and Tenkasi, 1995, p. 358). An example of perspective taking is when
collaborating scientists exchange representations of their understandings for example at a techni-
cal conference. Another example of perspective taking is when project team members participate
in a debriefng of the lessons learned following the completion of a specifc project. Perspective
taking is a learning activity at the group and organizational level. A basis for perspective taking
is the existence of boundary objects, a visible representation of an individuals knowledge that is
made visible for analysis and communication (Boland and Tenkasi, 1995). Examples of boundary
objects include cognitive maps, images, and the sort that help in sense making and taking other
perspectives into account.
According to Boland and Tenkasi (1995), fve new classes of electronic communications should
be included in KMS to improve the knowledge processes of the organization of the future (Boland
and Tenkasi, 1995):
1. Task Narrative Forums, which enable learning vicariously via video and audio without the
constraints of space and time. These serve primarily as perspective taking at the group and
organizational level because they allow groups and organizations to share their insights
via distributed collaboration media.
2. Knowledge Representation Forums, which highlight the knowledge that lies behind
complex documents and serve as a sense-making forum. They also serve as perspective
making at the individual learning level for those creating them and perspective taking at
the group and organizational level for those who use them. In other words, individuals
that contribute documents engage in perspective making during the process of creating
such documents. Other group and organizational members who access those documents
are able to use these documents as boundary objects in order to engage in perspective-
taking activities with the contributing authors,
3. Interpretative Reading Forums, which help to refect on current understandings on a topic.
They also serve as perspective making at the individual learning level and perspective
taking at the group and organizational level, but are highly dependent on a supportive
organizational culture. Individuals contributing to the reading forums can document their
perspectives via perspective-making activities, which can be commented upon by other
members of the group and organization via perspective-taking activities.
4. Theory Building Forums, which enable different communities of knowing to articulate,
critique, extend, and explore the theories that guide their work. They also serve as per-
spective making at the individual learning level and perspective taking at the group and
organizational level, and involve the construction of simulation models and virtual reality
systems. This discussion is similar to the interpretative reading forums and knowledge
representation forums, but the boundary objects used here are simulation models rather
than complex documents.
5. Intelligent Agent and Expert System Forums, which include software where expertise
has been refected upon and enable sharing insights as well as how to interpret them
(Ferguson, Mathur, and Shah, 2005). In this case, boundary objects are intelligent agents
and expert systems.
The above representation of the KMS in support of the emergent knowledge-intensive organi-
zations represents a signifcant change, which is enabled to some extent through such emerging
technologies as instant messaging, blogs, and wikis, which focus not on capturing knowledge
itself but rather on the practices and output of knowledge workers (McAfee, 2006, p. 23). The
emergent KMS are expected to support the needs of perspective making at the individual level
as well as perspective taking at the group and organizational levels. Thus, there seems to be an
ongoing shift from the earlier focus of KMS on either perspective taking or perspective making
towards integrating both perspective taking and perspective making. Both activities are required for
communities to be effective in learning and thereby become a learning organization. Consequently,
and in contrast to the highly hierarchical organizations of the past, the emergent knowledge-in-
tensive organizations, which have been labeled as Enterprise 2.0 (McAfee, 2006), are expected
to have interdependent knowledge communities, interconnected via groupware, blogs, wikis, and
instant messaging (McAfee, 2006).
Increased Embedding of KM across Organizational Processes
The increased recognition of the importance of KM led to the creation of organizational struc-
tures and roles related to KM, such as the Chief Knowledge Offcer. However, academics and
practitioners have also recognized the risk of KM being perceived as an end itself, potentially
leading to an internal bureaucracy focused on activities related to KM (Ruggles and Little,
2000). With this recognition, KM is becoming more integrally enmeshed across all organi-
zational tasks and processes, such as through the organizational routines that simultaneously
rely on knowledge and generate information that can potentially lead to the discovery of new
knowledge (Grant, 1996).
The increased embedding of KM in all walks of organizational life may also be seen in the
emergence of knowledge reciprocity as a new driver of behavioral norms in organizations (Willett,
2000). In most organizations, it is no longer suffcient to do your own job in isolation. Instead,
individuals are commonly expected to help others to succeed, and individual performance ap-
praisals are consequently dependent on the use of personal expertise in the provision of needed
help to others. This has led to an increased emphasis on a knowledge sharing culture, such as in
Texas Instruments annual Best Practice Celebration and Sharing Day (ODell and Grayson,
1998).
The model knowledge-intensive organization of today is expected to have the necessary com-
puting, imaging, and communication devices that enable a seamless integration of multimedia
for the collection, storage, processing, and display of information. In terms of its organization,
interdependent knowledge communities are interconnected via groupware, and contrast the highly
hierarchical organizations of the past. In terms of organizational culture, individuals recognize the
need to dialog and adopt a perspective-taking orientation. In short, these model knowledge-intensive
organizations are envisioned to use advanced groupware facilities to conduct meetings, construct
multi-author documents, and coordinate their promises and deadlines, all with the capability to
access data and knowledge through a world-wide network of knowledge repositories (Boland
and Tenkasi, 1995, p. 366).
KNOWLEDGE MANAGEMENT SYSTEMS AND LEARNING
In earlier sections, we have examined four evolutionary shifts in KM. In this section, we focus on
how ICT and social mechanisms come together in KMS that support learning at the individual,
group, and organizational levels.
The SECI process (Nonaka, 1991; Nonaka and Takeuchi, 1995) views knowledge creation
as the outcome of a spiraling process of interactions between explicit and tacit knowledge, each
process resulting in new forms of knowledge via (1) socialization (when knowledge is converted
between one individuals tacit knowledge to anothers); (2) externalization (when an individuals
tacit knowledge is made explicit); (3) internalization (when explicit knowledge is converted into
an individuals tacit knowledge via learning); and (4) combination (when one individuals explicit
knowledge is converted into other more complex sets of explicit knowledge).
Furthermore, according to Nonaka and Konno (1998), knowledge creation requires certain
fundamental conditions, specifcally the establishment of an organizations ba, which translates
into the word place. Ba is a shared space that may be physical (such as an offce), virtual (such
as e-mail), or a combination of both, and it provides for the opportunity for the human interaction
necessary for knowledge creation. There are four types of ba corresponding to the four stages of
the SECI model: (1) originating ba, where individuals engaged in the socialization process create
knowledge by sharing feelings, emotions, experiences, and mental models; (2) interacting ba, where
individuals with specifc knowledge and capabilities externalize their tacit knowledge and exchange
ideas and skills to develop common mental models; (3) cyber ba is the place of interaction in the
virtual world and represents the combination phase, by combining new explicit knowledge with
existing information and knowledge to create new organizational knowledge; and (4) exercising
ba, which supports the internalization phase by allowing individuals to refne their knowledge via
on-the-job training and participation (Nonaka and Konno, 1998). In the next sections, we discover
how these concepts relate to learning at the individual, group, and organizational levels.
Knowledge Management Systems and Learning at the Individual Level
Individuals acquisition and use of knowledge depend on the perceptual flters they use to interpret
events and actions (Daft and Weick, 1984; Fiol, 1994; Starbuck and Milliken, 1988). Moreover,
knowledge in groups and in organizations depends on the individuals knowledge (Cohen and
Levinthal, 1990; Fiol, 1994). Knowledge fows between individuals, which may be facilitated by
KM processes and the tools supporting them, which are important because they provide the inte-
grative ability of knowledge, from individual to group, and from group to the organizational level
(Grant, 1996a). In the transition from the individual to the group, shared beliefs develop through
the synthesis of individuals knowledge and interpretations (Probst, Bchel, and Raub, 1998).
Nonaka (1994) views internalization as the process that converts explicit knowledge into tacit
knowledge. In internalization, the explicit knowledge may be embodied in action and practice,
so that the individual acquiring the knowledge can reexperience what others go through. Alterna-
tively, individuals could acquire tacit knowledge in virtual situations, either vicariously by reading
or listening to others stories or experientially through simulations or experiments. Learning by
doing, on-the-job training, learning by observation, and face-to-face meetings are some of the
internalization processes by which individuals acquire knowledge. Nonaka (1994) draws a parallel
between internalization and the traditional notion of learning.
In contrast, externalization involves the expression of tacit knowledge and its translation into
comprehensible forms that others can understand (Nonaka, 1994; Nonaka and Takeuchi, 1995).
Conventional learning methodologies require the externalization of the professors knowledge as
the initial step in the students learning process (Raelin, 1997). Moreover, externalization involves
techniques for expressing ideas or images as words, concepts, visuals, or fgurative language
(e.g., metaphors, analogies, narratives) and deductive/inductive reasoning or creative inference.
It enables the translation of personal or professional knowledge into explicit forms that are easier
to understand.
Internalization and externalization are both fundamental to individual-level KM; through ex-
ternalization the individual makes the knowledge more agreeable and understandable to others,
while through internalization the individual absorbs knowledge others hold (Maturana and Varela,
1992; von Krogh and Roos, 1995). Whether we focus on internalization, such as when individuals
acquire knowledge by observing or talking to others, or on externalization, such as when they try
to model their knowledge into analogies, metaphors, or problem-solving systems, the learning
processes are personal and individualized (Magalhes, 1998).
KMS that support the processes of externalization

and internalization have been shown to
support learning at the individual level (Sabherwal and Becerra-Fernandez, 2003). In Table 2.2,
we summarize the fndings from that study, that is, the effects of each of the knowledge conver-
sion modes on learning at the individual level. We also summarize in that table the effects on the
group and organizational levels as will be discussed in succeeding sections. Both internalization
and externalization unlock the knowledge held by individuals . . . internalization is intrinsically
related to learning and externalization is essential to knowledge articulation (Sabherwal and
Becerra-Fernandez, 2003, p. 246). In Figure 2.1, adapted from Alavi and Leidner (2001) and
Sabherwal and Becerra-Fernandez (2003), arrows A and B represent externalization and internal-
ization processes, respectively at the individual level.
24
T
a
b
l
e

2
.
2
K
M

P
r
o
c
e
s
s
e
s
,

M
e
c
h
a
n
i
s
m
s
,

a
n
d

T
e
c
h
n
o
l
o
g
i
e
s
K
M

P
r
o
c
e
s
s
e
s
K
M
S
K
M

S
u
b
p
r
o
c
e
s
s
e
s

(
s
e
e

F
i
g
s
.

2
.
1
,

2
.
2
,

a
n
d

2
.
3
)
L
e
a
r
n
i
n
g

L
e
v
e
l

I
m
p
a
c
t
I
l
l
u
s
t
r
a
t
i
v
e

K
M

M
e
c
h
a
n
i
s
m
s
I
l
l
u
s
t
r
a
t
i
v
e

K
M

T
e
c
h
n
o
l
o
g
i
e
s
K
n
o
w
l
e
d
g
e

d
i
s
c
o
v
e
r
y
K
n
o
w
l
e
d
g
e

d
i
s
c
o
v
e
r
y

s
y
s
t
e
m
s
C
o
m
b
i
n
a
t
i
o
n

(
a
r
r
o
w

I
)
G
r
o
u
p
/
o
r
g
a
n
i
z
a
t
i
o
n
a
l
M
e
e
t
i
n
g
s
,

t
e
l
e
p
h
o
n
e

c
o
n
v
e
r
s
a
t
i
o
n
s
,

a
n
d

d
o
c
u
m
e
n
t
s
,

c
o
l
l
a
b
o
r
a
t
i
v
e

c
r
e
a
t
i
o
n

o
f

d
o
c
u
m
e
n
t
s
D
a
t
a
b
a
s
e
s
,

W
e
b
-
b
a
s
e
d

a
c
c
e
s
s

t
o

d
a
t
a
,

d
a
t
a

m
i
n
i
n
g
,

W
e
b

p
o
r
t
a
l
s
S
o
c
i
a
l
i
z
a
t
i
o
n

(
a
r
r
o
w

F
)
G
r
o
u
p
E
m
p
l
o
y
e
e

r
o
t
a
t
i
o
n

a
c
r
o
s
s

d
e
p
a
r
t
m
e
n
t
s
,

c
o
n
f
e
r
e
n
c
e
s
,

b
r
a
i
n
s
t
o
r
m
i
n
g

r
e
t
r
e
a
t
s
,

c
o
o
p
e
r
a
t
i
v
e

p
r
o
j
e
c
t
s
V
i
d
e
o
-
c
o
n
f
e
r
e
n
c
i
n
g
,

e
l
e
c
t
r
o
n
i
c

d
i
s
c
u
s
s
i
o
n

g
r
o
u
p
s
,

e
-
m
a
i
l
O
r
g
a
n
i
z
a
t
i
o
n
a
l
J
a
m

s
e
s
s
i
o
n
s

(
l
a
r
g
e
-
s
c
a
l
e

e
m
p
l
o
y
e
e
-
t
o
-
e
m
p
l
o
y
e
e

g
l
o
b
a
l

b
r
a
i
n
s
t
o
r
m
i
n
g

a
n
d

c
o
l
l
a
b
o
r
a
t
i
o
n
)
J
a
m

s
e
s
s
i
o
n
s

(
l
a
r
g
e
-
s
c
a
l
e

e
m
p
l
o
y
e
e
-
t
o
-
e
m
p
l
o
y
e
e

g
l
o
b
a
l

b
r
a
i
n
s
t
o
r
m
i
n
g

a
n
d

c
o
l
l
a
b
o
r
a
t
i
o
n
)
K
n
o
w
l
e
d
g
e

c
a
p
t
u
r
e
K
n
o
w
l
e
d
g
e

c
a
p
t
u
r
e

s
y
s
t
e
m
s
E
x
t
e
r
n
a
l
i
z
a
t
i
o
n

(
a
r
r
o
w

A
)
I
n
d
i
v
i
d
u
a
l
/
g
r
o
u
p
M
o
d
e
l
s
,

p
r
o
t
o
t
y
p
e
s
,

b
e
s
t

p
r
a
c
t
i
c
e
s
,

l
e
s
s
o
n
s

l
e
a
r
n
e
d
E
x
p
e
r
t

s
y
s
t
e
m
s
,

c
h
a
t

g
r
o
u
p
s
,

b
e
s
t

p
r
a
c
t
i
c
e
s
,

a
n
d

l
e
s
s
o
n
s

l
e
a
r
n
e
d

d
a
t
a
b
a
s
e
s
I
n
t
e
r
n
a
l
i
z
a
t
i
o
n

(
a
r
r
o
w

B
)
I
n
d
i
v
i
d
u
a
l
L
e
a
r
n
i
n
g

b
y

d
o
i
n
g
,

o
n
-
t
h
e
-
j
o
b

t
r
a
i
n
i
n
g
,

l
e
a
r
n
i
n
g

b
y

o
b
s
e
r
v
a
t
i
o
n
,

a
n
d

f
a
c
e
-
t
o
-
f
a
c
e

m
e
e
t
i
n
g
s
C
o
m
p
u
t
e
r
-
b
a
s
e
d

c
o
m
m
u
n
i
c
a
t
i
o
n
,

A
I
-
b
a
s
e
d

k
n
o
w
l
e
d
g
e

a
c
q
u
i
s
i
t
i
o
n
,

c
o
m
p
u
t
e
r
-
b
a
s
e
d

s
i
m
u
l
a
t
i
o
n
s
K
n
o
w
l
e
d
g
e

s
h
a
r
i
n
g
K
n
o
w
l
e
d
g
e

s
h
a
r
i
n
g

s
y
s
t
e
m
s
S
o
c
i
a
l
i
z
a
t
i
o
n

(
a
r
r
o
w

F
)
G
r
o
u
p
E
m
p
l
o
y
e
e

r
o
t
a
t
i
o
n

a
c
r
o
s
s

d
e
p
a
r
t
m
e
n
t
s
,

c
o
n
f
e
r
e
n
c
e
s
,

b
r
a
i
n
s
t
o
r
m
i
n
g

r
e
t
r
e
a
t
s
,

c
o
o
p
e
r
a
t
i
v
e

p
r
o
j
e
c
t
s
S
e
e

a
b
o
v
e
E
x
c
h
a
n
g
e

(
a
r
r
o
w

G
)
G
r
o
u
p
M
e
m
o
s
,

m
a
n
u
a
l
s
,

l
e
t
t
e
r
s
,

p
r
e
s
e
n
t
a
t
i
o
n
s
T
e
a
m

c
o
l
l
a
b
o
r
a
t
i
o
n

t
o
o
l
s
,

W
e
b
-
b
a
s
e
d

a
c
c
e
s
s

t
o

d
a
t
a
,

d
a
t
a
b
a
s
e
s
,

a
n
d

r
e
p
o
s
i
t
o
r
i
e
s

o
f

i
n
f
o
r
m
a
t
i
o
n
,

b
e
s
t

p
r
a
c
t
i
c
e
s

d
a
t
a
b
a
s
e
s
,

l
e
s
s
o
n
s

l
e
a
r
n
e
d

s
y
s
t
e
m
s
,

a
n
d

e
x
p
e
r
t
i
s
e

l
o
c
a
t
o
r

s
y
s
t
e
m
s
K
n
o
w
l
e
d
g
e

a
p
p
l
i
c
a
t
i
o
n
K
n
o
w
l
e
d
g
e

a
p
p
l
i
c
a
t
i
o
n

s
y
s
t
e
m
s
D
i
r
e
c
t
i
o
n

(
a
r
r
o
w

C
)
I
n
d
i
v
i
d
u
a
l
T
r
a
d
i
t
i
o
n
a
l

h
i
e
r
a
r
c
h
i
c
a
l

r
e
l
a
t
i
o
n
s
h
i
p
s

i
n

o
r
g
a
n
i
z
a
t
i
o
n
s
,

h
e
l
p

d
e
s
k
s
,

a
n
d

s
u
p
p
o
r
t

c
e
n
t
e
r
s
C
a
p
t
u
r
e

a
n
d

t
r
a
n
s
f
e
r

o
f

e
x
p
e
r
t
s

k
n
o
w
l
e
d
g
e
,

t
r
o
u
b
l
e
s
h
o
o
t
i
n
g

s
y
s
t
e
m
s
,

a
n
d

c
a
s
e
-
b
a
s
e
d

r
e
a
s
o
n
i
n
g

s
y
s
t
e
m
s
;

d
e
c
i
s
i
o
n

s
u
p
p
o
r
t

s
y
s
t
e
m
s
R
o
u
t
i
n
e
s

(
a
r
r
o
w

H
)
G
r
o
u
p
/
o
r
g
a
n
i
z
a
t
i
o
n
a
l
O
r
g
a
n
i
z
a
t
i
o
n
a
l

p
o
l
i
c
i
e
s
,

w
o
r
k

p
r
a
c
t
i
c
e
s
,

a
n
d

s
t
a
n
d
a
r
d
s
E
n
t
e
r
p
r
i
s
e

r
e
s
o
u
r
c
e

p
l
a
n
n
i
n
g

s
y
s
t
e
m
s
,

m
a
n
a
g
e
m
e
n
t

i
n
f
o
r
m
a
t
i
o
n

s
y
s
t
e
m
s
S
o
u
r
c
e
:

A
d
a
p
t
e
d

f
r
o
m

B
e
c
e
r
r
a
-
F
e
r
n
a
n
d
e
z
,

G
o
n
z
a
l
e
z
,

a
n
d

S
a
b
h
e
r
w
a
l
,

2
0
0
4
,

p
.

4
1
.
Note that processes denoted by dotted arrows are represented in Alavi and Leidner (2001) but
not discussed in this paper.
In addition to internalization and externalization, the process of direction has also been found
to infuence KM at the individual level (Sabherwal and Becerra-Fernandez, 2005). Direction is
the process whereby an individual possessing certain knowledge directs the action of another in-
dividual without transferring the underlying knowledge to that individual (Grant, 1996b). While
direction is not an effective mechanism for knowledge transfer, it is effcient in transferring
instructions between individuals (Grant, 1996b). An example of direction is a customer calling a
vendors product support department to receive instructions regarding how to address problems
faced while using one of that vendors products. In Figure 2.1, arrow C represents the process of
direction between individuals. Individuals may issue directions verbally, through documents (such
as documents or memos), or virtually (such as via e-mail).
Referring back to the discussion on the organizations ba, interacting ba is associated with
the process of externalization, and it refers to a space where an individuals tacit knowledge is
Figure 2.1 KMS and Learning at the Individual Level
Individual As
Tacit Knowledge
Individual Bs
Tacit Knowledge
Individual As
Explicit Knowledge
Individual Bs
Explicit Knowledge
Knowledge
Application
Storage
(documents, email)
Storage
(documents, email)
Group 1s Semantic Memory
Group 1s Episodic
Memory
D
E
E
E
D D
A A
B B
C
A A
B B
A The process of externalization
B The process of internalization and learning at the individual level
C The process of direction
D The process of knowledge capture of individual explicit knowledge to group episodic memory and
vice versa
E The process of knowledge capture of individual tacit knowledge to group episodic memory and vice
versa
Source: Adapted from Alavi and Leidner, 2001, and Sabherwal and Becerra-Fernandez, 2003.
Note: Dotted arrows are given in Alavi and Leidner, 2001, but not discussed in this paper.
converted to explicit knowledge and shared with other individuals through dialogue and collabo-
ration (Alavi and Leidner, 2001). Furthermore, exercising ba involves the conversion of explicit
knowledge to tacit via the internalization process and entails a space for active and continuous
individual learning (Alavi and Leidner, 2001, p. 117) Therefore, both interacting ba and exercis-
ing ba support learning at the individual level.
Together, internalization, externalization, and direction support the KM processes of knowl-
edge capture and knowledge application (Becerra-Fernandez, Gonzalez, and Sabherwal, 2004).
Knowledge can be captured into organizational memories, which can be classifed as semantic
or episodic. Arrows D and E in Figure 2.1 represent the process of knowledge capture. Semantic
memory is general, explicit, or articulated knowledge (such as annual reports) (Alavi and Leidner,
2001; Stein and Zwass, 1995). Episodic memory is context-specifc, situated knowledge (such
as lessons learned). In particular, knowledge capture and knowledge application are KM activi-
ties that support learning and applying new knowledge, respectively, at the individual level. The
early development of intelligent systems such as expert systems and decision support systems
supported the KM needs of eliciting and capturing the knowledge of experts, in such a way that
other individuals could apply this knowledge. Capturing knowledge into organizational memo-
ries is the frst step in transferring knowledge between individuals, and from the individual to the
group, and then the organization. In the next section, we discuss the effect of KMS and learning
at the group level.
Knowledge Management Systems and Learning at the Group Level
The progression of knowledge from the individual continues to the group level. This progres-
sion takes place through the synthesis of individual knowledge via social interaction, as well as
through the use of beliefs shared across groups to create organizational routines (Probst, Bchel,
and Raub, 1998). Furthermore, knowledge in groups and in organizations depends on the indi-
viduals knowledge (Cohen and Levinthal, 1990; Fiol, 1994). Perceived effectiveness of KM at
the individual level has been shown to infuence perceived effectiveness of KM at the group level
(Sabherwal and Fernandez, 2003), which reinforces the views that aggregation of knowledge to a
higher hierarchical level is essential for knowledge growth (Holtshouse, 1998).
According to Nonaka (1994), socialization involves the sharing of tacit knowledge between
individuals. It helps exchange knowledge through joint activities, such as being together in the
same environment, rather than through written or verbal form (Nonaka and Takeuchi, 1995). In
Figure 2.2, arrow F represents the process of socialization between groups of individuals. By
transferring ideas and images, apprenticeships allow newcomers to see the way others think and
feel. Knowledge is produced in a group setting not only through mere acquisition of the individu-
als knowledge but also through the sharing of common understanding, which helps synergize the
individuals knowledge (Fiol, 1994). Social processes play an important role in the transition of
knowledge from the individual level to the group level (Walsh, 1995; Weick and Roberts, 1993).
Davenport and Prusak (1998) describe how conversations at the water cooler helped knowledge
sharing among groups at IBM. Socialization has traditionally relied on physical proximity and
joint action (Huseman and Goodman, 1999), although this might be changing due to the rapid
progress in ICT. KMS that support socialization have been shown to support learning at the group
level (Sabherwal and Becerra-Fernandez, 2003).
Externalization, represented by arrow A in Figure 2.2, was also found to have an impact on
group learning via the effects of individual learning on group learning (Sabherwal and Becerra-
Fernandez, 2003). Externalization enables individuals to express their knowledge such that the rest
of their group can more easily understand it. Externalization consequently increases the utilization
of knowledge by a wider circle of individuals (Hansen et al., 1999; Kogut and Zander, 1992).
Exchange is the transfer of explicit knowledge between individuals in a group (Grant, 1996b). In
Figure 2.2, arrow G represents the process of exchange between groups of individuals. Exchange de-
pends on the conversion of tacit knowledge into explicit form, by externalization. Exchange is similar
to direction in that they both involve communication of information, but while direction communicates
the solution, exchange transfers the explicated knowledge. Exchange has been found to require high
levels of common knowledge (Sabherwal and Becerra-Fernandez, 2005) such as that we expect to fnd
among members of a group. For that reason, the impact of exchange is primarily at the group level.
Routines involve the utilization of knowledge embedded in procedures, rules, and norms that
guide future behavior (Becerra-Fernandez et al., 2004). In Figure 2.2, arrow H represents the process
of routines between groups of individuals. Routines are effcient, in the sense that they economize
on communications more than direction, because they become embedded in procedures and tech-
nologies such as ERP systems. Routines take time to develop and rely on repetition (Grant, 1996a).
Routines, like direction, may be limited in terms of the opportunities for learning they provide.
Routines are effcient in transferring instructions at the group and the organizational level.
Figure 2.2 KMS and Learning at the Group Level
Individuals A, B, C
Tacit Knowledge
Individuals D, E, F
Tacit Knowledge
Individuals A, B, C
Explicit Knowledge
Individuals D,E,F
Explicit Knowledge
Storage
(documents, email)
Storage
(documents, email)
Group 1s Episodic
Memory
D
E
E
E
D
G
A
D
Group 1s Episodic
Memory
Group 2s Episodic
Memory
A
F
I
A The process of externalization
D The process of knowledge capture of group explicit knowledge to group episodic memory and vice
versa
E The process of knowledge capture of group tacit knowledge to group episodic memory and vice
versa
F The process of socialization
G The process of exchange among group members
H The process of routines
I The process of combination
Source: Adapted from Alavi and Leidner, 2001. and Sabherwal and Becerra-Fernandez, 2003.
Note: Dotted arrows are given in Alavi and Leidner, 2001, but not discussed in this paper.
For example, in a study of successful far-fung teams (virtual teams that are multiunit/multiorga-
nizational, multifunctional, and globally dispersed), it was found that distinct communication and
knowledge sharing norms emerge (Malhotra and Majchrzak, 2004). Four different types of ICT sup-
port are required for task coordination, external connectivity, distributed cognition, and interactivity
(Malhotra and Majchrzak, 2004). Far-fung teams use episodic and semantic memories to support
distributed cognition (in support of externalization and routines) and extensive communication (in
support of exchange). Arrows D and E in Figure 2.2 represent the process of capturing group knowl-
edge into semantic and episodic group memories, respectively. Instant messaging (IM) (in support
of exchange) is another technology that characterizes far-fung teams. But successful far-fung teams
also implement knowledge sharing norms, such as, When members felt their exchange with another
member had to be private, they used IM for communication (in support of socialization) (Malhotra
and Majchrzak, 2004). In fact, the same study showed that far-fung teams have to achieve a strategic
ft between task characteristics, team composition, and ICT support, to overcome diffculties and
barriers associated with knowledge sharing and knowledge creation.
Referring back to the discussion on the organizations ba, originating ba is associated with the
process of socialization, and it refers to a space where individuals share experiences primarily
through face-to-face interactions and being in the same place at the same time (Alavi and Leidner,
2001). Originating ba is the ba from which the organizational knowledge creation process be-
gins (Alavi and Leidner, 2001, p. 116) by supporting learning at the group level frst and then
transcending to the organizational level.
Together, socialization, exchange, routines, and externalization (via the effect of individual
learning on group learning) support the KM process of knowledge sharing and application (Becerra-
Fernandez, Gonzalez, and Sabherwal, 2004). As the needs for group collaboration became evident,
knowledge sharing systems supporting socialization, exchange, and externalization among team
members were deployed across organizations. This type of activity, the deployment of knowledge
repositories and lessons learned systems, marks the introduction of KM as a relevant discipline
in organizations (Alavi, 1997; Hansen and Davenport, 1998). In addition, organizations imple-
mented systems for collaboration such as chat groups and expertise locator systems (also known
as yellow pages) (Becerra-Fernandez, 2000, 2006), which make individuals knowledge more
accessible to the group (Huseman and Goodman, 1999). Advanced computer storage and col-
laboration technology, such as groupware and document management systems, enables groups to
share and combine their organizational memory (Alavi and Leidner, 2001). Arrow I in Figure 2.2
represents the process of combination, which enables groups to convert their explicit knowledge
into new and more complex forms of explicit knowledge. The process of combination impacts
OL at the group and the organizational level.
Knowledge Management Systems and Learning at the Organizational Level
Argote (1999) identifes prior research on knowledge fows across various groups in organizations, across
products or models of the same product, and across organizational units (e.g., shifts within a manufac-
turing plant or departments). Consequently, organizational knowledge arises from group knowledge
as well as from individual knowledge, as is also indicated by the recent literature on communities of
practice (e.g., Brown and Duguid, 1991; Boland and Tenkasi, 1995; Storck and Hill, 2000).
Combination processes involve the conversion of explicit knowledge into more complex sets
of explicit knowledge (Nonaka, 1994). Focusing on communication, diffusion, integration, and
systemization of knowledge, combination contributes to knowledge at the group as well as at
the organizational level (Nonaka and Takeuchi, 1995). Combination helps integrate knowledge
of group members, but the new knowledge generated through combination often transcends the
group (Nonaka and Konno, 1998). In fact, Nonaka and Takeuchi (1995) refer to organizational
knowledge creation as the single market differentiator for Japanese companies, which are able to
successfully combine new technological innovations into novel products throughout the organiza-
tion, bringing about continuous innovation, and thus competitive advantage. Innovative organiza-
tions seek to develop new concepts that are created, justifed, and modeled at the organizational,
and sometimes interorganizational, level. Zander and Kogut (1995) also suggest that rates of
transfer and imitation . . . (might depend on the ability) to recombine their knowledge to improve
the innovation (p. 77). Moreover, complex organizational processes require the cooperation of
various groups within the organization, and combination supports these processes by aggregating
technologies and know-how (Nonaka, 1994). Indeed, knowledge that is not shared has limited
organizational value (Alavi and Leidner, 1999, 2001).
KMS that support the process of combination have been shown to support learning at the or-
ganizational level (Sabherwal and Becerra-Fernandez, 2003). Combination processes across the
organization are depicted by arrow I in Figure 2.3, which enables groups to convert their explicit
knowledge into new and more complex forms of explicit organizational knowledge. Referring back
to the discussion on the organizations ba, cyber ba is associated with the process of combination
and refers to a virtual space of interaction (Alavi and Leidner, 2001).
In addition, routines are also effcient ways to disseminate directions at the organizational level,
although the contribution they make to OL may not be considered as substantial.
Figure 2.3 summarizes the transfer of knowledge between individuals, groups, and organiza-
tions. Knowledge transfer (via socialization, combination, and routines) across individuals may
Figure 2.3 KMS and Learning at the Organizational Level
Individuals A, B, C
Knowledge
Individuals D,E,F
Knowledge
Group 1s Episodic
Memory
E
H/F
G
Group 1s Episodic
Memory
Group 2s Episodic
Memory
I
Group 3s Episodic
Memory
Individuals G, H, I
Knowledge
I
I
H/F
H/F
F The process of socialization across the organization
H The process of routines across the organization
I The process of combination across the organization
Source: Adapted from Alavi and Leidner, 2001, and Sabherwal and Becerra-Fernandez, 2003.
lead to knowledge creation, as knowledge fows among individuals in a community of practice they
develop group knowledge, which they collectively store in their organizational memory (Alavi and
Leidner, 2001). Organizational knowledge processes consist in the summation of the individual
and group knowledge processes (Alavi and Leidner, 2001). Figure 2.3 also points out one of the
major challenges in KM: to make individual and group knowledge meaningful to others across
the organization (Alavi and Leidner, 2001). For example, lessons learned systems (LLS) (Weber,
Aha, and Becerra-Fernandez, 2001) were one of the frst types of KMS that gained acceptance
in organizations, and many consulting agencies quickly adopted the implementation of such sys-
tems, which later they proposed to their clients. As the volume of information represented in these
knowledge bases increased, information and knowledge overload coupled with a lack of awareness
about the lessons context became a signifcant problem (Davenport and Hansen, 1998).
One of the problems was that many documents failed to provide much information about the
context in which the insight and experience embodied in the document was generated . . . Imagine
. . . you start reusing the drawings for the development of an offce building and notice that the
elevators in the previous drawings were constructed to be very deep. You later discover that the
previous drawings were for a hospital, where elevators need to be deep to hold beds. Wouldnt
it have been helpful if the document had stated up front that the drawings were only relevant
for hospital elevators? Contextualizing . . . would involve . . . describing the context in which
the knowledge was generated and used, where it may be useful, and where it shouldand
should notbe used. (Davenport and Hansen, 1998, p. 9)
Lessons learned systems that include contextual components could augment a recipients abil-
ity in understanding the context in which the knowledge was generated and used, and therefore
help overcome the limitations described above. For example, when lessons are being collected,
attributes describing the identity, sensory, informational, and positioning components could be
detected in the background, aided by the sensor network. These attributes could then be combined
with the users cognitive model, which could be used to stamp the lesson learned. This stamp could
effectively describe where the lesson originated and under what physical and logical conditions.
Subsequently, when lessons are being retrieved, in order to ascertain their relevance to the current
context, the system could match the stamp of the retrieved lessons to the users current contextual
components. Thus, using the example above, the system could proactively inform the user that
the retrieved elevator design originated from the hospital architecture group and thus may not be
applicable to an offce building design (Becerra-Fernandez, Cousins, and Weber, 2007). In sum-
mary, OL requires improving the linkages between individuals and between groups (Alavi and
Leidener, 2001) that enable the contextualization of knowledge across the organization (Becerra-
Fernandez, Cousins, and Weber, 2007).
Together, combination and routines support the KM process of knowledge discovery and ap-
plication at the organizational level (Becerra-Fernandez et al., 2004). Perhaps is in this area that
we are witnessing the latest wave in the deployment of KMS in support of OL.
THREE GENERATIONS OF KNOWLEDGE MANAGEMENT SYSTEMS
FOR PERSPECTIVE MAKING AND PERSPECTIVE TAKING AND
THEIR IMPACT ON LEARNING
Next, we build on the above discussion of shifts in KM and the nature of KMS to describe three
generations of KMS.
First-Generation Knowledge Management Systems and Their Impact on
Individual Learning
We tie the frst generation of KMS in support of knowledge capture and knowledge application at
the individual level to the development of the feld of artifcial intelligence (AI), which dates back
to the 1950s. The feld of AI is grounded on the realization that intelligence is tightly intertwined
with knowledge: Knowledge is associated with the symbols we manipulate. We can informally
describe human intelligence as our innate ability to learn and manipulate knowledge in order to
communicate or to solve a problem (Becerra-Fernandez et al., 2004, p. 102). The early emphasis
in AI research was on game playing, machine translation of natural languages, and systems to aid
problem solving by searching for an answer in a solution space.
By the 1970s, AI researchers realized that in order to build knowledge-based or expert sys-
tems, what was required was specifc knowledge about a particular narrow application domain of
interest. AI researchers then turned their attention to the perception-making processes of eliciting
and representing the tacit knowledge that resided in domain experts. These knowledge elicita-
tion processes focused on aiding experts in externalizing their knowledge. Knowledge engineers
who developed the knowledge-based systems attempted to internalize the experts knowledge in
order to accurately represent it in computerized models. In this process, learning occurred at the
individual level, primarily for the knowledge engineer. But the learning gains to other users in
the organization may have been limited, primarily because for many of them the utilization of
these intelligent systems may have offered more direction than opportunities for internalization
of the experts knowledge. In general, knowledge-based systems did not provide opportunities for
perspective taking, thereby limiting the potential for learning.
Second-Generation Knowledge Management Systems and Their Impact on
Individual and Group Learning
The early deployment of KMS in the 1980s and 1990s marks what we describe as the second
generation of KMS. The majority of the systems deployed during this second generation are knowl-
edge sharing systems known as knowledge repositories. Knowledge repositories were designed
to enable users to share explicit knowledge, such as documents at the group level. The standard
communications medium upon which knowledge repositories are based is the World Wide Web,
which facilitates the exchange of information, data, multimedia, and even applications among
multiple distinct computer platforms.
At the core of the knowledge repository is an electronic storage medium for documents with
multiple access points. A knowledge repository unifes an aggregate of relevant documents through
a common interface, typically Web based. This repository can be centralized or it can be distrib-
uted. Document management builds on the repository by adding support to the classifcation and
organization of information, unifying the actions of storage and retrieval of documents over a
platform-independent system. Documents are typically organized or indexed following a standard
hierarchical structure or classifcation taxonomy, much like the index catalog is used to organize
books in a library. Frequently, portal technologies are used to build a common entry into multiple
distributed repositories, using the analogy of a door as a common entry into the organizations
knowledge resources. In addition to document repositories, knowledge sharing systems may also
be classifed as incident report databases, alert systems, best practices databases, LLS, and expertise
locator systems (Becerra-Fernandez, Gonzalez, and Sabherwal, 2004).
Knowledge sharing systems support the processes of perspective making for the contributors
to the knowledge repository and perspective taking for the users of the knowledge repository.
For example, a review of knowledge repositories such as LLS reveals that in general they focus
on the needs of specifc groups within an organization, such as the consultants or the auditors
within an organization (Davenport and Hansen, 1998). For example, it is not unusual for organi-
zations to support more than one knowledge repository. Typically, knowledge repositories work
well within one particular group context, but oftentimes problems manifest themselves as other
groups within an organization seek to reuse the captured knowledge without proper awareness of
its context (Davenport and Hansen, 1998). Frequently, project groups will fund the deployment
of a knowledge repository to serve the needs of the project team, and the challenges of integration
are left for others to fgure out. Consequently, many times these knowledge repositories are not
integrated. As a result, knowledge repositories support perspective making at the individual level,
and perspective taking at the group level, thereby supporting learning at the individual and group
levels but not at the organizational level.
Third-Generation Knowledge Management Systems and Their Impacts on
Individual, Group, and Organizational Learning
Knowledge discovery systems support the processes to create new knowledge. Knowledge discovery
systems facilitate the process of socialization. Socialization enables the synthesis of tacit knowledge
across individuals and the integration of multiple streams for the creation of new knowledge, usually
through joint activities rather than via written or verbal instructions. For example, one mechanism
for socialization is research conferences, which enable researchers to develop new insights through
sharing their own fndings. Also, when friends brainstorm and do back-of-the-napkin diagrams,
leading to the discovery of new knowledge that did not exist individually before the group activity,
knowledge is created or discovered by the team. Socialization processes best support the processes of
perspective making and perspective taking for knowledge discovery. As summarized by Simon (1991),
It is not their research products that we value, but their engagement in research which guarantees
their attention to the literatureto the new knowledge being produced elsewhere (p. 130).
In addition, new explicit knowledge is discovered through the process of combination, wherein
the multiple bodies of explicit knowledge (and/or data and/or information) are synthesized to
create new, more complex sets of explicit knowledge. In order for knowledge to be useful at the
organizational level, existing explicit knowledge must be recontextualized to produce new explicit
knowledge, for example, during the creation of a new offce building design that is based upon
existing prior designs. Referring back to the prior example about the reuse of elevator drawings,
combining the information about the elevator drawing with contextual cues that may defne that
the design originates from the hospital design group may help users infer that the particular draw-
ings are not applicable for offce buildings.
Furthermore, most KM applications assume that knowledge workers are stationary and carry
out the KM processes of knowledge discovery, creation, sharing, and application (Becerra-Fer-
nandez, Gonzalez, and Sabherwal, 2004; Derballa and Pousttchi, 2004) using systems that are
accessed through fxed devices and wired networks. Even though it is more and more common
for knowledge to be created and used on the move by knowledge workers, not much attention
has been given to the development of nomadic context-aware KM applications (Lehmann et al.,
2005). Evidently, the next step in the evolution of KMS involves the move toward nomadic con-
text-aware KM environments, where computing becomes an inseparable part of the environment
and work of the knowledge worker, as they carry out their knowledge-intensive activities on the
move. Finally, contextual cues in nomadic computing environments can be combined with the
existing KMS content to enhance the systems and overcome their current contextual limitations
(Becerra-Fernandez, Cousins, and Weber, 2007).
Knowledge discovery systems may also be supported by data mining technologies, which may
be used to uncover new relationships among explicit data, which in turn can serve to develop
models that can predict or categorizehighly valuable assets in business intelligence. Therefore,
knowledge discovery mechanisms and technologies can facilitate socialization and combination
within or across organizations, and therefore impact OL.
For example, IBM has recently introduced a new infrastructure for global Intranet collabora-
tion via jam events (Fontaine, Millen, and ODriscoll, 2003). Jam events support very large-scale
employee-to-employee global brainstorming and collaboration. Jam topics are selected around
operational, transformational, and social topics with relevant and provocative issues and ques-
tions around a central theme. A jam marketing campaign with strong executive sponsorship is
directed to target participants around the globe. Typically, a behind-the-scene facilitator provides
around-the-clock support to the jam event, energizes the discussion with provocative comments,
and identifes the most important ideas. Jam events enable IBM employees to share their collective
wisdom around the world to identify appropriate solutions to their most signifcant challenges.
In this manner, jam events support the processes of perspective making and perspective taking at
the organizational level, thereby promoting the goals of OL.
Another emerging technology in this space is the use of wikis to support global collaboration. A
wiki is a web application that allows users to add content, as on an Internet forum, but also allows
anyone to edit the content (Wikipedia, September 25, 2005). Wikis allow users to write documents
collectively via Web browsers via marking-up pages that are interconnected via hyperlinks. Wikis
support the process of combination by enabling iterations of perspective making and perspective
taking for contributors and mark-up reviewers, thereby contributing to the goals of OL.
CONCLUSIONS
The goals of KM and OL are intertwined. KMS support the goals of learning at the individual,
group, and organizational levels: KMS and OL tie together via the development of the Internet
and other collaboration technologies that provide opportunities for organization-wide socializa-
tion. KMS support the processes of perspective making and perspective taking at the individual,
group, and organizational levels.
Early development of intelligent systems provided opportunities for externalization, internal-
ization, and direction, thereby supporting the needs of individual learning. We describe these as
the frst-generation KMS. The second generation of KMS is marked by the wide deployment of
knowledge repositories, which provided support for socialization and exchange. But limitations
placed by the importance of context in the applicability of knowledge repositories narrowed the
possibilities for their impact, thereby supporting learning only at the group level.
A third generation of KMS is emerging, as technologies for global brainstorming and collabo-
ration such as jam events and wikis are being introduced in organizations. These technologies are
providing opportunities for perspective making and perspective taking at the organizational level by
supporting the processes of combination and socialization, thereby promoting the goals of OL.
REFERENCES
Adler, P.S. 1989. When knowledge is the critical resource, knowledge management is the critical task. IEEE
Transactions on Engineering Management, 36, 2, 8794.
Alavi, M. 1997. KPMG Pete Marwick, U.S.: One giant brain. Boston: Harvard Business School Publishing.
Case No. 9-397-108, April.
Alavi, M., and Leidner, D. 1999. Knowledge management systems: Issues, challenges, and benefts. Com-
munications of the Association for Information Systems, 1,7.
. 2001. Knowledge management and knowledge management systems: Conceptual foundations and
research issues. MIS Quarterly, 25, 1, 107136.
Anderson, J.C. 1987. An approach for confrmatory measurement and structural equation modeling of orga-
nizational properties. Management Science, 33, 4, 525541.
Anderson, J.C., and Gerbing, D.W. 1988. Structural equation modeling in practice: A review and recom-
mended two-step approach. Psychological Bulletin, 103, 3, 411423.
Argote, L. 1999. Organizational Learning: Creating, Retaining and Transferring Knowledge. Norwell, MA:
Kluwer Academic Publishers Group.
Argote, L., and Ingram, P. 2000. Knowledge transfer: Basis for competitive advantage in frms. Organizational
Behavior and Human Decision Processes, 82, 1, 150169.
Argyris, C. 1977. Organizational learning and management information systems. Accounting, Organizations,
and Society, 2, 2, 113123.
Argyris, C., and Schon, D.A. 1974. Theories in practice: Increasing Professional Effectiveness. San Fran-
cisco: Jossey-Bass.
Armbrecht, F.M.R.; Chapas, R.B.; Chappelow, C.C.; Farris, G.F.; Friga, P.N.; Hartz, C.A.; McIlvaine, M.E.;
Postle, S.R.; and Whitwell, G.E. 2001. Knowledge management in research and development. Research
Technology Management, 44, 4, 2848.
Arrow, K.J. 1974. The Limits of Organization. New York: Norton.
Babin, B.J., and Boles, J.S. 1998. Employee behavior in a service environment: A model and test of potential
differences between men and women. Journal of Marketing, 62, 2, 7791.
Baker, J.; Parasuraman, A.; Grewal, D.; and Voss, G.B. 2002. The infuence of multiple store environment
cues on perceived merchandise value and patronage intentions. Journal of Marketing, 66, 120141.
Bagozzi, R.P., and Yi, Y. 1988. On the evaluation of structural evaluation models. Journal of Academy of
Marketing Science, 16 (Spring), 7494.
Baum, J.A.C., and Ingram, P. 1998. Survival-enhancing learning in the Manhattan hotel industry, 18981980.
Management Science, 44, 9961016.
Baum, J.R.; Locke, E.A.; and Smith, K.G. 2001. A multidimensional model of venture growth. Academy of
Management Journal 44, 2, 292303.
Becerra-Fernandez, I. 2000. The role of artifcial intelligence technologies in the implementation of people-
fnder knowledge management systems. Knowledge Based Systems, 13, 5, 315320.
. 2006. Searching for experts on the Web: A review of contemporary expertise locator systems. ACM
Transactions on Internet Technology, 6, 4, 333355.
Becerra-Fernandez, I.; Cousins, K.; and Weber, R. 2007. Nomadic context-aware knowledge management
systems: Applications, challenges, and research problems. International Journal of Mobile Learning
Organisation, 1, 2, 103121.
tions, and Technologies. Upper Saddle River, NJ: Prentice-Hall.
Becerra-Fernandez, I., and Sabherwal, R. 2001. Organizational knowledge management processes: A con-
tingency perspective. Journal of Management Information Systems, 18, 1, 2355.
Bentler, P.M., and Bonnett, D.G. 1980. Signifcance tests and goodness-of-ft in the analysis of covariance
structures. Psychological Bulletin, 88 (November), 588606.
Boland, R.J., and Tenkasi, R.V. 1995. Perspective making and perspective taking in communities of practice.
Organization Science, 6, 350372.
Bollen, K.A. 1989. Structural Equations With Latent Variables. Chapel Hill, NC: John Wiley and Sons.
Brown, J.S., and Duguid, P. 1991. Organizational learning and communities-of-practice: Toward a unifed
view of working, learning and innovation. Organization Science, 2, 1, 4057.
. 2001. Knowledge and organization: A social-practice perspective. Organization Science, 12, 2,
198213.
Brusoni, S.; Prencipe, A.; and Pavitt, K. 2001.Knowledge specialization, organizational coupling, and the boundaries
of the frm: Why do frms know more than they make? Administrative Science Quarterly, 46, 597621.
Buenger, V.; Daft, R.L.; Conlon, E.J.; and Austin, J. 1996. Competing values in organizations: Contextual
infuences and structural consequence. Organization Science 7, 5, 557576.
Burrell, G., and Morgan, G. 1979. Sociological Paradigms and Organizational Analysis. London: Heinemann.
Carlile, P.R., and Rebentisch, E.S. 2003. Into the black box: The knowledge transformation cycle. Manage-
ment Science, 49, 9, 11801195.
Carlson, J.R., and Zmud, R.W. 1999. Channel expansion theory and the experiential nature of media richness
perceptions. Academy of Management Journal, 42, 2, 153170.
Carmines, E.G., and McIver, S.P. 1981. Analyzing models with unobserved variables: Analysis of covariance
structures. In G.W. Bohrnstedt and E.F. Borgatta (eds.), Social Measurement: Current Issues. Beverly
Hills, CA: Sage Publications, pp. 65115.
Chin, W.W., and Newsted, P.R. 1995. The importance of specifcation in causal modeling: The case of end-
user computing satisfaction. Information Systems Research, 6, 1, 7381.
Choudhury, V., and Sampler, J. 1997. Information specifcity and environmental scanning: An economic
perspective. MIS Quarterly, 21, 1, 2553.
Churchman, C.W. 1972. The Design of Inquiring Systems: Basic Concepts of Systems and Organizations.
New York: Bencis Books.
Cohen, W.M., and Levinthal, D.A. 1990. Absorptive capacity: A new perspective on learning and innovation.
Administrative Science Quarterly, 35, 128152.
Conner, K.R., and Prahalad, C.K. 1996. A resource-based theory of the frm: Knowledge versus opportunism,
Organization Science, 7, 5, 477501.
Craig, D.A., Moore, A.A.; Monnell, D.W.; and Simpkins, P.A. 2000. A status of the development and utiliza-
tion of NASAs advanced engineering environment. Working Paper, NASA.
Das, A. 2003. Knowledge and productivity in technical support work. Management Science, 49, 4, 416
431.
Davenport, T., and Hansen, M. 1998. Knowledge management at Andersen Consulting. Boston: Harvard
Business School Publishing. Case No. 9-499-032.
Davenport, T., and Prusak, L. 1998. Working Knowledge. Boston: Harvard Business School Press.
Davenport, T.H., and Harris, J.G. 2005. Automated decision making comes of age. MIT Sloan Management
Review, 46 (Summer), 8389.
Davy, J.A., and Shipper, F. 1993. Voter behavior in union certifcation elections: A longitudinal study. Academy
of Management Journal, 36, 1, 187199.
Dean, J.W., Jr.; Yoon, S.J.; and Susman, G.I. 1992. Advanced manufacturing technology and organization
structure: Empowerment or subordination? Organization Science, 3, 2, 203229.
Demsetz, B.H. 1998. The theory of the frm revisited. Journal of Law, Economics and Organization, 4
(Spring), 141161.
Derballa, V., and Pousttchi, K. 2004. Extending knowledge management to mobile workplaces. Proceedings
of the Sixth International Conference on Electronic Commerce ICEC04.
Dougherty, D. 1992. A practice-centered model of organizational renewal through product innovation. Stra-
tegic Management Journal, 13 (Summer), 7792.
Duncan, R., and Weiss, A. 1979. Organizational learning: Implications for organizational design. In B. Staw
(ed.), Research in Organizational Behavior, Volume 1. Greenwich, CT: JAI.
Ferguson, G., Mathur, S., and Shah, B. 2005. Evolving from information to insight. 2005. MIT Sloan Man-
agement Review, 46 (Winter), 5158.
Fiol, C.M. 1994. Consensus, diversity, and learning in organizations. Organization Science, 5, 3, 403420.
Fiol, C.M., and Lyles, M.A. 1985. Organizational learning. Academy of Management Review, 10, 4,
803813.
Fontaine, M.; Millen, D.; and ODriscoll, T. 2003. Driving intranet collaboration at IBM. KM Review, 6
(Nov/Dec), 2023.
Fornell, C., and Larcker, D.F. 1981. Evaluating structural evaluation models with unobservable variables and
measurement error. Journal of Marketing Research 18 (February), 3950.
Fulk, J. 1993. Social construction of communication technology. Academy of Management Journal, 36, 5,
921950.
Fulmer, R.; Gibbs, P.; and Keys, J. 1998. The second generation learning organizations: New tools for sus-
taining competitive advantage. Organizational Dynamics, (Autumn), 720.
Gerbing, D.W., and Anderson, J.C. 1992. Monte Carlo evaluations of goodness of ft indices for structural
equation models. Sociological Methods and Research, 21, 132160.
Grandori, A., and Kogut, B. 2002. Dialogue on organization and knowledge. Organization Science, 13, 3,
224231.
Grant, R.M. 1996a. Prospering in dynamically-competitive environments: Organizational capability as
knowledge integration. Organization Science, 7, 4, 375387.
. 1996b. Toward a knowledge-based theory of the frm. Strategic Management Journal, 17,
109122.
Gupta, A.K., and Govindarajan, V. 2000. Knowledge managements social dimension: Lessons from Nucor
Steel. Sloan Management Review, (Fall), 7180.
Hansen, M., and Davenport, T. 1998. Knowledge management at Andersen Consulting. Boston: Harvard
Business School Publishing. Case No. 9-499-032.
Hansen, M.; Nohria, N.; and Tierney, T. 1999. Whats your strategy for managing knowledge? Harvard
Business Review, (MarchApril), 106119.
Hansen, M.T. 1999. The search transfer problem: The role of weak ties in sharing knowledge across organi-
zation subunits. Administrative Science Quarterly, 44, 1, 82111.
Hayek, F.A. 1945. The use of knowledge in society. The American Economic Review, 35, 4, 519530.
Hedberg, B. 1981. How organizations learn and unlearn? In P.C. Nystrom and W.H. Starbuck (eds.), Hand-
book of Organizational Design. London: Oxford University Press.
Hedlund, G. 1994. A model of knowledge management and the N-form corporation. Strategic Management
Journal, 15, 7390.
Henderson, R.M., and K.B. Clark. 1990. Architectural innovation: The reconfguration of existing product
technologies and the failure of established frms. Administrative Science Quarterly, 35, 1, 930.
Holtshouse, D. 1998. Knowledge research issues. California Management Review, 40, 3, 277280.
Huber, G.P. 1991. Organizational learning: The contributing processes and the literatures. Organization
Science, 2, 1, 88115.
Hult, G.T.M. 1998. Managing the international strategic sourcing process as a market-driven organizational
learning system. Decision Sciences, 29, 1, 193216.
Hult, G.T.M.; Hurley, R.F.; Giunipero, L.C.; and Nichols, E.L., Jr. 2000. Organizational learning in global pur-
chasing: A model and test of internal users and corporate buyers. Decision Sciences, 31, 2, 293325.
Hurley, R.F. 1995. Group culture and its effect on innovative productivity. Journal of Engineering and Tech-
nology Management, 12 (July), 5775.
Hurley, R.F., and Hult, T.M. 1988. Innovation, market orientation, and organizational learning: An integration
and empirical examination. Journal of Marketing, 62 (July), 4254.
Huseman, R.C., and Goodman, J.P. 1999. Leading with Knowledge: The Nature of Completion in the 21st
Century. Thousand Oaks, CA: Sage Publications.
Hyman, J. 1999. How knowledge works. The Philosophical Quarterly, 49 (197), 433451.
Jensen, M.C., and Meckling, W.H. 1996. Specifc and general knowledge, and organizational structure.
In P.S. Myers (ed.), Knowledge Management and Organizational Design. Newton, MA: Butterworth-
Heinemann, pp. 1738.
Jreskog, K.G. 1978. Structural analysis of covariance and correlation matrices. Psychometrika, 43,
443487.
Jreskog, K.G., and Srbom, D. 2001. LISREL 8: Users Reference Guide (2nd ed.). Chicago: Scientifc
Software International, Inc.
Kock, N., and Davison, R. 2003. Can lean media support knowledge sharing? Investigating a hidden advantage
of process improvement. IEEE Transactions on Engineering Management, 50, 2, 151163.
Kogut, B., and Zander, U. 1992. Knowledge of the frm, combinative capabilities and the replication of
technology. Organization Science, 3, 3, 383397.
Lehmann, H., Berger, S., Remus, U. and Lehner, F. (2005). Towards a Functional Architecture for Mobile
Knowledge ManagementThe Example of a University Portal. Proceedings of the Eleventh Americas
Conference on Information Systems. Omaha NE, USA, August 11th14th 2005, pp. 17241732.
Liebeskind, J. P.; Oliver, A.L.; Zucker, L.; and Brewer, M. 1996. Social networks, learning, and fexibility:
Sourcing scientifc knowledge in new biotechnology frms. Organization Science, 7, 4, 428443.
Liebowitz, J. 2005. Introduction: Knowledge and learning in the new era. International Journal of Knowledge
and Learning, 1, 3, 183185.
Machlup, F. 1980. Knowledge: Its Creation, Distribution and Economic Signifcance, Volume 1. Princeton,
NJ: Princeton University Press.
Magalhes, R. 1998. Organizational knowledge and learning. In G. von Krogh, J. Roos, and D. Kleine (eds.),
Knowing in Firms: Understanding, Managing and Measuring Knowledge. Thousand Oaks, CA: Sage
Publications, pp. 87123.
Malhotra, A., and Majchrzak, A. 2004. Enabling knowledge creation in far-fung teams: Best practices for
IT support and knowledge sharing. Journal of Knowledge Management, 8, 4, 7588.
March, J. 1991. Exploration and exploitation in organizational learning. Organization Science, 2, 1,
7187.
Marcoulides, G.A., and Heck, R.H. 1993. Organizational culture and performance: Proposing and testing a
model. Organization Science, 4, 2, 209225.
Maturana, H.R., and Varela, F.J. 1992. The Tree of Knowledge. Boston: Shambhala.
McAfee, A.P. 2006, Enterprise 2.0: The dawn of emergent collaboration. MIT Sloan Management Review,
47 (Spring), 2128.
McAllister, D.J. 1995. Affect and cognition-based trust as foundations for interpersonal cooperation in
organizations. Academy of Management Journal, 38, 1, 2459.
Medsker, G.J.; Williams, L.J.; and Holahan, P.J. 1994. A review of current practices for evaluating causal models in
organizational behavior and human resources management research. Journal of Management, 20, 2, 4346.
Menon, A.; Bharadwaj, S.G.; and Howell, R. 1996. The quality and effectiveness of marketing strategy:
Effects of functional and dysfunctional confict in intraorganizational relationships. Journal of Academy
of Marketing Science, 24, 4, 299314.
Mentzer, J.T.; Flint, D.J.; and Hult, G.T.M. 2001. Logistics service quality as a segment-customized process.
Journal of Marketing, 65 (October), 82104.
Nahapiet, J., and Ghoshal, S. 1998. Social capital, intellectual capital, and the organizational advantage.
Academy of Management Review, 23, 2, 242266.
Nevis, E., et al. 1995. Understanding organizations as learning systems. Sloan Management Review (Winter),
7384.
Nonaka, I. 1994. A dynamic theory of organizational knowledge creation. Organization Science, 5, 1,
1437.
Nonaka, I., and Konno, N. 1998. The concept of ba: Building a foundation for knowledge creation. Cali-
fornia Management Review, 4, 3, 4054.
Nonaka, I., and Takeuchi, H. 1995. The Knowledge Creating Company: How Japanese Companies Create
Nunnally, J.C. 1978. Psychometric Theory. New York: McGraw-Hill.
ODell, C., and Grayson, C.J., Jr. 1998. If Only We Knew What We Know. New York: The Free Press.
Okhuysen, G.A., and K.M. Eisenhardt. 2002. Integrating knowledge in groups: How formal interventions
enable fexibility. Organization Science, 13, 4, 370386.
Podsakoff, P.M., and Organ, D.W. 1986. Self-reports in organizational research: Problems and prospects.
Journal of Management, 12, 4, 531544.
Polanyi, M. 1966. The Tacit Dimension. London: Routledge and Keoan.
Ponzi, L. 2002. The intellectual structure and interdisciplinary breadth of knowledge management: A bib-
liometric study of its early stage of development. Sciento Metrics, 55, 2, 259272.
. 2004. Knowledge management: Birth of a discipline. In M.E.D. Koenig and T.K. Srikantaiah (eds.),
Knowledge Management Lessons Learned. ASIST Monograph Series, Information Today, Inc., 926.
Postrel, S. 2002. Islands of shared knowledge: Specialization and mutual understanding in problem-solving
teams. Organization Science, 13, 3, 303320.
Probst, G.; Bchel, B.; and Raub, S. 1998. Knowledge as a strategic resource. In G. von Krogh, J. Roos, and
D. Kleine (eds.), Knowing in Firms: Understanding, Managing and Measuring Knowledge. Thousand
Oaks, CA: Sage Publications, pp. 87123.
Raelin, J. 1997. A model of work-based learning. Organization Science, 8, 6, 563578.
Rich, G.A. 1997. The sales manager as a role model: Effects on trust, job satisfaction, and performance of
salespeople. Journal of Academy of Marketing Science, 25, 4, 319328.
Ruggles, R., and Little, R. 2001. Enabling complex adaptive processes through knowledge management.
In D. Morey, M. Maybury, and B. Thuraisingham (eds.), Knowledge Management: Classic and Con-
temporary Works. Cambridge, MA: The MIT Press, pp. 239248.
Sabherwal, R., and Becerra-Fernandez, I. 2003. An empirical study of the effect of knowledge management
processes at individual, group, and organizational levels. Decision Sciences, 34, 2, 225260.
. 2005. Integrating specifc knowledge: Insights from the Kennedy Space Center. IEEE Transactions
on Engineering Management, 52, 3, 301315.
Sabherwal, R., and Robey, D. 1993. An empirical taxonomy of implementation processes based on sequences
of events in information system development. Organization Science, 4, 4, 548576.
Sabherwal, R., and Sabherwal, S. 2005. Knowledge management using information technology: Determinants
of impact on frm value. Decision Sciences, 36, 4, 531567.
Satzinger, J.; Garfeld, M.; and Nagasundaram, M. 1999. The creative process: The effects of group memory
on individual idea generation. Journal of Management Information Systems, 15, 4, 143160.
Saunders, R.M. 2000. Managing knowledge: How to make money with what you know. Harvard Manage-
ment Communication Letter, (June), 35.
Scott, S.G., and Bruce, R.A. 1994. Determinants of innovative behavior: A path model of individual innova-
tion in the workplace. Academy of Management Journal, 37, 3, 580607.
Seibert, S.E.; Kraimer, M.L; and Liden, R.C. 2001. A social capital theory of career success. Academy of
Management Journal, 44, 2, 219238.
Simon, H.A. 1991. Bounded rationality and organizational learning. Organization Science, 2, 1, 125134.
Skyrme, D. 2000. Developing a knowledge strategy: From management to leadership. In D. Morey, M.
Maybury, and B. Thuraisigham (eds.), Knowledge Management: Classic and Contemporary Works.
Cambridge, MA: MIT Press, pp. 6184.
Slater, S., and Narver, J. 1995. Market orientation and the learning organization. Journal of Marketing, 59
(July), 6374.
Spender, J.C. 1996. Making knowledge the basis of a dynamic theory of the frm. Strategic Management
Journal, 17 (Winter), 4563.
Stata, R. 1989. Organizational learningThe key to management innovation. Sloan Management Review,
(Spring), 6374.
Systems Research, 6, 2, 85117.
Storck, J., and Hill, P. 2000. Knowledge diffusion through strategic communities. Sloan Management
Review, 41, 2, 6374.
Szulanski, G. 1996. Exploring internal stickiness: Impediments to the transfer of best practice within the
frm. Strategic Management Journal, 17 (Winter), 2743.
Takeishi, A. 2002. Knowledge partitioning in the interfrm division of labor: The case of automotive product
development. Organization Science, 13, 3, 321338.
Teece, D.J. 1998. Capturing value from knowledge assets: The new economy, markets for know-how, and
intangible assets. California Management Review, 40, 3, 5581.
Thomas-Hunt, M.C.; Ogden, T.Y.; and Neale, M.A. 2003. Whos really sharing? Effects of social and expert
status on knowledge exchange within groups. Management Science, 49, 4, 464477.
Tiwana, A., and Bush, A. 2005. Continuance in expertise-sharing networks: A social perspective. IEEE
Transactions on Engineering Management, 52, 1, 85101.
Tsoukas, H. 1996. The frm as a distributed knowledge system: A constructionist approach. Strategic Man-
agement Journal, 17 (Winter Special Issue), 1125.
Vandenbosch, B., and Ginzberg, M.J. 19961997. Lotus notes and collaboration: Le plus a change. Journal
of Management Information Systems, 13, 3, 6582.
Venkatraman, N. 1989. Strategic orientation of business enterprises: The construct, dimensionality, and
measurement. Management Science, 35, 8, 942962.
Venzin, M.; von Krogh, G. and Roos, J. 1998. Future research into knowledge management. In G. von Krogh,
J. Roos, and D. Kleine (eds.), Knowing in Firms: Understanding, Managing and Measuring Knowledge.
Thousand Oaks, CA: Sage Publications, pp. 2666.
Vera, D., and Crossan, M. 2003. Organizational learning and knowledge management: Toward an integrative
framework. In M. Easterby-Smith and M. A. Lyles (eds.), The Blackwell Handbook of Organizational
Learning and Knowledge Management. Malden, MA: Blackwell Publishing, pp. 122141.
von Krogh, G., and Roos, J. 1995. A perspective on knowledge, competence and strategy. Personnel Review,
24, 5676.
Walsh, J.P. 1995. Managerial and organizational cognition: Notes from a trip down memory lane. Organiza-
tion Science, 6, 3, 280321.
Wayne, S.J., and Liden, R.C. 1995. Effects of impression management on performance ratings: A longitudinal
study. Academy of Management Journal, 38, 1, 2459.
Weick, K.E. 1991. The nontraditional quality of organizational learning. Organization Science, 2, 1,
116124.
Weick, K.E., and K. Roberts. 1993. Collective mind in organizations: Heedful interrelating on fight decks.
Willett, C. 2001. Knowledge sharing shifts the power paradigm. In D. Morey, M. Maybury and B. Thurais-
ingham (eds.), Knowledge Management: Classic and Contemporary Works. Cambridge, MA: The MIT
Press, pp. 249260.
Williams, L.J., and Hazer, J.T. 1986. Antecedents and consequences of satisfaction and commitment in
turnover models: A reanalysis using latent variable structural equation methods. Journal of Applied
Psychology, 71, 219231.
Zander, U., and Kogut, B. 1995. Knowledge and the speed of the transfer of imitation of organizational
capabilities: An empirical test. Organization Science, 6, 1, 7692.
Zhu, D.; Prietula, M.J.; and Hsu, W.L. 1997. When processes learn: Steps toward crafting an intelligent
organization. Information Systems Research, 8, 3, 302317.
40
ChaptEr 3
KNOWLEDGE MANAGEMENT AND
ORGANIZATIONAL CULTURE
dorothy E. lEidnEr and timothy r. Kayworth
Abstract: Recent advances in knowledge management (KM) and KM-related systems (KMS) have
sparked interest in examining the organizational effects of such technology-based innovations.
Drawing from a variety of organizational paradigms, researchers have attempted to answer such
questions as: How do KMS potentially enhance a frms ability to create, store, and share its
intellectual resources? What factors infuence a frms effectiveness in using KM technologies?
How do KMS affect organizational performance? In addressing such questions, organizational
culture theory is one such paradigm that has received limited attention in explaining KM system
uses and its organizational impacts. This chapter examines the work related to organizational
culture and KM and suggests directions for further research. To accomplish this, we briefy trace
information-based systems advancements and the dominant organizational paradigms used to
investigate the organizational effects of such systems. Next, the chapter discusses organizational
culture and its implications for KMS implementation. We do this through frst presenting an orga-
nizational culture framework and then by discussing the various themes in KM-culture research.
The chapter then concludes with an examination of potential directions for future research in
KM-organizational culture research.
Keywords: Knowledge Management, Organizational Culture, Technology Impacts
INTRODUCTION
The mission of the knowledge management (KM) movement in organizations has followed the realiza-
tion that organizations house an abundance of knowledge and yet individuals have diffculty locating the
knowledge and putting it to effective use. When asked why the organization was building a worldwide
Intranet and KM system, the chief knowledge offcer of a large multinational consulting frm replied,
We have 80,000 people scattered around the world that need information to do their jobs effectively.
The information they needed was too diffcult to fnd and, even if they did fnd it, often inaccurate.
Our intranet is meant to solve this problem (D.E. Leidner, 1998, personal interview with the CIO of a
Fortune 500 frm). The KM story normally begins with organizations facing a problem similar to that
just described. To combat the diffculty of organizing and locating knowledge, organizations implement
knowledge repositories, thereby providing a central location and search point for relevant knowledge.
As the popularity of the system increases, so too does the volume of knowledge. And as the volume of
knowledge increases, so too does the diffculty of fnding high-quality, relevant knowledge to address
a specifc problem (Garud and Kumaraswamy, 2005). Organizations must then fnd ways to counter
the overabundance of information in their KM systems (KMS).
KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL CULTURE 41
Like many stories, this one sounds familiar. In fact, it was roughly two decades ago that ex-
ecutive information systems (EIS) emerged as a means of combating information that is widely
dispersed, inconsistent, and too general to be relevant. Case studies of organizations implement-
ing EIS suggested that the major reasons behind these systems were a need for timely, accurate,
and consistent information and to help managers cope with the problem of information overload
(Houdeshel and Watson, 1987; Rockart and DeLong, 1988). And even while EIS were geared
toward providing relevant information to executives, studies showed that these systems helped
improve executives mental models and understanding of their businesses (Leidner and Elam,
1995; Vandenbosch and Higgins, 1996). Thus, while not knowledge systems, they served an
important purpose for learning and understanding.
Even earlier than organizational venturing into EIS, management information systems (MIS)
were intended to provide relevant information for managerial control and planning. However,
MIS proved unable to provide timely, complete, accurate, and readable data of the type executives
needed for strategic decision making. Claimed Ackoff in a now-famous article, I do not deny
that most managers lack a good deal of information that they should have, but I do deny that this
is the most important information defciency from which they suffer. It seems to me that they suf-
fer from an overabundance of irrelevant information (Ackoff, 1967, p. 147). Two decades later,
Courtney, Croasdell, and Paradice (1997) lament the pervasiveness of unimportant information
on corporate information.
What these similarities demonstrate is that the need for systems to help organizations cope with
information needs and subsequent information problems remains challenging. Despite advances
in available technologies, organizations continue to struggle to effectively allow professionals and
managers to locate critical information or knowledge. Whether termed information or knowledge
repositories, systems that allow organizations to manage the storage and dissemination of un-
structured information can quickly become irrelevant if appropriate mechanisms are not in place
to maintain the integrity and relevance of the information (Garud and Kumaraswamy, 2005).
One of the most critical challenges faced by frms in ensuring the success of such KMS is that
of organizational culture. While organizational culture has been shown to have an important role
in sundry aspects of information systems (IS) development, implementation, and use (Leidner
and Kayworth, 2006), it has also taken a central role in discussions of KMS implementation
and use. This chapter will examine the work related to organizational culture and KM and
suggest directions for further research. First, however, the chapter briefy traces information-
based systems advancements and the dominant organizational paradigms used to investigate
the organizational effects of such systems. The chapter then discusses organizational culture
and its implications for KMS implementation. This is accomplished through frst presenting an
organizational culture framework and then discussing the themes in KM-culture research. The
chapter concludes with an examination of potential directions for future research in KM-orga-
nizational culture research.
ADVANCES IN INFORMATION SYSTEMS
Information systems (IS) can be classifed according to their broad function, to the organizational
function they serve, to the underlying technologies, or to the organizational level at which they are
used (Laudon and Laudon, 2000). This chapter considers IS by broad function because much of
the information technology (IT) literature focuses on particular systems classifed in this manner,
such as decision support systems (DSS), expert systems, and electronic mail. In particular, we
are interested in systems designed to provide information to managers and professionals at any
42 LEIDNER AND KAYWORTH
organizational level. Hence, we focus primarily on MIS and EIS (as both systems aim to supply
managerial information) in addition to KMS.
Management Information Systems and the Structuring of Organizations
As noted in Somogyi and Galliers (1987), as frms began to computerize in the 1950s, the frst
applications were in the area of transaction processing. Transaction processing systems are com-
puterized systems that perform and record the daily routine transactions necessary to the conduct
of business such as payroll, sales order entry, shipping, order tracking, accounts payable, and
material movement control (Laudon and Laudon, 2000). These systems were designed to facilitate
data collection and to improve the effciencies of organizational transactions. Soon thereafter,
with advances in programming languages, databases, and storage, systems emerged that were
oriented toward providing performance information to managers (Somogyi and Galliers, 1987).
Management information systems are computer-based IS that provide managers with reports and,
in some cases, with online access to the organizations current performance and historical records.
Management information systems primarily serve the functions of planning, controlling, and
decision making at the management level. Generally, they condense information obtained from
transaction processing systems and present it to management in the form of routine summary and
exception reports.
Simon (1977) predicted that computers, namely MIS, would recentralize decision making, shrink
line organizational structures, decrease the number of levels, and result in an increase in the number
and size of staff departments. It was believed that IT would enable greater centralization of author-
ity; clearer accountability of subordinates; a sharper distinction between top management and staff,
and the rest of the organization; and a transformation of the planning and innovating functions. The
organizational theory used to evaluate the effect of MIS on organizations was contingency theory
of organizational structure, technology, and the environment. Research prior to 1970 indicated
that IT provided a means of collecting and processing large amounts of data and information, thus
enabling a small number of persons to effectively control authority and decision making; hence, IT
was said to facilitate centralization (Klatzky, 1970; Stewart, 1971; Whisler, 1970). Research after
1970 seemed to fnd that, by enabling organizations to gather and process information rapidly, IT
facilitated decentralizing decision making (Carter, 1984; Foster and Flynn, 1984). For example,
Carter (1984) theorized that as the extent of computer utilization increased in subunit applications,
the locus of decision-making authority would become more decentralized in the organization
and the division of labor, as refected by functional diversifcation, functional specialization, and
functional differentiation, would increase. Carter found in her study of newspaper organizations
that as computers became the predominant technology, upper management was released from the
day-to-day encumbrances of centralized decision making, fostering a decentralized organizational
structure. In other cases, IT appeared to have had no effect when changes were expected (Franz,
Robey, and Koeblitz, 1986). Considering the weak relationships found when using technology as
an independent variable, other researchers used technology either as a moderator variable between
the environment and structure or as a dependent variable. Robey (1977) found that IT supported
an existing decentralized structure in organizations with uncertain environments but that, in more
stable environments, IT strengthened a centralized authority structure.
In summary, early research on the impact of IT, namely MIS, on organizations focused on the
effect of IT on organizational structures. The results were highly mixed, leading to an emergent
imperative that argued that the particular effects of IT were dependent on a given organizations con-
text and, hence, were not predictable or systematic across organizations. An alternative perspective
was that certain inherent limitations of MIS prevented predictable improvements to organizational
effectiveness. Among the limitations of MIS are that they have highly limited analytical capabilities;
they are oriented almost exclusively to internal, not environmental or external, events; and that
the information content is fxed and not tailored to individual users (Laudon and Laudon, 2000).
Decision Support Systems, Executive Information Systems, and Organizational
Decision Making
Decision support systems and EIS aimed to provide what MIS were unable to: specifc online
information relevant to decision makers in a fexible format. Decision support systems are inter-
active model-oriented systems that are used by managers and knowledge workers, analysts, and
professionals whose primary job is handling information and making decisions (Keen and Morton,
1982; Sprague and Carlson, 1982). Decision support systems assist management decision mak-
ing by combining data, sophisticated analytical models, and user-friendly software into a single
powerful system that can support semistructured or unstructured decision making (Keen and Mor-
ton, 1982; Sprague and Carlson, 1982). Decision support systems tend to be isolated from major
organizational IS and tend to be stand-alone systems developed by end-user divisions or groups
not under central IS control (Hogue, 1987). Executive information systems are computer-based
IS designed to provide managers access to information relevant to their management activities.
Originally designed for senior managers, the systems quickly became popular for managers at
all levels. Unlike DSS, which are tied to specifc decisions and which have a heavy emphasis on
models, EIS focus on the retrieval of specifc information, particularly daily operational informa-
tion that is used for monitoring organizational performance.
Features distinguishing EIS from such systems as MIS and DSS include a nonkeyboard inter-
face, status access to the organizational database, drill-down analysis capabilities (the incremental
examination of data at different levels of detail), trend analysis capabilities (the examination of
data across desired time intervals), extensive graphics, the providing of data from multiple sources
(including external data), and the highlighting of the information an executive believes is critical
(Kador, 1989; Mitchell, 1988). Whereas the traditional focus of MIS was on the storage and pro-
cessing of large amounts of information, the focus of EIS is on the retrieval of specifc information
about the daily operational status of an organizations activities as well as specifc information
about competitors and the marketplace (Friend, 1986).
Huber (1990) advanced a theory of the effects of advanced decision and information-providing
technologies, such as DSS and EIS, on organizational decision making. While he also made propo-
sitions concerning the effect of such systems on organizational design and structure, the dominant
paradigm for examining the organizational effects of information technology was turning toward
decision making. Huber and McDaniel (1986) argued that decision making was the most critical
management activity and that the effectiveness of IS rested more in facilitating organizational
decision making than in enabling structural responses to environmental uncertainty. A wide body
of research emerged that examined organizational decision making and the decision-making conse-
quences of IS. However, most of the IS literature focused on the individual level of analysis, which
was reasonable given that DSS were designed in most cases for individual decision makers, and
most of the EIS research also supported individual rather than organizational improvements.
While some of Hubers propositions have been substantiated (Leidner and Elam, 1995; Molloy
and Schwenk, 1995), the organizational-level effects have received little substantiation and have
been overshadowed by the individual-level effects (Elliott, 1992). Moreover, research on DSS
showed that decision makers used the tools in such a manner as to reduce time but not necessarily
to increase quality (Todd and Benbasat, 1991), but in the cases where the systems did appear to
increase quality, the decision makers seemed not to subjectively perceive this improvement (Le
Blanc and Kozar, 1990). Empirical evidence has shown that EIS enable faster decision making,
more rapid identifcation of problems, more analysis before decision making, and greater under-
standing of the business (Elliott, 1992; Leidner and Elam, 1995). Evidence also suggests that
EIS allow single- and double-loop learning (Vandenbosch and Higgins, 1996). Other promises
for EIS, which have not been empirically substantiated, involved helping companies cope with
reduced staff levels (Applegate, 1988; Applegate and Osborn, 1988), substantial monetary sav-
ings (Holub, 1988), power shifts and a change in business focus (Applegate and Osborn, 1988),
and improving service (Holub, 1988; Kador, 1989; Mitchell, 1988). Interestingly, these promises
sound reminiscent of the promises that were made for MIS and that are now being made for KMS,
as discussed later.
Among the most serious challenges to EIS implementation involved overcoming information
problems, namely organizational subunits feeling ownership of information that was suddenly ac-
cessed by senior managers who previously had relied on these subunits to summarize and analyze
their own performance in periodic reports. Such ownership problems led to system failure in some
cases, when subunits consciously and covertly altered data to be more favorable to the unit and
thereby rendered the EIS inaccurate (Elliott, 1992). Other weaknesses of EIS are the diffculty
of pulling information from multiple sources into a graphical PC-based interface, justifying the
costs of the systems given the unclear payoff, and ensuring that the information remains relevant
as the needs of managers change (Elliott, 1992). In summary, DSS and EIS research adopted an
organizational decision-making paradigm as a reference theory for determining the organizational
impacts of these systems. While the systems have well documented individual-level benefts, the
organizational-level benefts have been less lucid.
Knowledge Management Systems and the Culture of Organizations
Knowledge management systems emerged as a line of systems that compensates for some of the
limitations of its information-based predecessors. Knowledge includes the insights, understand-
ings, and practical know-how that employees possess. Knowledge management is a method of
systematically and actively managing ideas, information, and knowledge of employees. Knowl-
edge management systems refer to the use of modern information technologies (e.g., the Internet,
intranets, extranets, browsers, data warehouses, software flters and agents, and collaboration
tools) to systematize, enhance, and expedite intrafrm and interfrm KM (Alavi and Leidner,
1999). Knowledge management systems are intended to help organize, interpret, and make widely
accessible the expertise of an organizations human capital to help the organization cope with
turnover, rapid change, and downsizing. Knowledge management systems are being built in part
from increased pressure to maintain a well-informed, productive workforce.
Early research suggested that managers did not distinguish between knowledge and information
or dataindeed, that one persons knowledge is anothers data (Alavi and Leidner, 1999, p. 11).
The perspective held of KMthat it was a means of providing readily accessible information,
real-time information, and actionable informationis not readily distinguished from the perspec-
tive held of other information-based systems. Likewise, the early perspective of KM technologies
included such systems as data mining and warehouses, EIS, expert systems, and intelligent agents,
each of which is associated with separate systems and lines of inquiry. This partially explains the
slowness with which KM research took off among IS researchers.
Beginning with the early research on KMS, organizational culture has emerged with a central
theme to explain cultures potential infuence on the abilities of frms to undertake successful KM
initiatives. Much of the research indicates that KM efforts are often seen to clash with corporate
culture and, as a result, have limited impact (DeLong and Fahey, 2000; ODell and Grayson, 1998).
For example, one study by Alavi and Leidner (1999) found that managers considered culture to
account for 80% of a KM initiative. Another by Ernst & Young identifed culture as the biggest
impediment to knowledge transfer, citing the inability to change peoples behaviors as the biggest
hindrance to managing knowledge (Watson, 1998). Finally, in a study of 453 organizational KM
initiatives, Ruggles (1998) found that over half perceived that organizational culture was a major
barrier to success in their KM initiatives. The importance of organizational culture is also clear
from consulting frms such as Ernst & Young, which reports that a major aspect of KM initia-
tives involves working to shape organizational cultures that hinder their KM programs (KPMG
Management Consulting, 1998). The resulting solutions by these frms typically lead to incentives
that attempt to promote behaviors, even if they do little to change values. Consequently, much
of the literature proposes that a fundamental managerial role in KM initiatives is to nurture the
organizational culture necessary to facilitate KM efforts (Davenport, DeLong, and Beers, 1998;
Greengard, 1998; Hargadon, 1998). Regarding the importance of culture, DeLong and Fahey
(2000) remark: Obviously, there is a set of tools such as Lotus Notes, intranets, etc. which you
need to be knowledge based. But technology is only 20% of the picture. The remaining 80% is
people. You have to get the culture right (p. 113).
Knowledge management efforts may confict with organizational culture for a variety of
reasons. First, these efforts may require behaviors that clash with frm members embedded
values. For example, effective KM practice requires cultural values that foster and reward the
creation and use of knowledge as well as its sharing among individual members and groups
(Davenport, DeLong, and Beers, 1998; Leonard and Sensiper, 1998; ODell and Grayson,
1998). In reality, however, companies may inadvertently cultivate organizational cultures where
individual expertise and performance are highly rewarded but mentoring and assisting others
are not (Leonard and Sensiper, 1998). In such environments, there may be active discourage-
ment or self-censoring from participation in knowledge sharing activities and individuals may
actually be rewarded for information hoarding practices. In one example of this, engineers at a
company had little motivation to share knowledge and expertise with others due to a very strong
corporate value system embracing individual effort and accomplishment (DeLong and Fahey,
2000). Thus, the very cultural values required to inculcate knowledge sharing among the frms
members was at odds with the prevailing culture of individual performance and expertise. One
engineers comments helps to illustrate this cultural confict: In divisional reviews, the senior
manager says; show me something Ive never seen before. So the whole goal is to blow their
socks off. Nobody ever says; show me where youve worked together with another business
unit (p. 118).
Second, effective KM practices may be undermined due to cultural differences across business
subunits within the same organization (Hofstede, 1998; Sackman, 1992) leading to a lack of trust
across subunits (Bloor and Dawson, 1994; DeLong and Fahey, 2000). In one example of this, a
hospitals staff was found to be extremely reluctant to share critical insights with physicians who
were perceived to be members of a dominant subculture (Bloor and Dawson, 1994).
Finally, KM initiatives may be hindered by organizational cultures that are highly formalized
and depend heavily on standard operating procedures (SOPs), rules, and regulations as templates
for decision making (Hargadon, 1998; Huber, 1991; von Krogh, 1998). These rules may stife the
creation of new knowledge as members attempt to address novel problems with fxed patterns of
thinking that may no longer be appropriate.
These examples and others (Hasan and Gould, 2001; Schultze and Boland, 2000) help dem-
onstrate the profound impact that organizational culture may have on KM practice and of the
crucial role of senior management in fostering cultures conducive to these practices (Brown
and Duguid, 2000; Davenport, DeLong, and Beers, 1998; DeLong and Fahey, 2000; Gupta
and Govindarajan, 2000; Hargadon, 1998; Leonard and Sensiper, 1998; von Krogh, 1998).
Furthermore, they suggest that culture in some companies may be a core rigidity that actually
constrains innovation.
Summary
New classes of IS for managers and professionals are continuing to emerge, yet the perennial
problem of obtaining systematic benefts from such systems remains. Information system re-
searchers have attempted to explain the impact of IS on organizations by considering the effect
of IS on organizational structure and decision making. The former line of research led to mixed
fndings and the latter, fndings more at the individual than organizational level. With the changes
in systems, summarized in Table 3.1, the role of the user has progressed from involvement in
system design (MIS) to, in many cases, system designer (DSS), to interactive system user (EIS),
to information content provider as well as user (KMS). This shift in the role of the user requires
a concomitant shift in our conceptualization of IS with less emphasis on the systems aspect
and more on the information aspect, namely the users view of information as an individual or
corporate asset. Information has been classifed according to its accuracy, timeliness, reliability,
completeness, precision, conciseness, currency, format, accessibility, and perceived usefulness
(DeLone and McLean, 1992). While previous systems design focused on these aspects as the
foundation of information quality, the emergence of KMS has introduced the importance of
Table 3.1
Summary of Information-based Systems
MIS DSS EIS KMS
Purpose Provide summarized
performance reports
to management
Provide tools,
models and data
for aid in decision
analysis
Provide online
access to real-
time fnancial
and operational
information
Provide online
access to
unstructured
information
and knowledge
throughout the
organization
Users Managers at various
levels
Analysts and
middle managers
Senior and middle
managers
Professionals
and managers
throughout the
organization
Role of
users
Participation in
design
Participation in
design, active user
Participation in
design, active user
Participation in
design, active user,
content provider
Information
strategy
One-for-all One-for-all One-for-one Anyone, anytime,
anywhere
Interpretive
framework
Organizational
structure
Organizational
decision making
Organizational
decision making
Organizational
culture
the notion of information ownership and highlighted the infuence of organizational culture on
KM initiatives.
Moreover, we seem to have moved from a one for all to a one for one to an anyone any-
time anywhere information provision strategy as we have advanced from MIS to DSS and EIS,
to KMS. The latter strategy requires greater horizontal and vertical integration of information in
an organization. It is arguable that the potential impact of systems is greater when a larger part
of the organization is affected, such as with systems integrated organization-wide, or even across
organizations. And yet, systems requiring both vertical and horizontal integration will create the
greatest cultural challenges for organizations because they will bring to the surface both structural
or control issues. From here, we will provide a brief overview of organizational culture, followed
by a review of the literature on KM and organizational culture. We will then provide directions
for future research on KM and culture.
ORGANIZATIONAL CULTURE FRAMEWORK
While organizational culture is a rich construct, its very richness creates challenges in using it as an
interpretive framework for studying organizational phenomena such as KM. Much of this diffculty
lies in the numerous ways culture has been defned by organizational theorists. While some have
defned culture in terms of implicit sets of beliefs, basic assumptions, shared sets of core values,
important understandings, and the collective will (Sackman, 1992), others have conceptualized
culture in terms of more observable cultural artifacts such as norms and practices (DeLong and
Fahey, 2000; Hofstede, 1998), symbols (Burchell et al., 1980), as well as language, ideology,
rituals, myths, and ceremony (Pettigrew, 1979). As a result, these varying conceptualizations of
culture may give rise to a great deal of confusion and misunderstanding in organizational culture
research due to their sheer range (Kroeber and Kluckhohn, 1952). To address this issue, we draw
from Scheins (1985) three-level model that depicts culture in terms of basic assumptions, values,
and artifacts.
Scheins Three-Level Model of Organizational Culture
According to Schein, culture exists at three levels: basic assumptions, values, and artifacts. At
the deepest level, culture consists of basic assumptions. These assumptions or beliefs represent
interpretive schemes that people use to perceive situations and to make sense of ongoing events,
activities, and human relationships, thereby forming the basis for collective action (Van Maanan
and Barley, 1985). They form over time as members of a particular group develop strategies to
cope with problems and pass along the strategies to new members (Van Maanan and Barley, 1985).
Specifc cultures exist when groups, regardless of size, embrace similar interpretive schemes.
At the next level, values represent a more visible manifestation of culture that signify es-
poused beliefs identifying what is important to a particular cultural group. While assumptions
may be preconscious and invisible, values are more visible, even debatable with individuals
having a greater awareness of them. Even so, values alone are merely a refection of underlying
cultural assumptions (Schein, 1985). These values provide a means through which organiza-
tional members interpret signals, events, and issues and represent an enduring belief that a
given norm of behavior is preferable over an opposite norm (Bansal, 2003). Thus, values can
be seen as a set of social rules or norms that govern acceptable forms of social interaction
and communication among given culturally distinct groups (DeLong and Fahey, 2000; Nadler
and Tushman, 1988). These social norms have an impact on subsequent behaviors of frm
members through acting as a form of social control that defnes which behaviors and attitudes
are appropriate for members to display (OReilly and Chatman, 1996). Thus, a consistent
theme in the organizational literature has been that organizational values drive social group
behaviors (Nadler and Tushman, 1988).
Most contemporary theories of culture embrace this values perspective of organizational
culture. Thus, we fnd for example, in the IS feld, that most empirical research examining the
intersection of IT and culture, does so almost exclusively using a values-based approach to orga-
nizational culture (Alavi, Kayworth, and Leidner, 2006). This is generally true also of the general
management literature. Table 3.2 summarizes some of the most cited values-based approaches
to organizational culture. Empirical research drawing from one or more of these values-based
theories typically try to categorize specifc groups according to one or more sets of values and
then examine how differences in these values may lead to different organizational outcomes. For
example, in the management literature, Denison and Mishra (1985) examined how certain types
of organizational values were more closely related to organizational effectiveness than others. In
the IS literature, Dub and Robey (1999) have investigated how software developers may perceive
software development process improvements differently based on their cultural values. Cooke and
Laffertys Organizational Culture Inventory has been perhaps the most drawn from categorization
of organizational culture values.
At the third, and most visible level, culture is manifested through various artifacts such as art,
technology, visible and audible behavior patterns, myths, heroes, language, rituals, and ceremony
(Pettigrew, 1979). Thus, technology artifacts (e.g., a frms KM system) could be conceived of as
a type of cultural artifact that is the realization of a given set of cultural values that, in turn, are
manifestations of underlying assumption about information technology. Furthermore, various
work practices or behaviors (e.g., employees commonly working late after normal business hours)
could be conceived as more visible manifestation of underlying cultural values and implicit as-
sumptions. In one prominent work, Hofstede and colleagues (1990) have developed a taxonomy
of organizational subcultures based upon empirical data that reveals different cultural groups
shared perceptions of daily work practices. These work practice (behavioral) dimensions isolated
by Hofstede et al. (1990) are identifed below in Table 3.3.
Table 3.2
Organizational Culture: A Values Approach
Author(s) Organizational Culture Values Dimensions
Blake and Mouton (1964) Concern for people versus concern for production
Cook and Lafferty (1987) Values emphasizing collaboration and support (constructive), ap-
proval (passivity), and power (aggression). Organizational culture
inventory
Denison and Mishra (1995) Values emphasizing sense of ownership (involvement), individual
conformity (consistency), adaptability, and mission (sense of
purpose)
Early (1993) Values emphasizing self-achievement and autonomy (individualistic)
versus loyalty to ones group (collectivism)
Goffee and Jones (1996) Values emphasizing solidarity versus sociability
Jones (1983) Values emphasizing procedure (production)
Wallach (1983) Firm values tending toward bureaucracy, innovation, or supportiveness
Organizational Culture Diversity
A fnal issue relevant to our examination of KM and culture has to do with the question of whether
organizations have uniform, homogenous values or instead, various local cultures, each with their
own distinctive values. This is an important question since, depending on ones views, there could
be either multiple local cultures at work infuencing KM practices within a frm or instead, a single
dominant corporate culture driving KM choices, decisions, and outcomes.
Meyerson and Martin (1987) draw this distinction with their discussion of the integration and
differentiation perspectives of organizational culture. The integration perspective regards organi-
zational culture as a homogeneous set of values acting as an integrating mechanism or social or
normative glue that holds together a potentially diverse group of organizational members (p. 624).
Such a view is characterized by consistency across cultural manifestations, consensus among
cultural members, and usually a focus on leaders as culture creators (p. 625). Others holding this
view (Deal and Kennedy, 1982; Martin, 1985) see culture as something that can be shaped through
developing management strategies to create unifying organizational cultures (Martin, Sitkin, and
Boehm, 1985). One of the drawbacks to the integrationist perspective is its inability to explain
cultural confict and ambiguity in values (Myerson and Martin, 1987).
In stark contrast, the differentiation perspective depicts organizational culture as an assort-
ment of various local cultures each with their own distinctive values (Denison and Mishra, 1995;
Dougherty, 1992; Louis, 1985; Rose, 1988). So, while there may be an underlying dominant or-
Table 3.3
Organizational Culture Artifacts: Work Practices
Work Practice Dimension Description
Process versus results
orientation
A focus on improving the means by which organizational goals
are achieved (process) as opposed to a focus on the attainment
of goals.
Employee versus job orientation Employee orientation suggests a concern for people, whereas
a job orientation refers to a concern over performing tasks
effectively.
Parochial versus professional A parochial orientation suggests that individuals are loyal to their
organization, whereas a professional orientation suggests that
individuals are loyal to their profession. In parochial subcultures,
individuals get their identity from the company they work for,
whereas individuals from a professional subculture obtain their
sense of identify from the type of work they are involved in.
Open versus closed system In open cultures, the organization considers itself open to
outsiders and new employees with little time needed for new
employees to feel at home. In contrast, closed cultures are
typically secretive and very wary of outsiders as well as insiders.
Loose versus tight control The control dimension refects the degree of internal structuring,
with loose organizations having few written or unwritten codes of
behavior and tight organizations having strict unwritten and written
policies.
Normative versus pragmatic Pragmatic units are market driven and customer oriented,
whereas normative units are product oriented. Thus, in normative
cultures, great emphasis is placed on following procedures as
opposed to the achievement of results.
ganizational culture, various local cultures may exist within the frm. Rose (1988) notes that this
view of organizational culture may be more realistic particularly in large complex organizations
where the ongoing recruitment of personnel from the outside, the introduction of new technology,
and the existence of departmental and other group perspectives all make a unitary culture unlikely.
According to this differentiation view, organizations are umbrellas for collections of subcultures
where the content of such cultures may be infuenced by the task or technology used by employees,
by the constraints of the organizations stage in its lifecycle, or by external factors such as major
changes in a frms environment (Martin, Sitkin, and Boehm, 1985, p. 101).
We believe that this question of cultural diversity versus cultural homogeneity is an important
one. If one adopts the integration perspective of culture, then this would suggest that frms will
experience very similar results and perceptions of KM initiatives due to similarity in values. How-
ever, the alternate view (the differentiation perspective) suggests that as frms unfold their various
KM initiatives, they may experience very divergent behaviors and perceptions of KM within the
same frm due to conficting and divergent value systems.
KNOWLEDGE MANAGEMENTCULTURE RESEARCH THEMES
The next section describes research that has focused on some aspect of culture and KM. We or-
ganize the section into the major themes addressed by the KM-culture research.
Culture and Knowledge Sharing
A clear theme in early research on organizational values and KM has been that organizational
values are important to effective knowledge sharing practices among frm members (Barrett et
al., 2004; Davenport, DeLong, and Beers, 1998; DeTiene and Jackson, 2001; Janz and Prasarn-
phanich, 2003; Knapp and Yu, 1999; Levinthal and March, 1993; Miles et al., 1997). Two studies in
particular highlight this relationship between culture and knowledge sharing. In the frst, DeLong
and Fahey (2000) identifed specifc value orientations believed to facilitate or hinder knowledge
sharing. They found that value orientations such as trust and collaboration led to greater willingness
among frm members to share insights and expertise with each other. In contrast, value systems that
emphasized individual power and competition among frm members led to knowledge hoarding
behaviors. Consequently, they argue that frms should seek to reinforce and mold those cultural
values most consistent with knowledge sharing behaviors.
Second, Jarvenpaa and Staples (2001) draws from Goffee and Jones (1996) dimensions of
culture to defne the relationship between organizational culture and perception of organizational
ownership of information and knowledge and the resulting relationship of this construct with
knowledge sharing practices. Their results suggest that organizations with values emphasizing
the pursuit of shared objectives (solidarity) will tend to have a higher perception of organizational
ownership of information and knowledge produced by its individual members. As a result, this
perception should presumably lead to greater levels of organizational knowledge sharing. Their
study concludes that a propensity to share and perceived organizational ownership of information
leads to greater use of collaborative media to share information.
Culture and Knowledge Creation
Other research focuses on the related concept of knowledge creation (as opposed to sharing) and
its relationship to various values. For example, Lee and Choi (2003) examined various enablers
of knowledge creation, among them the organizational values of collaboration, trust, and learning.
They found support for their hypothesis of a positive relationship of organizational culture (defned
by collaboration, trust, and learning) and knowledge creation processes and conclude that shap-
ing an organizations cultural factors is key to a frms ability to manage knowledge effectively.
In a similar vein, Lee and Cole (2003) found that the culture of the extended Linux community
was important in regulating the norm of open sharing, in addition to providing a quality control
mechanism. They discovered that culture acted as a social control mechanism to manage commu-
nity members and to sanction those who deviated from norms. The freedom to express criticism
was found to be a signifcant underpinning of the development process that enabled knowledge
to expand. Likewise, the study by Styhre, Roth, and Ingelgard (2002) of a major pharmaceutical
company found that values emphasizing caring relationships facilitated a greater level of knowl-
edge creation within the organization.
Knowledge Management and Information Culture
Information culture refers to the particular values and beliefs pertaining to information itself and
what constitutes appropriate or inappropriate behaviors regarding information processing activi-
ties (Jarvenpaa and Staples, 2001). Davenport (1997) describes information culture as varying
along several dimensions, including open versus closed, factually oriented versus rumor versus
intuition-based, internally versus externally focused, controlling or empowering. Similar informa-
tion cultures can exist in dissimilar organizational cultures. In other words, information culture
is not a simple subset of organizational culture such that given organizational cultures have given
information cultures. Another way that information culture varies is in whether individuals associ-
ate information (and knowledge) with the organization or with themselves, (e.g., organizational
versus individual ownership of information). In any system where individuals are expected to
voluntarily contribute their personal knowledge, information culture will play a signifcant role
in the quality and quantity of information exchanged.
Culture and Knowledge Management Effectiveness
Gold, Malhotra, and Segars (2001) demonstrate the relationship between certain organizational
values, KM capabilities and subsequent KM effectiveness. They conclude that organizations with
more open and supportive value orientations are predisposed towards constructive knowledge
behaviors such as frm members sharing insights with others. These values, they argue, form part
of the frms knowledge infrastructure capability which may infuence organizational abilities to
innovate, to respond rapidly to change, and to be responsive to new market demands.
In a more recent study, Baltahazard and Cooke (2003) use the Organizational Culture Inventory
(OCI) instrument to examine the infuence of culture on KM success. Their study reveals that con-
structive-type values (achievement, self-actualization, encouraging, and affliative) have a positive
impact on certain organizational factors (role clarity, communication quality, organizational ft,
creativity, and job satisfaction) believed to promote KM success. One implication of this study for
practice is that managers can conduct cultural assessments of various organizational subgroups
prior to undertaking KM initiatives. Such assessments can help managers detect potential imple-
mentation problems for KM projects and to design the appropriate organizational interventions.
Another study focuses on how organizational culture infuences the impact of KM on individuals
as well as communities (Leidner, Alavi, and Kayworth, 2006). This paper found that even while
senior management may give enthusiastic support to a global KM effort, it is the local efforts that
seem to have the larger impact. The degree to which KM is used at a local level varies consider-
ably depending upon the subunits culture.
Culture and Knowledge Management Evolution
Some studies consider how a KM initiative evolves in an organization over time. In a recent study,
Alavi, Kayworth, and Leidner (20052006) examined how the various subcultures within the same
large IT services company led to different organizational approaches to KM as well as to varying
patterns of use for the same KM tools across business units. Specifcally, they found that the or-
ganization-level values of formalization and bureaucracy (Wallach, 1983) resulted in a top-down
repository-based approach to KM characterized by formalized procedures to identify, extract, and
structure intellectual capital for inclusion in central repositories. However, over time, new grass-
roots approaches to KM emerged based upon the various subcultures within the organization.
For example, one business unit with values emphasizing collaboration (cooperation, support, and
sharing) began to experiment with more ad-hoc community-based approaches to KM through the
use of collaborative technologies. Furthermore, their study indicated divergent patterns of KM
use based upon the cultural values embraced by various subgroups within the same organization.
For example, groups valuing personal expertise used KM tools for making connections to oth-
ers while subgroups with values embracing formalization used KM tools for developing and
accumulating intellectual capital. Individuals embracing innovation values (valuing a sense of
progress) used KM tools for purposes of collaboration and learning. This study is one of the frst
to draw from the differentiation perspective of organizational culture (Myerson and Martin, 1987)
to examine its effect on KM practice.
Another recent study examines the role of organizational subunit cultures in the success of an
organizational KM initiative. The paper draws upon Martin and Siehls (1983) characterization
of organizational subculture to understand how different subunits respond in different fashion to
a well-aligned organizational KM strategy (Ravishankar, Pan, and Leidner, 2006). The research
suggests that as a KM effort evolves from a central, global system to subunits, the degree to which
the subunits identify with the organization and share the dominant culture infuences the degree to
which the KM penetrates the subunit. Thus, the evolution of the KM is predicated on supportive
cultures at the subunit level.
Summary
While culture is referred to in many KM studies, it is the primary focus of only a small number.
Table 3.4 summarizes some of the key empirical research where the subject of culture has been a
central theme in examining KM approaches, practices, and outcomes.
As the above review indicates, the initial studies of KM and culture have focused predominantly
at the individual level of analysis, asking the question of how organizational culture infuences
individuals predisposition toward positive KM related behaviors. The underlying theme of these
early studies has been that certain types of organizational values will lead to different types of KM
behaviors and that these behaviors will lead to varying outcomes. Thus, certain cultural values,
such as sharing, openness, and trust, will lead to positive KM behaviors (e.g., knowledge contribu-
tion and sharing), which will lead to innovation and effciencies, while other values will lead to
dysfunctional KM behaviors (e.g., information hoarding) and, hence, undesirable outcomes such
as ineffciencies. Therefore, organizations should seek to promote and build the types of cultural
values that support their specifc KM objectives (DeLong and Fahey, 2000). The early studies
Table 3.4
Empirical Research: KM and Culture
Authors Theme Cultural Perspective Findings
Alavi et al.
(2006)
KM evolution
and culture
Values: formalization,
collaboration,
autonomy, expertise,
and innovativeness
Different subgroup values within the same
organization led to divergent patterns of
KM use as well as approaches to KM.
Baltahazard
and Cook
(2003)
KM
effectiveness
and culture
Value: Organizational
Culture Inventory
(OCI)
Results of study show that constructive
organizational cultures will be positively
associated with organizational outcomes
that promote KM success.
Delong and
Fahey (2000)
Culture and
knowledge
sharing
Values: trust and
collaboration
versus power and
competition
Values related to trust and collaboration
will lead to greater levels of knowledge
sharing.
Gold et al.
(2001)
KM,
effectiveness,
and culture
Values: trust and
openness (von
Krogh, 1998)
Supportive, encouraging organizational
values are a key factor that determines
a frms KM infrastructure capability and
subsequent effectiveness at KM practice.
Jarvenpaa and
Staples (2001)
Culture and
knowledge
sharing
Values: solidarity
versus sociability
(Goffee and Jones,
1996)
Organizational values emphasizing the
pursuit of shared objectives will lead
to greater levels of knowledge sharing
through its impact on individuals
perception of the ownership of
organizational information and knowledge.
Lee and Choi
(2003)
Culture and
knowledge
creation
Values: trust,
collaboration,
learning
Knowledge creation is associated with
the organizational culture values of trust,
collaboration, and learning.
Lee and Cole
(2003)
Culture and
knowledge
creation
Values: openness Values emphasizing openness to criticism
facilitated knowledge creation in an online
open source community.
Leidner et al.
(2006)
Culture
and KM
effectiveness
Values: bureaucratic,
supportive,
innovative
Global KM efforts will have different
results in different organizations, in part
depending upon the dominant culture.
Ravishankar
et al. (2006)
Culture and
KM evolution
Martin and Siehls
typology of subunit
culture
KM efforts started at the global effort
will only evolve at the local level to the
extent that the local culture embraces the
organization culture.
Styhre et al.
(2002)
Culture and
knowledge
creation
Values: caring
relationships
Values emphasizing caring relationships
facilitate greater levels of knowledge
creation.
have mostly focused on either the creation or sharing aspects of KM and have not yet delved into
the knowledge sourcing (Gray and Meister, 2004) or application aspects. In addition, with the
exception of Alavi, Kayworth, and Leidner (2005 through 2006), little research has specifcally
addressed the choice of technologies in support of KM and how the technology choices may
be infuenced by organizational culture. Finally, early studies have taken fairly narrow aspects
of organizational culture and attempted to understand how certain aspects of a culture (such as
whether it is open versus closed) infuence the KM behaviors. Subsequent studies have focused at
the organizational level of analysis and attempted to uncover how organizational culture infuences
an organizations approach to KM as well as the evolution of the KM initiative. Subsequent stud-
ies have also adopted an organizational subculture perspective to understand how the same KM
can be adopted and applied differently in different organizational subunits. However, in full, the
KM-culture research is still nascent and many fruitful avenues for future research remain. Next,
we discuss two promising themes for future research.
KNOWLEDGE MANAGEMENTCULTURE RESEARCH:
FUTURE DIRECTIONS
Multilevel Cultural Analysis
One avenue of research consists of crossing levels of analysis to come to a more complex under-
standing of KM behaviors. As an example, we can intersect organizational culture, disposition,
and information cultures to better understand individual KM behaviors. Earley (1993) argued
that organizations could also be thought of as being dominantly individualistic or collectivist.
Organizations encouraging individuals to pursue and maximize individuals goals and rewarding
performance based on individual achievement would be considered as having an individualistic
culture, whereas organizations placing priority on collective goals and joint contributions and
rewards for organizational accomplishments would be considered collectivist (Chatman and
Barsade, 1995).
On an individual level, Chatman and Barsade (1995) propose that workplace coopera-
tionthe willful contribution of employee effort to the successful completion of interdependent
tasksis as much dependent on individual disposition as organizational culture. They suggest
that individuals with cooperative dispositions place priority on working together with others
toward a common purpose, while persons with a low cooperative disposition place priority on
maximizing their own welfare irrespective of others. Cooperative persons are more motivated
to understand and uphold group norms and expect others to cooperate, whereas individualistic
people are more concerned with personal goals and expect others to behave in like manner.
Chatman and Barsade (1995) proposed that people who have a high disposition to cooperate and
who work in a collectivistic organizational culture will be the most cooperative, while people
who have a low disposition to cooperate and who work in an individualistic culture will be the
least cooperative. This may suggest that individualistic cultures are results oriented and tend to
be closed, whereas cooperative cultures are process oriented and tend to be open. It might be
that cooperative people in a cooperative culture could be more willing to share tacit knowledge
than are individualistic individuals in a cooperative culture or cooperative individuals in an
individualistic culture. Figure 3.1 illustrates the relationships between individual culture (dis-
position), organizational culture (individualistic versus collectivist), and KM sharing, sourcing,
and perceived value.
Figure 3.1 suggests that the central issues facing a KM initiative vary across organizational
cultures and individual dispositions. An organizational culture that is individualistic combined
with an individual who has an individual disposition will result in closed sharing and sourcing.
1

A central issue would be trustthe individualistic individual in an individualistic organization
would assume that other individuals would act in a self-interested manner and withhold their best
knowledge. Helping individuals trust that the knowledge they share will not be used against them
and that their own value in the frm will increase, not decrease, as a result of sharing and sourcing
would be the major challenge.
The individualistically disposed individual in a collectivist organization would be selective in
his or her sourcing and sharing, doing so when there were clear benefts to him or her and avoiding
so when the costs of sourcing or sharing (namely time, but also reputation) were greater than the
benefts. In such an environment, the reward system becomes a central issue. In order to encour-
age the individualistically disposed individuals to contribute their best knowledge consistently
and to reuse knowledge (without negatively associating knowledge reuse with lack of individual
creativity), the organization has to have a reward system in place that encourages employees to
conform to group norms for knowledge sourcing or sharing.
The cooperatively disposed individual in an individualistic organization would be more likely
to engage in random sharing and sourcing of knowledge. By nature of his or her disposition, such
an individual will be willing to share his or her knowledge with others. However, due to overriding
individualistic values of the organization, others will not be willing to share or seek out knowledge
unless they are frst convinced of the value of doing so. They will only do so randomly as they
perceive that they may personally beneft from others knowledge (sourcing) or from sharing their
insights with others. In such an environment, the central issue is one of perceived value where
organizational members need to be convinced of the intrinsic value to both themselves and the
organization of both seeking and sharing knowledge.
Finally, the cooperatively disposed individual in a collective organization would be more
likely to engage in the open sharing and sourcing of knowledge. This will occur because such an
individual will be predisposed toward both seeking (sourcing) and sharing knowledge consistent
with the team-oriented environment characterized by the organizations collectivist value system.
Given such openness to knowledge sharing and sourcing, the organization will most likely experi-
ence a very high volume of both types of activities. In an environment like this, a central issue for
organization becomes one of setting up quality control systems to ensure that the organization is
Figure 3.1 Relationship between Individual Personality, Organizational Culture, and KM
Closed sharing
and sourcing.
Trust a central
issue.
Selective
sharing and
sourcing.
Reward systems
a central issue.
Random sharing
and sourcing.
Perceived value
a central issue.
Open sharing
and sourcing.
Quality control a
central issue.
Individualistic
Individual Disposition
Cooperation
Individualistic Collectivistic
Organizational Level Culture
not overwhelmed with irrelevant, hard-to-fnd knowledge. Rather, quality controls need to be set
up to ensure the smooth fow of accurate, timely, easy-to-locate knowledge.
Longitudinal Perspective
We did not encounter studies that attempted to determine whether and, if so, how a KM initiative
altered organizational or subunit culture or individual disposition. However, several studies of
IT at the organizational level suggest that the use of IT can result in a change in organizational
culture (Leidner and Kayworth, 2006). One study by Doherty and Doig (2003) examined the
infuence of a new data warehousing capability on organizational culture. The authors found that
improvements in frms data warehousing capabilities led to changes in customer service, fexibility,
empowerment, and integration values. In the other study, Doherty and Perry (2001) examined the
infuence of a new workfow management system (WMS) on organizational culture. They found
that implementation of the WMS strengthened organizational culture values related to customer
orientation, fexibility, quality focus, and performance orientation.
While these two studies do not specifcally address the KM context, their fndings do suggest
that the application of KM could, over time, infuence cultural values. Furthermore, there are
arguments from a theoretical standpoint that would suggest that information technology, over
time, will infuence cultural values. Most notably is the Structuration Theory (Orlikowski, 1992),
which advances the notion that a given group will enact a given technology based upon the struc-
tural properties of the group (e.g., cultural values). However, over time, the groups values (e.g.,
structure) may be altered by the actual use of the technology. This explains why various groups
may adapt and structure the same information and communication technologies differently and
why, as some speculate, IT may change culture gradually over time.
Given that KM initiatives are often intended to help create a knowledge-oriented culture, it
would indeed be surprising if KM did not have some impact on organizational or subunit culture.
To study this question, researchers would need to take a longitudinal approach and trace the de-
velopments and evolution of a KM over time to discover the ways in which the values associated
with information, knowledge, and IT gradually changed.
CONCLUSION
Early efforts at designing IS to support the information needs of managers focused frst on MIS
and later on DSS/EIS technologies to provide timely, accurate, and relevant information to decision
makers, More recently, weve seen the emergence of KMS as core technologies critical to manag-
ing the information and knowledge intensive needs of large, often globally dispersed organiza-
tions. While earlier technologies used organizational structure and decision-making interpretive
frameworks, we believe that an understanding of organizational culture and its relationship to KM
is the key to understanding how frms can effectively deploy organization-wide KMS to leverage
their intellectual assets for competitive advantage. At this juncture, there are several critical areas
of inquiry to better understand cultures relationship with KM and ultimately to better understand
how frms can be more effective at KM. First, there is a need for analysis of KMculture interac-
tions at multiple levels to include individual disposition, organizational-level culture, and subunit
cultures. This will help move us from the simplistic unitary view of culture to a more complex one
where KM approaches, outcomes, and uses within the same organization may vary due to complex
interactions of multiple subcultures values. Second, we need to examine how KM use may, over
time, infuence organizational culture values and even how cultural change may be engineered
through targeted use of KM. Finally, while not addressed specifcally here, further work needs to
be done examining how national culture might interact with organizational as well as individual
cultures to shape KM initiatives and outcomes.
NOTE
1. Knowledge sourcing is defned by Gray and Meister (2004, p. 821): the extent to which an individual
accesses other employees expertise, experience, insights, and opinions.
REFERENCES
Ackoff, R.L. 1967. Management misinformation systems. Management Science, 14, 4, 147156.
Alavi, M.; Kayworth, T.; and Leidner, D.E. 20052006. An empirical examination of the infuence of orga-
nizational culture on knowledge management practices. Journal of Management Information Systems,
22, 3, 191224.
Alavi, M., and Leidner, D. 1999. Knowledge management systems: Issues, challenges, and benefts. Com-
munications of the AIS, 1, 7, 127.
Applegate, L.M. 1988. Lockheed-Georgia Company: Executive information systems. Boston: Harvard Busi-
ness School Publishing. Case No. 9-187-147.
Applegate, L.M., and Osborn, C.S. 1988. Phillips 66 Company: Controlling a company through crisis. Boston:
Harvard Business School Publishing. Case No. 9-189-006.
Baltahazard, P.A., and Cooke, R.A. 2003. Organizational culture and knowledge management success: as-
sessing the behavior-performance continuum. In Proceedings of the 37th Hawaii International Conference
on Information Systems, Honolulu, HI.
Bansal, P. 2003. From issues to actions: The importance of individual concerns and organizational values in
responding to natural environmental issues. Organization Science, 14 (Sep/Oct), 510527.
Barrett, M.; Capplemann, S.; Shoib, G.; and Walsham, G. 2004. Learning in knowledge communities: Man-
aging technology and context. European Management Journal, 22 (February), 111.
Blake, D.R.S., and Mouton, J.S. 1964. The Managerial Grid. Houston, TX: Gulf Publishing.
Bloor, G., and Dawson, P. 1994. Understanding professional culture in organizational context. Organization
Studies, 15, 2, 275295.
Brown, S.J., and Duguid, P. 2000. Balancing act: How to capture knowledge without killing it. Harvard
Business Review, 78, 3 (MayJune), 7380.
Burchell, S.; Clubb, C.; Hopwood, A.G.; Hughes, J.; and Nahapiet, J. 1980. The roles of accounting in
organizations and society. Accounting, Organizations, and Society, 5, 1, 527.
Carter, N.M. 1984. Computerization as a predominate technology: Its infuence on the structure of newspaper
organizations. Academy of Management Journal, 27 (June), 247270.
Chatman, J.A., and Barsade, S.G. 1995. Personality, organizational culture, and cooperation: Evidence from
a business simulation. Administrative Science Quarterly, 40 (September), 423443.
Cooke, R.A., and Lafferty, J.C. 1987. Organizational Culture Inventory (Form III). Plymouth, MI: Human
Synergistics.
Courtney, J.; Croasdell, D.; and Paradice, D. 1997. Lockean inquiring organizations: guiding principles
and design guidelines for learning organizations. In Proceedings of the 1997 Americas Conference on
Information Systems, Indianapolis, IN.
Davenport, T.H. 1997. Information Ecology, Oxford: Oxford University Press.
Davenport, T.H.; DeLong, D.W.; and Beers, M.C. 1998. Successful knowledge management projects. Sloan
Management Review, 39 (Winter), 4357.
Deal, T.E., and Kennedy, A.A. 1982. Corporate Cultures: The Rites and Rituals of Corporate Life. Reading,
MA: Addison-Wesley.
DeLone, W.H., and McLean, E.R. 1992. Information systems success: The quest for the dependent variable.
Information Systems Research, 3 (March), 6095.
DeLong, D.W., and Fahey, L. 2000. Diagnosing cultural barriers to knowledge management. Academy of
Management Executive, 14 (November), 113127.
Denison, D.R., and Mishra, A.K. 1995. Organizational culture and effectiveness. Organization Science, 6,
2, 204223.
DeTiene, K.B., and Jackson, L.A. 2001. Knowledge management: Understanding theory and developing
strategy. Competitiveness Review, 11, 1, 111.
Dougherty, D. 1992. Interpretive barriers to successful product innovation in large frms. Organizational
Science, 3, 2, 179202.
Doherty, N.F., and Doig, G. 2003. An analysis of the anticipated cultural impacts of the implementation of
data warehouses. IEEE Transactions on Engineering Management, 50 (February), 7888.
Doherty, N.F., and Perry, I. 2001. The cultural impact of workfow management systems in the fnancial
services sector. The Services Industry Journal, 21 (October), 147166.
Dub, L., and Robey, D. 1999. Software stories: Three cultural perspectives on the organizational con-
text of software development practices, Accounting Management and Information Technologies, 9, 4,
223259.
Earley, P.C. 1993. East meets West meets Mideast: Further explorations of collectivist and individualistic
work groups. Academy of Management Journal, 36 (April), 319348.
Elliott, D.G. 1992. Executive information systems: Their impact on executive decision making. Doctoral
dissertation, The University of Texas at Austin, May.
Foster, L.W., and Flynn, D.M. 1984. Management information technology: Its effects on organizational form
and function. MIS Quarterly, 8 (December), 229236.
Franz, C.R.; Robey, D.; and Koeblitz, R.R. 1986. User response to an online information system: A feld
experiment. MIS Quarterly, 10 (March), 2842.
Friend, D. 1986. Executive database: Accounting historys link to the future. The Magazine for Financial
Executives, 2 (October), 68.
Garud, R., and Kumaraswamy, A. 2005. Vicious and virtuous circles in the management of knowledge: The
case of Infosys Technologies. MIS Quarterly, 29 (March), 933.
Goffee, R., and Jones, G. 1996. What holds the modern company together? Harvard Business Review, 74
(NovemberDecember), 133148.
Gold, A.H.; Malhotra, A.; and Segars, A.H. 2001. Knowledge management: An organizational capabilities
perspective. Journal of Management Information Systems, 18, 1, 185214.
Gray, P.H., and Meister, D.B. 2004. Knowledge sourcing effectiveness. Management Science, 50 (June),
821834.
Greengard, S. 1998. Storing, shaping and sharing collective wisdom. Workforce, 77 (October), 8288.
Gupta, A.K., and Govindarajan, V. 2000 Knowledge managements social dimension: Lessons from Nucor
Steel. Sloan Management Review, 42 (Fall), 7180.
Hargadon, A.B. 1998. Firms as knowledge brokers: Lessons in pursuing continuous innovation. California
Management Review, 40 (Spring), 209227.
Hasan, H., and Gould, E. 2001. Support for the sense-making activity of managers. Decision Support Sys-
tems, 31 (May), 7186.
Hofstede, G. 1998. Identifying organizational subcultures: An empirical approach. Journal of Management
Studies, 35, 1, 112.
Hofstede, G.; Neuijen, B.; Ohayv, D.D.; and Sanders, G. 1990. Measuring organizational cultures: A qualita-
tive and quantitative study across twenty cases. Administrative Science Quarterly, 35 (June), 286316.
Hogue, J.T. 1987. A framework for the examination of management involvement in decision support systems.
Journal of Management Information Systems, 4 (Summer), 96110.
Holub, A. 1988. What happens when information regarding the quality of a banks services becomes visible?
EIS Conference Report, 1 (July).
Houdeshel, G., and Watson, H.J. 1987. The management information and decision support (MIDS) system
at Lockheed-Georgia. MIS Quarterly, 11 (March), 126140.
Huber, G.P. 1990. A theory of the effects of advanced information technologies on organizational design,
intelligence, and decision making. Academy of Management Review, 15, 1, 4771.
. 1991. Organizational learning: The contributing processes and the literatures. Organization Science,
2, 1, 88115.
Huber, G.P., and McDaniel, R.R. 1986. The decision-making paradigm of organizational design. Manage-
ment Science, 32 (May), 572589.
Janz, B.D., and Prasarnphanich, P. 2003. Understanding the antecedents of effective knowledge management:
The importance of a knowledge-centered culture. Decision Sciences, 34 (Spring), 351384.
Jarvenpaa, S.L., and Staples, S.D. 2001. Exploring perceptions of organizational ownership of information
and expertise. Journal of Management Information Systems, 18 (Summer), 151183.
Jones, G.R. 1983. Transaction costs, property rights, and organizational culture: An exchange perspective.
Administrative Science Quarterly 28 (September), 454467.
Kador, J. 1989. Executive support systems keep execs in control. Planner, 11 (Spring), 15.
Keen, P.G.W., and Morton, M.S.S. 1982. Decision Support Systems: An Organizational Perspective. Read-
ing, MA: Addison-Wesley.
Klatzky, S.R. 1970. Automation, size, and the locus of decision making: The cascade effect. Journal of
Business, 43, 2, 141151.
Knapp, E., and Yu, D. 1999. How culture helps or hinders the fow of knowledge. Knowledge Management
Review, 2, (March/April), 1621.
KPMG Management Consulting. 1998. Knowledge Management: Research Report.
Kroeber, A.L., and Kluckhohn, C. 1952. Culture: A Critical Review of Concepts and Defnitions. Cambridge,
MA: The Museum.
Laudon, K.C., and Laudon, J.P., 2000. Management Information Systems, 6th ed. Upper Saddle River, NJ:
Prentice-Hall.
Le Blanc, L.A., and Kozar, K.A. 1990. An empirical investigation of the relationship between DSS usage
and system performance: A case study of a navigation support system. MIS Quarterly, 14 (September),
263277.
Lee, H., and Choi, B. 2003. Knowledge management enablers, processes, and organizational performance:
An integrative view and empirical examination. Journal of Management Information Systems, 20, 1,
179228.
Lee, G.K., and Cole, R.E. 2003. From a frm-based to a community-based model of knowledge creation: The
case of the Linux kernel development. Organization Science, 14, 6, 633649.
Leidner, D.; Alavi, M.; and Kayworth, T. 2006. The role of culture in knowledge management: A case study
of two global frms. The International Journal of Electronic Collaboration, 2, 1, 1740.
Leidner, D.E., and Elam, J.J. 1995. The impact of executive information systems on organizational intelligence
and decision making. Organization Science, 6 (NovemberDecember), 645665.
Leidner, D.E., and Kayworth, T.R. 2006. A review of culture in information systems research: Towards a
theory of information technology culture confict. MIS Quarterly, 30, 2, 357399.
Leonard, D., and Sensiper, S. 1998. The role of tacit knowledge in group innovation. California Management
Review, 40 (Spring), 112132.
Levinthal, D., and March, J. 1993. The myopia of learning. Strategic Management Journal, 14 (Winter),
95112.
Louis, M.R. 1985. Sourcing workplace cultures: Why, when and how. In Kilmann, R.H., Saxton, M.J., and
Serpa, R. (eds.), Gaining Control of the Corporate Culture, San Francisco, CA: Jossey-Bass, 126136.
Martin, J., 1985. Can organizational culture be managed? In P.J. Frost, L.F. Moore, M.R. Louis, C.C. Lund-
berg, and J. Martin (eds.), Organizational Culture. London: Sage Publications, pp. 9598.
Martin, J., and Siehl, C. 1983. Organizational culture and counterculture: An uneasy symbiosis. Organiza-
tional Dynamics, 12, 2, 5265.
Martin, J.; Sitkin, S.B.; and Boehm, M. 1985. Founders and the elusiveness of a cultural legacy. In P.J. Frost,
L.F. Moore, M.R. Louis, C.C. Lundberg, and J. Martin (eds.), Organizational Culture. London: Sage
Publications, pp. 99124.
Meyerson, D., and Martin, J. 1987. Cultural change: An integration of three different views. Journal of
Management Studies, 24, 623647.
Miles, R.; Snow, C.; Matthews, J.; Miles, G.; and Coleman, H. 1997. Organizing in the knowledge age:
Anticipating the cellular form. Academy of Management Executive, 11 (November), 724.
Mitchell, P. 1988. An EIS is good for you. EIS Conference Report, 1 (July).
Molloy, S., and Schwenk, C.R. 1995. The effects of information technology on strategic decision making.
Journal of Management Studies, 32 (May), 283311.
Nadler, D., and Tushman, M. 1988. Strategic Organization Design. Glenview, IL: Scott Foresman and
Company.
ODell, C., and Grayson, C.J. 1998. If only we know what we know: Identifcation and transfer of best
practices. California Management Review, 40 (Spring), 154174.
OReilly, C.A.; Chatman, J.; and Caldwell, D.F. 1996. Culture as social control: Corporations, cults, and
commitment. Research in Organizational Behavior, 18, 157200.
Orlikowski, W.J. 1992. The duality of technology: Rethinking the concept of technology in organizations.
Organization Science, 3 (August), 398427.
Ravishankar, R.; Pan, S.L.; and Leidner, D.E. 2006. KM Strategy and Cultural Alignment. Working paper,
National University of Singapore.
Pettigrew, A.M. 1979. On studying organizational cultures. Administrative Science Quarterly, 24, 4,
570581.
Robey, D. 1977. Computers and management structure: Some empirical fndings re-examined. Human Rela-
tions, 30 (November), 963976.
Rockart, J., and DeLong, D. 1988. Executive Support Systems: The Emergence of Top Management Computer
Use. Homewood, IL: Dow-Jones-Irwin.
Rose, R.A. 1988. Organizations as multiple cultures: A rules theory analysis. Human Relations, 41,
139170.
Ruggles, R. 1998. The state of the notion: knowledge management in practice. California Management
Review, 40 (Spring), 8089.
Sackmann, S.A. 1992. Culture and sub-cultures: An analysis of organizational knowledge. Administrative
Science Quarterly, 37, 140161.
Schultze, U., and Boland, R. 2000. Knowledge management technology and the reproduction of knowledge
work practices. Journal of Strategic Information Systems, 9, 23, 193213.
Schein, E.H. 1985. How culture forms, develops and changes. In R.H. Kilmann, M.J. Saxton, R. Serpa, et al.
(eds.), Gaining Control of the Corporate Culture. San Francisco, CA: Jossey-Bass, pp. 1743.
Simon, H.A. 1977. The New Science of Management Decision. Englewood Cliffs, NJ: Prentice-Hall.
Somogyi, E.K., and Galliers, R.D. 1987. Applied information technology: From data processing to strategic
information systems. Journal of Information Technology, 2 (March), 3041.
Sprague, R.H., and Carlson, E.D. 1982. Building Effective Decision Support Systems. Englewood Cliffs,
NJ: Prentice-Hall.
Stewart, R. 1971. How Computers Affect Management. Cambridge, MA: MIT Press.
Styhre, A.; Roth, J.; and Ingelgard, A. 2002. Care of the other: Knowledge-creation through care in profes-
sional teams. Scandinavian Journal of Management 18 (December), 503522.
Todd, P., and Benbasat, I. 1987. Process tracing methods in decision support systems research: Exploring
the black box. MIS Quarterly, 11 (December), 492512.
. 1991. An experimental investigation of the impact of computer-based decision aids on decision
making strategies. Information Systems Research, 2 (June), 87115.
Van Maanen, J., and Barley, S.R. 1985. Cultural organization: Fragments of a theory. In P.J. Frost, L.F. Moore,
M.R. Louise, C.C. Lundberg, and J. Martin (eds.), Organizational Culture. London: Sage Publications,
pp. 3154.
Vandenbosch, B., and Higgins, C. 1996. Information acquisition and mental models: An investigation into
the relationship between behavior and learning. Information Systems Research, 7 (June), 198214.
Von Krogh, G. 1998. Care in knowledge creation. California Management Review, 40 (Spring), 133153.
Wallach, E.J. 1983. Individuals and organizations: The cultural match. Training and Development Journal,
37 (February), 2835.
Watson, S. 1998. Getting to aha! Companies use intranets to turn information and experience into knowl-
edgeand gain a competitive edge. Computer World, 32 (January 26), 15.
Whisler, T. 1970. The Impact of Computers on Organizations. New York: Praeger Publishers.
PART II
THE ROLE OF INFORMATION
TECHNOLOGY IN
63
ChaptEr 4
SOCIAL NETWORKS AND
INFORMATION TECHNOLOGY
Evolution and New Frontiers
maryam alaVi and GErald C. KanE
Abstract: The purposes of this chapter are to trace the development of knowledge sharing research
in the social network tradition, identify key features of knowledge sharing networks that have
emerged from this research stream, and suggest future directions for research. A key component
in the evolution of social network research on knowledge sharing has been the recognition that
information technology (IT) can play an integral role in knowledge sharing networks. A number
of studies have found that the contributions of the social network perspective can be extended
effectively into IT-based knowledge sharing relationships. We rely upon these observations to
suggest that the continued evolution of social network research on knowledge sharing can and
should incorporate the IT artifact more intentionally and robustly into its examination of knowledge
sharing activities of social networks. In doing so, we argue that the social network perspective
on knowledge sharing represents a useful framework and new frontier for information systems
research.
Keywords: Social Networks, Knowledge Sharing, Embedded Knowledge, Communication Support
Capabilities, Information Management Capabilities, Nodes, Ties
INTRODUCTION
Social network research has become an increasingly popular and important lens in organizational
research in recent years. Unlike traditional research, which focuses on the characteristics of
individual actors and the effects these characteristics have on particular outcomes, social net-
works focus on the relationship between actors, the characteristics of these relationships, and
their outcomes. With its roots in the Harvard school of sociology in the 1920s and 1930s, the
social network approach has been applied extensively in sociology, anthropology, psychology,
and epidemiology, among other felds (Scott, 2000). Social networks have also been a major fo-
cus of study in the management literature over the past 2030 years (Baker and Faulkner, 2002;
Borgatti and Foster, 2003; Gulati, Dialdin, and Wang, 2002; Monge and Contractor, 2000; Raider
and Krackhardt, 2002).
Despite its long and rich history, research on social networks has grown exponentially in recent
years (Borgatti and Foster, 2003). Some of this growth may be attributable to an increase in com-
puting power and specialized research software, such as UCINet, NetMiner, and Pajek, enabling
64 ALAVI AND KANE
the visualization and statistical analysis of large networks. Journals (e.g., Social Networks) and
professional organizations (e.g., INSNA) have emerged to serve as an outlet and a resource for
research focused on social networks. Most recently, commercial applications that use this research
and allow people to monitor, track, and develop their own social networks (e.g., linkedin.com,
facebook.com) have brought the concepts of social networks into popular awareness.
A signifcant emphasis of this emerging stream of social network research has related to orga-
nizational knowledge sharing (Chan and Liebowitz, 2006). Social network research on knowledge
sharing tends to focus on informal aspects of knowledge sharing, as opposed to its formal aspects
(Cross and Parker, 2004). Whereas formal initiatives establish channels and structures and provide
tools through which individuals are supposed to or able to share knowledge, informal approaches
tend to emphasize how individuals actually work within these channels and other mechanisms to
share knowledge in a given context.
The purposes of this chapter are to trace the development of knowledge sharing research in the
social network tradition, identify key features of knowledge sharing networks that have emerged
from this research stream, and suggest future directions for research. A key component in the evo-
lution of social network research on knowledge sharing has been the recognition that information
technology (IT) can play an integral role in knowledge sharing networks. A number of studies
have found that the contributions of the social network perspective can be extended effectively
into IT-based knowledge sharing relationships. We rely upon these observations to suggest that the
continued evolution of social network research on knowledge sharing can and should incorporate
the IT artifact more intentionally and robustly into its examination of knowledge sharing activities
of social networks. In doing so, we argue that the social network perspective on knowledge sharing
represents a useful framework and new frontier for information systems (IS) research.
Specifcally, this chapter will proceed as follows. First, we address the foundations of social
network researchthe defnition of knowledge, the essential components, a brief history of social
network research, and the role of IT in knowledge sharing networks. Second, we examine the features
of social networks that researchers have found critical in infuencing knowledge sharing outcomes.
Third, we consider future directions for research in knowledge sharing within the social network
paradigm, particularly as it relates to new ways that IT can be incorporated into its analyses.
FOUNDATIONS OF SOCIAL NETWORK RESEARCH
In this section, we address foundations of social network research that are important for under-
standing the potential role of social network research in knowledge sharing, particularly as it
relates to IS research. First, we review the defnition of knowledge most frequently adopted by
social network researchers. Second, we identify the key components and terminology of social
networks that form the core of social network analysis. Third, we trace the evolution of knowledge
sharing research in the social network tradition, particularly as it relates to the introduction of IT
into social networks. Finally, we look at how IT has been integrated into social network research,
specifcally addressing its unique role in knowledge sharing networks.
Defnition of Knowledge
The debate surrounding the defnition of knowledge can be traced to the ancient Greeks and con-
tinues into the present day (Alavi and Leidner, 2001). The defnition of knowledge is signifcant
because how one defnes it likely infuences how one seeks to manage and study it (Earl, 2001).
If the organization views knowledge as a stock, they are likely to focus on efforts that store and
SOCIAL NETWORKS AND INFORMATION TECHNOLOGY 65
organize knowledge; if the organization views knowledge as a fow, they are more likely to focus
on developing effective knowledge processes. The researcher and the research method will also
be somewhat related to the view of knowledge, but it is not clear whether the researchers view of
knowledge infuences the chosen method, or vice versa. For instance, the researcher may choose
particular research sites and appropriate methods based on their own understanding of knowledge.
Conversely, the researcher may fnd an appropriate knowledge management initiative to study,
adopting the view of knowledge embodied by the initiative and selecting appropriate research
methods accordingly. The view of knowledge and the research method are related to one another,
but the causal direction of the infuence is unclear.
Knowledge management research in the IS literature frequently adopts the data-information-
knowledge hierarchy, in which the three constructs are defned in relation to one another (Alavi
and Leidner, 2001). The primary level of this hierarchy is data, which consist of raw symbols.
Information is created from data by either processing it or combining it with other data to create
a meaning. Knowledge is formed as individuals interact with information and personalize it for
ones own use. Because different researchers have argued for different relationships among these
constructs (Spiegler, 2003; Tuomi, 1999), some researchers have noted that the hierarchy often
leads to as much obfuscation as clarifcation (Gray, 2002).
Social network researchers adopt a slightly different defnition of knowledge, one also shared
by IS researchersthe embedded view of knowledge. Unlike the data-information-knowledge
perspective, knowledge in the embedded view is not an object with a discrete identity of its own.
Rather, it exists as an inherent part of the social relationships and other environmental factors
in which it is created and applied, and knowledge cannot be understood apart from these social
interactions (Granovetter, 1985). In the embedded view, knowledge is localized as individuals
develop knowledge by dealing with specifc problems they address in their environment; knowl-
edge is embedded in practice in which individuals have diffculty articulating knowledge apart
from the way they use it in action; and knowledge is invested in practice as individuals develop
more effcient and effective ways of interacting based on this knowledge-generating experience
(Carlile, 2002).
The embedded view of knowledge has a rich history in organizational research. Kogut and
Zander (1992, 1996) argue that knowledge is developed within the frm through the patterns of
knowledge combination and exchange engaged in by its employees and, thus, knowledge should
be considered as embedded in these relationships. Otherwise, if knowledge was held only in the
minds of employees, then frms could gain new competencies simply by hiring new employees.
Spender (1996) extended this argument by suggesting that knowledge could also be embedded in
the patterns of combination and exchange between people and artifacts. Drawing upon the work
of Callon (1998, 1986) and LaTour (1987, 1996), Spender suggested that knowledge is embedded
in a frms actor-networks, where humans and artifacts are regarded as equivalent entities. Thus,
the embedded view of knowledge suggests that knowledge is embedded in individuals minds, in
the relationships between individuals, and in the relationships between individuals and artifacts,
such as databases and computer fles. This conceptualization of embedded knowledge is attractive
to IS researchers because it defnes knowledge in a way that accounts for the role of both people
and IS in knowledge sharing (Nidumolu, Subramani, and Aldrich, 2001).
Components of Social Networks
Because the embedded view of knowledge focuses on the knowledge held by entities and the
relationships between them, social network research analyzes these relationships by focusing
66 ALAVI AND KANE
on two essential components of a networknodes and ties. The term nodes denotes any entity
in a network (e.g., people or IS), whereas ties refers to the relationships between these nodes.
Most commonly, social network research investigates different characteristics of the ties in a
networksuch as the number of ties, types of tie (friendship, advice, professional), the strength
of the tie (strong or weak), or the density of ties in a network (high or low). Network researchers
may also investigate the characteristics of the nodes, but these analyses occur primarily in regard
to how the nodes are positioned in relation to one another as a result of their ties (Brass, 1995;
Scott, 2000; Wasserman and Faust, 1994).
Nodes may represent any entity in a network (individuals, business units, companies, etc.), so the
level of analysis in network research is described somewhat differently than in traditional research.
Researchers can examine a network at three different levels of analysis: the dyad, the ego network,
or the whole network (Raider and Krackhardt, 2002). Dyadic analysis focuses on characteristics
of a single relationship between two nodes. Dyadic research typically analyzes the effect of one
type of relationship on other types of relationships. Ego network analysis focuses on the portfolio
of relationships maintained by a single node and whether the characteristics of that portfolio have
particular effects on or benefts to that node. Whole network analysis addresses all of the nodes and
ties among them, examining either the environmental characteristics that lead to a particular structure
(Kogut, 2000) or the benefts resulting from a structure (Watts, 1999). Social network researchers
have studied knowledge sharing in networks at each of these levels of analysis.
Researchers have also noted that the structures and features of networks can be modifed in two
different ways: active and passive (Watts 1999). The active approach suggests that organizations can
intentionally engineer certain features and characteristics of the network. For instance, Cross and
Prusak (2002) suggested that individuals play certain specifc roles as central connectors, boundary
spanners, information brokers, and peripheral specialists within the knowledge network. Thus, if
a particular organization or team is in need of particular characteristics within a given knowledge
network, it can simply move an employee accustomed to playing that particular role in the orga-
nization into or out of the network. The passive perspective on knowledge sharing suggests that
the network forms in response to organizational conditions and that the organization does not seek
to intentionally reshape a network but, instead, guides the development of a network by changing
the conditions under which it operates. Kogut (2000) identifed these environmental conditions as
a networks generative rules and illustrated how the incentive structures in organizations shaped
how individuals shared knowledge within the organization. A number of additional factors, such
as organizational culture, importance of networks for success, and compensation schema, are
considered important to an individuals willingness to develop and/or participate in knowledge
sharing networks (Collins and Clark, 2003; Hoegl, Parboteeah, and Munson, 2003).
History of Knowledge Sharing in Social Networks
The social network approach to knowledge sharing has a rich history, and this stream of research
has evolved over the years by examining knowledge sharing relationships in even fner detail. The
earliest studies addressed knowledge implications of general social networks, suggesting that the
position of a node in the social network had distinct knowledge benefts or liabilities. Later stud-
ies began to distinguish knowledge sharing relationships from other types of relationships in the
social network, examining the distinct impact of these relationships on network outcomes. Most
recently, studies have recognized that knowledge sharing relationships are multivalent and focus
on the multiple facets of knowledge sharing relationships and different channels across which
these relationships can occur.
The frst studies involving knowledge sharing in social networks addressed the knowledge
outcomes of general social relationships. Some of these studies examined particular types of
social relationships maintained by individuals, suggesting that the nature of these relationships
may infuence the amount and type of knowledge one could access through those relationships.
This approach can be traced back to Granovetters (1973) work on individuals seeking jobs, in
which particular types of relationships were found to convey particular knowledge benefts for
the job seeker. Others have noted that ones position within a larger network of relationships also
conveyed particular knowledge benefts. Burt (1992) noted that people who were central in the
network of social relationships enjoyed certain knowledge benefts above those who were more
peripheral.
From this earliest research on knowledge sharing, social network researchers began to examine
the knowledge sharing relationship independently from other network relationships. Krackhardt
and Hanson (1993) explored the difference between knowledge sharing relationships and purely
social friendship networks, suggesting that the knowledge sharing network could reveal the experts
in the organization. Ibarra (1993) compared the position of an individual within the knowledge
sharing network to their formal position within the organization. The position of the individual
in the informal knowledge sharing network was more signifcantly related to outcomes than their
formal position in the organization. In addition, researchers do not necessarily assume that all
knowledge sharing relationships are benefcial; indeed, some relationships can actually distrib-
ute misinformation or ineffective knowledge (Schultze and Orlikowski, 2004). Sparrowe et al.
(2001) distinguished between benefcial knowledge sharing relationships and those that hindered
knowledge sharing tasks, fnding that both explained knowledge outcomes in different ways. In
each of these studies, the knowledge sharing relationship is isolated and examined independently
from other relationships in the social network.
Knowledge sharing research has continued the evolution of examining the knowledge sharing
relationship more closely, examining the various facets of the network knowledge sharing relation-
ships. Hansen has focused on the content of knowledge sharing relationships, recognizing that tacit
and explicit knowledge are shared differently via network interactions (1999) and examining how
specifc content may infuence the nature of knowledge sharing relationships (2002). Other network
researchers have also begun to explore the multifaceted nature of knowledge sharing in networks.
Cross, Borgatti, and Parker (2002) argue that the knowledge sharing in networks actually consists
of a portfolio of networks, distinguishing between the communication network, the information
network, the problem-solving network, and the access networks all to describe different essential
facets of knowledge sharing networks. Similarly, Borgatti and Cross (2003) survey members of
a biotechnology and fnancial services frm to determine which of several factors infuence the
likelihood of knowledge sharing among individuals in the frms. They found that access to an
individual, awareness of what that person knows, and valuing what that person knows were all
important in explaining knowledge sharing in organizations.
Using Information Technology to Support Knowledge Sharing in Social Networks
Although the social network perspective on knowledge sharing has evolved in the detail of its
analysis regarding interpersonal knowledge sharing relationships, it has also evolved in terms of
the context of the relationships under analysis. Researchers recognized that the social network
approach to knowledge sharing relationships would be uniquely applicable in studying knowl-
edge sharing relationships that occur over the Internet and IT-enabled communication support
capabilities (e.g., e-mail) (Wellman, 2001). IT-enabled communication support capabilities (CSC)
68 ALAVI AND KANE
have proved to be a valuable contribution to knowledge sharing in organizations. Mainstream
IS research has tended to focus on the more formal characteristics of these systems, such as the
richness capability by the system (Daft, Lengel, and Trevino, 1987) and their ft with appropriate
knowledge sharing tasks (Goodhue and Thompson, 1995). In contrast, the social network approach
examines the informal aspects of how individuals actually employ these CSC for knowledge shar-
ing, examining the characteristics of the emergent knowledge sharing relationships that occur via
these CSC (Table 4.1).
Markus (1994) was among the frst to expand the understanding of CSC-based knowledge
sharing beyond its conceptualization as purely a function of media richness and task ft. She found
that, whereas managers perceptions of the appropriateness of the task for CSC use were somewhat
relevant, managers also tended to use e-mail for equivocal tasks such as problem defnition and
dispute resolution, tasks for which e-mail richness was not ideally suited. Markus concluded that
an individuals choice to use CSC for knowledge sharing may be partially explained by the ft
with the knowledge sharing task but also had explanations that could be traced to interpersonal
norms, personal preferences, and organizational culture. This fnding is consistent with the rich
tradition of research that has traced technology diffusion in organizations along existing social
networks (Table 4.2).
Zack and McKenney (1995) explored the more complex relationships of an entire network
of knowledge sharing relationships that occurred via CSC. They compared the knowledge shar-
ing patterns of two newspaper editorial boards that used identical CSC for the same knowledge
sharing tasks. They found that these two boards used the CSC in signifcantly different ways,
despite the fact that they shared nearly identical tasks, tracing the differences in usage to their
existing cultures and interpersonal knowledge sharing relationships. Thus, they concluded that
entire knowledge sharing networks could indeed occur via communication support technologies,
Table 4.1
Previous Studies Involving CSC-based Knowledge Sharing Networks
Year Author Research Findings
1992 McKenney and Zack E-mail weak ties used for different tasks than face-to-face
strong ties.
1995 Pickering and King Relationships via e-mail categorized as weak ties. More likely to
be adopted early by frms that rely on such ties for professional
mobile workforce.
1995 Zack and McKenney Identical work structures, different social structures used
collaborative technologies differently.
1996 Constant, Sproull,
and Keisler
Examined factors leading to value obtained in e-mail advice
seeking.
1999 Ahuja and Carley Social networks via CSC share characteristics of traditional
social structures. Found evidence of hierarchy and
centralization.
2001 Butler Network size and communication activity mediated value of
online community groups.
2003 Ahuja, Galetta, and
Carley
Centrality was an important predictor of performance, more
than functional role, status, or communication role.
2005 Wasko and Faraj Centrality was a predictor of an individuals willingness to share
knowledge in an electronic network of practice.
and Zack and McKenney were among the frst to use the tools of social network analysis to begin
exploring these relationships.
FEATURES OF KNOWLEDGE SHARING NETWORKS
Network research on knowledge sharing has found that particular features of social networks are
important for understanding the outcomes of knowledge sharing relationships. These features
infuence network outcomes in both interpersonal and CSC-based knowledge sharing networks in
similar ways. First, the strength of the relationships, or ties, between individuals has been shown to
infuence the ease with which particular types of knowledge is shared between two people. Second,
the portfolio and confguration of ties that comprise a knowledge networkoften referred to as a
networks densityhas also been shown to be important, both in conjunction with and indepen-
dent of tie strength. Third, the position of an individual within a network of otherswhether that
individual is central or peripheral to the larger network of which he or she is a parthas proved
infuential in understanding the outcomes of the knowledge network. Fourth, other factors such as
trust and shared background knowledge have also proved important to understanding knowledge
sharing in social networks.
Tie Strength
Network research has long examined how the nature of ties between individuals infuences the
knowledge sharing behavior of the network (Table 4.3). Tie strength is defned as the amount of
Table 4.2
Previous Studies Involving Technology Diffusion in Social Networks
1990 Barley Introduction of technology changed social network within the
hospital but changed it in different ways in different hospitals.
1990 Burkhardt and Brass Early adopters of IT increased their centrality and power more
than late adopters.
1990 Papa Networks comprised of actors from diverse departments
and hierarchical levels were more fexible and productive in
responding to the implementation of a new information system.
1992 Nohria and Eccles Recognizes that electronic networks have place, but place is a
complement of face-to-face interaction.
1991 Rice and Aydin Relational and positional proximity more important to attitudes
towards new technology than usage of spatial proximity.
Basically, what you think others think about an information
system will infuence your perception of it, beyond your own
usage statistics.
1994 Burkhardt Network proximity affects feelings of personal mastery of
technology, structural similarities affect beliefs and attitudes
towards new technology.
1997 Abrahamson and
Rosenkopf
Demonstrated technology diffusion through core-periphery
structure.
2004 Schultze and Orlikowski Human versus Web-based ties. Replaced human intermediaries
with Web interface in insurance frm.
70 ALAVI AND KANE
time, emotional intensity, intimacy, or reciprocal services between two nodes in a network (Brass
1995). Ties are usually described in terms of their strength, either strong or weak; two facets of the
ties are used to determine its strength: frequency of interaction and depth of interaction (Hansen,
1999; Mardsen and Campbell, 1984). Although tie strength has traditionally been examined as a
dichotomous variable (weak or strong), recent research has begun to examine tie strength along
an interval scale (Rubenstein-Montano et al., 2001).
Perhaps the earliest study to investigate the effects of tie strength on knowledge sharing was
Granovetters (1973) classic work on individuals searching for job prospects. He found that weak
ties were more likely to result in a successful job hunt because they allowed individual access to
different types of knowledge. These weak ties were more likely to connect the individual with
a greater variety of others and, in turn, a greater variety of knowledge. The strong ties were not
as effective in the job seeking environment because they were more likely to connect the job
seekers with similar others and, thus, redundant knowledge. In contrast, others have recognized
that strong ties have other benefts for knowledge sharing. Stevenson and Gilly (1991) found that
managers tended to break from the formally established knowledge sharing relationships when
solving certain types of problems, turning instead to individuals with whom they had strong ties
for knowledge and advice. Whereas weak ties provide individuals access to a greater variety of
knowledge, individuals are more likely to rely on strong ties in certain situations because they
provide greater confdence and trust in the knowledge shared between the parties.
Uzzi (1997) extended these fndings on strong ties. He found that managers who maintained
Table 4.3
Network Features: Tie Strength
1973 Granovetter Job seekers relied on weak ties, not strong ties to gain
knowledge of job opportunities.
1988 Coleman Individuals with strong ties less likely to drop out of high
school.
1992 Krackhardt Organizational initiative failed because they did not account
for effects of strong ties.
1992;
1997
Burt Value of weak ties is contingent upon ability to bridge diverse
networks and knowledge.
1996 Constant, Sproull, and
Keisler
Diversity of information leads to more useful responses via
e-mail queries.
1997 Uzzi Strong ties among CEOs both strengthened and weakened
frm position in different contexts.
1997 Podolny and Baron Weak ties better predicted job promotion.
1999 McEvily and Zaheer Nonredundancy of ties, not tie strength, is important for pre-
dicting performance.
1998 Kumar, VanDissel, and
Bielli
IT implementation failed because did not account for strong
ties among users.
2002 Hansen Strong ties between groups are better for tacit knowledge
transfer.
2003 Reagans and Zuckerman Strong ties better for all knowledge transfer, once controlled
for network structure.
2003 Perry-Smith and Shalley Weak ties are more important than strong ties for creativity.
strong ties were more likely to engage in a greater degree and deeper level of knowledge sharing
than managers who had a higher number of weak ties. Nevertheless, Uzzi recognized that these
strong ties might also make the managers more vulnerable under certain conditions. As individuals
develop stronger and more effcient knowledge sharing relationships, they depend more heavily
on these relationships. If a key player in this knowledge sharing network is removed, the entire
network is placed at risk because they have not cultivated other knowledge sharing relationships
upon which to draw. Furthermore, new knowledge was less likely to enter these strong-tie net-
works because their members were less likely to maintain a wide variety of relationships outside
this network. They also were less likely to be aware of knowledge regarding new innovations that
occurred outside their strong-tie network.
Drawing upon and extending this earlier research, Hansen (1999, 2002) suggested that the
relationship between tie strength and knowledge sharing was somewhat paradoxical. Weak ties
are better suited for searching for required knowledge by business units; weak ties are also better
for transferring knowledge that could be readily documented, what Hansen called codifed knowl-
edge. Strong ties, on the other hand, proved to be better for transferring noncodifed knowledge,
which is often regarded as a very valuable type of knowledge to share within an organization.
Hansen also found that strong ties were actually detrimental when seeking to transfer codifed
knowledge because individuals needed to expend energy to maintain the effcacy of these strong
ties for knowledge sharing. This expended energy detracted from performance when noncodifed
knowledge sharing was not necessary for project success.
Tie Strength in Communication Support CapabilitiesSupported Social Networks
Because of its ability to connect a large number of individuals through relatively lean channels of
communication (Daft, Lengel, and Trevino, 1987), researchers have often suggested that knowledge
sharing via CSC introduces weak ties into a network. Pickering and King (1995) were among the
frst to apply this weak tie interpretation to knowledge sharing interactions conducted via CSC,
suggesting that technologically enabled CSC shared many of the same characteristics as tradi-
tional weak ties, such as connecting a wide variety of individuals less frequently and through less
emotionally laden interactions. Constant, Sproull, and Kiesler (1996) also adopted the weak-ties
interpretation of CSC, although their results demonstrate more support of the value of network
density, discussed in the next section, than of tie strength, per se. The weak-ties interpretation also
seems consistent with other fndings showing that CSC provides breadth of knowledge sharing
above depth (Miranda and Saunders, 2003).
Because this weak-ties interpretation of CSC largely rests on the assumption that CSC-based
knowledge sharing is a relatively lean mechanism, the work of Carlson and Zmud (1999) on
channel richness cautions against overemphasizing the weak-ties interpretation of CSC. They
found that as people gained experience communicating via a given CSC, they could expand the
richness by developing new communication genres and communication patterns that could then
transmit richer meaning via the same channel. Furthermore, individuals may use the CSC knowl-
edge sharing relationships to augment other relationships, resulting in a multichannel relationship
of face-to-face and CSC-based channels that may be stronger than those possible via traditional
face-to-face relationships alone (Gergen, 1991; Robey and Boudreau, 1999).
In summary, we can draw several observations from this collection of studies on tie strength
and knowledge sharing. First, strong and weak ties have different effects on knowledge sharing
relationships and tie strength is critical for understanding the effectiveness of knowledge sharing
relationships. Strong ties appear better for cultivating trust and reliability, whereas weak ties ap-
72 ALAVI AND KANE
pear better for accessing or searching for different types of knowledge. Second, each type of tie
has its drawbacks as well as its benefts. Strong ties are costly to maintain and may also make an
individual vulnerable to the dissolution of those ties or to the opportunity costs of not accessing
new knowledge. Weak ties are less effective at transferring noncodifed knowledge, which is often
regarded as a valuable type of shared organizational knowledge. Summary of characteristics of
weak/strong ties is displayed in Table 4.4.
Density
Many researchers have expanded the study of tie strength to include not only the dyadic knowl-
edge sharing relationship between two individuals but also the other relationships that exist as a
part of the wider network in which the single knowledge sharing relationship is situated (Table
4.5). This examination of the wider network of relationships usually involves a consideration of
a networks density, defned as the ratio of actual ties to the number of possible ties in a network
(Brass, 1995).
Because tie strength infuences the number of relationships maintained by a node, tie strength
is somewhat related to network density. Nevertheless, recent research has cautioned against using
tie strength as a proxy for network density (Argote, McEvily, and Reagans, 2003). In their study
of knowledge sharing in a research and development frm, Reagans and McEvily (2003) found
that network density has a distinct infuence on organizational knowledge sharing. Although tie
strength continued to be an important factor in explaining knowledge sharing outcomes, network
density explained outcomes over and above tie strength. Further, when network density was intro-
duced into the analysis, previously signifcant relationships between knowledge codifablity and
tie strength disappeared. They cautioned against interpreting these results as evidence against a
connection between knowledge codifability and tie strength but underscored the importance of
including network density in any analysis of knowledge sharing.
Precisely how density infuences knowledge sharing is a matter of some debate among network
researchers, the origin of which can be traced to the work of two researchers who argued for dif-
ferent effects of density on social network outcomes. Drawing upon his research within local high
schools, Coleman (1988) suggested that dense networks were more benefcial than sparse networks.
Table 4.4
Summary of Characteristics between Weak/Strong Ties
Weak ties Access greater variety of knowledge (Granovetter, 1973)
Less dependent upon/vulnerable to (Uzzi, 1997)
Better for knowledge searching (Hansen, 1999)
Characteristic of electronic communication (Pickering and King, 1995)
Short-term interactions (Carlson and Zmud, 1999)
Low cost to maintain (Hansen, 2002)
Knowledge sharing breadth (Miranda and Saunders, 2003)
Strong ties Cultivates trust (Rice and Aydin, 1991)
More dependent on/vulnerable to (Uzzi, 1997)
Better for knowledge transfer (Hansen, 1999)
Characteristic of face-to-face communication (Pickering and King, 1995)
Developed over time (Carlson and Zmud, 1999)
High cost to maintain (Hansen, 2002)
Cultivates knowledge sharing depth (Miranda and Saunders, 2003)
Students were less likely to drop out of schools in which the individuals maintained a dense network
of relationships. Burt (1992, 1997), in contrast, studied knowledge sharing relationships within a
large corporation and argued that sparser networks were superior because they gave individuals
greater access to more diverse knowledge sources. Individuals in a sparse network are less likely
to be connected to the same set of others, so the knowledge provided by these sources is less likely
to be redundant with knowledge received. He found that the value of a network was contingent
upon the degree to which it connected an individual to unique sources of others; sparser networks
enabled individuals to tap into a greater number of unique sources of knowledge.
Whether high-density or low-density knowledge sharing networks perform better has been and
continues to be a key area of interest for network researchers, primarily because researchers have
continued to fnd evidence to support both sides of the argument (Raider and Krackhardt, 2002).
Ingram and Roberts (2000) explored the effects both of direct tie strength and network density in
relation to strong knowledge sharing ties in the Sydney Hotel industry. They found that strong ties
with competitors improved knowledge fows between the disparate competitors in the hotel industry,
thereby improving the performance of the hotels. The effects of these strong ties were enhanced,
however, when they occurred in the context of a high-density network of other strong ties because
the network of strong ties encouraged trust between competitors by bringing normative expectations
to bear on the various managers. Cummings (2004) found nearly the opposite result in a study of
182 work groups in a Fortune 500 telecommunications frm. He found that low-density networks,
characterized by structural diversity of the group, led to increased value of external knowledge shar-
ing within the group. Structural diversity represented individuals ability to tap into different sources
of knowledge. Groups that were more structurally diverse, tapping into more diverse knowledge,
enjoyed greater performance benefts as the result of knowledge sharing outside the group.
Table 4.5
Network Features: Density
1997 Walker, Kogut, and Shan Closure networks perform better than bridging net-
works in biotech alliances.
2000 Dyer and Nobeoka Closure networks better for exploitation of knowl-
edge, whereas bridging networks better for explora-
tion of knowledge.
2000 Ahuja Infuence of weak ties moderated by number of
strong ties in network.
2001 Regans and Zuckerman Closure networks perform better within teams, bridg-
ing networks perform better across teams.
2002 Raider and Krackhardt Distinction between and characteristics of bridging
versus bonding networks key issue facing network
research.
2004 Hoegl, Weinkauf, and Gemuenden Both weak and strong ties are important, demon-
strate independent infuences.
2004 Cummings Value of external knowledge sharing higher when
groups ties are diverse.
2005 VanAlstyne and Brynjolfsson Electronic communication channels may lead to
higher or lower density networks, dependent on
tendency to interact with similar others.
74 ALAVI AND KANE
Clearly both high- and low-density networks have value, and research has sought to understand
the conditions under which each operates. Researchers have noted that three different network
features may infuence the effects of density on knowledge outcomesenvironmental context,
network context, and knowledge task. First, researchers have suggested that environmental context
in which the network is situated appears to partially explain the effects of density on performance,
noting the difference in Colemans (1988) and Burts (1992) research settings: network density
may behave differently in a high school than in a corporation. Walker, Kogut, and Shan (1997)
examined the effects of interorganizational knowledge sharing on the performance of biotechnology
startups. They found that the dense network advocated by Coleman better explained the long-term
performance of these companies than did sparse networks in this environment that depended so
heavily on trust between frms.
Second, network context in which the interactions occur also appear relevant to understanding
the effects of density. Networks are typically not evenly distributed, exhibiting characteristics of
clustering in which nodes are self-organized into subgroups (often referred to in network terminol-
ogy as cliques), and the nature of this clustering can infuence the impact of density on knowledge
sharing. Reagans and Zuckerman (2001) used the terminology of global and local networks to
talk about network context. Global networks occur across an entire organization (across cliques),
whereas local networks are smaller subnetworks that may occur within an organization (within
cliques). They found that higher density within local networksthose within cliqueswere as-
sociated with higher performance. On the other hand, lower density in global networksthose
occurring across cliques in the organizationwere also associated with improved performance.
Thus, how the density is situated in terms of network context also appears to infuence knowledge
sharing outcomes.
Third, knowledge sharing goals may also be relevant to understanding the infuence of den-
sity. Rowley, Behrens, and Krackhardt (2000) suggested that differences between high- and
low-density networks were dependent on the knowledge sharing goals of the network, defned in
terms of exploration and exploitation. Exploration refers to a networks objective to create new
knowledge through search and experimentation, whereas exploitation refers to the objective of
applying existing knowledge to known problems for improved results (March, 1991). They argued
that sparse networks were better for knowledge exploration, while dense networks were better for
knowledge exploitation.
Density in Communication Support CapabilitiesSupported Social Networks
Research into knowledge sharing explored not only how density of a social network infuences
knowledge sharing effectiveness but also how network density infuences CSC-supported networks.
VanAlstyne and Brynjolfsson (1996, 2005) theorize about the likely social structures that would
develop as a result of knowledge sharing via CSC. They noted that, although researchers often
touted the theoretical ability of technologically enabled capabilities to broaden the knowledge
sharing network to virtually anyone in the world using CSC, personal search patterns employed
by individuals could functionally shrink the knowledge sharing networka process they called
cyberbalkanization. Although individuals may be able to share knowledge with a greater number
of others via CSC, the process of selecting which knowledge to search for and which others to
share knowledge with may actually reduce the diversity of knowledge an individual can access.
Thus, CSC may either lead to increased or decreased density in networks, and researchers
have long noted these potentially dialectical effects of IS in organizations (Robey and Boudreau,
1999). Constant, Sproull, and Keisler (1996) found evidence for similar dialectical effects in an
empirical setting. In examining the perceived value of knowledge obtained via CSC, they found
that the value of knowledge received was not related to the number of responses that an individual
received to questions posed but it was related to a greater diversity of responses. This suggests
that, like in face-to-face situations, the density of ones knowledge sharing enabled by CSC con-
tributes to its overall value.
Butler (2001) pursued similar issues regarding the nature of network density in CSC. In an
examination of knowledge sharing occurring via a listserv, he examined the effects of network
density on knowledge sharing in a virtual environment. Butler concluded that the density of CSC-
based knowledge networks had an inverted U-shaped relationship between density and outcomes.
Networks that were too sparse did not hold value for its members and these networks soon ceased
to exist. On the other hand, networks that were too dense were too diffcult to participate in and
extract needed knowledge from, and these networks soon experienced attrition as well.
The research on network density and knowledge sharing complements and extends the earlier
discussions on tie strength. Both high- and low-density networks may be valuable for knowledge
sharing, but their performance largely depends on the environmental context, the network context,
and knowledge tasks in which they operate. Further, the value of a knowledge network depends
on the number of different sources of knowledge it enables the individual to tap into, and CSC
are most valuable to the degree that they can allow individuals to tap into different sources of
knowledge.
Centrality
Knowledge sharing researchers have also suggested that the position of an individual within a
network is also important to knowledge sharing effectiveness (Table 4.6). Usually, network posi-
tion is described in one of two wayscentral or peripheral. Although social network analysis
has developed a number of different ways to defne centrality within a networkbetweenness
centrality, closeness centrality, degree centrality, and eigenvector centralitythe general construct
of centrality seeks to capture the same characteristic: whether an individual occupies a central
position within the relational network (Brass 1995).
Network researchers have suggested that centrality can infuence outcomes at both the indi-
vidual and at the whole network level of analysis. The effects of centrality have been most heavily
researched at the individual level. Burt (1992) noted that individuals in central positions within the
network possessed a number of knowledge sharing benefts in terms of timing, access, and referral
due to their role as knowledge broker within the network. Timing suggests that position allows
central individuals to receive knowledge before others since they are more likely to be privy to
the knowledge fows through the network. Access notes that central individuals have connections
to a greater number of individuals, both directly and indirectly, allowing the central individual to
tap into the knowledge in the network more readily than others. Referral describes the fact that
others in the network are more likely to know what knowledge interests the central individual
and they are more likely to pass along relevant knowledge to the central individual without that
central individual having to request it.
Although centrality has been shown to be an important variable, the performance benefts of
centrality can be moderated by other characteristics of the individual node. Mehra, Kilduff, and
Brass (2001) found that high self-monitors (people prone to the active construction of public
personas for social ends) were more likely to be aware of their position as the central node in the
network and leverage that position for benefts than were low self-monitors. Tsai (2001) found
that business units that were more central in the knowledge sharing network enjoyed greater
76 ALAVI AND KANE
performance as a result of their position but that the benefts of the units central position were
moderated by that units absorptive capacityits ability to assimilate new knowledge. Thus,
although centrality does beneft individual nodes, certain nodes are better served than others by
being in central positions.
While centrality is typically used to describe individual-level outcomes (Brass 1995), centrality
has also been used as a metric to describe outcomes at the whole network level of analysis. First,
the central node in a network is often regarded as the networks knowledge broker (Cross and
Prusak, 2002). Controlling central positions in the network, these nodes are frequently responsible
for conveying knowledge from one side of the network to the other. Thus, the effectiveness with
which the central node can broker knowledge across the network infuences the performance of the
network as a whole. Rulke and Galaskiewicz (2000) found empirical evidence for this argument in
an experimental setting. Controlling whether individuals in a network generalized or specialized
knowledge, they found that groups where the central individuals in the network had generalized
knowledge regarding the task at hand performed better than did groups where central individuals
had specialized knowledge. These results suggest that the whole network performed better when
the central nodes were better able to serve the role of knowledge broker. These network-level
effects obviously depend somewhat on the overall network structure and the degree to which the
network depends on the knowledge broker.
Second, a unique characteristic of knowledge networks is that centrality may not primarily be
refective of characteristics of the central node itself, but it rather captures the degree to which the
rest of the network values the knowledge held by those nodes. Perry-Smith and Shalley (2003)
argued that, as a result of a nodes decentralized position, peripheral nodes often have qualitatively
different knowledge than the rest of the network. If others value this different knowledge in the
network, then others in the network will seek to access that knowledge more frequently. In this
case, rather than an individual occupying a central position within a network, the network forms
Table 4.6
Network Features: Centrality
1992 Burt Central individuals enjoy information benefts in terms of
timing, access, and referral.
2000 Rulke and Galaskiewicz Teams with generalist knowledge perform better in
centralized networks.
2001 Mehra, Kilduff, and Brass Centrality predicted performance for high self-monitoring
individuals.
2001 Tsai Central units perform better than peripheral units,
moderated by absorptive capacity.
2003 Perry-Smith and Shalley Peripheral individuals more likely source of creativity,
centrality a result of network confguring around those with
valuable knowledge.
2003 Ahuja, Galetta, and Carley Centrality strong predictor of performance in virtual teams.
2003 Thomas-Hunt, Ogden, and Neale Central individuals performed differently than peripheral
individuals in knowledge sharing and value of knowledge
shared.
2004 Owen-Smith and Powell Benefts conferred by central position depends on
characteristics of network.
around the individual with the valued knowledge. Thus, a nodes central position is, at least in
part, a function of the value the network places on the knowledge held by that node, and one can
expect the performance of a whole network will depend somewhat on which knowledge it values
most heavily in accomplishing its task.
Centrality in Communication Support CapabilitiesSupported Social Networks
Early theoretical research suggested that knowledge sharing via CSC would lead to decentralized
and nonhierarchical knowledge sharing. Ahuja and Carley (1999) examined the CSC knowledge
sharing patterns of a virtual group in order to assess the validity of these theoretical assumptions.
What they found was that these so-called virtual organizations do demonstrate certain elements
of hierarchy and centralization that are task-based rather than authority-based, as in traditional
organizations. Furthermore, they found that ones position within this knowledge sharing network
led to perceptions of improved individual performance but not necessarily to objective measures
of performance. Ahuja and Carley (1999) were among the frst to identify complex social struc-
tures within knowledge sharing relationships via CSC and to suggest that the structure of these
relationships led to similar performance improvements as they did in face-to-face knowledge
sharing networks.
In a follow-up study to this earlier work, Ahuja, Galletta, and Carley (2003) analyzed the
knowledge sharing patterns in a CSC network. In this study, they examined the effects of individual
characteristicssuch as organizational role, status, and communication roleof individuals in
the knowledge sharing network, in relation to their position within the network. They found that
not only did centrality within the CSC knowledge sharing network mediate the individual-level
effects of the individual characteristics on performance but also network position explained more
of the variance in performance than any of the individual characteristics. Centrality mediated the
effects of individual characteristics on knowledge network outcomes.
Wasko and Faraj (2005) synthesized a number of these fndings into a theoretical model of
knowledge exchange in CSC-mediated knowledge networks. They suggest that both the structure
of the network and the strength of ties within the network will predict the value of knowledge
sharing relationships in a network. Further, they found empirical evidence that ones centrality
in the network infuenced the degree to which individuals were willing to share knowledge in a
CSC-based knowledge sharing network. Just as in the fndings on tie strength and density, cen-
trality seems to have many of the same effects in CSC networks as in interpersonal knowledge
sharing networks.
Other Network Features
Network researchers have investigated the role of additional features of the knowledge network
that infuence knowledge network outcomes, primarily trust and shared background knowledge.
Levin and Cross (2004) highlighted the benefts of trust on knowledge exchange. They found that
two types of trustbenevolence-based and competence-based trustmoderated the effects of
knowledge transfer in strong tie relationships. Further, they found that while holding the levels of
trust constant in their analysis, the benefts of weak ties as sources of nonredundant knowledge
became more apparent. Clearly, the level of trust present among actors in the network infuences
both the likelihood and willingness of individuals to share knowledge with one another, and other
research into knowledge sharing has highlighted the infuence of trust on knowledge sharing (Lee
and Choi, 2003).
78 ALAVI AND KANE
Trust has also emerged as an important factor for knowledge sharing in CSC networks. The
ability to trust the CSC for knowledge sharing is an important factor in deciding whether to
share knowledge via CSC. CSC-based knowledge sharing introduces additional possibilities of
knowledge sharing failure, which may be critical in certain settings. Drozdova (2002) examined
the knowledge sharing patterns within terrorist networks, noting that these networks rarely made
use of CSC technology for knowledge sharing. The primary reason she found for this tendency
involved the networks inability to trust CSC as a knowledge sharing mechanism. CSC provide
for effective and effcient knowledge sharing between individuals, but communication using these
technologies runs a certain risk of either not being delivered to the appropriate recipient in a timely
manner (i.e., power outages, misspelled e-mail addresses, etc.) or being intercepted by an outside
party. Drozdova notes that knowledge sharing networks that cannot accept the risks inherent to
CSC are not likely to use these means of knowledge sharing. The knowledge network must be
able to trust that the CSC will function reliably for their purposes.
In addition to trust, researchers have highlighted the need for shared background knowledge
(i.e., knowledge held in common) in order to facilitate knowledge sharing in a social network.
At one level, this shared knowledge may be an awareness of who knows what in a network, a
situation also known as transactive memory (Wegner, 1987). To effectively share knowledge with
one another, individuals must know whom to turn to when specifc knowledge is needed. Shared
background knowledge often enables individuals to share knowledge between one another more
quickly and easily than if there is little background knowledge (Borgatti and Cross, 2003). Both
transactive memory (knowing who knows what) and background knowledge may be understood
together to represent shared knowledge between individuals in the knowledge network.
The inherent leanness of the CSC-based interactions compared to face-to-face interaction
introduces problems in groups that have not established the mutual knowledge and common lan-
guages required for effective knowledge sharing. Cramton (2001) identifed that groups engaging
in knowledge sharing exclusively via CSC encountered several problems: they often failed to retain
contextual information, possessed unevenly distributed knowledge, had diffculty communicating
and understanding the salience of information, had differing access to information, and had trouble
interpreting the meaning of silence. Shared background knowledge is both extremely critical and
diffcult to maintain in CSC-based knowledge sharing. Thus, although CSC offers a number of
advantages as a knowledge sharing medium, it can also introduce a number of challenges. The
establishment of a shared cognitive context is a key challenge and critical feature of knowledge
sharing effectiveness via CSC.
FUTURE RESEARCH
Despite this rich stream of research involving knowledge sharing via CSC, signifcant opportunities
still exist for integrating the IT artifact into knowledge sharing research from the social network
perspective. Knowledge management researchers have noted that IT can introduce two distinct
types of knowledge sharing capabilities when introduced into an organizationCSC and infor-
mation management capabilities (Goodman and Darr, 1998; Robey, Boudreau, and Rose, 2000).
Whereas the purpose of CSC is to aid the interpersonal transmission of knowledge between two
individuals, the purpose of information management capabilities is for the IT artifact to serve as
an entity to store and organize knowledge (Ackerman, 1998).
We have shown that researchers have given considerable attention to the role of CSC in knowl-
edge sharing networks and that these studies have yielded promising results. At the same time, the
role of information management capabilities (IMC) has gone largely unaddressed. This distinction
between CSC and IMC in knowledge sharing networks is particularly important, because the dif-
ferent roles the IT artifact plays means that CSC and IMC are integrated differently into social
networks. By enabling electronic communication channels between individuals, CSC primarily
serve as ties within the network. IMC, on the other hand, serve as an independent source of codi-
fed knowledge and serve as a node within the network.
This distinction between IS as nodes versus IS as ties in a network is partially related to the
personalization versus codifcation debate in the knowledge management literature (Hansen,
Nohria, and Tierney, 1999). When IS is viewed as a tie in a network, it favors a knowledge per-
sonalization approach in which individuals use IS primarily to assemble a personal network of
other individuals from which to draw knowledge. When IS is viewed as a node in the network, it
favors a knowledge codifcation approach, in which individuals are interacting with knowledge
repositories in conjunction with a network of other people. Thus, social network research can be
used to examine knowledge codifcation and knowledge personalization, depending on how IS is
integrated into the social network.
Because IS can be integrated into social networks in these two very different ways, this inte-
gration of IMC as nodes in the social network represents one next step in the evolution of social
network research on knowledge sharing and also represents an exciting frontier for IS research.
Information systems researchers have typically addressed the human-IS relationship in terms of
the user-system dyad, where researchers isolate particular characteristics of the interaction between
a single user and a single system (Lamb and Kling, 2003). Although attention to the user-system
dyad has been invaluable to the cumulative IS research tradition, focused attention on the isolated
user-system relationship may obscure the critical infuences that wider and more complex interac-
tions may have on understanding the impact of IMC on knowledge sharing. Integrating IMC as
a node in the knowledge sharing network gives IS researchers the opportunity to examine these
complex interactions between people and IMC for knowledge sharing tasks that may prove critical
in understanding the impact of IS on knowledge outcomes (Dourish, 2001).
First, individuals do not interact with IMC independently of one another, but their social
interactions signifcantly infuence the role the information system plays in knowledge sharing
(Hutchins, 1995). Interpersonal relationships infuence whether individuals are willing to adopt
a new information system (Abrahamson and Rosenkopf, 1997; Burkhardt, 1994; Burkhardt and
Brass, 1990), and interpersonal relationships also infuence an individuals ongoing and every-
day interactions with IMC for knowledge sharing (Lamb and Kling, 2003). Further, the role of
interpersonal relationships is particularly critical for understanding the role of IMC in knowledge
sharing because these relationships may directly affect the quality of knowledge available to the
system. Individuals in an organization generate new knowledge through the process of combining
knowledge, exchanging knowledge, and socializing with one another (Carlile, 2002; Kogut and
Zander, 1992, 1996). Effective interpersonal knowledge sharing relationships means that more
valuable knowledge is generated and available for storage in and access through an information
system, assuming that knowledge is able to be codifed. Thus, the impact of an information system
on knowledge sharing is critically dependent on the interpersonal knowledge sharing relationships
that surround it (Kumar, van Dissel, and Bielli, 1998; Zack and McKenney, 1995).
Second, knowledge sharing in contemporary organizations does not simply involve whether
and how an individual chooses to interact with one information system, but it also involves how
people manage a collection of multiple IMC to accomplish knowledge sharing tasks (Lamb and
Kling, 2003). As computers have become more ubiquitous in recent years, individuals have access
to multiple IMC with which to share knowledge (Vertegaal, 2003). Interacting with multiple IMC
is a nontrivial matter, because these systems are embedded with distinct social assumptions by
80 ALAVI AND KANE
their designers (Desanctis and Poole, 1994; Schultze and Orlikowski, 2004). These assumptions
may complement or confict with one another, making it more or less diffcult for individuals to
use these systems together. Even simple differences such as password schema can signifcantly
infuence the ease with which systems can be used in conjunction with one another.
Third, IMC do not support knowledge sharing only through an individuals direct interaction
with the systems, but they can also beneft knowledge sharing through their indirect infuence as
well. Whether due to specialist knowledge, personal ability, or available time, people often interact
with IMC on others behalf. One person may access knowledge held by a particular information
system and then share it with others via interpersonal or electronic channels, potentially adding
value to the knowledge in its transmission (Carlile, 2002; Majchrzak, Cooper, and Neece, 2004).
Conversely, individuals may also contribute knowledge to IMC indirectly through others. A fre-
quent practice in many organizations is for knowledge to be vetted by subject matter experts before
uploading the knowledge into the relevant system, and people submit knowledge to appropriate
parties who then contribute that knowledge to the systems (Davenport and Glaser, 2002; Markus,
2001). In both of these instances, IMC support knowledge sharing of people in the organization,
even when the people who beneft from or contribute to that knowledge most do not directly in-
teract with those systems. The impact of IS on knowledge sharing can be a function of both the
direct and indirect interactions with those systems.
Nahapiet and Ghoshal (1998) suggest that three dimensions of the social network are important
for understanding knowledge sharing outcomesthe structural, relational, and cognitive dimen-
sions. These dimensions have rough analogues in the IS literature, suggesting a fruitful direction
for future research. The structural dimension focuses on the position of nodes in relation to one
another. The structural dimension of networks could examine how individuals interact with
knowledge repositories directly, indirectly through others, and in conjunction with other members
of the network. The relational dimension focuses on the affective relationships between nodes.
How individuals in a network feel about the ISboth positively and negativelymay infuence
their willingness and effectiveness in both contributing knowledge to and drawing knowledge
from repositories. The cognitive dimension addresses the shared knowledge or shared language
between nodes. Individuals in a network may better understand how to interact with knowledge
repositories than others in a network, which may infuence the ease with which they share knowl-
edge via repositories. Each of these network dimensions offers promising avenues of knowledge
sharing research from a network perspective.
Using IS to support knowledge sharing in contemporary organizations, therefore, is more
complex than the user-system dyad alone. It also involves interactions between multiple people,
with multiple IMC, and in both direct and indirect ways. Thus, the role of IS in knowledge sharing
cannot be examined solely in isolation from these wider relationships but should also be examined
as a part of a wider network of people and IMC. IS research needs a lens through which to begin
exploring this network of complex human-IS interactions to more fully assess the role of IS in
knowledge sharing. In addition, IS research requires a social network perspective that integrates
IMC as a node into the network. This approach would allow IS researchers to examine the wider
impacts of IS in organizations and better understand how complex interactions infuence the
impact of IS in organizations.
CONCLUSION
In this paper, we have traced the evolution of knowledge sharing research in the social network
tradition. First, we identifed the foundations of social network researchthe key assumptions
and features that characterize social network research approaches to knowledge sharing. Second,
we analyzed key features of social networks that have proved important for both face-to-face and
CSC-mediated knowledge sharing relationships. Finally, we addressed future research directions
regarding social networks in IS research. We argued that the continued evolutions of knowledge
sharing research in the social network tradition will likely lead to the incorporation of IT-enabled
information management capabilities as distinct nodes in knowledge sharing networks. Because this
frontier provides a rich lens to examine how individuals actually interact with IMC for knowledge
sharing, it is likely to be a valuable enterprise for both social network and IS literature.
REFERENCES
Abrahamson, E., and Rosenkopf, L. 1997. Social network effects on the extent of innovation diffusion: A
computer simulation. Organization Science, 8, 3, 289309.
Ackerman, M. 1998. Augmenting organizational memory: A feld study of answer garden. ACM Transactions
on Information Systems, 16, 3, 203224.
Ahuja, G. 2000. Collaboration networks, structural holes, and innovation: A longitudinal study. Administra-
tive Science Quarterly, 45, 3, 425455.
Ahuja, M.K., and Carley, K.M. 1999. Network structure in virtual organizations. Organization Science, 10,
6, 741757.
Ahuja, M.K.; Galletta, D.F.; and Carley, K.M. 2003. Individual centrality and performance in virtual R&D
groups: An empirical study. Management Science, 49, 1, 2138.
Alavi, M., and Leidner, D.E. 2001. Review: Knowledge management and knowledge management systems:
Conceptual foundations and research issues. MIS Quarterly, 25, 1, 107136.
Argote, L.; McEvily, B.; and Reagans, R. 2003. Managing knowledge in organizations: An integrative
framework and review of emerging themes. Management Science, 49, 4, 571582.
Baker, W.E., and Faulkner, R.R. 2002. Interorganizational networks. In J.A.C. Baum (ed.), Blackwell Com-
panion to Organizations. Oxford: Blackwell Publishers, pp. 520540.
Barley, S.R. 1990. The alignment of technology and structure through roles and networks. Administrative
Science Quarterly, 35, 1, 61103.
Borgatti, S., and Foster, P.C. 2003. The network paradigm in organizational research: A review and typology.
Journal of Management, 29, 6, 9911013.
Borgatti, S.P., and Cross, R. 2003. A relational view of information seeking and learning in social networks.
Management Science, 49, 4, 432445.
Brass, D.J. 1995. A social network perspective on human resources management. Research in Personnel and
Human Resources Management, 13, 3979.
Burkhardt, M.E. 1994. Social-interaction effects following a technological-changeA longitudinal investi-
gation. Academy of Management Journal, 37, 4, 869898.
Burkhardt, M.E., and Brass, D.J. 1990. Changing patterns or patterns of changeThe effects of a change in
technology on social network structure and power. Administrative Science Quarterly, 35, 1, 104127.
Burt, R.S. 1992. Structural Holes: The Social Structure of Competition. Cambridge, MA: Harvard University
Press.
. 1997. The contingent value of social capital. Administrative Science Quarterly, 42, 2, 339365.
Butler, B.S. 2001. Membership size, communication activity, and sustainability: A resource-based model of
online social structures. Information Systems Research, 12, 4, 346362.
Callon, M. 1998. The Laws of the Markets. Oxford/Malden, MA: Blackwell Publishers/Sociological Re-
view.
Callon, M.; Law, J.; and Rip, A. 1986. Mapping the Dynamics of Science and Technology: Sociology of Sci-
ence in the Real World. Basingstoke, UK: Macmillan.
Carlile, P.R. 2002. A pragmatic view of knowledge and boundaries: Boundary objects in new product devel-
opment. Organization Science, 13, 4, 442455.
Carlson, J.R., and Zmud, R.W. 1999. Channel expansion theory and the experiential nature of media richness
perceptions. Academy of Management Journal, 42, 2, 153170.
Chan, K., and Liebowitz, J. 2006. The synergy of social network analysis and knowledge mapping: A case
study. International Journal of Management and Decision Making, 7, 1, 1935.
82 ALAVI AND KANE
Coleman, J.S. 1988. Social capital in the creation of human capital. American Journal of Sociology, 94,
95120.
Collins, C.J., and Clark, K.D. 2003. Strategic human resource practices, top management team social net-
works, and frm performance: The role of human resource practices in creating organizational competitive
advantage. Academy of Management Journal, 46, 6, 740751.
Constant, D.; Sproull, L.; and Kiesler, S. 1996. The kindness of strangers: The usefulness of electronic weak
ties for technical advice. Organization Science, 7, 2, 119135.
Cramton, C.D. 2001. The mutual knowledge problem and its consequences for dispersed collaboration.
Cross, R.; Borgatti, S.P.; and Parker, A. 2002. Making invisible work visible: Using social network analysis
to support strategic collaboration. California Management Review, 44, 2, 2535.
Cross, R., and Parker, A. 2004. The Hidden Power of Social Networks: Understanding How Work Really
Gets Done in Organizations. Boston: Harvard Business School Press.
Cross, R., and Prusak, L. 2002. The people who make organizations goor stop. Harvard Business Review,
80, 6, 104112.
Cummings, J.N. 2004. Work groups, structural diversity, and knowledge sharing in a global organization.
Daft, R.L.; Lengel, R.H.; and Trevino, L.K. 1987. Message equivocality, media selection, and manager
performance: Implications for information systems. MIS Quarterly, 1, 3, 354366.
Davenport, T.H., and Glaser, J. 2002. Just-in-time delivery comes to knowledge management. Harvard
Business Review, 80, 7, 107114.
DeSanctis, G., and Poole, M.S. 1994. Capturing the complexity in advanced technology useAdaptive
structuration theory. Organization Science, 5, 2, 121147.
Dourish, P. 2001. Where the Action Is: The Foundations of Embodied Interaction. Cambridge, MA: MIT
Press.
Drozdova, E. 2002. Dealing with low-tech terrorist communications in the hi-tech age: Toward a theory of
fault-intolerant network organizations. International Conference on Information Systems, Barcelona,
Spain.
Dyer, J.H., and Nobeoka, K. 2000. Creating and managing a high-performance knowledge-sharing network:
The Toyota case. Strategic Management Journal, 21, 3, 345367.
Systems, 18, 1, 215233.
Gergen, K.J. 1991. The Saturated Self: Dilemmas of Identity in Contemporary Life. New York: Basic
Books.
Goodhue, D.L., and Thompson, R.L. 1995. Task-technology ft and individual performance. MIS Quarterly,
19, 2, 213236.
Goodman, P.S., and Darr, E.D. 1998. Computer-aided systems and communities: Mechanisms for organiza-
tional learning in distributed environments. MIS Quarterly, 22, 4, 417440.
Granovetter, M. 1973. The strength of weak ties. American Journal of Sociology, 78, 6, 13601380.
. 1985. Economic action and social structure: The problem of embeddedness. American Journal of
Sociology, 91, 3, 481510.
Gray, P.H. 2002. Knowledge sourcing effectiveness. Unpublished doctoral dissertation, Queens University,
Kingston, Ontario.
Gulati, R.; Dialdin, D.A.; and Wang, L. 2002. Organizational networks. In J.A.C. Baum (ed.), The Blackwell
Companion to Organizations. Oxford, UK: Blackwell Publishers, pp. 281303.
Hansen, M.T. 1999. The search-transfer problem: The role of weak ties in sharing knowledge across orga-
nization subunits. Administrative Science Quarterly, 44, 1, 82111.
. 2002. Knowledge networks: Explaining effective knowledge sharing in multiunit companies. Or-
ganization Science, 13, 3, 232248.
Hansen, M.T.; Nohria, N.; and Tierney, T. 1999. Whats your strategy for managing knowledge? Harvard
Hoegl, M.; Parboteeah, K.P.; and Munson, C.L. 2003. Team-level antecedents of individuals knowledge
networks. Decision Sciences, 34, 4, 741770.
Hoegl, M.; Weinkauf, K.; and Gemuenden, H.G. 2004. Interteam coordination, project commitment, and
teamwork in multiteam R&D projects: A longitudinal study. Organization Science, 15, 1, 3855.
Hutchins, E. 1995. Cognition in the Wild. Cambridge, MA: MIT Press.
Ibarra, H. 1993. Network centrality, power, and innovation involvementDeterminants of technical and
administrative roles. Academy of Management Journal, 36, 3, 471501.
Ingram, P., and Roberts, P.W. 2000. Friendships among competitors in the Sydney hotel industry. American
Journal of Sociology, 106, 2, 387407.
Kogut, B. 2000. The network as knowledge: Generative rules and the emergence of structure. Strategic
Management Journal, 21, 3, 405425.
Kogut, B., and Zander, U. 1992. Knowledge of the frm, combinative capabilities, and the replication of
technology. Organization Science, 3, 383397.
. 1996. What frms do? Coordination, identity, and learning. Organization Science, 7, 5, 502518.
Krackhardt, D. 1992. The strength of strong ties: The importance of Philios in organizations. In N. Nohria
and R.G. Eccles (eds.), Networks and Organizations. Cambridge, MA: Harvard Business School Press,
pp. 216239.
Krackhardt, D., and Hanson, J.R. 1993. Informal networksThe company behind the chart. Harvard Busi-
ness Review, 71, 4, 104111.
Kumar, K.; van Dissel, H.G.; and Bielli, P. 1998. The merchant of PratoRevisited: Toward a third rationality
of information systems. MIS Quarterly, 22, 2, 199226.
Lamb, R., and Kling, R. 2003. Reconceptualizing users as social actors in information systems research.
MIS Quarterly, 27, 2, 197235.
Latour, B. 1987. Science in Action: How to Follow Scientists and Engineers Through Society. Cambridge,
MA: Harvard University Press.
. 1996. Aramis, or, The Love of Technology. Cambridge, MA: Harvard University Press.
Lee, H., and Choi, B. 2003. Knowledge management enablers, processes, and organizational performance:
An integrative view and empirical examination. Journal of Management Information Systems, 20, 1,
179228.
Levin, D.Z., and Cross, R. 2004. The strength of weak ties you can trust: The mediating role of trust in ef-
fective knowledge transfer. Management Science, 50, 11, 14771490.
Majchrzak, A.; Cooper, L.P.; and Neece, O.E. 2004. Knowledge reuse for innovation. Management Science,
50, 2, 174188.
March, J.G. 1991. Exploration and exploitation in organizational learning. Organization Science, 2, 1,
7187.
Mardsen, P.V., and Campbell, K.E. 1984. Measuring tie strength. Social Forces, 63, 482501.
Markus, M.L. 1994. Electronic mail as the medium of managerial choice. Organization Science, 5, 4,
502527.
. 2001. Toward a theory of knowledge reuse: Types of knowledge reuse situations and factors in reuse
success. Journal of Management Information Systems, 18, 1, 5793.
McEvily, B., and Zaheer, A. 1999. Bridging ties: A source of frm heterogeneity in competitive capabilities.
Strategic Management Journal, 20, 12, 11331156.
McKenney, J.; Zack, M.; and Doherty, V.S. 1992. Complementary communication media: A comparison of
electronic mail and face-to-face communication in a programming team. In N. Nohria and R.G. Eccles
(eds.), Networks and Organizations. Cambridge, MA: Harvard Business School Press.
Mehra, A.; Kilduff, M.; and Brass, D.J. 2001. The social networks of high and low self-monitors: Implica-
tions for workplace performance. Administrative Science Quarterly, 46, 1, 121146.
Miranda, S.M., and Saunders, C.S. 2003. The social construction of meaning: An alternative perspective on
information sharing. Information Systems Research, 14, 1, 87106.
Monge, P.R., and Contractor, N.S. 2000. Emergence of communication networks. In F.M. Jablin and L.L.
Putnam (eds.), The New Handbook of Organizational Communication: Advances in Theory, Research,
and Methods. Thousand Oaks, CA: Sage Publications, pp. 440502.
Nahapiet, J., and Ghoshal, S. 1998. Social capital, intellectual capital, and the organizational advantage.
Nidumolu, S.R.; Subramani, M.; and Aldrich, A. 2001. Situated learning and the situated knowledge web:
Exploring the ground beneath knowledge management. Journal of Management Information Systems,
18, 1, 115150.
Nohria, N., and Eccles, R.G. 1992. Face-to-face: Making network organizations work. In N. Nohria and R.G.
Eccles (eds.), Networks and Organizations. Cambridge, MA: Harvard Business School Press.
Owen-Smith, J., and Powell, W.W. 2004. Knowledge networks as channels and conduits: The effects of
spillovers in the Boston biotechnology community. Organization Science, 15, 1, 521.
84 ALAVI AND KANE
Papa, M.J. 1990. Communication network patterns and employee performance with new technology. Com-
munication Research, 17, 3, 344368.
Perry-Smith, J.E., and Shalley, C.E. 2003. The social side of creativity: A static and dynamic social network
perspective. Academy of Management Review, 28, 1, 89106.
Pickering, J.M., and King, J.L. 1995. Hardwiring weak tiesInterorganizational computer-mediated commu-
nication, occupational communities, and organizational change. Organization Science, 6, 4, 479486.
Podolny, J.M., and Baron, J.N. 1997. Resources and relationships: Social networks and mobility in the
workplace. American Sociological Review, 62, 5, 673693.
Raider, H., and Krackhardt, D.J. 2002. Intraorganizational networks. In J.A.C. Baum (ed.), The Blackwell
Companion to Organizations. Oxford: Blackwell Publishers, pp. 5874.
Reagans, R., and McEvily, B. 2003. Network structure and knowledge transfer: The effects of cohesion and
range. Administrative Science Quarterly, 48, 2, 240267.
Reagans, R., and Zuckerman, E.W. 2001. Networks, diversity, and productivity: The social capital of corporate
R&D teams. Organization Science, 12, 4, 502517.
Rice, R.E., and Aydin, C. 1991. Attitudes toward new organizational technologyNetwork proximity as a
mechanism for social information-processing. Administrative Science Quarterly, 36, 2, 219244.
Robey, D., and Boudreau, M.C. 1999. Accounting for the contradictory organizational consequences of
information technology: Theoretical directions and methodological implications. Information Systems
Research, 10, 2, 167185.
Robey, D.; Boudreau, M.-C.; and Rose, G.M. 2000. Information technology and organizational learning:
A review and assessment of research. Accounting, Management and Information Technologies, 10,
125155.
Rowley, T.; Behrens, D.; and Krackhardt, D. 2000. Redundant governance structures: An analysis of structural
and relational embeddedness in the steel and semiconductor industries. Strategic Management Journal,
21, 3, 369386.
Rubenstein-Montano, B.; Liebowitz, J.; Buchwalter, J.; McCaw, D.; Newman, B.; and Rebeck, K. 2001. A
systems thinking framework for knowledge management. Decision Support Systems, 31, 1, 516.
Rulke, D.L., and Galaskiewicz, J. 2000. Distribution of knowledge, group network structure, and group
performance. Management Science, 46, 5, 612625.
Schultze, U., and Orlikowski, W.J. 2004. A practice perspective on technology-mediated network relations:
The use of Internet-based self-serve technologies. Information Systems Research, 51, 1, 87106.
Scott, J. 2000. Social Network Analysis: A Handbook, 2nd ed. Newbury Park, CA: Sage Publications.
Sparrowe, R.T.; Liden, R.C.; Wayne, S.J.; and Kraimer, M.L. 2001. Social networks and the performance of
individuals and groups. Academy of Management Journal, 44, 2, 316325.
Spender, J.C. 1996. Making knowledge the basis of a dynamic theory of the frm. Strategic Management
Journal, 17, 4562.
Spiegler, I. 2003. Technology and knowledge: bridging a generating gap. Information & Management,
40, 6, 533539.
Stevenson, W.B., and Gilly, M.C. 1991. Information-processing and problem-solvingThe migration of prob-
lems through formal positions and networks of ties. Academy of Management Journal, 34, 4, 918928.
Thomas-Hunt, M.C.; Ogden, T.Y.; and Neale, M.A. 2003. Whos really sharing? Effects of social and expert
status on knowledge exchange within groups. Management Science, 49, 4, 464477.
Tsai, W.P. 2001. Knowledge transfer in intraorganizational networks: Effects of network position and absorptive
capacity on business unit innovation and performance. Academy of Management Journal, 44, 5, 9961004.
Tuomi, I. 1999. Data is more than knowledge: Implications of the reversed knowledge hierarchy for knowl-
edge management and organizational memory. Journal of Management Information Systems, 16, 3,
103117.
Uzzi, B. 1997. Social structure and competition in interfrm networks: The paradox of embeddedness. Ad-
ministrative Science Quarterly, 42, 1, 3567.
Van Alstyne, M., and Brynjolfsson, E. 1996. Electronic communities: Global village or cyberbalkanization?
International Conference on Information Systems, Cleveland, OH, 1996.
. 2005. Global village or cyber-balkans? Modeling and measuring the integration of electronic com-
munities, Management Science, 51, 6, 851868.
Vertegaal, R. 2003. Attentive user interfaces. Communications of the ACM, 46, 3, 3033.
Walker, G.; Kogut, B.; and Shan, W.J. 1997. Social capital, structural holes and the formation of an industry
network. Organization Science, 8, 2, 109125.
Wasko, M.M., and Faraj, S. 2005. Why should I share? Examining social capital and knowledge contribution
in electronic networks of practice. MIS Quarterly, 29, 1, 3557.
Wasserman, S., and Faust, K. 1994. Social Network Analysis: Methods and Applications. New York: Cam-
bridge University Press.
Watts, D.J. 1999. Networks, dynamics, and the small-world phenomenon. American Journal of Sociology,
105, 2, 493.
Wegner, D.M. 1987. Transactive memory: A contemporary analysis of the group mind, In B. Mullen and
G.G. Goethals (eds.), Theories of Group Behavior. New York: Springer-Verlag, pp. 185208.
Wellman, B. 2001. Computer networks as social networks. Science, 293, 20312034.
Zack, M.H., and McKenney, J.L. 1995. Social-context and interaction in ongoing computer-supported man-
agement groups. Organization Science, 6, 4, 394422.
86
ChaptEr 5
THE EVOLUTION OF KNOWLEDGE
MANAGEMENT TECHNOLOGY
From Explicit Rules to Implicit Profles
UlriKE sChUltZE
Abstract: This chapter explores knowledge managements evolution, particularly the evolution of
its assumptions about knowledge and knowledge work, by studying a specifc aspect of the knowl-
edge management movementnamely, the technologies designed to solve knowledge problems.
In analyzing one knowledge management system from each decade starting in the 1970s, this
chapter provides support for the contention that the concept of knowledge management made the
role of knowledge, which was typically considered a natural and implicit aspect of organizing,
an explicit asset that needed to be managed as a separate organizational activity. However, over
time, knowledge management was expected to recede into the background and once again become
an implicit, natural aspect of organizing.
Keywords: Organizing Vision, Expert System, Community of Practice, Groupware, Intelligent
Agent
INTRODUCTION
Research into the origins of knowledge management highlights that knowledge and efforts to man-
age it are not recent phenomena (e.g., Grant, 2000; Wiig, 2000). The desire to create new knowledge
and to transfer and preserve it has been central to human development throughout civilization. Even
in the business context, the problem of managing knowledge has motivated the development of
organizational practices such as scientifc management (circa 1916) and continuous explorations
with different forms of organizing (e.g., matrix organizations). However, until the early 1990s,
the role of knowledge in most of these organizational efforts remained largely implicit. In other
words, knowledge was treated as an inherent, natural, and necessary aspect of organizational work.
It was not considered or managed separately from organizing and managing.
The use of knowledge management as a term, a consulting practice (Prusak, 2001), and an organiz-
ing vision (Swanson and Ramiller, 1997) articulated and made explicit what role knowledge plays in
organizations (Wiig, 2000). By expressing and explaining in clear language the importance of knowledge
as an object to be managed and knowledge management as a specifc activity, the knowledge manage-
ment movement has had a signifcant effect on organizations. For instance, it has led to the introduction
of knowledge offcers (e.g., CKO), the identifcation of knowledge processes (e.g., expertise location),
and the installation of new information technologies (e.g., groupware). However, according to Grant
THE EVOLUTION OF KNOWLEDGE MANAGEMENT TECHNOLOGY 87
(2000, p. 43), its primary contribution has been to look broadly at the entirety of knowledge and
knowledge processes within frms in order to extend the scope of knowledge management activities
to types of knowledge and knowledge activities that have been undermanaged by frms, and also to
achieve increasing integration across different knowledge management activities.
Regarding the future of knowledge management, Wiig (1997) suggests that the management of
knowledge as an explicit and standalone management practice will disappear as it is once again
absorbed into everyday organizational work. Thus, the management of knowledge in organizations
will again become implicit and invisible as it is reembedded in organizational work. However, the
contributions of the knowledge management movement are not expected to disappear. Instead,
practices such as knowledge sharing, team work, and collaboration; always ascertaining that the
best knowledge and practices are used; capturing and organizing research results, innovations,
lessons learned; and continually renewing knowledge wherever and in whatever manifestation
it is found (Wiig, 1997, p. 8), which the knowledge management movement brought into the
foreground, will remain, albeit in the background of organizational work. While Prusak (2001)
considers this the more desirable of knowledge managements potential trajectories, he does leave
open the possibility that the knowledge management movement could suffer a fate similar to that
of reengineering. Reengineering was hijacked by a host of downsizing initiatives that not only
hurt organizations but also discredited the ideas associated with it.
A historical review of knowledge management highlights that all management is, in effect,
about managing knowledge (Grant, 2000, p. 28). However, the concept of knowledge manage-
ment was brought into high relief in the 1990s (Prusak, 2001), during which time it raised aware-
ness around the knowledge-related aspects of organizational work. In the 1990s, consulting frms
formed knowledge management practices and software frms developed and marketed knowledge
management technologies. In other words, knowledge management became an organizing vision
(Swanson and Ramiller, 1997), that is, a vision for organizing that embeds and utilizes informa-
tion technology in organizational structures and processes (p. 460, emphasis as per original).
An organizing vision is a social representation about the organizational implications of using an
information technology.
More recently, however, the term knowledge management appears to be attached less frequently
to technology solutions. For instance, in 2004 the CEO of Tacit Knowledge Systems Inc., David
Gilmour, rejected the term knowledge management to describe the frms ActiveNet product
(Claburn, 2004); he instead referred to it as part of the on-demand enterprise infrastructure.
Interestingly, the same David Gilmour (2003) had previously described ActiveNet as a way of
fxing knowledge management. Furthermore, Davenport and Glaser (2002) extol the virtues
of baking knowledge into the systems that support standard business processes so that extant
knowledge cannot be ignored in everyday work. For instance, knowledge about drug interactions
can be embedded into a physicians drug prescription system. The system would alert the physi-
cian to problematic prescription decisions based on contraindicative interactions between drugs.
This suggests that knowledge management technology may indeed be becoming an invisible and
implicit part of the organizational infrastructure.
The purpose of this chapter is to explore knowledge managements evolution, particularly the
evolution of its assumptions about knowledge and knowledge work, by studying a specifc aspect
of the knowledge management movement, namely the technologies designed to solve knowledge
problems. These technologies are central to knowledge management as an organizing vision
(Swanson and Ramiller, 1997). In analyzing one knowledge management system from each de-
cade starting in the 1970s, this chapter will highlight the nature of the knowledge problem that
each system attempted to solve, how it solved the problem, and the meaning of knowledge and
88 SCHULTZE
its management implicit in the systems design. The four knowledge management technologies
that will be reviewed in this chapter are MYCIN, an expert system (1970s); a listserv used by a
community of practice (CoP) in the development of the Common LISP language (1980s); grape-
VINE, a Lotus Notesbased knowledge management application (1990s); and a portal that used
Autonomys agent-based technology to build a competence location system (2000s).
While all four systems included in this analysis clearly support a knowledge management process
(e.g., creating, sharing, capturing, and applying knowledge), only one of the systems in this analysis,
namely grapeVINE, was specifcally referred to and marketed as a knowledge management solution.
Furthermore, the technologies chosen for this analysis were selected for their location within each
decade rather than some other features (e.g., support for a specifc work task, profession, or industry).
This not only makes a controlled comparison diffcult but also means that this analysis cannot be viewed
as a comprehensive history of knowledge management technology. Rather, it should be viewed as a
thinbut hopefully rich and deeply variedslice of the evolution of knowledge management.
The description of each of the four systems is derived from academic publications that document
the organizational knowledge problem they addressed, how they addressed them, and what the
implications were. The selection of the technologies was signifcantly infuenced by the availability
of detailed descriptions of the systems and their implementation. Thus, no claims of representative
sampling can be made. Furthermore, the language used to describe each system is more refective
of academic discourse than the designers and users framing of the technology.
Despite these research limitations, this chapters analysis highlights a number of trends that
support Wiigs (1997) contention that knowledge management will once again become an implicit,
natural part of organizing. For instance, early knowledge management systems were more focused
on supporting decision-making tasks using narrow, domain-specifc knowledge, whereas more
recent solutions focus on the sharing of common knowledge. Representations of knowledge in the
early systems were explicit and explainable (e.g., keywords were clearly defned using everyday
language), whereas knowledge representations in later systems are increasingly implicit and invis-
ible (e.g., patterns dynamically derived from data and embedded in the algorithms of intelligent
agents and collaborative fltering technologies). Furthermore, early representations of knowledge
took the form of an explicitly stated body of rules or formal expressions of lessons learned from
past experience. As such, they were synthesized through conscious, effortful refection on the part
of the expert or knowledge worker. In contrast, more contemporary technologies develop profles or
outlines of selected characteristics that refect the users preferences or knowledge implicitly derived
from the users behavior. As the representations of knowledge become more implicit, vast amounts
of data drawn from users behaviors are used as indicators and surrogates of interests, knowledge and
competence. Thus, the boundaries between business intelligence (which relies extensively on such
profling techniques and data mining tools) and knowledge management appear to be blurring.
This chapter will commence with a brief history of knowledge management as presented in prior
research. This will be followed by discussions of the four technologies. For each, the assumptions
about knowledge work, the nature of knowledge, and what it means to manage it, implicit in the
knowledge management solution will be highlighted. The chapter will conclude with a summary
of trends in the evolution of knowledge management, particularly as they relate to the meaning
of knowledge and its management.
AN ABBREVIATED HISTORY OF KNOWLEDGE MANAGEMENT
The historical research on knowledge management suggests that the importance of knowledge
and the need to manage it remained implicit in organizations until the mid-1970s, when Chaparral
Steel developed a corporate strategy and internal organizational structure that relied on the explicit
management of knowledge (Wiig, 1997). In the 1980s, a few organizations implemented expert
systems to automate complex decision processes. For instance, in 1980, DEC installed XCON,
a knowledge-based system whose rules had been gleaned from engineers. These rules specifed
how the various computer components of a VAX computer needed to be confgured in order to
satisfy customers computing needs (Dhar and Stein, 1997).
Despite these early examples of organizations explicit efforts to manage knowledge, it was
only in the 1990s that knowledge management became an organizing vision. Organizing visions
are developed in an interorganizational context, by a community that is made up of heterogeneous
players including vendors, consultants, IT professionals, and early adopters, who are united in their
interest to provide a public interpretation of the IT innovation (Swanson and Ramiller, 1997). They
are the product of community discourse, which may or may not be based on material processes
and technologies.
The purpose of organizing visions is to make a new technology understandable and legitimate
and to mobilize its adoption (Swanson and Ramiller, 1997). Organizing visions reduce the per-
ceived uncertainty of the innovation by interpreting the technologys implications for organizations.
They provide a rationale for investing in the innovation by linking it to business issues that are of
particular interest to the community. Last, they structure and mobilize market forces, e.g., vendors
and service providers, that support the materialization of the innovation.
Even though it is unclear when and by whom the term knowledge management was coined,
according to Prusak (2001, p. 1003), perhaps a good milestone to mark the beginning of the
knowledge management timeline is a conference held in Boston in early 1993 that several col-
leagues and I organizedthe frst conference specifcally devoted to knowledge management.
While others may disagree with Prusaks claim, there is general consensus that the emergence of
knowledge management and its meaning resulted from the confuence of multiple trends, includ-
ing globalization, digitization, ubiquitous computing, and a knowledge-centric view of the frm
(Grant, 2000; Prusak, 2001; Wiig, 1999).
It is also noteworthy that the organizing vision of knowledge management followed on the heels
of the reengineering movement, leading many to view knowledge management as an invention
by consultants looking for a proftable new subject to replace an expiring one (Prusak, 2001,
p. 1002). However, Davenport (1995) identifed a different relationship between reengineering
and knowledge management. Labeling reengineering as the fad that forgot people, Davenport
highlighted that the need for knowledge management initiatives was the result of the earlier re-
engineering movement, which relied heavily on IT to not automate, but obliterate (Hammer,
1990) extant organizational processes and structures. As a result, middle management, which
had traditionally been the human receptacle of an organizations knowledge, was so dramatically
downsized that organizations lost this crucial asset.
Even though reengineering, increasing globalization, and a new conceptualization of econo-
mies and organizations as more knowledge intensive certainly raised awareness about the need
to manage knowledge in organizations, Grant (2000) maintains that the increasing digitization of
knowledge played a defning role in the development of the knowledge management movement.
Digitization facilitates the transfer of knowledge that has traditionally been stored in people and
capital equipment to databases and to software. For instance, with the automation of processes
and workfows, both tacit and explicit knowledge are codifed and embedded in software. Also,
best practices identifed and documented in one part of the organization can readily be shared with
other parts of the organization via groupware databases.
Thus, technology plays a central role in the evolution of knowledge management and thus
90 SCHULTZE
promises to serve as a useful guide through the evolution of the knowledge management movement.
The next four sections will be devoted to describing four knowledge management technologies
spanning from the 1970s to today.
EXPERT SYSTEMS: MYCIN (1970s)
Expert systems are knowledge-based systems that achieve expert-level performance through the
use of artifcial intelligence techniques such as symbolic representation, inference, and heuristic
search (Benfer, Brent and Furbee, 1991). They are frequently described as decision support sys-
tems (Dhar and Stein, 1997; Davenport, 2005), as they supportrather than automatedecision
making. The knowledge represented in expert systems is typically narrow and domain specifc.
This not only represents the specialized nature of expertise, but also reduces the complexity of
the problem space.
MYCIN was one of the most well-known, experimental expert systems in the area of medical
computing. The application of computers promised to address a number of problems in health
care. These included skyrocketing costs, the pace with which new medical discoveries were made
and the challenges that physicians faced with regard to assimilating the new knowledge, the un-
even geographic distribution of physicians, and the increasing demand on the physicians time
(Shortliffe, 1976). Thus, the knowledge management problem that expert systems sought to solve
was the effcient communication and use of expertise in order to improve individual physicians
diagnostic decision making.
The development of MYCIN began in 1972, at Stanford University. Based on concerns that
antibiotics were prescribed too often, the purpose of the system was to assist physicianswho were
not experts in the feld of antimicrobialswith diagnostic and therapy selection tasks (Shortliffe,
1976). MYCIN helped physicians decide whether the patient had a serious infection, determine
the likely identity of the offending organism, decide which drugs were most effective, and choose
the most appropriate drug. The system not only was supposed to provide the physician reliable
advice in an effcient manner, but it was also supposed to educate the physician (through instruc-
tions and explanations) when so prompted by the user.
Within the domain of antimicrobials, MYCIN focused specifcally on the process of therapy
selection for patients with bacteria in the blood. In particular, MYCIN helped physicians distinguish
between different forms of meningitis (Clancey, 1987). Guided by information engineers, experts
in the area of antimicrobials collaboratively discussed representative case histories to elicit their
knowledge. This knowledge took the form of deductive and inferential statements, which represented
heuristic, probabilistic relationships between phenomena. These relationships were then encoded
as 150
1
decision rules in MYCINs initial knowledge base (Shortliffe, 1976, p. 72) (see Figure 5.1).
These rules, which are represented in deductive ifthen logicwhere the if part refects one or
more premises and the then part refects the conclusion or the action to be takenwere then used
to generate the questions that MYCINs consultation system asked the user and the explanations
that MYCINs explanation module gave in response to the users queries. The user interface took
the form of an interactive, Question and Answer dialog: MYCIN asked questions and waited for the
users input, and the user could ask for clarifcation of the question (i.e., why a question was being
asked or what a valid input would be), as well as explanations (i.e., the rules and probabilities that
had led to the answer).
MYCIN was also designed such that the rule base could be updated in an interactive manner
by the end user. The user had to enter the rule in English, using an ifthen format, and MYCIN
would translate it into LISP.
Even though in trials MYCIN outperformed medical students and practicing physicians and
was on par with experts in bacterial diseases, it was never put into production (McCarthy, 1984).
This was, in part, because of concerns about the programs inability to know its own limitations.
Due to its narrow domain knowledge, MYCIN would make a treatment recommendation that
was correct but incomplete because it only took the bacteria in the blood into account. Thus, the
effective use of MYCIN required a user (physician) who had a clear understanding of MYCINs
limitations and who had a broader knowledge of symptoms, therapies and treatment plans (i.e.,
a sequence of therapies that take the effects of earlier therapies into account). For patient safety,
the physician was required to display the kind of common knowledge about medical treatment,
that is, rules of thumb that the average physician learns through experience, which MYCIN lacked
(McCarthy, 1984).
As an expert system, MYCIN highlights the following assumptions about knowledge work,
knowledge and what it means to manage it:
Knowledgeworkersareindividualsengagedinmakingcomplex, yet repeatable decisions.
Thegoalofknowledgemanagementistheautomation (or guiding) of reasoning through the
application or reuse of extant knowledge.
Knowledgeistheheuristic,probabilisticrelationshipbetweenphenomena;thisrelationship
can be represented as an explicit rule; defning these rules requires a knowledge engineering
process that extracts and codifes the tacit and explicit knowledge of experts into program-
mable rules.
Narrowlyfocused,domain-specifcexpert knowledge is fundamentally different from com-
mon knowledge (i.e., the rules of thumb that the average person learns through his or her
experience in the world).
LISTSERVS: THE COMMON LISP PROJECT (1980s)
Even though e-mail only became commercially available in 1988,
2
the scientifc community was using
e-mail via the U.S. Department of Defenses ARPANET since the early 1970s. The listserv features
of the ARPANET e-mail system supported distributed group work among scientists. A specifc appli-
cation of the ARPANET email system is the Common LISP listserv documented by Orlikowski and
Yates (1994). The Common LISP (CL) project was initiated when the Department of Defense, which
had funded most of the LISP development work, put pressure on the [LISP] designers to produce a
Figure 5.1 Schematic of MYCIN
92 SCHULTZE
standard LISP language so that programs written in that language would be portable across computer
types (p. 550). The LISP designers were geographically dispersed computer scientists, located at both
universities and companies. However, they were part of the same occupational community and most
of them knew each other either from interactions at professional conferences or from their training at
either MIT or Stanford University. They were regular users of electronic mail and decided to do most
of their work on this project electronically. Throughout the two-and-a-halfyear project (April 1981
through December 1983), which ultimately produced a reference manual entitled Common LISP:
The Language, the participants had two face-to-face meetings at conferences.
Even though Orlikowski and Yates (1994) do not present the listserv as a knowledge management
system and do not emphasize the role of knowledge and its management in their case descrip-
tion, there is little doubt that the CL group represents a community of practice (CoP). According
to Wenger (2000), CoPs are defned by the following three characteristics: (a) members share a
sense of joint enterprise in which individual members are invested, e.g., an interest in a topic, (b),
relationships and mutual engagement that bind members together into a social entity, and (c) a
shared repertoire of communal resourcese.g., routines, a vocabulary or artifactsthat members
have developed over time through their mutual engagement. All three of these characteristics ap-
ply to the CL group. Indeed, Orlikowski and Yates (1994: 563) highlight that as members of the
same occupational community, the CL participants shared background knowledge, experiences,
values and interpretive schemes, including knowledge of the projects task (designing computer
languages), its domain (LISP dialects), and its process (the practice of language design and norms
about using electronic mail). This highlights that the community was able to adopt, without ap-
parent discussion or dissent, a set of practices, rules, norms, and conventions of interaction based
on their shared a priori knowledge and assumptions about how their community of practice orga-
nized around and worked on language design (Orlikowski and Yates, 1994, p. 563). This shared
knowledge was largely implicit and did not require explicit discussion on the listserv.
Furthermore, Orlikowski and Yates (1994) highlight that documentation was clearly an un-
derlying purpose of the entire electronic exchange, as suggested by the archiving of all messages
(p. 555). This suggests that the listserv served a knowledge repository function as it allowed
participants to review discussions that had gone before.
The CL group also refects the apprenticeship model inherent in CoPs (see Figure 5.2), where
highly experienced masters are located at the center of the community and apprentices are on the
periphery. Masters are more infuential and powerful with regard to group decision making. In the
CL community, participation was not evenly distributed and voting was not democratic. Of the 117
CL members, there were 17 participants who contributed 1332 of the 2000 listserv postings. Of
these 17, there were 5 members who were particularly active. They were regarded as key players
by the community. When it came to voting on proposals, the results were weighted according to
the expertise and experience of members, as well as the cogency of the arguments they made.
Given the listservs support of the LISP CoP and its documentation function, it is conceivable
that the CL listserv would have been labeled a knowledge management system if this label had
been popularized in the early 1980s. For the purpose of our analysis here, the fact that the CL list-
serv qualifes as a knowledge management system in todays terms but that its description was not
overshadowed by the knowledge management label is quite useful. Not only does the description
of the knowledge management solution focus more on the nature of knowledge work, especially
in the context of a geographically distributed CoP, but it also shows the dynamic adaptation of the
technology to a given knowledge management challenge. Thus, rather than focusing on the success
or failure of a particular knowledge management solution, the Common LISP listserv case provides
us insights into the development of a CoPs emergent knowledge management practices.
Orlikowskis and Yatess (1994) analysis of the CL listserv identifes four communication genres
(i.e., typifed communicative acts) that refected the groups knowledge work. These are the memo,
the dialog, the proposal, and the ballot system (see Figure 5.2). The memo genre served a general
communicative purpose in the distributed organization. The dialogue genre refected a conversa-
tion. It embedded text from a previous message in the posting, thus highlighting the continuity
and interdependence between the participants and their contributions. As its name implies, the
proposal genre contained specifc LISP syntax that was proposed by an individual member for
consideration by the rest of the community. The ballot genre supported a group voting process
that helped determine the degree of group consensus on decisions that the group needed to make.
It was used to identify settled and contentious issues, reveal areas in which further work was
needed, and indicate whose concerns had to be attended to (Orlikowski and Yates, 1994, p. 561).
The ballot genre system consisted of three interrelated messages: (1) a ballot questionnaire with a
list of issues, (2) responses containing votes and comments, and (3) a summary of ballot results.
The genre perspective that Orlikowski and Yates (1994) apply to examine the communicative
practices of the CL CoP highlights the role that implicit process knowledge plays in the smooth
functioning of distributed work. Even though the listserv archive contained only explicit knowledge
(i.e., documented discussions, proposals and ballots), this form of knowledge work was made
possible by the implicit knowledgesuch as norms around e-mail communicating and the nature
of the computer-language development taskembedded in the community. This highlights the
Figure 5.2 Schematic of the Common LISP Listserv
94 SCHULTZE
importance of implicit process knowledge, which is typically well established in such domain-
specifc, occupational communities as the CL community.
The knowledge-related insights that we can take away from this listserv technologys applica-
tion to distributed knowledge work are as follows:
Knowledgeworkersareseenasmembers of a community of practice (CoP), collaboratively
engaged in the creation of new knowledge (i.e., standard LISP).
Eventhoughthelistservisexpectedtoserveasarepository,itsprimaryroleliesinsupporting
the collaborative knowledge creation process of the distributed members; as a consequence,
the knowledge management solution focuses on the mechanisms (i.e., genres) for coordinat-
ing group decision making.
Knowledge is socially constructed and therefore requires the development of agreement
among group members; the ballot genre is designed to move a proposal (low level of agree-
ment) to a standard (high level of agreement).
Whilethelistservcapturedandstoredonlyexplicit knowledge, it relied on the CoP members
implicit, shared knowledge of the projects task, its domain and its process for its smooth
operation.
LOTUS NOTES GROUPWARE: GRAPEVINE (1990s)
Of all the technologies analyzed in this chapter, grapeVINE is the only one that was labeled a
knowledge management technology by its developer and, therefore, by its users. It mimicked the
informal organization (i.e., the human grapevine) that operates in most organizations (Schultze,
1997). The human grapevine is essentially a manual information fltering and processing mecha-
nism: a member of the organization takes information (e.g., news or a rumor) that she or he thinks
is so important that others should know about it and passes it on to others, whoin turnpass it
along to still others in their (informal) network. As it moves through the grapevine, the information
does not remain unchanged; instead, the person passing it on typically adds his or her interpreta-
tion and signifcation. In this way, the information is situated in the local context, thus becoming
more actionable and therefore more knowledge like.
The designers of grapeVINE represented this informal knowledge management process in terms
of an alertassessescalate model, which they also referred to as gatekeeping (Schultze,
1997). Gatekeeping implied that a person who either had an interest in a topic area or who was
considered a subject matter expert would consistently scan the environment for interesting
news and information (alert). The gatekeeper would then ascertain the informations credibility
and relevance to the organization (assess). If the information was found to be of signifcance to
the organization, it would be passed on to others together with an explanatory commentary or
interpretation (escalate).
The technology worked as follows (see Figure 5.3): The frm would identify source databases
to be profled. These typically included external newsfeeds or WWW pages, but could also include
internal document databases, e.g., discussion or best practices databases. The source documents
were then profled using the Knowledge Chart. The Knowledge Chart was centrally administered,
typically by the grapeVINE project manager or a corporate librarian, and it comprised the keywords
that had to be present in a source document for it to be included in the frms grapeVINE system.
For instance, the names of the frms competitors were typically part of the Knowledge Chart, as
were keywords related to the frms products and industry-specifc processes.
Individual users would then put together their Interest Profle by selecting keywords from
the Knowledge Chart. For each keyword in their Interest Profle, users would add the escalation
level at which a message should be before passing into his or her Lotus Notes in-tray. Thus,
an analyst serving as a gatekeeper for news about the frms competitors would set his or her
message escalation level at 1-routine, the default of any message that entered the system. In
contrast, a senior manager would set his or her escalation level for news about competitors at
level 4-critical, which ensured that only a message that a gatekeeper and/or the community of
grapeVINE users considered so signifcant that they had escalated it, would fnd its way to the
senior manager (Schultze, 1997).
The approach of grapeVINE to knowledge management was expected to fulfll the 1990s promise
of knowledge management, namely that the right person, would get the right information at
the right time (Schultze, 1997). In other words, users would neither miss important information,
nor would they struggle with information overload, as only information relevant to them would
make its way into their Lotus Notes in-tray. Furthermore, gatekeepers who were responsible for
environmental scanning would not have to proactively search for information. Instead, grapeVINE
would flter information from external news feeds and automatically route the relevant messages
to the gatekeepers in-tray. In that way, gatekeepers would not miss important news.
Another knowledge management promise that the grapeVINE design intended to fulfll related
to the conversion of tacit knowledge into explicit knowledge. The assumption was that in assessing
the relevance of a news item, the gatekeeper would confront his or her tacit assumptions regarding
the organization, its competitors, etc. If the information was deemed relevant, the grapeVINE sys-
tem prompted the gatekeeper to make these assumptions explicit, thus converting tacit knowledge
into explicit knowledge that could be shared and stored.
A case study of three frms attempts to implement grapeVINE (Schultze, 1997) highlights
several challenges with this technology. The frms studied were Hewlett Packard, National Semi-
Figure 5.3 Schematic of grapeVINE
96 SCHULTZE
conductor, and ABC Company (a pseudonym). The frst challenge was the lack of formality around
the work of gatekeeping. For instance, at Hewlett Packard most knowledge workers recognized the
value of staying abreast of events both inside and outside the frm. However, regularly scanning the
environment and assessing the signifcance of events constituted signifcant work and represented
substantial responsibility, for which employees wanted recognition. Thus, as the technology insti-
tutionalized the informal grapevine, thus increasing the formality and visibility of the knowledge
management process, gatekeeping needed to be acknowledged as a formal part of an employees
job, complete with suitable incentives, evaluation criteria, and metrics.
Second, gatekeeping and the escalation of messages only make sense when there is critical mass
of users on the grapeVINE system. In all three frms, there was a lack of critical mass, which meant
that gatekeepers copied news items they thought relevant and pasted them into e-mail messages,
whose distribution they could control. In other words, gatekeepers wanted to make sure that the
people they thought needed to be alerted to the news would indeed receive it. There was no way
of ensuring this in the grapeVINE system.
Third, grapeVINEs keyword-based fltering mechanism was considered too crude by the admin-
istrators of the Knowledge Chart. For instance, the Knowledge Chart at National Semiconductor,
which was already unwieldy with its 1,700 keywords, was still generating an unwanted number
of hits and thus overwhelming the gatekeepers with messages. For instance, some gatekeepers
at National Semiconductor reported receiving over 1,000 news alerts a day. Thus, grapeVINEs
promise to cure the information overload problem did not materialize.
Fourth, the problems that each of the three frms had with implementing grapeVINE highlighted
that the knowledge management model embedded in the technology did not ft the gatekeepers
work practices. This was particularly apparent in ABC Company, where the competitive intelligence
analysts, who were the primary users of the technology, found that the technology democratized
access to external information and thereby undermined their ability to present themselves as value-
adding knowledge workers in the frm.
The implementation challenges indicate that the formalization and partial automation of the
organizational grapevine interfered with some of the elements that make the informal organiza-
tion work. For instance, the grapeVINE design assumed that the interpersonal relationships
that make up the organizations social fabricand the personal trust, reciprocity, and social
capital associated with themcould be replaced by a programmatic, generalized form of reci-
procity that ignored the social dimension of generating and sharing knowledge. Traditionally,
the sharing of knowledge and information had formed part of the ongoing negotiation for social
recognition, power, and infuence within the (informal) organization. The more formal and
mechanistic knowledge management process of grapeVINE implied a more democratic and
less situated informing environment that did not ft well with the social nature of knowledge
networks in organizations.
Regarding the assumptions about knowledge work, knowledge, and knowledge management
underlying the grapeVINE solution, we can conclude the following:
Knowledgeworkimpliesindividual gatekeeping in a community-based system of environ-
mental scanning; gatekeepers are identifed through their explicitly declared interest and/or
expertise in a subject matter.
The management of knowledge requires the conversion of tacit knowledge embedded in
individuals and groups into explicit knowledge that can be shared further.
Knowledgeisinformation made actionable by signifying and situating it in a specifc tem-
poral and/or spatial context.
COMPETENCE MANAGEMENT SYSTEMS: VOLVOS VIP
PROTOTYPE (2000)
Core competences are defned as the collective knowledge and capabilities that are embedded in
the organization and that are central determinants of the organizations competitiveness due to their
centrality to customer value, their resistance to imitation, and their ability to extend to new busi-
ness applications (Hamel and Prahalad, 1994). At an operational level, competence management
involves the specifcation of an organizations competence needs, the identifcation of competence
gaps (between needed and actual competence), competence sourcing, competence development
through training and coaching, and the staffng of projects (Baladi, 1999). This typically renders
competence management the purview of human resources departments, which seek to manage the
competence of individual members of the frm. At the level of the individual, human resources is
concerned with job competence, which is defned as possessing skills that are critical for the indi-
vidual to master if he or she is to achieve high performance in the completion of a task (Boyatzis,
1982). Thus, individual competence is an underlying characteristic of a person, which results
in effective and/or superior performance in a job (Boyatzis, 1982). The personal characteristics
that facilitate high performance (and that are therefore part of individual competence) include
motivation, disposition, self-image, values, moral standards, norms of social behavior, and traits,
as well as communication, general reasoning, and learning capabilities (Bergenhenegouwen, ten
Horn, and Mooijman, 1996; Rothwell and Lindholm, 1999).
Competence management systems (CMS) represent a class of information systems specifcally
designed to assist with these competence management tasks (Lindgren, Henfridsson, and Schul-
tze, 2004). In an action research study of commercial CMS, conducted between July 1999 and
December 2001, Lindgren et al. (2004) identifed a number of shortcomings with respect to these
systems. These included that users were unable to see the competences that others in the organi-
zation had. In this way, expertise location was not a CMS feature to which users throughout the
organization had access. This curtailed the systems usefulness with regard to knowledge transfer
in the organization. Furthermore, the commercial CMS focused on past (i.e., competence-in-stock)
and formal (e.g., a degree earned, a skill mastered at a certain level of expertise) competences. In
other words, they accounted for neither competence-in-use (competence that users were applying
and developing in the process of doing work) nor competence-in-the-making (competence that
employees were interested in developing). This was problematic in that the organizations com-
petence records were always out of date, and because individual users would purposely hide their
competence in areas in which they no longer wanted to work. For instance, expert FORTRAN
programmers who wanted to learn other programming languages might be inclined to hide their
past programming experience.
In an effort to develop IT design principles for effective CMS, Lindgren et al. (2004) developed
and empirically assessed the usability of a CMS prototype that addressed these shortcomings.
Their prototype was developed as a complement to the commercial CMS that Volvo IT, one of
the frms participating in the action research study, was using. The prototype was called the Volvo
Information Portal (VIP), and it used Autonomys AgentWare platform as the engine for advanced
pattern-matching of textual data. Autonomys search engine goes beyond matching on keywords.
Instead, its algorithm relies on Bayes probability theorem, Shannon and Weavers information
theory, and neural networks to support natural language searches and meaning-based matching
of textual information. As such, Autonomys technology supported concept rather than keyword
searches.
The VIP prototype worked as follows (Figure 5.4): VIP allowed the users (i.e., knowledge
98 SCHULTZE
workers) to defne intelligent agents that searched a database for intranet documents matching the
users interests. By defning one or more agents, VIP users were thus able to monitor the corporate
intranet for items that matched their interests. The users defned their interests in a free-text natural
language format from which the system then created an internal, digital representation.
The search results from each agent were displayed in a simple list similar to those generated
by search engines. By clicking on the hyperlinks generated by the agent, users could retrieve
the documents matching their interests. When users had read and identifed one or more of the
returned documents as relevant, they could provide the agent with explicit feedback through a
retrain feature, which adjusted the agents search criteria based on the characteristics of the re-
trieved document.
The VIP prototype also provided a community feature, which allowed users to fnd other users
with a similar interest profle to their own, thus enabling information sharing and collaboration
across Volvo ITs many offces. When invoking this feature, the interest profles embedded in a
users agents were matched with the profles of all other agents. Matches were displayed, listing the
name and contact information of the identifed users. Finally, the VIP prototype supported searches
for users with specifc interests. VIP users could enter a search for an interest in natural-language
format and VIP would return all users whose agents contained the specifed interest.
Figure 5.4 Schematic of VIP
Users subsequent evaluation of the prototype highlighted that some appreciated the real-time
capture of either their competence-in-use or their competence-in-the-making. They perceived it as
an improvement over their extant CMS, which required the explicit formulation and recording of
a users competence. They also appreciated the ability to identify others in the organization who
shared their competences and interests as this facilitated knowledge sharing and the development
of communities of interest.
However, some users expressed concerns about the transparency of their search agent. The
accessibility of such highly personal information as their interestsrather than their formal
competencies such as degreesmade them feel vulnerable and anxious. For instance, the system
did not include a feature through which search agents could be made private and excluded from
another users search for others competence. Some users were also uncomfortable with the lack
of control that they had over how their interests were derived and represented in VIP. The profle
of a users interests became increasingly implicit (i.e., derived from user behavior), especially if
they used the systems retrain feature. Thus, as the system supposedly became better and better
at inferring the users interests from the users ratings of specifc hits, the user was also losing
control over the representation of his or her interests.
Regarding the assumptions about knowledge work, knowledge, and knowledge management
embedded in the application of Autonomys AgentWare technology in CMS, we can say the fol-
lowing:
Knowledgeworkimpliesindividual learning or competence development with the expecta-
tion that individuals competence is made available to the organizational community.
Knowledgeisdynamic and continuously evolving through the interaction of competence-in-
stock, competence-in-use, and competence-in-the-making.
Intelligentsearchagents,whichbuildimplicit competence profles based on individual us-
ers information seeking and evaluating behaviors, represent individuals knowledge and the
organizations knowledge (collectively).
Themanagementofknowledgerequirescompetence location, that is, identifying individuals
with the required competences.
Themanagementofknowledgerequiresstrategic competence development, which implies
that individuals competence interests should ideally be aligned with the organizations
strategic direction for (core) competence development.
DISCUSSION
The purpose of this chapter was to explore knowledge managements evolution by studying a
specifc aspect of the knowledge management movement, namely the technologies that lie at the
heart of the organizing vision. Analyzing four knowledge management systems, this chapter sought
to develop insight into how assumptions about knowledge, knowledge work, and knowledge man-
agement evolved over the decades. By comparing and contrasting the meaning of knowledge and
its management underlying the design of MYCIN (1970s), the Common LISP listserv (1980s),
grapeVINE (1990s), and an agent-enhanced CMS (2000s), this chapter sought to identify some
trends in the evolution of knowledge management.
Before discussing the trends suggested in this brief historical review of knowledge management
technology, it is important to acknowledge some of the characteristics of the sample of technologies
on which this review is based. Even though all four systems included in this analysis clearly support
knowledge management processes such as capturing and disseminating knowledge, only grapeVINE
100 SCHULTZE
was specifcally referred to and marketed as a knowledge management solution. Furthermore, the
technologies chosen for this analysis were selected for their location within each decade rather than
some other features (e.g., support for a specifc work task, profession, or industry). While this makes a
controlled comparison diffcult, it is noteworthy that the four technologies represent the three classes
of knowledge management strategy identifed by Wasko and Faraj (2000). In particular, MYCIN
is an example of a knowledge repository and thus indicative of a strategy that manages knowledge
as an object. While the Common LISP listserv is an example of a technology that treats knowledge
as embedded in a community, both grapeVINE and the VIP CMS seem to manage knowledge as
embedded in individuals primarily. This suggests that the small sample of technologies reviewed in
this chapter nevertheless encompasses the different classes of knowledge management solutions.
The selection of the technologies analyzed was signifcantly infuenced by the availability of
detailed descriptions of the systems and their implementation in academic publications. Thus, no
claims of representative sampling can be made. Indeed, it would be problematic to assume that the
agent-based technology that VIP leveraged to locate other people in the organization was widely
used in expert location systems today. Instead, most contemporary expert location systems rely on
explicit profles that require users to do a self-assessment of their expertise (Becerra-Fernandez,
2006). Nevertheless, given the shortcomings of these expertise location systems (Becerra-Fernan-
dez, 2006), there is little doubt that we will increasingly see knowledge management systems that
rely on agent technology and business intelligence tools to build implicit profles based on users
search and communication behavior (e.g., scanning e-mail content).
Despite these apparent limitations, we can nevertheless identify signifcant shifts over the
decades with regard to the way in which knowledge is defned and the knowledge management
problems are framed. For instance, it is interesting to note the distinction between expertise and
competence, that is, the conceptualization of knowledge used in the frst (expert systems1970s)
and last (CMS2000s) system in this chapters review. Expertise is defned as judgmental knowl-
edge (Shortliffe, 1976) related to a specifc feld. Expertise is therefore closely associated with the
ability to reason logically and to make decisions. Competence, in contrast, implies a broader, more
encompassing defnition of knowledge. It is defned as possessing skills and characteristics that are
critical for effective and/or superior performance of a task (Boyatzis, 1982). The skills and charac-
teristics implied in competence include motivation, self-image, values, moral standards, and norms
of social behavior. Thus, while knowledge viewed as expertise seems to emphasize the cognitive
and reasoning aspects of knowledge work in the design of knowledge management technologies,
knowledge as competence embraces emotional, ideological, and social aspects as well.
This distinction between expertise and competence is also refected in the design of the knowledge
management systems. For instance, MYCIN focused on narrow, domain-specifc knowledge, whereas
the other systems reviewed in this chapter sought to encompass both domain-specifc and common
knowledge (i.e., rules of thumb learned through an individuals experience of everyday life). Benfer
et al. (1991) highlight that the knowledge problem expert systems seek to solve is the representation
and replication of logical reasoning, whereas the knowledge problem most other knowledge systems
seek to solve is the sharing of primarily common knowledge. Using Zuboffs (1988) terminology,
Wenger (1987) suggests that this signifes a shift in the use of IT from automate to informate.
As the discussions of both the Common LISP listserv and grapeVINE highlight, the management
of common knowledge requires consideration of both the explicit and implicit/tacit components.
While Orlikowski and Yates (1994) identify implicit process knowledge (e.g., unexpressed norms
about e-mail communication) as playing a key role in the smooth functioning of distributed knowl-
edge work, this knowledge was not made explicit, although some of it would have been designed
into the communication genres developed by the community of practice. In contrast, one of the
objectives of grapeVINE was to support the conversion of tacit knowledge into explicit knowl-
edge. Through the alertassessescalate (or gatekeeper) model, knowledge that organizational
participants relied on to interpret news and information, was supposed to be documented and made
explicit. As the case studies summarized in this chapter illustrate, however, grapeVINEs intended
knowledge explication process failed to materialize in practice.
Thus, while the domain-specifc knowledge in MYCIN was rather successfully elicited and
converted into an explicit set of rules through a knowledge engineering process, the grapeVINE
case demonstrated that making tacit, common knowledge explicit without a knowledge engineer
proves to be challenging. In contrast, the VIP systems solution was to keep tacit knowledge im-
plicit, in the form of digital profles over which users had no direct control. In other words, instead
of forcing users to articulate their knowledge and competence through keywords or the escalation
of information items as in grapeVINE, the VIP system relied on the users search behavior and
their feedback to the search agent through the retrain feature to develop a digital representation
of their competence-in-use and competence-in-the-making, which were taken to be proxies for
actual competence (i.e., competence-in-stock).
Based in part on the systems reviewed here and in part on more recent developments in the area
of knowledge management technologies, it would therefore appear that there is a movement towards
relying on vast amounts of behavioral data for the discovery of patterns, which remain implicit
and embedded in the technology. For instance, both Autonomy (autonomy.com) and Tacit (tacit.
com) use routine communications, such as e-mail, as source data from which individuals implicit
knowledge profles are developed (Gilmour, 2003). Similarly, collaborative fltering technologies,
which form the basis of recommendation system that are frequently used in e-commerce settings,
also rely on user behavior (e.g., customer purchases) as implicit inputs for personalization (Schafer,
Konstan, and Riedl, 2001). These implicit inputs may or may not be augmented by explicit inputs,
e.g., customers rating their purchases. Comparing these implicit profling technologies with the
early expert systems that relied exclusively on explicit rules, it is noteworthy that MYCIN relied
on 150 to 400 rules, whereas collaborative fltering technologies typically rely on vast amounts
of behavioral data stored in high-capacity data warehouses.
Nevertheless, expert systems are forming part of contemporary organizations knowledge
management infrastructures. For instance, the prescription decision support system that Daven-
port and Glaser (2002) describe seems to display some expert system features. Furthermore, the
CYC project, which aims to develop computer programs capable of common sense reasoning, is
another example of expert system development today. Since 1984, Doug Lenat, the founder of
CYC, has been working on a knowledge base that contains common knowledge, that is, rule of
thumb that the average person learns through his or her experience in the world. For instance, the
average person knows from personal experience that when two objects collide, they typically make
a noise; that when a glass of water falls from a table, the water will spill and the glass is likely to
break; and that if you are carrying something in a container thats open on one side, you should
carry it with the open end up (Anthes, 2002). Even though most people instinctively know this,
they would not express this heuristic with the degree of formalism required by an expert system.
Since 2002, Cycorp (www.cyc.com) has made the CYC database public, allowing the general
population to propose rules for addition to the knowledge base.
CONCLUSION
Knowledge management is not a new phenomenon. While all of organizing implies the management
of knowledge, it has traditionally been implicit. As an organizing vision, knowledge management
102 SCHULTZE
emerged in the 1990s and it highlighted the importance of managing knowledge, thereby turning
knowledge into an object and its management into a discernable activity. In this way, knowledge
and its management was made explicit. However, the meaning of knowledge and what it means
to manage it is malleable and emergent. This chapters review of four knowledge management
technologies highlights that early knowledge management systems were more focused on sup-
porting decision-making tasks using narrow, domain-specifc knowledge. In contrast, more recent
solutions focus on the sharing of common knowledge. Representations of knowledge in the early
systems were explicit and visible (e.g., rules and keywords), whereas knowledge representations
in later systems are increasingly implicit and invisible (e.g., profles dynamically derived from
patterns of behavior and then embedded in the algorithms of intelligent agents and collaborative
fltering technologies). Furthermore, as the representations of knowledge become more implicit,
vast amounts of data drawn from users behaviors are used as indicators (and surrogates) of users
interests, knowledge and competence.
NOTES
1. Eventually, MYCINs rule base increased to 200 rules (Shortliffe, 1976: 86). By 1980, MYCIN was
comprised of 400 rules (Clancey, 1987).
2. MCI offered the frst commercially sanctioned access to the Internets messaging service in 1988.
Compuserv e-mail was launched in 1989, and in 1993 AOL and other access providers connected their
proprietary e-mail systems to the Internet.
REFERENCES
Anthes, G.H. 2002. Computerizing common sense. ComputerWorld (April 8).
Baladi, P. 1999. Knowledge and competence management: Ericsson Business Consulting. Business Strategy
Review, 10, 4, 2028.
Becerra-Fernandez, I. 2006. Searching for experts on the Web. ACM Transactions on Internet Technology,
6, 4 (November), 333355.
Benfer, R.A.; Brent, E.E.J.; and Furbee, L. 1991. Expert Systems. Newbury Park: Sage.
Bergenhenegouwen, G.J.; ten Horn, H.F.K.; and Mooijman, E.A.M. 1996. Competence developmentA
challenge for HRM professionals: Core competences of organizations as guidelines for the development
of employees. Journal of European Industrial Training, 20, 9, 2935.
Boyatzis, R.E. 1982. The Competent Manager: A Model for Effective Performance. New York: John Wiley
and Sons.
Claburn, T. 2004. Lockheed fnds a new way to connect questions to answers. Information Week (June 14).
Clancey, W.J. 1987. From GUIDON to NEOMYCIN and HERACLES in twenty short lessons. In A. Van
Lamsweerde and P. Dufour (eds.), Current Issues in Expert Systems. London: Academic Press.
Davenport, T.H. 1995. The fad that forgot people. FastCompany (November 1), 70.
. 2005. Thinking for a Living: How to Get Better Performance and Results from Knowledge Workers.
Boston: Harvard Business School Press.
Davenport, T.H., and Glaser, J. 2002. Just-in-time delivery comes to knowledge management. Harvard
Dhar, V., and Stein, R. 1997. Intelligent Decision Support Methods: The Science of Knowledge Work. Upper
Saddle River, NJ: Prentice-Hall.
Gilmour, D. 2003. How to fx knowledge management. Harvard Business Review, 81, 10, 1617.
Grant, R.M. 2000. Shifts in the world economy: The drivers of knowledge management. In C. Despres and D.
Chauvel (eds.), Knowledge Horizons: The Present and the Promise of Knowledge Management. Boston:
Butterworth Heinemann, pp. 2753.
Hamel, G., and Prahalad, C.K. 1994. Competing for the Future. Boston: Harvard Business School Press.
Hammer, M. 1990. Reengineering work: Dont automate, obliterate. Harvard Business Review, 68, 4,
104112.
Lindgren, R.; Henfridsson, O.; and Schultze, U. 2004. Design principles for competence management systems:
A synthesis of an action research study. MIS Quarterly, 28, 3, 435472.
McCarthy, J. 1984. Some expert systems need common sense. Available at http://www-formal.stanford,edu/
jmc.
Orlikowski, W.J., and Yates, J. 1994. Genre repertoire: The structuring of communicative practices in orga-
nizations. Administrative Science Quarterly, 39, 541574.
Prusak, L. 2001. Where did knowledge management come from? IBM Systems Journal, 40, 4, 10021007.
Rothwell, W.J., and Lindholm, J.E. 1999. Competence identifcation, modeling and assessment in the USA.
International Journal of Training and Development, 3, 2, 90105.
Schafer, J.B.; Konstan, J.A.; and Riedl, J. 2001. E-commerce recommendation applications. Data Mining
and Knowledge Discovery, 5, 115153.
Schultze, U. 1997. Knowledge management and technology: Three companies experience with grapeVINE.
In J. Elam (ed.), Emissary Prentice Hall Custom Cases.Upper Saddle River, NJ: Prentice-Hall.
Shortliffe, E.H. (ed.). 1976. Computer-Based Medical Consultation: MYCIN. New York: Elsevier.
Swanson, E.B., and Ramiller, N.C. 1997. The organizing vision in information system innovation. Organiza-
Wasko, M., and Faraj, S. 2000. It is what one does: Why people participate and help others in electronic
communities of practice. Journal of Strategic Information Systems, 9, 2/3, 155173.
Wenger, E. 1987. Artifcial Intelligence and Tutoring Systems: Computational and Cognitive Approaches to
the Communication of Knowledge. Los Altos, CA: Morgan Kaufman Publishers.
. 2000. Communities of practice: The structure of knowledge stewarding. In C. Despres and D.
Chauvel (eds.), Knowledge Horizons: The Present and Promise of Knowledge Management. Boston:
Butterworth Heinemann, pp. 205223.
Wiig, K.M. 1997. Knowledge management: Where did it come from and where will it go? Expert Systems
. 2000. Knowledge management: An emerging discipline rooted in a long history. In C. Despres and
D. Chauvel (eds.), Knowledge Horizons: The Present and Promise of Knowledge Management. Boston:
Butterworth-Heinemann, pp. 326.
Zuboff, S. 1988. In the Age of the Smart Machine. New York: Basic Books.
104
ChaptEr 6
A FOUR-LAYER MODEL FOR INFORMATION
TECHNOLOGY SUPPORT OF KNOWLEDGE
MANAGEMENT
mattEo BonifaCio, thomas franZ, and stEffEn staaB
Abstract: Strategies for knowledge management as well as strategies for information technology
(IT) systems supporting knowledge management have often been conceived to serve one particu-
lar knowledge management paradigm. We argue that in practice there is not one paradigm that
suits all needs of all organizations. Considering knowledge management as a measure to further
communication in an organization, we fnd that different ways of organizing communication
existsome of which are not well supported by current IT systems for knowledge management. In
particular, we investigate two dimensions of communication: frst, the vocabulary or knowledge
structures used to organize (i.e., annotate, store, and retrieve) the knowledge using means such as
Semantic Web technology and, second, the way of organizing access to the knowledge in a more
or less distributed manner involving technologies like distributed databases or agent-based or
peer-to-peerbased systems. Based on these dimensions, we overview existing knowledge sharing
systems and elaborate on novel knowledge sharing models and tools. Thus, we derive a four-layer
model for IT support of knowledge management, where only one layer for centralized knowledge
sharing is well understood, where the layers of individual and decentralized knowledge manage-
ment are currently intensively researched, and where the forth layer of evolutionary knowledge
management is just appearing on current research agendas.
Keywords: Knowledge Management, Knowledge Management Systems, Information Technology,
Semantic Web
THE ORGANIZATION IS A NETWORK
The core objective of knowledge management (KM) is the creation of value for an organization
by establishing a principal way of handling knowledge in the organizations. Over the last two
decades, a large number of organizational measures have been proposed or rediscovered to meet
this objective. Organizational means for this purpose range from creating coffee corners, over
discovering and nourishing communities of interest, to establishing new comprehensive, orga-
nizational learning procedures. The essence of all these organizational experiments and changes
was that the knowledge fabric of the organization is not depicted by the organizational chart and
it is not found in the established business processes, but rather it is constituted by the network of
people communicating their knowledge along and beyond established organizational domains and
processes. Organizational measures toward KM harvest added value from such networks.
A FOUR-LAYER MODEL FOR IT SUPPORT OF KM 105
Intriguingly, though, information technology (IT) support for KM has hardly kept pace with
the richness of knowledge communication established in such human networks. Either, it has just
been used as is, like e-mail, in order to facilitate human communication without considering spe-
cial needs for KM. Knowledge management support here would have considered knowledge life
cycle support: creation, organization, sharing, use, and evaluation of knowledge. Or, it included
knowledge life cycle support but was mostly bound to one dominating model of knowledge com-
munication: centralized sharing.
The centralized sharing model as depicted in Figure 6.1 is quite simple. Knowledge is evaluated
by the individual. It is then contributed to a central information repository, such as a knowledge
database. It is organized into central categories (cf. Figure 6.1) and can be retrieved by a desig-
nated audience. Typically, there are curators who take care of initial flling, organize community
assessments, and maintain the basic structures. The centralized model of sharing knowledge is
widespread, it is powerfulbut it covers the wide variety of KM needs in human networks only
to a very limited extent.
In this chapter, we investigate the IT needs of organizations with explicit KM from a human
network perspective. Furthermore, we assume that IT structures work best if they do not get into
confict with organizational structures but rather refect as closely as possible the underlying
organizational needs and practices. For this purpose, we start with a consideration of various
organizational practices and relate them to KM efforts in The Organization Evolves. This
perspective lets us resolve the one model of centralized sharing of knowledge into a layered
model with varying degrees of knowledge sharing as discussed in more detail in Knowledge
Management System Classifcation by Degrees of Networking: Requirements and Use Cases.
In Evolving Knowledge Management Systems and Technologies, we use the layered and
more fne-grained model to survey the space of established, new, and forthcoming IT solutions
and technologies for KM.
1
Figure 6.1 Centralized Sharing of Knowledge
A
B C
D
106 BONIFACIO, FRANZ, AND STAAB
THE ORGANIZATION EVOLVES
The very notion of organization and its related feld of inquiry emerged when, after the so-called
scientifc management experience, researchers and practitioners felt the need to overcome the one-
best-way approach. According to this view, the administration of an organization could be done, as
much as a laboratory experiment, according to scientifc principles. Moreover, management was
seen as the function that centralized organizational knowledge; by means of scientifc principles,
managers could organize work optimally because they could possess all the needed knowledge.
Individuals knowledge and experience were very important but in a particular sense: through
the systematic measurement and observation of working practices, managers could collect best
practices and spread them to the overall organization. Employees were asked to actively engage in
such process because those productivity gains obtained through knowledge reuse would become
a beneft for all the companys stakeholders. Knowledge was a core asset that was systematically
codifed and disseminated by means of methods and procedures.
The organizational debate rose when scholars started questioning this monolithic model.
The debate produced a wide range of positions, options, and alternatives that will not be ana-
lyzed here. For the purpose of this chapter, we want to underline some common points that can
provide a useful framework to formulate our argument on KM. First, although from different
perspectives, researchers agreed that the main goal of an organization science should have
been the one of studying differentiation rather than conformity: namely, the fact that different
models of organizations exist and, further, that within the same organization different parts or
units may work according to different models. Second, the various models proposed hovered
along a continuum that went from the formal, centralized, and specialized model proposed by
the scientifc management (the so-called mechanic model), to more decentralized, unstructured,
and despecialized models (the so-called organic models). Third, choices along this continuum
should be judged in terms of appropriateness, rather than in terms of ideological opposition.
That is, each model is appropriate under particular conditions. Fourth, the models express dif-
ferent solutions in gathering and processing knowledge about the world, whatever these terms
(knowledge and environment) mean. From this perspective, while the mechanic orientation seems
to be more suitable when the outside is more stable (and knowledge more standardizable),
the organic one seems to be more appropriate the more the outside is uncertain (and the more
knowledge becomes contextual). As a corollary, when the environment changes, the organiza-
tion (or part of it) may evolve accordingly.
Interestingly, the recent KM debates have had a similar origin but did not follow a similar
evolution. In fact, on the one hand, it started assuming a one-best-way approach that focused
on the centralized codifcation, standardization, and reuse of knowledge. On the other hand,
when alternative models have been proposed, these were intended to function more in terms of
contraposition than in terms of appropriateness. As an example, the communities of practice
approach (that is quite consistent to the organic approach) has been proposed as a view on
organizational knowledge dynamics that is alternative to the centralized one (similar to the
mechanic). In this chapter, we propose that KM, instead of arguing for or against a particular
approach, should follow the more fruitful route that was originally taken by the organiza-
tional debatethat is, to view the various models of KM as responses that are to be judged
as appropriate to different conditions and, as a corollary, to view KM models as part of an
evolution that occurs within and across different organizations. We will refer to this approach
as evolutionary KM.
In particular, we propose that the typical contraposition between decentralized KM approaches,
that privilege organic issues such as informality and lack of structure, and centralized ones, that
privilege mechanic issues such as formality and standardization, should be reinterpreted in terms
of appropriateness. From this perspective, local knowledge is needed to generate appropriate solu-
tions to heterogeneous and changing environments. Centralized knowledge is needed to share these
solutions among different contexts. Both levels should be preserved; without locality, conformity
might surmount. Without centrality, each wheel might be reinvented. A process is needed in order
not just to preserve the two levels but rather to connect them as phases of an evolutionary process:
what is locally valuable knowledge should be consolidated centrally when a new domain stabilizes
to sustain economies of reuse; what is centralized should be relocalized when what was known
comes into crisis and locally reelaborated. KM technologies should be consistent with these chal-
lenges and need to act as a bridge between centralized and localized technologies and must allow
for evolution of the organization.
KNOWLEDGE MANAGEMENT SYSTEM CLASSIFICATION BY
DEGREES OF NETWORKING: REQUIREMENTS AND USE CASES
The model for centralized sharing of knowledge constitutes only one point in the space of possible
solutions for KM. In particular in professional working environments, it focuses on one kind of
communication that comes with a considerable set of assumptions on the characteristics of the
communication:
Communicationismostlyunidirectional: more people read than write.
Sharedknowledgeismature knowledge: immature knowledge generated on the fy is mostly
not dealt with. Shared knowledge is expected to be evaluated locally before it is published
and further examined by dedicated authorities, such as moderators or editors.
Sharingofknowledgeisanactive act: the sharing of knowledge requires that one actively
writes and publishes a memo, a Web page, or a data entry.
Knowledge management happens in a semiopen domain: access is not restricted to one
individual but is extended to a group such as specifed by corresponding access rights.
Knowledgemanagementhappensina semiclosed domain: the sharing of knowledge is exactly
confned to the group with formal access rights. The vocabulary used is shared knowledge
by the audience.
All these assumptions may be valid, but they may be questioned, too. Seeing KM through
the glasses of KM as communication in human networks, an extended model will topicalize the
individual node in the network as well as the topological structure of the network and the way it
evolves. A corresponding picture of knowledge sharing may then distinguish (at least) four layers
of knowledge sharing as depicted in Figure 6.2 and explained in the following.
A Layered Model of Sharing Knowledge
The layering of our model shown in Figure 6.2 should not be interpreted in its strictest sense, that
is, in a sense that each lower layer is a premise for the next layer. However, it is common that one
has to manage individual knowledge frst, before it comes to sharing of knowledge. Second, cen-
tralized sharing is a special case for, third, decentralized sharing (and hence, we argue, precedes
it). Fourth, the evolutionary model presumes that there is some minimal model of (de)centralized
sharing in order that evolution may have a signifcant basis and may be useful.
Individual Knowledge Management
At the lowest layer, people have found over the last years that the individual needs more powerful
IT systems to share knowledge with himself or herself in the sense that the individual organizes
his information for later retrieval by himself or herself. There are plenty of digital knowledge
creation and communication systems that the individual uses: e-mail, instant messaging, group
chats, document repositories, portals, disk drives, bookmarks, databases (from local address books
to large ERP databases), Web pages, text documents, spreadsheets, PDF documents, images, vid-
eos, calendars, SMS, voice-over-IP, fax, etc. While large companies have worked hard to digitize
their major business processes starting with scanning all incoming papers in order to avoid media
ruptures (e.g., for managing automobile insurance claims), such smoothness is often not found
when it concerns more ad hoc work of less regularity and when it concerns the individual desktop
leading to redundancy of data (e.g., telephone numbers on different media, such as mobile tele-
phone book versus PC address book), low quality of data (e.g., outdated telephone numbers), and
the diffculty or impossibility to retrieve the data easily (e.g., to search for a telephone number
by Smith with one search in multiple sources). This led to a recent increase of research and
development in the areas of Personal Information Management (PIM) and Desktop Search
2
(see
further explanations below, pp. 112113).
A typical use case for individual KM lies in organizing and retrieving project communications
from ones desktop. Such project communication may involve mails, appointments, or documents.
There are partial solutions for integrating some of these concerns, such as Microsofts Outlook for
organizing mail and calendar or Google Desktop Search for searching mail and desktop, but these
need to further develop to fully cover individual KM needs (also discussed on pp. 112113).
While individual KM benefts the individual, the organization wants to integrate processes of
local KM (e.g., knowledge creation) into organizational procedures: frst, to exploit the benefts
of individual KM efforts and, second, to ensure appropriate integration by some corporate process
instead of accidental sharing. Hence, means such as apprenticeship have been invented long
ago and KM as an organizational topic (and also coined organizational learning) has required and
will always require the next layer.
Centralized Sharing of Knowledge
The validity of this model for the appropriate set of use cases is not jeopardized by the need for
other models. The centralized sharing of knowledge is facilitated by plenty of established software
Individual Knowledge Management
Centralized Sharing of Knowledge
Decentralized Sharing of Knowledge
Evolutionary Model of Sharing Knowledge
4
3
2
1
Figure 6.2 A Layered Model of Sharing Knowledge
systems that are server based and typically organized around documents, sometimes organized
around folder structures, taxonomies, or similar kind of metadata. Folder structures, vocabular-
ies, taxonomies, ontologies, etc. denote differently sophisticated and complex means to defne
knowledge and organize information. Throughout this chapter, we use the more general notion of
knowledge structure (KS) to summarize any such means.
A typical use case for this model is demonstrated through the creation and maintenance of best
practice knowledge bases. Here, business practices are reconsidered along the core business process
(e.g., at project touchdowns), distilled for interesting experiences, and then fed into a centralized
best-practices knowledge base. Such a knowledge base is exactly characterized by the dominance
of asynchronous write-read, mature knowledge, active sharing, and semi-open, and semi-closed
domains as elaborated earlier.
This approach encounters its limits when it has to seek a compromising position, where the
compromise runs counter to the structures in which the human communication network works
most productively. For instance, when researchers develop their own new way of structuring
knowledge, it may be counterproductive or at least of little help to make them shape their thoughts
into predefned templates or vocabulariessuch as required by the model of centralized sharing
of knowledge. Thus, at this point, the view of the individual and the organizational compromise
get into confict and require a next layer of knowledge sharing.
Decentralized Sharing of Knowledge
The next layer is about the decentralized sharing of knowledge as indicated in Figure 6.3. This
model circumvents a knowledge server bottleneck. Instead it allows for sharing all the knowledge
(one wants to share) that one has on ones own desktop. It allows for a peer-to-peer communica-
Figure 6.3 Decentralized Sharing
A
B C
D
A
B C
D
A
B C
D
A
B C
D
tion structure that refects individual expertise and knowledge needs and standard organizational
processes (Bonifacio, Camussone, and Zini, 2004). It requires from the system that it allows for
distributed authorization and authentication of information access and for distributed search and
knowledge organization. Typical system infrastructures are (but need not be exclusively) peer-to-
peer systems or agent infrastructures.
A typical use case for decentralized sharing arises during project work spanning several orga-
nizations. For a centralized system in such a setting, an important question would be who controls
the server, pays the associated costs, and, after project touchdown, has access to the knowledge.
In a peer-to-peer organization, new project members are just added to the peer-to-peer network,
giving them rights to provide and search knowledge. Each project member benefts from the
advantages of individual KM as she personally controls and organizes the shared knowledge. In
comparison to centralized sharing environments, decentralized sharing allows for a desktop-like
user experience where the burden of uploading and downloading information, common in central-
ized systems, is not applicable.
This approach requires the location of knowledge, and it still expects either that the various
knowledge structures of different participants are suffciently similar (cf. Figure 6.3) or that they
can be translated back and forth (a problem that the centralized sharing of knowledge never has,
because it allows only one way of structuring knowledge). It hits its limits when vocabularies and
the meaning of a vocabulary change too fastwhich often happens when also the social and the
communication structures change very quickly. Then, the need for a new layer of sharing knowledge
arises that reconciles the centralized and the decentralized model of sharing knowledge.
Evolutionary Model of Sharing Knowledge
The evolutionary model of sharing knowledge draws from communication structures, knowledge
structures, and content in order to propose revisions to the communication in the network and
revisions to the structures and vocabularies used for organizing and communicating knowledge
(cf. Figure 6.4). Thus, it needs to reconcile between the individual views of layer 1, preexisting
standards such as established for layer 2, and it needs to use the communication structures such as
established for layer 3. In doing so, the evolutionary model of sharing knowledge does not abolish
the need for layers 1 to 3. Rather it is to be used when more traditional means (if layer 3 can at
all be categorized into this phrase) fail to provide the appropriate solution. Figure 6.4 indicates
the evolutionary model of sharing, showing different knowledge structures at each participant as
well as direct and mediated (dotted lines) paths between participants.
A typical use case for evolutionary sharing exists with scientifc research in computer science.
Scientifc research can be characterized by the need to share a lot of knowledge (e.g., about lat-
est research papers), by the need to cope with large and quickly evolving vocabularies, and by
the need to share and cooperate far beyond organizational boundaries based on mutual trust and
previously successful cooperations. Thereby the communication network structure is in constant
fux, because the interests of people move in different directions and the need to fnd and use new
methods and tools is ubiquitous. Not accidentally, the need for IT tools such as grid environments
(Foster and Kesselmann, 2004), which are about sharing data and computational resources and
building virtual organizations, has been pushed by academic institutions.
Therefore, there is the need to manage trust by the individual rather than to have it managed
by an overarching organization. The evolutionary model requires complex computations at each
peer as vocabularies must be reconciled, learned, and mapped. Correctness and completeness of
match-making are not guaranteed, and thus may reduce the quality of the shared data.
Dimensions for Knowledge Sharing in Human Networks
We have now provided a set of criteria that lay out the space of IT solutions for KM based on
different organizational models: local, centralized, decentralized, and evolutionary.
To better capture the differences between various systems and tools, we need to distinguish at a
more fne-grained level what is actually local, centralized, decentralized, or evolving. When com-
municating knowledge via IT systems, we see two main dimensions that vary along these values.
First, the knowledge structures (including the vocabulary) used to organize knowledge may be
available (only) locally. There may have been a kind of (simple) standardization effort to agree on a
common, centralized knowledge structure, there may be a plurality of such knowledge structures that are
decentralized and are mapped to each other, or there may be a continuously evolving set of knowledge
structures the overview of which needs to be automatically changed and adapted all the time.
Second, the ways to actually access knowledge may vary. Knowledge may be accessible only if
it is local. Or, it may be accessible if it lies on a central server. Or it may be accessible in a well-
defned decentralized mode. Finally, it may be available even if one needs to bridge several trusted
relationships in an evolving network. Thus, we have a classifcation scheme for two dimensions
with four different possible attribute values each as summarized in Table 6.1.
The reader may note that the dichotomy between these dimensions has also been explored in
related work, such as in Semantic Web and Peer-to-Peer (Staab and Stuckenschmidt, 2005),
where the Semantic Web (SW) accounts for ways to organize knowledge structures and peer-to-
peer for ways to organize knowledge access. The editors there, however, have focused on the two
layers of centralized and decentralized systems. Semantic Web research deals with methods to
build a World Wide Web containing machine-interpretable information. The development of the
Resource Description Framework (RDF) that provides a machine-interpretable, semistructured
language to record, interlink, and distribute metadata is one outcome of SW research. RDF has
been further extended toward ontologies that can be, among others, expressed by the Web Ontol-
A
B C
D
A
B C
D
F
C
G
B
D
E
A
Figure 6.4 Evolutionary Sharing
ogy Language (OWL), which adds knowledge representation and reasoning to RDF based on
Description Logics (DL). The advances in knowledge representation, exchange, and reasoning
as building blocks for the SW also leverage IT support for KM in general. Several of the systems
described in the following section are based on and contribute to that work.
EVOLVING KNOWLEDGE MANAGEMENT SYSTEMS
AND TECHNOLOGIES
In this section, we contrast actual KM as refected by current KM systems with novel KM ap-
proaches. We discuss KM systems with respect to the organizational model to provide a classifca-
tion for these systems and to pinpoint the current status of KM. The latter defnes the starting point
from which we progress toward the evolutionary KM approach that refects human networking
and dwell on recent research and technologies that foster progress toward this model.
Knowledge Management Today
Local
Within the past years, KM entered the local desktop in the form of personal information man-
agement (PIM) applications that are aiming at better user support for common tasks such as
Table 6.1
Dimensions of Knowledge Sharing Systems
communication, time management, and data management. PIM systems combine and integrate
multiple knowledge sources to improve availability, management, and reuse of information. Most
PIM systems combine an address manager, an e-mail client, and a scheduler so that, for example,
an appointment can be seamlessly sent via e-mail or an entry in the address book can be used to
denote a participant of a meeting or the recipient of an e-mail. Such functionality is provided by
software like the Mozilla Suite, Microsoft Outlook, and KDEs Kontact.
Improved management of information stored on the local fle system has also been addressed
lately. Enhanced search tools such as Googles desktop search, Gnomes search tool Beagle, and
Copernic support the user in fnding and locating information stored on the fle system. These
systems are often superior to common search features provided by the operating system as they are
able to search the content of various proprietary fle formats so that a search may include e-mails,
bookmarks, contacts, images, and several text document types (PDF, Microsoft Word, etc.).
Increased popularity of enhanced search tools indicates that the common organizational model
of folder hierarchies meets its borders with the increasing amount and heterogeneity of data stored
on computers today. Folder hierarchies are simple and resemble real world fle storage, thus be-
ing easy to grasp by novice users. A shortcoming of folder hierarchies is that content can only
be classifed following a single view disregarding that content usually has multiple contexts. As
a simplistic example, consider some text documents that need to be stored in a folder structure:
One could distinguish the documents by the type of the document such as Contract, Invitation,
Invoice, etc.; by the person who created the document; or by the time it was created. Either clas-
sifcation provides different context and is differently suitable depending on the task that needs
to be established with these documents.
Improved fle systems as well as tools that address the shortcomings of conventional fle systems
are already available or close to release: Microsofts new fle system WinFS,
3
the Gnome Storage
4

project, Apples Spotlight,
5
and the database fle system project (DBFS
6
) are samples of such novel
approaches. The conceptual e-mail manager (CEM) (Cole and Stumme, 2000) addresses the same
shortcoming in the specifc context of e-mailing, enabling multiple categorizations for e-mails and
offering different views onto the so-classifed e-mails.
A model for content classifcation that is less restrictive than the taxonomic model employed
by folder hierarchies, however more complex to maintain, is known as tagging and recently gained
popularity in centralized information sharing systems
7
but is also applicable locally.
Additional local KM systems are Haystack (Huynh, Karger, Quan, and Sinha, 2003) and Gnowsis
(Sauermann, 2005), which are research prototypes aiming at more advanced management features.
The Haystack system combines several knowledge sources and applications in a single PIM system
and supports the user in relating arbitrary information from any such knowledge source to feature
enhanced management and utilization that goes beyond the features of the popular market players
mentioned before. Among others, Haystack manages bookmarks, e-mails, contacts, appointments,
and local fles and includes a Web browser, e-mail client, scheduler, and contact manager. Any
information managed through Haystack is accompanied by metadata that can be extended and
modifed by the user, such as a local fle, can be related to an e-mail or a contact so that the ad-
ditional context can be exploited when searching and browsing for information.
The Gnowsis system follows a different approach, as it does not provide any application-specifc
user interface. Instead, Gnowsis provides a single data store to integrate arbitrary information and
offers an interface that can be used by regular desktop applications to contribute their data. Similar
as with Haystack, all information managed with Gnowsis can be related to each other to enable
enhanced information management features.
In all systems mentioned so far, knowledge is defned utilizing individual knowledge structures
such as individual hierarchies on the fle system, in e-mail clients, or by user tags, while no other
than local access is offered. Such desktop applications commonly offer high usability and simplify
ad hoc knowledge creation due to high personalization of the knowledge structures and knowledge
processes. It can also be noticed that the formerly separated retrieval tasks such as searching and
browsing merge with the rise of novel data management tools as presented.
Centralized KM Systems
In contrast to systems implementing a local model that omits networked information access,
systems that store knowledge in a single centralized repository are the most popular ones at the
time of this writing. The centralized model has two signifcant advantages:
1. Centralization permits knowledge sharing.
2. Centralization simplifes access control.
Most widely used are systems that centralize both knowledge structures and information ac-
cess. Centralized knowledge structures conform to predefned standards that are agreed on by the
knowledge community and ensure that knowledge can be easily shared within that community. A
multitude of standard groupware systems, including document management systems (DMS) and
content management systems (CMS), follow this organizational model. Nearly all such systems
provide a browser-based user interface that offers entry, manipulation, and retrieval options for
the content provided by the system. Groupware solutions usually offer similar features as PIM
systems do for the local desktop, while DMSs often resemble a local fle system characterized
by a folder hierarchy where each folder contains documents. Compared to DMS, CMS offer
more fexibility to present information as they do not provide only a centralized storage but also
a presentation system.
Centralized systems that use tagging for information organization represent a hybrid approach
between centralized access and evolutionary vocabularies. Flickr is such a system where users
can share and organize images using taggings.
Another example of a partly centralized KM application is semantic e-mail processes (Mc-
Dowell, Etzioni, and Halevy, 2004). An example that illustrates semantic e-mail processes is the
organization of a meeting: In a common scenario, the initiator of a meeting frst needs to invite
potential participants to the meeting and afterward has to take appropriate actions on incoming
replies that may confrm or dismiss the proposed meeting date. Semantic e-mail processes reduce
the effort of the meeting initiator who usually needs to act on positive (a guest confrms the invita-
tion) and negative (e.g., a guest rejects the proposed meeting date) replies by automating decision
processes such as taking an appropriate action on incoming replies to the meeting proposal. Deci-
sions can be automatically taken based on a prior defnition of rules and a semantic specifcation
of the particular task, such as a meeting organization. Semantic e-mail processes rely on e-mail
servers, which represent a centralized IT infrastructure. The knowledge structure used for the
defnition of a semantic e-mail process, however, is not shared as it is only locally available by
the creator who defned the process.
Decentralized Samples
Peer-to-Peer (P2P) systems became popular through fle sharing applications such as Napster
8

or Kazaa
9
and represent the next group of KM systems we identify. The most signifcant differ-
ence to the organizational models mentioned before is the decentralization of information access.
Information is not available at a single centralized repository but is kept by multiple so-called
peers, each of which provides a portion of the information available in the network that consists
of all participating peers. Next to P2P systems, agent-based systems have also been proposed for
establishing a decentralized KM infrastructure (van Elst, Dignum, and Abecker, 2003, Dignum
and Dignum, 2003).
The Bibster system (Haase et al., 2006) is a P2P system that provides a decentralized repository
for bibliography data established by multiple local repositories. While Bibster allows each user
to maintain his own bibliography database, bibliographic information maintained by other users
is also available through a single user interface. If a user searches for a particular bibliography
entry, the local database as well as the network of connected Bibster clients is searched to retrieve
the entry. For the organization of entries, Bibster employs an additional taxonomy of research
topics that can be used to classify bibliography data by associating bibliography entries to one
or multiple entries in that taxonomy. The taxonomy and the fle format for bibliographical data
represent the knowledge structures used in Bibster. As the whole network agrees on these knowl-
edge structures, the organizational model for the knowledge structures is centralized. Compared
to systems that centralize access, Bibster refects a common local organizational model although
being networked and decentralized.
Xarop (Tempich et al., 2004) is an example of a completely decentralized knowledge sharing
system. In contrast to Bibster, where a fxed taxonomy is used to organize information, Xarop al-
lows the user to organize information with individually defned taxonomies. Thus, the knowledge
structures describing the shared knowledge are not centralized; however, they follow a globally
agreed meta model (the SWAP common model) that defnes typical concepts for shared knowledge
sources such as the concept of a fle, folder, and so on. Based on this shared meta model, Xarop
implements means to map individual taxonomies to enable the retrieval of information that is not
available locally and thus not organized by the known local taxonomy. A similar example is the Kex
P2P system (Bonifacio, Bouquet, Mameli, and Nori, 2002) that supports document search within
the network utilizing multiple individual contexts at each peer. A context is a topic-hierarchy used
to classify local content and follows a commonly agreed schema that is exploited to implement con-
text matching and utilize context information to represent perspectives onto the shared knowledge.
Both Xarop and Kex allow for defning individual knowledge structures within the boundaries of
a given meta model, thus providing higher personalization than centralized models.
K-Trek (Busetta et al., 2003) is an architecture that builds upon P2P technology to provide
knowledge for mobile users. Based on the location of mobile devices carried by the users, different
devices are available for knowledge sharing and retrieval.
Peer-to-Peer clients in the domain of fle-sharing have proved the abilities of decentralized
systems. Such systems often feel local (i.e., providing similar ease of use as systems following
a local organizational model) and combine this degree of personalization with networking.
Shortcomings of Current KM Systems
All of the systems introduced before address some challenges of KM such as personalized KM,
effciency of knowledge work or knowledge sharing, and distribution of knowledge. The overview
reveals that a generally accepted KM model that fulflls the multitude of requirements imposed on
KM does not exist and explains the near-continuous appearance of new systems that tackle various
issues in KM. Table 6.2 visualizes the classifcation of the presented systems given by the over-
view. Note that the entry for fully evolutionary approaches, Tagora, points to an integrated project
of the European Union under which a system with evolutionary access models and knowledge
structures is going to be developed. No systems that follow this model are currently available as
more research is needed to enable such systems. In the following, we enumerate shortcomings of
available KM systems that result from the organizational model used by these systems.
Omitting Knowledge Sharing
Local PIM systems often improve time management and communication by integrating data and
applications that are related to these tasks. Integration of further knowledge sources such as the
documents stored on the local fle system and additional communication channels (instant mes-
saging, blogging) is partly addressed by research projects such as Haystack or Gnowsis. The value
and potentials of knowledge sharing, however, are completely disregarded by these approaches
and denote their main shortcoming.
Single Sourced
Centralized systems lack the integration of multiple knowledge sources even more than localized
ones. They are completely disconnected from knowledge sources other than the single centralized
one disregarding additional available knowledge and the relations existing between the information
Table 6.2
Knowledge Management Systems
stored at the central repository and available at various additional sources. Knowledge of the type
write once, read never as well as process knowledge that could be captured locally can hardly
be shared with centralized models even though it may be valuable for others.
Lacking Personalization
Centralized knowledge structures prevent users from contributing knowledge as it requires them to think
in different structures than they are used to, or even in structures they reject, increasing their effort of
contribution while decreasing their motivation, their belief in and their utilization of the KM system.
Imposing Security Issues
Although a decentralized model supports the combination of multiple knowledge sources while
enabling networked knowledge sharing, the decentralized systems we introduced previously are
usually bound to one specifc knowledge source. In comparison to centralized models, access
control and knowledge sharing are costly tasks as locally defned knowledge cannot be shared as
easily and access control needs to be implemented and coordinated at each peer.
Toward Evolutionary KM
Standardization of knowledge structures in traditional knowledge sharing environments collides
with the individual needs of the user to organize knowledge according to his individual knowledge
structures. Knowledge sharing should be seamlessly integrated without requiring users to invest
signifcant additional effort into converting their work to a representation that is different from or
even contradicts their mental models. Minimizing user effort can certainly increase acceptance
and quality of the KM environment but imposes several challenges dealing with the generation,
sharing, and utilization of knowledge. The following paragraphs elaborate on such challenges and
demands toward evolutionary KM.
Integration of Multiple Knowledge Sources
To maximize the amount of knowledge that can beneft the community, any available knowledge
source should possibly contribute (e.g., fle system hierarchies provide classifcations of the
content stored in folder hierarchies, replies and responses in e-mail threads can include process
knowledge and exchange frequency can indicate user-to-user relations, bookmarks can indicate a
users expertise in certain felds, and group calendars contain knowledge about the presence and
location of persons). Furthermore, the Web represents a vast knowledge source that can be used as
indicated by the Flink project (Mika, 2005) explained in the Research and Technologies section.
While new knowledge is not necessarily benefcial as it may impose contradictions and overload,
the integration process needs to include the assessment of the quality of the integrated knowledge
and the manageability of the resulting enhanced knowledge base.
Support for the Individual
Knowledge work should not confne the knowledge worker in how she accomplishes her tasks. To
encourage the knowledge worker and increase ease of use, user acceptance, and support for knowledge
work, everyone should be able to defne knowledge as desired by using individual knowledge structures.
Most of the systems introduced previously lack the ability of providing this level of personalization in
combination with knowledge sharing support. For example, DMS often provide a folder hierarchy to
organize containing documents. The hierarchy, however, is shared by all users that share documents
with the system. Different users prefer different categorizations for their data depending on which tasks
they accomplish with that data or based on their personal preferences and mental models.
Interconnecting Information
Human memory heavily relies on mnemonics that tie various information together. Todays systems
often focus on only one knowledge source and prohibit defning and exploiting relations between
information existing in other places. Consider User A who received a text document from User B
via e-mail, saved that document on his hard disk, and edited it. At a later time, User A might want
to further edit the fle or send it back to User B but has forgotten where the fle is stored. However,
User A knows who sent the fle, but the association between User B, the fle, and its location is
often lost. Any content, whether available in local sources or on the network, should possibly
be interconnected to support human memory as best as possible to improve reuse, effciency of
search, retrieval, and perception of knowledge.
Dynamic Access Control
Access control is an important requirement in KM to secure knowledge of businesses and to ensure
privacy for the individual. Static access control becomes unsuitable when arbitrary knowledge
sources are used and when participants cannot be centrally managed and should have different
access rights for different information. Real world information access is based on trust between
people where people implicitly infer trust between two persons when they transfer information
originating from or being about one person to another person. The evolutionary KM model requires
similar features to automate access control on each peer.
The evolutionary KM model defnes a new KM scenario that deals with the dynamicity of vo-
cabularies, users, information, and access. Evolutionary KM systems improve usability allowing
ad hoc knowledge creation, providing high personalization while reducing overhead, participat-
ing in a networked KM environment by propagating or providing local knowledge, and being
capable of searching, importing, and combining knowledge available locally and on the network.
The next section explains research approaches and results that tackle the challenges raised by the
evolutionary model.
Research and Technologies
In the following, we discuss various aspects of evolutionary KM and mention several tools as
listed in Table 6.3 that represent samples of state-of-the-art research and technologies dealing with
several challenges of this KM model.
Automating the Creation of Individual Ontologies
Information management is commonly alleviated by a categorization of the information that pro-
vides a context and can be exploited in tasks such as information retrieval. A simple form of an
ontology, a taxonomy, is often used for this purpose, such as fle system hierarchies, mail folders,
and catalogs. Ontology learning denotes the generation of a knowledge structure (ontology) based
on the analysis of some content. Tools like Text-To-Onto (Maedche and Staab, 2004) analyze text
documents to automate the extraction of contained concepts and concept relations that can be used
to build an ontology that structures the information represented by such documents. Ontological
descriptions can then be exploited to improve knowledge retrieval and to combine knowledge.
Ontology learning plays an important role for knowledge preparation helping to overcome cold
start problems (i.e., to ease the transition from existing); however, ontology learning does not
achieve rich descriptions that can be exploited in exchange and reuse.
Knowledge Sharing with Individual Ontologies
Centralized systems are based on the standardization of knowledge representation in order to ensure
shareability and consistency. In the case of distributed and evolutionary KM, sharing knowledge
with others is challenged by the deviation between individual knowledge structures upon which
the formalizations of the knowledge of each participant are based. In order to establish interoper-
ability between KM systems under such circumstances, individual knowledge structures have to be
matched. Figure 6.5 illustrates one such matching scenario where two different ontologies partly
Table 6.3
Methods and Tools
overlap. Multiple research works deal with tasks referred to as ontology matching, semantic coor-
dination, ontology mapping, ontology merging, and ontology alignment. Given that two individual
ontologies have been successfully aligned, the knowledge defned with any of the ontologies can
be transferred. Quick Ontology Mapping (QOM) (Ehrig and Staab, 2004), CtxMatch (Bouquet,
Serafni, and Zanobini, 2003), GLUE (Doan et al., 2004) and PROMPT (Noy and Musen, 2000)
are samples of algorithms and applications dealing with this challenge. Different from a subsequent
alignment of ontologies, the DILIGENT methodology (Pinto, Staab, and Tempich, 2004) can be
seen as an approach to align ontologies at the time they are engineered. DILIGENT considers the
distributed engineering of ontologies and the evolvement of ontologies over time.
Combination of Knowledge
Given that multiple knowledge sources are combined by appropriate means (e.g., ontology align-
ment tools as mentioned), the knowledge contained in multiple sources needs to be combined in
order to derive new knowledge and to suffciently present it to the user. An example of knowledge
combination is given by Flink (Mika, 2005), a Web portal that aggregates information about
researchers from e-mail lists, Web pages, publication archives, and user profles. The conglomer-
ated informatione.g., researcher A cooperated with B to publish a paper P at conference Cis
exploited by social network analysis to compute new knowledge. For instance, visualizations of
the social network of a researcher are offered, and rankings of the importance of a researcher are
given. Next to Flink, further exploitations of knowledge about social networks have also been
surveyed in (Staab et al., 2005).
Web portals are Web pages that contain information from multiple sources. They commonly
provide users with a single entry-point for information that is effectively distributed. SEAL
(Maedche et al., 2003) stands for semantic portal and is an approach to improve the development
and maintainance of web portals by exploiting the advantages of the semantics and structure given
by Semantic Web sources.
Supporting Dynamic Individual Knowledge Structures
The evolutionary KM model respects individual knowledge structures that can be constantly
modifed by the user to suit his needs. Knowledge sharing under such circumstances is partly
Files
Images Papers
Rome
Files
Images Documents
Rome Papers
?
Figure 6.5 Ontology Matching Example
addressed by the semantics aware messenger (SAM) (Franz and Staab, 2005) that allows two
communication partners to classify messages utilizing user-defned tags while chatting. The
taggings beneft the retrieval of messages; for example, they allow retrieval of messages from
conversations with different persons that all deal with the same topic. SAM utilizes the taggings
of multiple users so that a user can browse messages based on the views (represented by the
taggings) of others.
Trust to Control Access
Similarly to the dynamics of knowledge structures, dynamics of access control are imposed by
distributed and evolutionary KM models: Such a networked KM infrastructure may have an
unrestricted number of participants as no centralized membership and access control exists. Con-
sequently, the network consisting of multiple local KM systems needs to defne who has access
to which knowledge and how much credibility is given to knowledge from a certain participant.
Ziegler and Golbeck (2005) developed an ontology for defning trust in general but also more
fne-grained options to defne trust bound to certain topics as one often trusts a particular person
concerning one issue but may distrust the same person on other issues. Given such trust informa-
tion, a trust network can be established that defnes how much one should trust another. Further
research toward automating such trust ratings to cope with dynamic access control is required to
establish evolutionary KM in the dimension of information access.
Automating access control and alignment of knowledge defned by individual knowledge
structures are the main challenges in dealing with the dynamics imposed by evolutionary KM.
While we here aimed to formulate our vision from a users point of view, the Semantic Web
crowd as a technical community recently proposed the Networked/Social Semantic Desktop,
which exhibits several similarities to an individual but networked KM system as rendered here.
The Networked Semantic Desktop (Decker and Frank, 2005) is a computer desktop that provides
many of the features required for local, centralized, decentralized, and evolutionary knowledge
organization combining social networking, P2P infrastructure, and local KM to establish a
novel working environment that naturally supports collaboration, knowledge exchange, and
access control (by computing trust in people and credibility of information) in a networked
environment.
CONCLUSION
We developed a model for distinguishing different IT-based KM approaches and applied this model
to state-of-the-art systems. Based on this model, we analyzed challenges, benefts, and shortcomings
of common KM systems and envisioned novel evolutionary KM approaches. Different research
contributing to such novel KM environments has been sketched subsequently.
Networking and seamless local organization are two sides of the same coin. A knowledge habitat
must provide means for evolutionary growth of smoothly organizing and accessing knowledge.
The next step, now, we believe, must come from the organizational side. This wealth of new
technologies will not easily be absorbed by current management strategies that focus on a lot of
control over knowledge fows while decentralized and evolutionary KM systems will radically
shift knowledge organization toward the edges. Currently, companies often refuse peer-to-peer
sharing, because they fear that it gives too much power to the individual and too little control to
the company, but we conjecture that the ones that frst learn to harvest the benefts of such edge
computing systems will outperform their competitors knowledge creation values.
ACKNOWLEDGMENTS
This work has been partially supported by the integrated projects X-Media (EU IST FP626978)
and Tagora (EU IST FP6034721).
NOTES
1. Being aware of the different notions of KM, we would like to clarify that the approaches described
in this chapter represent codifcation-based KM systems that enable the exchange and personalization of
information.
2. For example, Copernic or Google desktop search.
3. http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnwinfs/html/winfs03112004.asp.
4. http://www.gnome.org/~seth/storage/.
5. http://www.apple.com/macosx/features/spotlight/.
6. http://dbfs.sourceforge.net/.
7. http://fickr.com, http://del.icio.us.
8. http://www.napster.com/.
9. http://www.kazaa.com.
REFERENCES
Bonifacio, M.; Bouquet, P.; Mameli, G.; and Nori, M. 2002. KEx: A peer-to-peer solution for distributed
knowledge management. In Proceedings of the Fourth International Conference on Practical Aspects of
Knowledge Management (PAKM02), December 23, Vienna, Austria.
Bonifacio, M.; Camussone, P.; and Zini, C. 2004. Managing the KM trade-off: Knowledge: Centralization
versus distribution. Journal of Universal Computer Science 3 (10).
Bouquet, P.; Serafni, L.; and Zanobini, S. 2003. Semantic coordination: A new approach and an ap-
plication. In Proceedings of the 2nd International Semantic Web Conference, Sanibel Island, FL,
pp. 130145.
Busetta, P.; Bouquet, P.; Adami, G.; Bonifacio, M.; and Palmieri, F. July 2003. K-Trek: A peer-to-peer ap-
proach to distribute knowledge in large environments. Second International Workshop on Agents and
Peer-to-Peer Computing at AAMAS03, Melbourne, Australia.
Cole, R., and Stumme, G. 2000. CEMA conceptual E-mail manager. International Conference on Con-
ceptual Structures. LNCS, Springer, pp. 438452.
Decker, S., and Frank, M. N.D. The social semantic desktop. Available at http://www.deri.at/publications/
techpapers/documents/DERI-TR-20040502.pdf. Accessed September 2007.
Dignum, V., and Dignum, F. 2003. The knowledge market: Agent-mediated knowledge sharing. In V. Marik,
J. Muller, and M. Pechoucek (eds.), Proceedings of CEEMAS03, Prague, June 1618. LNAI, Springer,
pp. 168179.
Doan, A.; Madhavan, J.; Domingos, P.; and Halevy, A.Y. 2004. Ontology matching: A machine learning ap-
proach. In S. Staab and R. Studer (eds.), Handbook on Ontologies. Berlin: Springer, pp. 385404.
Ehrig, M., and Staab, S. 2004. QOMquick ontology mapping. In F. van Harmelen, S. McIlraith, and D.
Plexousakis (eds.), Proceedings of the 3rd International Semantic Web Conference (ISWC2004), Hiro-
shima, Japan. LNCS, Springer..
Foster, I., and Kesselmann, C. 2004. The Grid: Blueprint for a New Computing Infra-structure. San Fran-
cisco: Morgan Kaufmann.
Franz, T., and Staab, S. 2005. SAM: Semantics aware instant messaging for the networked semantic
desktop. Proceedings of the Workshop on the Semantic Desktop at the ISWC, Galway, Ireland, No-
vember 610.
Haase, P.; Broekstra, J.; Ehrig, M.; Menken, M.; Mika, P.; Plechawski, M.; Pyszlak, P.; Schnizler, B.; Siebes,
R.; Staab, S.; and Tempich, C. 2004. Proceedings of the 3rd International Semantic Web Conference
(ISWC2004), Hiroshima, Japan, November.
Haase, P.; Schnitzler B.; Broekstra, J.; Ehrig, M.; van Harmelen, F.; Menken, M.; Mika, P.; Plechawski, M.;
Pyszlak, P.; Siebes, R.; Staab, S.; Tempich, C. 2006. BibsterA semantics-based bibliographic peer-to-
peer system. In S. Staab and H. Stuckenschmidt (eds.), Semantic Web and Peer-to-Peer. Berlin: Springer,
pp. 349364.
Huynh, D.; Karger, D.R.; Quan, D.; and V. Sinha. 2003. Haystack: A platform for creating, organizing and
visualizing semistructured information. In Proceedings of the Eighth International Conference on Intel-
ligent User Interfaces, 323.
Maedche, A.; Stojanovic, N.; Studer, R.; and Y. Sure. 2003. SEmantic portALThe SEAL approach. In
Creating the Semantic Web. Cambridge, MA: MIT Press.
Maedche, A., and Staab, S. 2004. Ontology learning. In Steffen Staab and Rudi Studer (eds.), Handbook on
Ontologies. Berlin: Springer, pp. 173190.
McDowell, L.; Etzioni, O.; and Halevy, A.Y. 2004. Specifying semantic E-mail processes. WWW Workshop
on Application Design, Development and Implementation Issues in the Semantic Web, New York.
Mika, P. 2005. Flink: Semantic Web technology for the extraction and analysis of social networks. Journal
of Web Semantics, 3, 2.
Noy, N.F., and Musen, M.A. 2000. PROMPT: Algorithm and tool for automated ontology merging and
alignment. In Proceedings of the Seventeenth National Conference on Artifcial Intelligence and Twelfth
Conference on Innovative Applications for Artifcial Intelligence. Austin, TX. AAAI Press/The MIT
Press, 450455.
Pinto, S.; Staab, S.; and Tempich, C. 2004. DILIGENT: Towards a fne-grained methodology for distributed
loosely-controlled and evolving engineering of ontologies. In Proceedings of the Sixteenth European
Conference on Artifcial Intelligence. Valencia. IOS Press.
Sauermann, L. 2005. The Gnowsis semantic desktop for information integration. In Proceedings of WM
2005: Professional Knowledge Management. DFUI, Kaiserslantern, 3942.
Staab, S., and Stuckenschmidt, H. (eds.). 2006. Semantic Web and Peer-to-Peer. Berlin: Springer.
Staab, S.; Ding, L.; Domingos, P.; Finin, T.; Golbeck, J.; Joshi, A.; Mika, P.; Nowak, A.; and Vallacher, R.
2005. Social networks applied. IEEE Intelligent Systems, Trends & Controversies (March/April).
Tempich, C.; Ehrig, M.; Fluit, C.; Haase, P.; Marti, E.L.; Plechawski, M.; and Staab, S. 2004. XAROPA
midterm report in introducing decentralized semantics-based knowledge application. Practical Applica-
tions of Knowledge Management. LCNS, Springer.
Tsui, E. Technologies for personal and peer-to-peer knowledge management. CSC report. Available at http://
www.knowledgeboard.com/doclibrary/knowledgeboard/tsui_fnal_p2pkm.pdf
van Elst, L.; Dignum, V.; and Abecker, A. 2003. Towards agent-mediated knowledge management. Agent-
mediated Knowledge Management. LNCS, Springer, 130.
PART III
THE ROLE OF KNOWLEDGE
MANAGEMENT SYSTEMS IN
THE ORGANIZATION
127
ChaptEr 7
MOBILIZING KNOWLEDGE IN
A YU-GI-OH! WORLD
yoUnGjin yoo
Abstract: In the postindustrial, experience economy, simply delivering better and cheaper prod-
ucts no longer guarantees success in the market place. Companies need to fnd ways to deliver
unique, powerful, and multilayered experiences to the customers. This requires companies to
shift their attention from nouns to verbs, to deal with multiple meanings of products, to recognize
the importance of interactions and connections among different resources, and to be relentlessly
customer-centric. These changes will likely challenge much of the institutionalized management
practices that have emerged in response to the challenges of the industrial economy. Traditional
knowledge management approaches in the form of codifcation strategy and the communities of
practices that developed in the context of the institutionalized management practices are likely
to be insuffcient to support companies to design unique and multilayered experiences to custom-
ers. Drawing on the feldwork on the design practices of Frank O. Gehry and a leading design
frmIDEO, I propose interaction design as a way going forward for knowledge management
practice and research in the experience economy.
Keywords: Knowledge, Design, Experience, Experience Economy, Interactive Design, Innovation
EPIPHANY
In the summer of 2004, during my visit to Hong Kong with my family, I took my two boys to a
street market where you could fnd all types of products. They were looking for an Asian version
of Yu-Gi-Oh! cards, their favorite trading card at the time. To their delight, they found one of the
rarest cards at a local store. Little did they know that they were not real Yu-Gi-Oh! cards from the
original manufacturer. As soon as we returned to our fat, my younger boy excitedly ran to his room
and opened the deck. Within a few minutes, however, he was very upset, ready to throw away all
the cards he bought, saying, They are not real! I was a little bit surprised by how quickly my son
discovered the authenticity of the cards (or the lack thereof) and asked him how he found out. After
all, to me those cards looked just as good as the real ones that I bought for him here back home. My
son showed me a series of small numbers printed on the lower left corner of the card. According
to him, he could unlock a hidden powerful monster (one of the main characters in the Yu-Gi-Oh!
game) in his Nintendo Gameboy pack if he typed in the right combination numbers. These hidden
monsters were not known to the public when the video game packs were frst released. But, as the
company introduced a new television series or movie that featured certain monsters, they introduced
new card decks with certain cards that could unlock these monsters in the Gameboy. As my 8-year
old son was explaining the rules of the game and why he needed to gather these powerful monsters
128 YOO
in his card deck, I realized the complex and multiladen nature of Yu-Gi-Oh! experiences. They were
not just playing cards, nor just playing video games. The experiences of watching a daily television
show, playing with Gameboy videogames, dueling with each other using card decks, and download-
ing the latest cheat sheets and facts about monsters from the Internet were all interconnected. At the
heart of the commercial success of Yu-Gi-Oh! was the companys ability to mobilize and integrate
diverse knowledge resources that reside in heterogeneous communities in order to design and deliver
experiences in such a unique and multilayered manner.
KNOWLEDGE MANAGEMENT IN TROUBLE
The feld of knowledge management (KM) has emerged as an important feld for both scholarly
research and management practice over the past decade. It is not hard to fnd organizations both
small and large that have a group of people and systems dedicated to support various activities
under the umbrella of KM. The popularity of KM has gone beyond traditional for-proft organiza-
tions, reaching to government organizations at various levels (including NASA, the U.S. Army,
the Federal Reserve Bank, and state governments), international institutions (the World Bank
and the United Nations) or smaller not-for-proft organizations such as community foundations.
Refecting the growing importance of KM practice in organizations, there has been an impressive
body of scholarly work on various aspects of KM (Schultze and Leidner, 2002). Leading academic
journals, including MIS Quarterly, Organization Science, and Organizational Behavior and Human
Decision Processing, had special issues on the topic.
Despite the continuing interest among researchers and practitioners on KM, however, KM still
remains a mystic to senior executives flled with obscure terminologies and is often treated as a
marginal periphery activity at best, if not dismissed outright as irrelevant academic ideas at the
door of boardrooms. In my own experiences of working with companies, the mere mentioning
of the term knowledge management often causes disdain for many senior executives, although
they excitedly embrace the very idea behind the KMas long as it is not called KM. What has
caused such dissonance?
In this chapter, I argue that the early conceptualizations of KM are not well suited to deal with
the problems of the experience economy. While those early efforts helped companies save costs and
improve effciency, which were important virtues in management during the industrial economy,
they failed to provide new ideas and inspiration for management in the experience economy, the
Yu-Gi-Oh! world. As a result, the reality of KM initiatives based on these early conceptualizations
has not lived up to its potentialthe ideal of creating an organization in which knowledge fows
freely from one corner of the organization to another in order to create new and novel products and
solutions that delivers powerful and unique experiences to customers. In what follows, I discuss
key characteristics of KM challenges in the experience economy, followed by a brief discussion
on the historical evolution of KM. I then present two vignettes that illustrate key characteristics
of KM in the experience economy from my feldwork on work practices of one of the most in-
novative product design frmsIDEO and a world-renowned architect, Frank O. Gehry. Drawing
on the observations from these two companies, I conclude the paper by suggesting designing of
interactions as a way of going forward for the feld of KM in the experience economy.
CHALLENGES IN THE EXPERIENCE ECONOMY
The demise of the industrial economy and the emergence of a postindustrial economy has been
predicted and discussed extensively by many authors (Bell, 1973; Giddens, 1990; Pine and Gilm-
MOBILIZING KNOWLEDGE IN A YU-GI-OH! WORLD 129
ore, 1999; Zuboff and Maxmin, 2002) and I do not intend to repeat them here. Instead, I attempt
to characterize several major shifts in the economy that have signifcant implications on the way
organizations manage knowledge.
First, a key shift that defnes the experience economy is a shift from products to experiences as
the main source of value creation. As in the case of the Yu-Gi-Oh! example, what distinguishes a
product is not just its physical and material features but how the product is used to create unique
experiences. Recently, the word experience has been used extensively in order to describe
the success of innovative companies like Starbucks or Apple. It is often used in the context
of product design, corporate identity, and branding. As a result, the emphasis on experience
often leads to a discussion of aesthetic aspects of product design or corporate identity. While
these are important and essential elements of the shift toward experiences, they fail to surface
a much deeper transformation that needs to take place in management practice in order to fully
embrace the idea of experiences. As philosopher John Dewey (1934) puts it, Experience occurs
continuously, because the interaction of live creature and environing conditions is involved in
the process of living (p. 35). Experiences, then, are to be understood in terms of interaction
between an individual and products. An experience is an outcome of an action taken by an in-
dividual with an intention toward an object(s). As such, the shift from products to experiences
requires an emphasis on actions in a verb form rather than things in a noun form as the focus of
management practice (Ciborra, 1996; Weick, 1998; Yoo, Boland, and Lyytinen, 2006).
Institutionalized management practice and theory have been preoccupied with nouns as com-
pleted and whole things, instead of dealing with verbs of unfolding activities and processes. In the
industrial economy, the primary goal of organizations was to produce better products with lower
cost compared to their competitors (Porter, 1980). To the contrary, in the experience economy,
the key is not to have a better product but to create unique and powerful experiences using the
products. Thus, what organizations offer should not be the product as a fnished good but as an
affordance or action possibilities (Gibson, 1986). In such cases, what companies focus on is
not the products themselves but what customers do with them. An experience, then, is not some-
thing that is designed ahead of time and handed over to the customers. Instead, it emerges from
the interactions between the products and the customers (Heskett, 2002). In this sense, when
the management focus shifts from noun to verb, products are not presented as a fnal completed
thing. A product is a platform of potential activities. It gives birth to its meanings through users
activity in their practices.
Second, the emphasis on verbs implies another key characteristic of the experience economy.
That is, products and services have multiple meanings. Because the experience is something that
emerges out of users activities, the same products are laden with multiple meanings. For example,
portable digital music players like Apples iPod are associated with many different user activities
(listening, walking, running, driving, resting, and so on) that collectively make up the portable digital
music experience. Sometimes, new forms of activities emerge independent of a companys original
plan. iPod and other portable digital music players, combined with RSS (real simple syndication)
protocol, can be also used as a time-shifting device for content that are otherwise time dependent,
creating a completely new verb, podcast. Multiple meanings arise not only because of the different
activities associated with the product but also different contexts in which those activities take place
in practice. The same activity (say, driving) can have different meanings depending on when it is
done in the fow of activities (say, morning commute or weekend leisure driving on a countryside
road). Also, the same activity can carry different meanings depending on who else is involved in
the activity along with the focal users. For example, shopping can take drastically different mean-
ings when the shopper is alone or if she is a young mother with two active toddlers. Understanding
130 YOO
subtle differences in meaning can lead to important innovations. For example, clean clothes can
mean several different things, each of which points to a new innovation possibility. For some us-
ers, it meant no marks or dirt, but for others, it meant a fresh smell, and yet for others, it meant a
soft feeling of the fabric. For busy professionals, it meant looking sharp and ready for important
meetings. All of these insights into the different meanings of clean clothes can lead companies to
introduce several different innovative detergent products.
Third, the value of connections overshadows the value of possessions in the experience
economy. As users are more concerned about experiences than products, the ability to identify
and connect to the resources that are essential to the experiences becomes more important than
the mere ownership of the resources. For example, in the case of Yu-Gi-Oh!, it was the seamless
connections of several different facets of experiencing the trading card game that created its own
unique experiences. Similarly, the users of the Internet derive its value not by owning Web pages
that contain useful information but by being able to connect to the site. The rapid growth of net-
work technology and digital convergence (Lyytinen and Yoo, 2002) underpin the importance of
connections as a primary economic value proposition. The advancement of network technology
has enabled organizations and consumers alike to access knowledge resources that are not owned
by them. Furthermore, digitized resources can be easily disintegrated, modifed, and reintegrated
with other digitized resources, opening up the possibilities of novel experiences. Organizations
can create new value by connecting activities and knowledge resources that were previously
separated. Again, Apples iPod succeeded not just because of its superior industrial design and
engineering but also because of its ability to seamlessly bring together different resources that
were previously separated (portable digital music players, desktop software, and an inexpensive
and comprehensive digital music store). In fact, Apple outsources much of the key hardware
components of the iPod, including the MP3 chips. So, the key to successful innovations is not so
much about the possession of knowledge resources but the ability to bring their connections to
bear for seamless consumptions by the users.
Finally, the experience economy is relentlessly customer focused (Pine and Gilmore, 1999;
Zuboff and Maxmin, 2002). The notion of mass market with standardized products is being rapidly
replaced by mass customization. Companies like IBM, UPS, and GE now emphasize combining
products with high-valued services in order to deliver solutions, such as an automation systems,
a telecommunication network, or a logistic system, rather than products to solve a customers
problem (Foote, Galbraith, Hope, and Miller, 2001; Sawhney, Balasubramanian, and Krishnan,
2004). More than 50% of IBMs revenue originates from delivering custom solutions. The radical
importance of customers in the experience economy requires much deeper, holistic and focused
knowledge about individual customers. Instead of focusing on the competencies of the frm and
its internal operational effciencies, companies need to focus on fulflling unique needs of indi-
vidual customers. This demands a radical departure from the organizing logic in the industrial era
where organizations attempted to minimize the variety, and thus complexity, in order to maximize
the internal operational effciency. A key managerial challenge that emerges from this relentless
customer focus is the tension between the logic of variety and the logic of unity. The question
here is how to sustain a constant fow of customized solutions, while maintaining the identity and
coherence of the organization (Davies and Brady, 2000; Yoo, Boland, and Lyytinen, 2006).
These four characteristics of the experience economythe shift from noun to verb, multiple
meanings of products, the importance connections, and customer-centric focusare not meant
to be exhaustive and comprehensive. Instead, these are key characteristics that have signifcant
implications for KM activities in organizations. They require KM that supports a dynamic con-
fguration and reconfguration of various forms of knowledge resources that belong to different
communities of practices. They require KM practices that do not attempt to reduce different
perspectives, but instead celebrate the diverse perspectives. They also require KM systems and
practices that support knowledge sharing not only among those who belong to the same com-
munities of knowledge and share identity but also among those who do not share such identity.
The richness of the experiences of products in the experience economy comes from the dialectic
transformation of diverse perspectives from different communities of knowledge and KM systems
and practices needed to support such dialectic and transformative process (Carlile, 2002). Finally,
KM practice will have to allow managers to manage tensions between variety and unity. In the
next section, I review the existing approaches to KM and how they fail to address these emerging
needs in the experience economy.
A BRIEF HISTORY OF KNOWLEDGE MANAGEMENT
Because there are several excellent reviews of the KM literature (Alavi, 2000; Alavi and Leidner,
2001; Schultze and Leidner, 2002), I will not focus on that here. Instead, at the risk of being overly
simplistic, I will try to give a narrative account of the rise and the fall of different generations of
KM as they are refected in practice and research in order to point at the inherent limitations of these
earlier efforts in KM in dealing with complex knowledge problems in the experience economy.
One of the earliest pioneers of KM was Ikujiro Nonaka (Nonaka, 1991; Nonaka and Takeuchi,
1995). In his classic Harvard Business Review article, The Knowledge-Creating Company,
Nonaka argues that the superior performance of Japanese companies in the past can be attributed
to their ability to transform, transfer, and reuse tacit knowledge from one individual to another
individual within the frm. He draws on the work of Polanyi (1966) in order to argue that there
are two types of knowledge: explicit and tacit. Although such an attempt to separate explicit and
tacit knowledge was not consistent with Polanyis original idea of tacit and explicit being two
fundamental dimensions of knowledge, the idea of transforming tacit knowledge into explicit
knowledge through a codifcation process became quickly popular. Many frms implemented a large
knowledge repository, and information technology vendors and consultants introduced a variety of
tools that were aimed at addressing these needs. Often, frms KM practices were wrapped around
those repositories as the central enabling technology, encouraging employees to document best
practices and share them through these repositories (Zack, 1999).
In addition to the problematic conceptualization of tacit knowledge, the codifcation strategy
of KM raised several other challenges for companies that tried to implement it. First, it was
much easier to create demand for knowledge than supply. Several leading companies, including
consulting frms who were the main proponents of this approach, responded by creating various
incentive systems designed to encourage employees to contribute to the repository. Second, the
use of powerful incentive systems led to information overload and employees struggled to fnd
the golden nuggets of knowledge that could really help them solve their problems (Hansen and
Haas, 2001). Third, the codifcation approach inevitably emphasizes the use of a noun rather than
a verb, as it focuses on things that are stored in the repository. Finally, because the codifcation
approach emphasizes the reuse of existing knowledge (Markus, 2001), it was not well suited for
a situation where new knowledge needs to be created (Hansen, Nohria, and Tierney, 1999; Yoo
and Torrey, 2002).
Drawing on the sociology of knowledge and practice theory, a group of scholars have noted the
social nature of knowledge (Brown and Duguid, 1991; Lave and Wenger, 1991; Wenger, 1998).
They point out that knowledge resides in a community of practice and thus, companies need to
nurture and support communities of practice in order to leverage the knowledge resources in the
132 YOO
organization. Story telling, boundary objects, and boundary spanners are often recognized as key
mechanisms for knowledge sharing and transfer in organizations. They emphasize the importance
of identity, language, and work practices as essential elements of sharing knowledge. Around the
time when companies began to lose their interest in codifcation-based KM approaches due to
the recognition of their limitations, the idea of communities of practice as a new way of manag-
ing knowledge rejuvenated the KM activities in companies. The same consulting companies that
advocated the combination of a knowledge repository and codifcation as KM solutions began to
introduce new KM solutions with the idea of virtual communities of practice. Virtual communi-
ties of practices are often created by connecting employees who share similar interests through
various forms of electronic communication tools such as intranet and portal. This new approach
to KM centered on the communities of practice overcame many of the limitations of the codifca-
tion-based approaches to KM. However, its inherent emphasis on communities as the main focus
of KM limits the usefulness of such approaches for the companies who need to design and deliver
novel products and services that refect the emergent needs of the experience economy.
A strong identifcation with a focal community can facilitate the knowledge sharing and
learning within the community. However, strong communities of practice can be potential ob-
stacles for creating products laden with multiple meanings that can provide dynamic and unique
experiences to the customer, as it will be harder to share and coordinate knowledge and expertise
across the boundaries of these communities that defne language, norms and rules of each com-
munity. Furthermore, communities of practice often develop an internal focus, which makes it
diffcult for the members to appreciate what the customers truly want from the company (Zuboff
and Maxmin, 2002). What is necessary here is KM practicesa combination of tools, methods,
and attitudesthat enable organizations to discover customers unique needs and deliver solu-
tions that address those needs by mobilizing knowledge resources from different communities
of practices. Therefore, the tools and methods that support such KM practice should not attempt
to reduce multiple and often conficting perspectives from different communities onto a singular
framework. Instead, these new tools and methods should bring out such differences in perspectives
and dialectically transform them in order to design and deliver unique products and experiences
that are laden with multiple meanings. In this sense, organizations that embrace new KM practices
are always designingdesigning new solutions for the customers and designing confgurations
of knowledge resources that are necessary to deliver the solutions. And, these two designs are not
separated from each other. Instead, they are reciprocally constitutive. The dynamic and recipro-
cal processes that connect these two aspects of design are the essence of new KM practice that I
propose for the experience economy.
In the following section, I describe vignettes of two companies that follow such KM practices.
The frst vignette is from IDEO, a leading frm of industrial designers, who are known for their
innovative products such as the original Apple Mouse and the Palm Pilot V, as well as their in-
novative management culture and processes (Kelley and Littman, 2001). Recently, IDEO began
helping its clients in their own efforts to improve innovations, applying the same methods and
principles that it uses for product design (Coughlan and Prokokopoff, 2004). I followed their
projects on and off for more than a year through ongoing conversations with the senior principals
of the frm who lead its organization design practice, site visits, and examinations of archival
data. I report here on how they work with large hospitals that are trying to refocus their energy to
provide better patient service.
The second vignette is from Gehry Partners, LLC, a leading architecture frm, that is known
for the dramatic designs of its principal designer, Frank O. Gehry, which include the Guggenheim
Museum in Bilbao, Spain, Peter B. Lewis Building at Case Western Reserve University in Cleveland,
Ohio, and the Disney Concert Hall in Los Angeles. I have been studying Frank Gehrys design
practice for 4 years, tracing the innovations in construction and engineering associated with his
unique building projects, which use three-dimensional digital representations in their design and
construction.
1
I report here how Gehry and his associates mobilize knowledge resources for each
project differently in order to achieve their purposecreating a building that provides unique
experiences to those who live in and see it.
TWO VIGNETTES
IDEOFrom Organs to Patient Journeys
IDEO is a leading design frm, well known for its innovative product designs as well as its in-
novative design process (Kelley and Littman, 2001). Recently, IDEO began helping its clients
in organization design by using its methods for product design. According to Peter Coughlan
and Ilya Prokopoff, who lead the organization design practice for innovation in the frm, IDEO
initially became interested in helping their clients to organize for innovations when they noticed
that some of their clients were not able to successfully use the products that IDEO helped them
design because of the ineffective organization design of their clients (interview, October 4, 2002).
When IDEO designs new products, much of their attention is given to the customer experience of
the product, rather than the product itself. They found that, however, their clients often failed to
deliver such experiences because of the functional silos in their organizations. Customer experi-
ences include physical, cognitive, emotional, and social elements, and it takes a multidisciplinary
approach to effectively deliver such a totality of customer experiences. Coughlan and Prokopoff
found that many organizations found it diffcult to deliver customer experiences from such a ho-
listic perspective due to the way the client organizations are structured.
The IDEO design process, called Deep Dive, consists of several tools and methods including
empathic observations, brainstorming, rapid prototyping, visualization, and narratives among oth-
ers (IDEO, 2001; Kelley and Littman, 2001). These tools are organized into a few iterative steps
that include understand, observe, synthesize, visualize, realize, evaluate, refne, and communicate.
These design methods are designed to focus on customer experiences and identify necessary
knowledge resources in order to deliver such experiences.
Recently, IDEO began extending the application of these methods and tools to their clients,
particularly in the health care industry, to help them fnd ways in which they can mobilize diverse
knowledge resources to address dynamic and unique patients needs. Such design work with their
clients involves several days of intensive workshops, using their Deep Dive process involving key
stakeholders of the organization as well as line workers (nurses, physicians, and administrative
staff in the case of health care organizations). Traditional hospitals are organized based on medical
specializations, which are primarily based on the human organs that each specialty is concerned
with. Yet, one thing that is nowhere to be found in that organization design is the patient. Just as
their product design practices begin and end with customer experiences, IDEOs approach to ad-
dressing the hospitals problems also starts and ends with patient experiences.
The process begins with participants of the workshop brainstorming the known issues, chal-
lenges, and opportunities. These ideas are organized into what they call a Patient Journey Map,
which shows a patients experience of moving through the hospital for key processes, along with
many touch points with different specialties in the hospital from the patients perspective. This also
identifes other actors who are involved in the journey, such as family and friends of the patients,
whose experiences signifcantly infuence those of patients. After the journey mapping, the Deep
134 YOO
Dive participants immerse themselves in the actual performance of the organization by following
patient journeys with empathic observations that they hope will reveal implicit assumptions and bias
built in the organizational routines and at the interfaces across different functional silos. In the next
step, the participants in the workshop synthesize the ideas and observations that have emerged in
the previous stages, and use them as a guiding logic in designing new roles, relations, interactions,
facilities, and technological artifacts. Combining rapid prototyping, visualization, and role-playing,
they narrativize the envisioned future in a concrete context. They experiment with these design
ideas by trying them out in a real setting and use feedback from such experimentation to continue
refning the design. Several observations about KM can be drawn from their approach.
First, IDEO helps its clients radically focus their attention back to the purpose of their existence,
providing services to their customer. In the case of health care, it is the experience of patients that
they bring to the attention of their clients. Their use of conceptual tools such as Patient Journey is
a good example. This customer-experienceoriented approach to design enables the organization
members to reexamine the roles and relationship among various functions and encourages them
to try out new ways of confguring those interactions in order to create a holistically functional
customer journey. Often this would require new roles, relations, and tools be designed in order to
mobilize knowledge resources in providing a holistic experience for the customer. Existing orga-
nizational rules, norms, and processes that used to sustain the specialized skills and knowledge
are replaced or complemented with new rules, roles, and artifacts that provide a customer-centric
perspective. In this sense, things that are designedroles, relations, and toolsare not the fnal
completed objectives of their design. Instead, these artifacts are designed in order to shape certain
actions and interactions among users and the members of different communities. IDEO found
that bringing the focus of designing back to patients becomes an important source of emotional
energy for those who work with them. Even though the people who work in a client hospital are
often frustrated by the broken system and the lack of trust they feel, they continue to work in
the hospital because of their personal and professional passion for the patients, which is offcially
dormant in the highly specialized and compartmentalized work of health care. It seems as though
a radical focus on the patient can revitalize this dormant passion among health care professionals,
which in turn becomes the ongoing source of mobilizing knowledge transcending the boundaries
of professional communities in health care, even after the IDEO Deep Dive process has ended.
Second, IDEOs approach to innovation explicitly contains a reciprocal cycle between unity
through global and standardized knowledge and variety through situated and localized knowl-
edge. Often, organization design efforts are driven by top management, along with a few external
consultants and organization development experts. Coughlan makes the point that even though
he repeatedly works with hospitals, he makes a conscious effort not to impose his own ideas and
experiences from previous engagements onto the current project. Although he and his team may
follow the same process and method to facilitate their organization design efforts, the organization
design always emerges from the clients. Most important, the design ideas and the possibilities for
offering new and better experiences to a patient often come from the people who actually interact
with the patient during their journey. IDEO uses various contextualized observation tools, such as
shadowing, day-in-the-life, and photo (or video) journeys in order to stimulate new design ideas.
These ideas are then discussed in brainstorming sessions to become more powerful and crystallized
into design logic for mobilizing knowledge resources for the patient care services. Ideas identifed
early in the process are used to guide later phases of the design effort, including prototyping and
visualization. This represents a quick cycle of a reciprocal circle between unity through global
knowledge and variety through situated knowledge. These cycles of moving between local and
global logics continue throughout and, hopefully after, the process. This process is essential in
creating the products and services laden with multiple meanings as the global design logic that
emerges from the process contains the perspectives of different professional communities who
participated in the process.
Third, IDEO emphasizes the importance of prototyping and visualization in their design efforts
(see Figure 7.1). Drawing on observations and brainstorming ideas, client members working with
an IDEO team develop concrete scenarios of future customer experiences. This includes not only
the prototyping of physical artifacts but also the enactments of potential future processes to explore
possibilities for new interactions and relationships. This provides a concrete basis for evaluating
the effcacy of design ideas before being implemented. It is also important to note that they heav-
ily emphasize physical artifacts in organization design. They consciously use physical artifacts
such as information technology, building layouts, carts, furniture, clothing, and even signs in the
building in order to create the types of interactions with patients that lead to desired experiences
(interview, November 11, 2003). This is signifcantly different from traditional organization design
practices done at an abstract levelon an organizational chart. Instead, IDEO leverages physical
artifacts in order to reconfgure a social infrastructure. This again shows IDEOs focus on verbs,
not nouns, even as they design physical artifacts.
Using tools and methods developed for designing products that provide desired customer expe-
riences, IDEO hopes to shift the focus of clients from things to their ability to continue to design
desirable experiences for their customers. IDEO encourages its clients to think about design as a
way of giving form to a more desirable experience for its patients. It means giving birth to new
forms of interactions among different functions in the hospital as part of putting a desired patient
experience out into the world, one that is experienced not only by the patients themselves, but
also by their family and friends, because health care is a collective social experience. What the
patients experience is multilayered unique cares that are delivered through the mobilization of
knowledge resources that are previously disconnected. Therefore, this also means giving birth to
new forms of interactions internally to the hospital to mobilize diverse knowledge resources by
continuously searching for ways to create better customer experiences.
Frank O. Gehry: From Buildings to Architecture
Over the past 4 years, I have been privileged to study the design practice of Frank O. Gehry and
his frm, Gehry Partners, LLP. Their design practice is as unique as the buildings they create,
which are exemplifed by the Guggenheim Museum in Bilbao, Spain, the Peter B. Lewis building
at Case Western Reserve University, and, most recently, the Disney Concert Hall in Los Angeles.
His buildings create powerful and complex experiences to those who own, use, and visit the build-
ings. I will focus here on how they mobilize knowledge resources from the members from diverse
Figure 7.1 Prototypes by IDEO at various stages
136 YOO
and different trades each time they build buildings in order to draw observations with regard to
KM practices in the experience economy [see Yoo, Boland, and Lyytinen (2006) for a detailed
description of their design practices].
First, Gehry Partners do not design their buildings on paper but by making physical models.
They literally construct hundreds of models during every project. The models are made of what-
ever is at handwood, metal, foil, plastic, cloth, or paperand are purposefully made in a rough
unfnished manner to emphasize that they are tools for thinking and ways of exploring ideas, not
representations of well-formed ideas (see Figure 7.2). The multiple models they create during
the design of a building project are not intended as things, or design as a noun, but as a physi-
cal artifact that enables a process. That is, they are part of ongoing design as a verband each
model is a snapshot of their designing as an ongoing unfolding action. These models are tools
for designing, not just the fnal representations of his design ideas. Gehry uses multiple physical
models with different scales and materials in order to explore different facets of their approach to
the design problem, because each model reveals different characteristics of the emerging design.
Contrary to modern management practices that divide the human experience into segmented areas
of functional operation, and conveniently reduces them into abstract, decontextualized, and partial
representations, Gehrys design practice centers on keeping the totality of the human experience.
Multiple physical models, drawings, sketches, and three-dimensional computer models are all part
of his efforts to evoke and respond to these human experiencescognitive, cultural, functional,
and emotional. In this way, his design approach allows for multiple voices to be heard, each voice
speaking to a different aspect of human experiences. Gehry and his associates are committed to
resist the temptation to collapse these multiple voices into a single one. Instead, they celebrate
these voices in their unique ways and let them speak about the functional requirements of the
design problem in their own way.
Focusing on design as a verb allows for playful interactions among different materials, mod-
els, ideas, and alternatives. It is this spirit of playfulness that brings the energy and emotion to
individuals involved in the process. At the same time, Gehrys design process seeks to realize the
possibilities of an idealized dream. In order to capture his dream image, he draws (see Figure
7.3). His equivocal and evocative sketches provide glimpses of the dream image he is searching
for, and by not bridling multiple voices or playfulness, the design emerges by drawing emotional
energy from them. For this reason, he continues to draw even during the construction process in
order not to lose his dream image.
Figure 7.2 Multiple Models and Scales

Second, Gehry is famous for his use of the most advanced three-dimensional software tools from
the aerospace industry to digitize their physical models and begin working out details of how their
complex structures might be built (see Figure 7.4). The software is built on a centralized database
that contains full-scale digital three-dimensional models of the building, using XYZ coordinates.
Using the software, the members of different trades can add different layers of the design of the
building with detailed local design specifcations. As all the layers are coordinated through the
same XYZ coordinate systems, the three-dimensional model with multiple layers allows multiple
perspectives to coexist in a single design (Yoo, 2005).
Third, Gehry Partners design process starts with an intensive period of working with a client
to understand the desired functionality for a building (its program), including the desired social
dynamics, sense of identity, and cultural feel, as well as the activities, tasks, and uses of building
spaces. Throughout the process, Gehry Partners are consciously aware that the client will have
Figure 7.3 An Early Sketch of the Lewis Building
Figure 7.4 Examples of 3D Models
138 YOO
some strongly held beliefs about what they want and what their building should be like but are
also aware that the client is likely carrying these ideas forward from their past experiences without
suffciently refecting on what it is that they are really trying to accomplish with the project, and
what is possible to invent beyond their current notions of how to best achieve their goals. Gehry
Partners are committed to the belief that our current ways of understanding what we are doing,
what is possible to do, and the ways of doing them are all open to radical improvement. As Jim
Glymph, Gehrys senior partner says, We always try to reinvent the wheel (interview, December
7, 2004), meaning that they consciously return to what appears to be well-established practices
and understandings in order to question them, as a child would. He continued, Fourteen out of
ffteen times we will fail to come up with something better, but that ffteenth time could be a real
breakthrough. Gehry himself refers to this as an attempt to keep the design problem liquid and
to resist allowing it to crystallize. In their practice, they encounter a constant pressure to crystal-
lize ideas and move on, and so they must willfully resist. Gehry speaks of purposefully trying to
make the models and pictures that will be viewed by a client as rough and unfnished as possible,
to emphasize the liquid state they are trying to maintain in their thinking (Gehry, 2004).
Finally, Gehry Partners organization structure is radically project based (Yoo, Boland, and
Lyytinen, 2006). Although project-based organization structures are not uncommon in the con-
struction industry (Gann and Salter, 2000), Gehry Partners is unique in that they have no standard
methods or forms for organizing their projects, allowing each one to emerge from the unique cir-
cumstances and actors involved. They emphasize the need for each project structure to uniquely
enable a genuine dialogue among all the parties. A genuine project dialogue is centered on form
giving, or placing the dream idea as a realitya remarkable buildinginto the world. The project
then becomes a source for their ongoing organizing and discovering their own identity, with each
element of the project organization being designed as part of the form giving process of dialogue
unique to that project and its resources, constraints, actors, and goals. As Glymph put it, The
organization structure does not cause the success of a project, instead, the success of the project
is the cause of its organization structure (interview, December 7, 2004).
At the same time, it is important to note that they do have a skeleton for organizing each project,
which guides the organizing process, and that skeleton is based on the premise that the project
should refect an ideal form of collaboration in which all parties are on an equal footing, able to
raise fundamental questions about the real purpose of the project, and how to best achieve it. An
important early project with three-dimensional software provides the frm with a concrete example
of the ideal form of collaboration they seek in each project. This project was the construction of a
large Fish Sculpture on the Olympic grounds of the 1992 Barcelona Olympics. The project was
completed successfully under extremely tight time and budget constraints, and enabled them to not
only gain a positive experience with using fully three-dimensional software throughout a project,
from beginning to end, but also experiment with a project organization in which the hands-on
actors drove all aspects of the project. According to Glymph, We literally agreed to attempt it
only if the project managers were left out of it and the whole management process was abandoned
because it was too slow. We made some deals with the city on approvals, a lot of deals were made
to operate outside the rules and basically, most of the deals had to do with peeling away layers of
oversight and management and getting down to just the people that are doing the work. And then
putting only the absolute minimum amount of management back on top of that to keep it from
going off the rail (interview, November 9, 2002). The successful experience of the fsh project
provides them an idealized form of organizing to seek on every projecta dialogue among the
actors engaged in doing the work as equals. Glymph notes, they havent done a paperless project
again, because weve not had that environment where we were committed to suspend all the rules.
And so that [Barcelona Fish Project] still remains the most effcient, the most successful use [of
3D digital tools] and it was the frst use (interview, November 9, 2002).
A WAY GOING FORWARD: IMPLICATIONS FOR
Work practices of IDEO and Gehry Partners show the possibility of radically different KM
practicesefforts to design unique and multilayered experiences through mobilizing knowledge
resources from diverse communities. In the world of Yu-Gi-Oh!, where the value is created not
through making better and cheaper products but rather through providing unique and multilayered
experiences to customers, their design practices provide powerful inspiration for KM practices
in the experience economy.
These are two exceptional organizations that invent ingenious products and services and
construct buildings that often seem to defy gravity. They consistently surprise their customers,
competitors, and themselves through their ability to produce novel and innovative products and
services that create powerful and unique experiences to those who encounter their products and
buildings. Such unique experiences are created as they are radically committed to designing new
forms of interactions each time they engage in their work. Both organizations design physical
artifacts, not as the fnal completed things but as a way to support actions and interactions that
will produce desirable experiences.
As human experiences come from interactions with the surrounding (Dewey, 1934; Lakoff
and Johnson, 1980), the core of the design focus of these two frms is interactions, which then
guides and shapes experiences of customers and those of the various members of the organiza-
tion. In this sense, I propose interaction design as a way going forward for KM practice in the
experience economy. Interaction design is an emerging interdisciplinary feld of inquiry to study
how computer systems and tools become ordinary everyday tools for social practices for living,
studying, working, playing, and learning (Buchanan, 2004; McCullough, 2004). Although it shares
parts of its intellectual root with the feld of humancomputer interaction that studies effective
user-interface design, the emerging feld of interaction design has a much broader scope and
somewhat different focus. The traditional approach to humancomputer interaction design based
on desktop computing technologies narrowly focuses on the computer screen design to improve
the effectiveness and effciency of computer usages. However, as computing and communication
technologies become increasingly mobile and embedded into everyday artifacts (such as cars,
music players, and clothes) and surrounding environments (Lyytinen and Yoo, 2002; Mitchell,
2003), the interactions between computer and users are increasingly taking place outside of the fat
screen and keyboard (Dourish, 2001). As such, the idea of human-interaction design has expanded
to understand the emergence and sustenance of everyday social practices and the role of physical
artifacts in them. Thus, the interaction design is also infuenced by architecture, communication
design, and product design. While the humancomputer interaction focuses on the design of arti-
facts, the discipline of interaction design focuses on the design of human experiences and actions
that were mediated by technology artifacts. It approaches the complex design challenges through
the lens of human experiences.
At the heart of their work practices, we fnd two aspects of interactions design mirroring each
other. When looking at these organizations from the inside out, one fnds their efforts to design
interactions without in order to deliver unique and multilayered experiences to the world. What
IDEO designs for its hospital clients are the interactions between the hospital (and its staff) and
its patients and their friends and family members. Buildings are ways in which Gehry interacts
140 YOO
with the owner and the people who live in and visit his buildings. These interactions in both cases
are instantiations of idealized forms of experiences that the designers (both IDEO and Gehry
Partners) envision. What they design as things are mediators of these interactions that shape and
create interaction possibilities.
At the same time, when looking from the outside in, one fnds that IDEO and Gehry Partners
constantly engage in designing interactions within among different communities of practice.
2
In
the case of IDEOs work with hospitals, it is the interactions among different professional com-
munities in medicine to save lives. In the case of Gehrys work, it is the interactions among the
members of different trades to build the building. Both frms refuse to accept standard operating
procedures, constantly seeking to reinvent the wheeleven though the previously designed wheel
might have won critical accolades.
In this sense, it is the deep commitment and passion to design new forms of interactions outside
to the world that provide cognitive, emotional, technical, and moral boost for the designing and
mobilizing interactions withinas it requires constant confgurations and reconfgurations of re-
lationships among many different communities of practice who do not share the same goal or the
same language. In fact, it is precisely because they do not share the same identity and perspectives
that these different communities are needed. Contrary to the traditional KM approaches that are
primarily focused on the relationship within a community or a frm, this new approach to KM always
starts with the designing of experiences for the customers that shapes and guides the interactions
within among different communities. Thus, designing interactions without and within needs to
be the core element of KM in a knowledge-based economy. Future KM tools and techniques then
need to support the reciprocal relationship between both sides of interaction designs.
An important conceptual development related to this is the idea of the pragmatic use of bound-
ary objects (Carlile, 2002). The use of interaction design approaches allows products of IDEO
and Gehry Partners often laden with multiple meanings. New roles and relationships designed
by IDEO for their hospital clients refect multiple and complex functions that they need to per-
form. Gehry is always dealing with multiple meanings of space as he designs his buildings. A
key strength of their approaches is their inherent ability to bring forth multiple perspectives and
voices in the design process, and their commitment to resist the temptation to reduce them into
a simple singular perspective and voice. For this purpose, they use many tools and techniques
that are common in the interaction design disciplineobservation, rapid prototyping, visualiza-
tion, narrativization of scenario, participatory design, multiple models, and three-dimensional
visualization tools, among others. Finally, both IDEO and Gehry demonstrate relentless focus on
customers as the starting point of their design inquiry. The engagement with the customers is a
way in which both organizations discover their identities. In this sense, their identity does not give
birth to their products. Instead, it is the very process of designing new products that gives birth to
their newly discovered evolving identity. Both organizations are keenly aware of the importance
of maintaining the reciprocal cycle between variety through localized and situated knowledge and
unity through global and standardized knowledge.
Another important element of knowledge that one can draw from IDEO and Gehry Partners is
the mobilization of diverse knowledge resources from different communities of practice through
the commitment to fght against the tyranny of or and to embrace the world of and. This is a
stark contrast to the typical KM approaches that emphasize the importance of shared perspective,
language, and identity. This also puts the current popular approach to KM using best practices on
its head. Future KM tools and techniques should be able to support multiple and often confict-
ing perspectives from different communities. The way Gehry Partners use the three-dimensional
visualization software with layers of models with a common coordinate system is a good example
of such tools. The conscious and deliberate use of multiple models by both frms is also another
good example of how different tools can be used for KM in the future. Emerging tools like WiKi
allows multiple perspectives to be incorporated within a single framework and can be used to sup-
port KM practices that celebrate the diversities of perspectives. No tools and techniques, however,
can have a fundamental effect on KM unless there is a commitment in the organization, preferably
from top management, to respect and celebrate different perspectives in the organizations.
Finally, the lessons from these two companies suggest the importance of a reciprocal cycle
between variety and unity. From the standpoint of KM, this is related to the development of meta-
knowledge. Meta-knowledge provides a critical backdrop of organizational unity, which enables
the dynamic and evolving variations in the foreground through bricolage, improvisation and
emergence. Various techniques such as narrativization of the future, prototyping of interactions
with physical artifacts, and story-telling are being used by IDEO and Gehry Partners in order to
develop and sustain meta-knowledge. Over time, such meta-knowledge will constitute a design
gestalt for both products and organization structure (Yoo, Boland, and Lyytinen, 2006).
In addition to these key characteristics, these two examples offer a set of methods that are
commonly used in the interaction design discipline (IDEO, 2001; Buchanan, 2004) including
observation, visualization, rapid prototyping, and narrativization of scenarios. These tools help
designers to focus on verbs and their meanings in the context. Most of them are unusual tools
for KM practices. And, by no means is this an exhaustive list of such tools. However, in order to
better understand the multiple meanings of users activities, companies must use these tools as
part of their KM practices.
IMPLICATIONS FOR FUTURE KNOWLEDGE
MANAGEMENT RESEARCH
The idea of KM practices focusing on the design of interactions within and without the frm also
suggests several new streams of research in the future. First, we need much better and refned
understanding about how we design interactions both within and without the frms boundary.
Such design activities are necessarily sociotechnical in their nature. We design technological and
organizational artifacts that will shape and mediate interactions among human actors. Such in-
teractions involve many different dimensions: physical, cognitive, communicative, technical, and
affective, just to name a few. Much more careful theorization and empirical works need to be done
in order to better understand the impact of different material characteristics of different artifacts
in KM practices. For example, emerging technologies based on Web-based service architecture,
often labeled as Web 2.0 technologies, is likely to play an important role in shaping interactions
among different actors from different communities, as these tools allow the storing and control
of contents in the local level, without compromising the ability to mobilize and coordinate these
distributed contents in order to address ad hoc knowledge needs. Yet, it is not clear how such
radically distributed technologies will be accepted in organizations that follow more centralized
organizing and coordinating logics.
Second, current interests in transactive memory in the KM literature (Hollingshead, 1998;
Moreland, 1999; Yoo and Kanawattanachai, 2001) are an important step in better understanding
how knowledge is mobilized across the boundaries of different communities. Also, the ongoing
research on boundary objects and its emphasis on translation of interests and representations can
shed important light on the process of knowledge mobilization (Star and Griesemer, 1989; Carlile,
2002). Yet, both of these streams focus on the cognitive aspect of knowledge mobilization. However,
the practice of knowledge mobilization is more like mobilization of collective actions for social
142 YOO
movements. This involves not only the cognitive aspect but also emotional and political aspects.
Future research on KM needs to develop a more refned theoretical model based on empirical study
examining the process of mobilization of knowledge resources in a more holistic way.
Third, of course, whether KM tools, methods, and practices based on principles of interaction
design will outperform more conventional KM approaches is an important empirical question to
examine. An important challenge for addressing this question is how to measure the outcome of KM
practices. Traditional KM initiatives were typically measured for their effectiveness in saving cost
and instituting consistency and reliability. Often these measures are internally focused, and studies
relying on such measures are likely inadequate to measure the impact of KM on the frms ability
to design and deliver creative products and solutions. As such, developing and validating metrics
of KM practices based on interaction design will be the necessary frst step in this direction.
Finally, if an organization decides to apply these ideas from interaction designs to change its
KM practices, one needs to study how such a change process can take place. Careful case stud-
ies of frms that are going through changes in order to deal with the challenges in the experience
economy would be an appropriate next step. In such case studies, it will be necessary to carefully
document how the frms strategic direction, internal organizing logic, and their use of information
technology refect their struggle to design and deliver new products and services in the experience
economy. An interesting challenge there is to study the mechanisms by which frms resolve the
contradictory challenges of variety and unity. Yoo, Boland, and Lyytinen (2006) suggest design
gestalt as one such mechanism by which organizations can handle such challenges. More research
examining other such organizing mechanisms to deal with KM challenges in the experience
economy is needed. The role of leadership, organizational culture, and history in the process of
organization change are also important topics that need to be examined.
The world of Yu-Gi-Oh! is here to stay. Its commercial success demonstrates the importance of
delivering experiences to the customer that are laden with diverse, yet interconnected, activities.
In order to design and deliver such experiences, companies must embrace interaction design as
a new approach to KM.
NOTES
1. This project was conducted in collaboration with Richard Boland and Kalle Lyytinen with support
from the National Science Foundation (No. IIS-0208963).
2. Here, interaction within includes not only the members of the frm but also other critical members of
production ecologyincluding suppliers, vendors, and other trading partnerswho contribute to the design
and delivery of the products and services.
REFERENCES
Alavi, M. 2000. Managing organizational knowledge. In R.W. Zmud (ed.), Framing the Domain of IT
Management: Projecting the Future from the Past. Cincinnati, OH: Pinnafex Educational Resources,
pp. 1528.
Bell, D. 1973. The Coming of Post-industrial Society: A Venture in Forecasting. New York: Basic Books.
Brown, J.S., and Duguid, P. 1991. Organizational learning and communities of practice: Toward a unifed
view of working, learning, and innovation. Organization Science, 2, 1, 4057.
Buchanan, R. 2004. Interaction pathways in organizational life. In R.J. Boland and F. Collopy (eds.), Manag-
ing as Designing. Stanford, CA: Stanford University Press, pp. 5463.
Carlile, P.R. 2002. A pragmatic view of knowledge and boundaries: Boundary objects in new product devel-
opment. Organization Science, 13, 4, 442455.
Ciborra, C.U. 1996. The platform organization: Recombining strategies, structures, and surprises. Organiza-
Coughlan, P., and Prokokopoff, I. 2004. Managing change, by design. In R.J. Boland and F. Collopy (eds.),
Managing as Designing. Stanford, CA: Stanford University Press, pp. 188192.
Davies, A., and Brady, T. 2000. Organisational capabilities and learning in complex product systems: Towards
repeatable solutions. Research Policy, 29, 931953.
Dewey, J. 1934. Art as Experience. New York: Perigee Books.
Dourish, P. 2001. Where the Action Is: The Foundation of Embodied Interaction. Cambridge, MA: MIT
Press.
Foote, N.W.; Galbraith, J.; Hope, Q.; and Miller, D. 2001. Making solutions the answer. The McKinsey
Quarterly, 3, 8493.
Gann, D.M., and Salter, A.J. 2000. Innovation in project-based, service-enhanced frms: The construction of
complex products and systems. Research Policy, 29, 955972.
Gehry, F.O. 2004. Refections on designing and architectural practices. In R.J. Boland and F. Collopy (eds.),
Managing as Designing. Stanford, CA: Stanford University Press, pp. 1935.
Gibson, J.J. 1986. The Ecological Approach to Visual Perception. Hillsdale, NJ: Lawrence Erlbaum.
Giddens, A. 1990. The Consequences of Modernity. Stanford, CA: Stanford University Press.
Hansen, M.T., and Haas, M.R. 2001. Competing for attention in knowledge markets: Electronic document
dissemination in a management consulting company. Administrative Science Quarterly, 46, 128.
Heskett, J. 2002. Toothpicks & Logos: Design in Everyday Life. New York: Oxford University Press.
Hollingshead, A.B. 1998. Communication, learning, and retrieval in transactive memory systems. Journal
of Experimental Social Psychology, 34, 5, 423442.
IDEO. 2001. IDEO Method Cards. Palo Alto, CA: IDEO.
Kelley, T., and Littman, J. 2001. The Art of Innovation: Lessons in Creativity from IDEO, Americas Learning
Design Firm. New York: Doubleday.
Lakoff, G., and Johnson, M. 1980. Metaphors We Live By. Chicago: University of Chicago Press.
Lave, J., and Wenger, E. 1991. Situated Learning. Cambridge, UK: Cambridge University Press.
Lyytinen, K., and Yoo, Y. 2002. The next wave of Nomadic computing. Information Systems Research, 13,
4, 377388.
Markus, M.L. 2001. Toward a theory of knowledge reuse: Types of knowledge reuse situations and factors
in reuse success. Journal of Management Information Systems, 18, 1, 5793.
McCullough, M. 2004. Digital Ground: Architecture, Pervasive Computing and Environmental Knowing.
Cambridge, MA: MIT Press.
Mitchell, W.J. 2003. Me++: The Cyborg Self and the Networked City. Cambridge, MA: MIT Press.
Moreland, R.L. 1999. Transactive memory: Learning who knows what in work groups and organizations. In
L.L. Thompson, J.M. Levine, and D.M. Messick (eds.), Shared Cognition in Organizations. Hillsdale,
NJ: Lawrence Erlbaum.
Nonaka, I. 1991. The knowledge-creating company. Harvard Business Review, 69 (NovemberDecember),
96104.
the Dynamics of Innovation. New York: Oxford University Press, pp. xi, 284.
Pine, B.J., and Gilmore, J.H. 1999. The Experience Economy. Boston: Harvard Business School Press.
Polanyi, M. 1966. The Tacit Dimension. New York: Doubleday.
Porter, M.E. 1980. Competitive Strategy: Techniques for Analyzing Industries and Competitors. New York:
The Free Press.
Sawhney, M.; Balasubramanian, S.; and Krishnan, V. 2004. Creating growth with services. MIT Sloan Man-
agement Review, 45, 2, 3444.
Schultze, U., and Leidner, D.E. 2002. Studying knowledge management in information systems research:
Discourses and theoretical assumptions. MIS Quarterly, 26, 3, 213242.
Star, S.L., and Griesemer, J.R. 1989. Institutional ecology, translations and boundary objects: Amateurs
and professionals in Berkeleys Museum of Vertebrate Zoology, 190739. Social Studies of Science, 19,
387420.
Weick, K. 1998. Improvisation as a mindset for organizational analysis. Organization Science, 9, 5,
543555.
144 YOO
Wenger, E. 1998. Communities of Practice: Learning, Meaning, and Identity. Cambridge, UK: Cambridge
University Press.
Yoo, Y. 2005. On artifact as default. In F. Collopy and J. Faust (eds.), Dialogues in the Realm of Managing
as Designing. Cleveland, OH: Cleveland Institute of Art, pp. 138149.
Yoo, Y.; Boland, R.J.; and Lyytinen, K. 2006. From organization design to organization designing. Organiza-
tion Science, 17, 2, 215229.
Yoo, Y., and Kanawattanachai, P. 2001. Developments of transactive memory systems and collective mind
in virtual teams. The International Journal of Organizational Analysis, 9, 2, 187208.
Yoo, Y., and Torrey, B. 2002. National cultures and knowledge management in a global learning organization:
A case study. In N. Bontis and C.W. Choo (eds.), The Strategic Management of Intellectual Capital and Or-
ganizational Knowledge: A Collection of Readings. New York: Oxford University Press, pp. 421434.
Zack, M.H. 1999. Managing codifed knowledge. Sloan Management Review, 40, 4, 4558.
Zuboff, S., and Maxmin, J. 2002. The Support Economy. New York: Viking.
145
ChaptEr 8
THE IMPACT OF COMPUTER-MEDIATED
COMMUNICATION ON KNOWLEDGE TRANSFER
AND ORGANIZATIONAL FORM
daVid G. sChwartZ and doV tEEni
Abstract: It is the transfer of knowledge that ultimately drives changes to organizational form.
Historically, much research has focused on the introduction of computer-mediated communication
(CMC) as an agent of change to organizational form. CMC has also received much attention as an
enabler for knowledge transfer within the context of knowledge management and organizational
learning. After examining these two streams of research in parallel, we suggest that (1) the per-
ceived relationship between CMC and organizational form runs through knowledge transfer and
(2) organizational form in turn drives additional changes in the use of CMC. In other words, the
impact of CMC on organizational form is not necessarily direct and can be better understood by
interjecting knowledge transfer as an intermediate step. In addition, changes in organizational
form can be seen as looping back to impact CMC thus creating a cycle. This paper describes and
justifes what we refer to as the CMC Impact Cyclethe fow of organizational change moving
from CMC to knowledge transfer to organizational form and back to CMC.
Keywords: Computer-Mediated Communication, Organizational Form, Knowledge Transfer,
Knowledge Management, CMC Impact Cycle
INTRODUCTION
Evolutionary theory tells us change occurs slowly, over time, as a result of a selection process in
which characteristics advantageous to the survival of an entity are reinforced by future generations
so that those characteristics become more apparent and established. Knowledge transfer (KT)our
abilities to share ideas, convince, infuence, advise, teach, and improve the decision-making
abilities of othershas been evolving since the advent of spoken and, later, written language.
We have developed expressive techniques, such as infection, sarcasm, humor, and emphasis. We
have developed modes of interaction, such as lectures, letters, discussions, and debates. We have
developed social tools such as mentoring, confding, team building, and networking. Each of these
has found its way into the fabric of our organizations, some as explicitly endorsed, and others
as implicitly expected, modes of interaction. Evolutionary theory also tells us that occasionally
there occurs a disruptive event that can split or lay to waste years of evolutionary process. In the
evolution of KT in organizations, the advent of computer-mediated communication (CMC) is just
such an event. To be clear, we are not referring to KT as memetics (Aunger, 2000; Dennett, 2002),
what many consider a worthy evolutionary theory modifed from Darwins original theories in
146 SCHWARTZ AND TEENI
an attempt to explain the survival and propagation of ideas, culture, and perhaps knowledge. We
use the term evolution in its soft metaphoric sense as a map of how certain ideas and conceptions
with respect to KT and CMC have adapted and must continue to do so.
The story of KT and CMC includes a third construct, namely, organizational form.
Organizational form (OF) is the framework in which the organization defnes how tasks are divided,
resources are deployed, and departments are coordinated (Daft, 1991). Two opposing archetypes
are the loose organic organization and the tight mechanistic organization. The loose organization is
characterized by shared tasks, relaxed hierarchies, horizontal (between teams) communication, and
informal communication. In contrast, the tight organization is characterized by specifed tasks, strict
hierarchies, vertical communication, few cross-functional teams, and centralized decision making.
Organizational forms are increasingly dynamic; however, it is the transfer of knowledge that
ultimately drives their change. Changing OFs are not only the result of advances in CMC (Fulk and
DeSanctis, 1995), but the new forms are themselves harbingers of change in the ways CMC is then
applied by actors in the mutated organization. In no functional area is the impact of this relationship
felt more than in that of KT. The relationship between CMC and KT has been intensely studied
(Majchrzak et al., 2000; Schwartz and Teeni, 2000; Huysman and de Witt, 2002; van den Hoof
and de Ritter, 2004), so, too, has the relationship between CMC and changing OFs (Orlikowski,
Yates, Okamura, and Fujimoto, 1995, Pickering and King, 1995; Reagens and McEvily, 2002;
Huber, 2003). Yet the interplay between these three remains largely unexplored. We will refer to
this interplay, to this cyclic chain of impact starting with CMC affecting KT, continuing with KT
affecting OF, and recursively with CMC again, as the CMC Impact Cycle (Figure 8.1).
To delve deeper into the CMC Impact Cycle, we rely on the communication perspective. The
communication perspective is a crucial aspect of organizations, especially virtual organizations.
Indeed, communication plays a pivotal role in organizations, and may even be seen as the founda-
Knowledge
Transfer
Organizational
Form
Computer
Mediated
Communication
Figure 8.1 The CMC, KT, OF Triumvirate
THE IMPACT OF COMPUTER-MEDIATED COMMUNICATION 147
tion for most organizational action (Galbraith, 1977; Weick, 1979). Perhaps more than any other
aspect, communication is traditionally seen as most tightly interacting with OF. Not only does
form follow function, but function, too, is constrained by form. We will show how understanding
and designing KT are intimately related to understanding and designing CMC, and that CMC acts
through KT to infuence OF, which in turn affects the use and development of CMC.
The purpose of this article is to examine the triumvirate of OFs, KT, and CMC and to explicate
the relationships and interactions between them. We begin by presenting the concepts central to
our discussion and a review of related literature.
CORE CONCEPTS AND RELATED WORK
The relationship between CMC, KT, and OF has been developing and evolving for over 30 years,
since the frst studies on the impact of e-mail use on team behavior were conducted beginning
with identifying communication networks within an organization (Allen, 1972). The three areas
of knowledge transfer, organizational form, and computer-mediated communication have each
been the subject of much diverse research and to a certain degree have developed independent
terminological bases. We briefy address the core concepts and key literature of each area.
Knowledge Transfer
Knowledge transfer is the focused, unidirectional communication of knowledge between individu-
als, groups or organizations such that the recipient of knowledge (a) has a cognitive understanding,
(b) has the ability to apply the knowledge, or (c) applies the knowledge (King, 2006, p. 254). In
his discussion of KT, King draws from Argote (1999) and Darr and Kurtzberg (2000), reinforcing
the approach that KT is the communication of knowledge from a source so that it is learned and
applied by a recipient. A key element in these defnitions is the term communication or, more spe-
cifcally, focused unidirectional communication. However, despite what one might expect based
on Kings defnition, the extant KT literature seems to take the communicative act for granted,
focusing instead on higher-level functions such as motivation, trust, ambiguity, and the like.
Computer-Mediated Communication
Organizational communication is the bidirectional exchange of information between people
with the intent of reaching mutual understanding and/or affecting the relationship between them
(Teeni, 2001). We are concerned primarily with CMC in the context of organizations. Computer-
mediated communication or, as Huber (2003) characterizes it, computer-assisted communication
technologies, has grown to include e-mail, instant messaging, chat rooms, and video and computer
conferencing, facilitating access to people inside and outside a given organization. It is safe to as-
sume that additional technological tools will be added to the CMC toolbox in the coming years.
We defne CMC in the broadest possible manner in order to cover the complete range of tools
that are being or may be used in organizational environments. CMC means the following:
Any form of communication between one or more people that uses some element of com-
puter technology to initiate, edit, annotate (re)transmit, store, retrieve, or view information.
Information, for the purpose of this defnition, includes text, static graphics, animation,
audio, video, and any other medium that can be represented in digital form for processing,
transmission, and storage on a networked computing platform. As such this defnition would
include e-mail, message boards and forums, instant messaging and chat rooms, voice-over
IP conversations, network-based videoconferencing, and RSS streams. We shall refer to all
of the above CMC tools as CMC modalities. The common denominator between all of these
modalities is the existence of:
a. One or more message sources or authors
b. One or more message recipients
c. A message comprised of one or more information elements
Note that this defnition entertains a number of aspects not considered in previous defnitions of
CMC. The frst is the possibility that the source and recipient are one and the same person. While
this type of interaction is in its infancy and has yet to be studied, there are indications from Web
sites such as www.futureme.org and www.mailtothefuture.com (Ewalt, 2005) that harnessing CMC
to communicate with oneself at a future point in time may indeed have measurable impacts and
become relevant in an organizational settingparticularly if such communication is role based.
For example, consider the current chief fnancial offcer of a company sending an e-mail message
to the chief fnancial offcer 1 fscal year later.
Second, under this defnition, information types can be expanded. For example, once digital
medium representational abilities are expanded to include olfactory stimuli (Harel, Carmel, and
Lancet 2003; Torodan, 2005), the defnition is predisposed to include digital scent as information
to be communicated. The advantages of having a defnition that can fexibly include different
combinations of participants as well as heretofore unconsidered information elements will become
apparent as our discussion takes us to consider new and developing OFs.
Examples of the study of CMC as an enabler for KT include Carley and Wendt (1991), Constant,
Sproull, and Kiesler (1996), Majchrzak, Rice, King, Malhotra, and Ba (2000a), Huysman and de
Witt (2002), van den Hooff and de Ritter (2004), and Schwartz and Teeni (2000).
Organizational Forms
We reformulate the idea of organizational form defned above in terms of relationships between
(human and nonhuman) actors. These relationships can be represented on fve dimensions: space,
time, knowledge, interpersonal relations, and functional interdependencies. Two of these dimensions,
namely time and space, have received considerable attention in the literature (Sproul and Kiesler,
1992). Other primary sources aggregating foundational work regarding the relationship between CMC
and OFs are Fulk and Steinfeld (1990) and DeSanctis and Fulk (1999). Additional treatments of the
relationship between CMC and OF can be found in Huber (2003), Reagens and McEvily (2003),
Orlikowsky, Yates, Okamura, and Fujimoto (1995), Pickering and King (1995), and Hinds and Kiesler
(1995), showing how CMC affects OF. Some of the earliest work is that of Conrath (1973), who
studied the relationship between communications environment and organizational structure.
In thinking of OF, we will use the following dimensions:
Space concerns the relationship in space between actors, e.g., the geographical distribution
of the organization.
Time concerns the distribution of work or actors along time, e.g., do several workers perform
work simultaneously, do projects consist of tasks organized on some time line, etc?
Knowledge concerns management and application of knowledge, assuming here that knowl-
edge is known by actors. It includes the strategic issues of what knowledge is stored, how
it is distributed among actors (including information systems), and how it is managed. It
also refers to how actors apply ithow actors should apply knowledge in routine actions,
in nonroutine actions, and in learning.
Interpersonal relations concern the relationships, attitudes, collegiality, and so on between
actors. This is usually an informal, and not intentionally designed, dimension of organization
but nevertheless may be instrumental in determining the feasibility and quality of learning.
Functional interdependencies characterize the organization of roles and allocation of tasks to
actors and the consequent interdependencies between tasks and actors. Some organizations
are tightly coupled, while others are not; some are hierarchically organized, others are fat,
and so on.
These fve dimensions of OF provide a richer understanding of its interplay with communica-
tion. We now turn to an initial view of the relationship between CMC and OF.
ORGANIZATIONAL FORM AND THE IMPACT OF CMC
The dimensions of OF may also be infuenced by expressive relationships that (Cross, Johnson-
Cramer, and Parise, 2005) include aspects such as friendship (Kilduff, 1992; Krackhardt, 1992;
Lincoln and Miller, 1979), personal or career support (Higgins and Kram 2001), and trust (Tsai and
Ghoshal, 1998). These, we note, are similar to the key elements for successful KT as discussed by
King (2006). Cross, Johnson-Cramer, and Parise (2005) also describe how the relationship between
networks and OF results in interaction patterns that can affect change via power as discussed by
Burkhardt and Brass (1990).
In order to fully capitalize on the use of CMC, one must understand the complex impact of CMC
on the specifc actions that actors perform in the organization, including their interdependencies.
But one must also understand the impact of CMC on the forms of the organization, which in turn
affect the actions. This dynamic picture complicates the analysis but is necessary to gain a more
realistic view of the multiple impacts of CMC.
One example of an interaction similar to the CMC Impact Cycle can be seen with respect to the
growing importance of virtual organizations, as discussed by Schwartz, Divitini, and Brasethvik
(2000). We believe that the extension of communication-organization models, such as Figure 8.2,
to explicitly consider the interplay with knowledge and functional interdependencies is called for
and begin by examining the relationship between KT and OF as it is affected by CMC.
As Figure 8.3 shows, there are multiple relationships between KT and OF. This follows Kings
(2006) summary of the key research results of KT:
1. An arduous relationship between the source and recipient negatively affects KT (e.g.,
Szulanski, 1996).
2. Shared understanding between the source and recipients is particularly important to suc-
cessful KT (Ko, Kirsch, and King, 2005).
3. Absorptive capacity has long been believed to be important in infuencing effective KT
(e.g., Galbraith, 1990).
4. Knowledge observability is important to successful KT (e.g., Birkinshaw, Frost, and
Ensign, 2002).
5. Intrinsic motivation may be more important than extrinsic motivation in KT.
6. Source credibility is important in KT, while the sources encoding competence may not
be so important (Ko, Kirsch, and King, 2005).
Figure 8.2
Knowledge
Transfer
Organizational
Form
Computer
Mediated
Communication
Ability to:
Communicate
Manage distributed
processes
Create dynamic teams
Ability to:
Acquire knowledge
Distribute knowledge
Store knowledge
+ +
Ability to
be
virtual
Demand for
KM
functionality
Figure 8.3 The Interplay Between Organizational Form, and the Computer-Mediated
Communication and Knowledge Transfer Layers
Interpersonal Relationships
Shared Understanding
Credibility
Knowledge Observability
Addressed in this study
Not addressed in this study
Absorptive Capacity
Knowledge Layer
Communication Layer
+
Computer
Mediated
Communication
Knowledge
Transfer
Organizational
Form
In the discussion that follows, we refer to the above elements as interpersonal relationship, shared
understanding, absorptive capacity, knowledge observability, intrinsic motivation, and credibility,
respectively.
Beginning from the Knowledge Transfer point in Figure 8.3, this means that the ways in which
the form of an organization evolves over time are directed and moderated by KT activities in that
organization. For example, management hierarchies have been shown to fatten through the introduc-
tion of problem diagnosis systems that enable lower-level management to act more autonomously
in the detection and correction of problems. This, it should be noted, is not a planned result from the
introduction of IT, which in most cases is process driven and directed at solving a specifc operational
problem. Nor is it a direct result of the IT, but rather a direct result of changes in KT. Changes in OF
resulting from the introduction of new IT can be unexpected. For example, van Dijk, Duysters, and
Beulens (2003) discuss unintended changes in the strategic roles and activities of actors in supply
chains and networks following the introduction of interorganizational systems in the food industry.
Conversely, a change in OF will necessitate a response and adaptation of the ways in which KT
occurs in that organization, which must be facilitated by new or modifed CMC. For example, the
introduction of virtual teams to an organizational structure has been shown to require the develop-
ment of new transfer norms and protocols (Walther et al. 2001). In cases such as this, KT evolves
out of necessity in response to structural changes dictated by the new OFs.
Direct impacts of IT on OF can be seen for each of the parameters of OF presented earlier:
SpaceCMC is stretching the space the organization spansat the group level, we see par-
tially distributed or even fully virtual teams that never meet face to face. At the organizational
level, CMC enables distributed organizations that span states, countries, and continents.
TimeIT helps overcome time-related constraints, e.g., using asynchronous CMC for com-
munication between actors operating in different time zones or working in different shifts.
KnowledgeIT changes our capacity to manage and apply knowledge. IT changes the
knowledge we can store and makes accessible implicit knowledge that was previously known
to individuals without others knowing who knows what. It can support an organization that
distributes its knowledge in repositories. It has the potential to enable learning.
RelationshipsIn varying studies, IT has been shown to have both positive effects and ad-
verse impacts on relationships in organizations. Whether positive or negative, the impact is
substantial.
InterdependenciesIT fattens the organization, makes it less hierarchical, but can often
replace old interdependencies with new ones.
These statements refect the impacts of IT on OF for each of its dimensions. More research on
each of these statements is certainly still needed. Once we explain our model in Figure 8.3, we
return to these dimensions of OF and examine the impact of technology by considering explicitly
the interaction with KT.
INTERJECTING KNOWLEDGE TRANSFER TO THE RELATIONSHIP
BETWEEN COMPUTER-MEDIATED COMMUNICATION AND
ORGANIZATIONAL FORM: A TWO-LAYERED MODEL
While retaining the focus on communication, we broaden the idea of OF as it pertains to KT. As
noted earlier, communication should itself be understood in its broadest sense; for example, it
should encompass verbal as well as nonverbal, textual as well as graphic, thought as well as feel-
ings, and so on, as KT needs to effectively deal with each nuance of communications. Examining
the specifc rubric of how information technologies affect KT brings us to CMC. As such, in order
to effectively provide support of KT, CMC should be appropriately designed to support the whole
gamut of communication types. Because KT is not limited to a restrictive OF, so, too, CMC must
be designed for multiple, dynamic, and differentiated OFs.
It is useful to regard the transfer of knowledge at two levels of abstraction: for the upper level,
we retain the term knowledge, and for the lower level, we use the label communication to connote
a focus on organizational communication that is computer-mediated, as shown in Figure 8.3. We
frst distinguish between the levels and then discuss the interaction between the levels and its
implications.
Knowledge (the upper level) is facilitated by communication (the lower level). The higher level
directs and constrains the lower level, and the lower level enables or inhibits the higher level.
Importantly, IT should be designed to support both levels in tandem and should enhance the
interaction and interplay between levels. As Figure 8.3 shows, it is not just CMC that changes
OFs, but there is a cycle in which new and changed OFs necessitate and drive new knowledge
management (KM) functionality, much of which is related to communication and KT.
In designing the organization of the future, we must extend the core concepts of organizational
communication to the knowledge level, for it is that level that ultimately affects the OF. We com-
bine these ideas to examine the interplay between OF, KT and CMC and how all are supported
and affected by IT (Figure 8.3). To examine the models design implications, both organizational
and technical, we develop the interaction between levels, frst looking at the role of communica-
tion in KT, then examining the role of KT in communications, and fnally revisiting the role of
CMC in OF.
THE ROLE OF COMMUNICATION IN KNOWLEDGE TRANSFER
Knowledge transfer has, at its core, the concept of knowledge communication. This is discussed in
depth by Eppler (2006), who consolidates no fewer than 36 studies dealing with the communica-
tion of knowledge. In his taxonomy, Eppler consolidates fve specifc KT problem categories that
hinder communication, namely:
1. Expert-caused diffculties
2. Manager-caused diffculties
3. Mutual behavior problems of both expert and manager
4. Interaction situation diffculties
5. Problems imposed by organizational context
What all of these categories have in common, according to Eppler, is inadequate communication
behavior patterns for KT. What they also share in common, unidentifed by Eppler, is that they
are each intensely affected by OF.
Jacobsons (2006) model of knowledge sharing focuses on the importance of communication
channels and modeling communication channels as a starting point for a more systematic and
scientifc approach (p. 512) to understanding knowledge sharing problems. The dyadic commu-
nication model espoused by Jacobson considers six factors:
1. Knowledge source
2. Message
3. Knowledge receiver
4. Channel
5. Feedback
6. Environment or organizational context
We point out that each of these elements can be affected by OF as well, an observation that
we return to later. It is clear that communication plays a central role in KT, but what about the
other direction?
THE ROLE OF KNOWLEDGE TRANSFER IN COMMUNICATION
To understand the central role of KM in supporting organizational communication, Teeni (2006)
concentrates on four concepts: context, levels of abstraction, adaptation, and organizational
memory (OM).
Context
Context refects most closely the link between knowledge and communication. We assume that in
any communication there is a core message that the sender wishes to convey to the receiver. Send-
ers add contextual information to the core message to increase the likelihood that the receiver will
understand their intentions. Whatever information receivers choose to use (from the information
available to them) in reasoning about the core message can be regarded as context. Part of this
context is in the receivers head or in other available sources, and part needs to be provided to the
receiver by the sender as contextual information to ensure mutual understanding.
Contextual information refers to several possible aspects of the core message: the situation
in which the message was produced, the situation in which it is anticipated to be received, an
explanation about a statement, an explanation how to go about executing a request for action, or
the underlying assumptions about an argument. Providing the contextual information to explain
the core message is a common communication strategy called contextualization. Contextual infor-
mation can be seen as layers of information around the core message, and contextualization can
be seen as the act of adding layers of information. Knowledge management techniques capable
of determining and identifying context, retrieving or generating the information, providing the
information in effective message forms and through effective media, and testing its impact may
play a crucial role in enriching communication with appropriate contextual information.
Contextualization, as a communications strategy, is discussed in detail by Teeni (2001, 2006)
and examined in terms of improving task performance (Newlands, Anderson, and Mullin, 2003).
It has also been studied with respect to many different CMC modalities such as chat (Herring,
2001). Much work has been done on the development and testing of modeling tools to facilitate
KT (Basque, Imbeault, Pudelko, and Lonard, 2004; Hdrich and Maier, 2006). Modeling the fow
of knowledge in the modern organization has been put forth as a crucial element of success, yet
the application of theory to fow management through IT has yet to be realized (Nissen, Kamel,
and Sengupta, 2000; Nissen, 2002). Chat, instant messaging, and e-mail technologies have all
been examined as possible channels for KT in context. Nardi, Whittaker, and Bradner (2000) and
Isaacs, Walendowski, Whittaker, Schiano, and Kamm (2002) study the nature of instant messag-
ing conversations within organizations. Ribak, Jacovi, and Soraka (2002) examine integrating
instant messaging as part of knowledge search processes. Schwartz and Sadan (2006) explore
the use of flters to derive context from instant messaging sessions in order to integrate with KM
systems. Using e-mail to create a shared knowledge context was introduced in Schwartz (1998)
and expanded in Schwartz (1999) and Schwartz and Teeni (2000).
Abstraction
Levels of abstraction in the core and contextual information communicated represent the third
way in which KM can support organizational communication. In thinking and communicating,
people represent action at multiple levels of abstraction, and at any one moment, one of these
levels is their focal level (Vallacher and Wegner, 1987; Berger, 1998). Moreover, people tend to
remain on higher rather than lower levels of abstraction but shift their attention to a lower level of
abstraction when communication complexity increases and breakdowns occur. As argued earlier,
we regard the relationship between CMC and KT as having both a higher knowledge layer and a
lower communication layer (Figure 8.3) so that in effect when complexity is encountered in KT,
an actor would revert to the communication layer.
KM techniques capable of identifying communication breakdowns and correcting them must
rely on knowledge of communication at all levels of abstraction (such knowledge may be modeled
as a multilevel model of communication analogous to the open systems interconnect seven-layered
protocol model). These KM techniques would be essential for ensuring effective communication
and correcting lower levels of communication in order to enable communicators to concentrate
on the higher-level functions of KT.
Adaptation
Another concept is that of adaptation in communication. Effective communicators match the
medium, the message form, and the communication strategies to the communication situation and
the dynamics of the dialog. For example, communication between heterogeneous communicators
should include more contextual information and may be more effective when richer, rather than
leaner, media is selected. Knowledge of the communication situation, such as the relationships
between communicators, as well as knowledge of how to communicate can be used to generate
more effective communication. Communication complexity can be seen as a systemic measure of
the communication situation and its susceptibility to communication breakdowns (Teeni, 2001). It
can therefore act as a sensor to trigger adaptation. Knowledge management techniques capable of
detecting the need to adapt and capable also of adapting the system parameters can play an important
role in facilitating communication support systems that provide tailored communication.
Organizational Memory
The last concept is the role of organizational memory in communication (Anand, Manz and Glick,
1998). In our view, OM is part of IT. OM is a historic repository whose purpose is to support
organizational effectiveness and future task performance (Schatz, 1991; Walsh and Ungson, 1991;
Burstein and Linger, 2006). Stein and Zwass (1995) defne a two-layer framework consisting
of memory functions at one level and scanning, sharing, control, and environmental functions
at another level. Organizational memory is a general term for the collection of information and
knowledge known to the organization, as well as the KM necessary to acquire, store, and utilize
this knowledge. OM therefore is essential for communication not only because it is a source of
contextual information but also because it embodies the knowledge of how to communicate ef-
fectively in the organization (e.g., who knows, or should know, what). In essence, many aspects
of OF are embedded and represented in the OM. Furthermore, the information known to the
organization is, in a substantial part, represented in organization communication on digital media
such as e-mail and bulletin boards. It follows that CMC can be a major source of the information
stored in the OM. In other words, communication is a major provider of knowledge as well as
being an essential enabler of KM.
Thus, the roles that KM plays in affecting communications within an organization are many.
THE ROLE OF COMPUTER-MEDIATED COMMUNICATION IN
ORGANIZATIONAL FORM
In this section, we review the central role that CMC plays in determining OF, frst discussing the
general relationship and role and then focusing on two specifc aspects: interpersonal relation-
ships and shared understanding.
ORGANIZATIONAL FORMS
The fve aspects of OF discussed earlier, namely space, time, knowledge, interpersonal relation-
ships, and interdependencies, are dealt with both explicitly and implicitly in much research that
focuses on the relationship between CMC and OF. In our discussion, we assume that communication
is vital for effective organizations whatever their form; for example, Sagie et al. (2002) showed
how information sharing mediates job satisfaction and commitment in both the loose and tight
organizations. Our goal is to understand the differential paths that lead to the same results.
Some of the earliest work on the impact of CMC on OF is that of Rice (1980, 1982) and Wil-
liams and Rice (1983), who pioneered the study of how new communications technologies change
the way individuals interact in an organizational setting. This work continued through the 1990s,
relating e-mail use and network structure to R&D activity (Rice, 1994), and through the current
decade (Rice and Gattiker, 2000; Majchrzak, Rice, Malhotra, King, and Ba, 2000).
Much of todays research into the CMC-OF dynamic continues to be based on the work of Rice
and colleagues as well as the infuential work of Fulk and DeSanctis (1995), which preceded a
watershed of development and adoption of new communications technologies. Katzy, Evaristo,
and Zigurs (2000) study the relationship between KM and virtual teams. They present a typol-
ogy of project interactions and suggest that the transfer of knowledge is handled largely by the
technological infrastructure. Majchrzak, Rice, Malhotra, King, and Ba (2000) study the use of
collaborative tools in support of KT between the members of virtual teams. Their study, however,
deals with dedicated CT applications that have been designed specifcally for team collaboration,
and not with CMC tools as discussed in this article.
Orlikowski, Yates, Okamura, and Fujimoto (1995), in a study of technology-use mediation,
take a detailed look at how the use of CMC in an organization can be shaped and structured. Their
conclusion is of particular importanceIn the new and fuid organizational forms now emerging
in the face of rapidly changing environments, contextualization of technologies will be a critical
mechanism for helping communication norms and work practices adapt. This conclusion serving
as a call for fnding formally structured organization and task-specifc adaptations of core CMC
technologies, such as KT. Reagans and McEvily (2003) show how network structure and ensuing
communication patterns affect knowledge structure in organizations.
The importance of these streams of research is increasing rather than diminishing over time. The
impact and interaction between CMC and OF, in fact, are refected in one of the future trends of the
impact of IT on the feld of KM as discussed by Becerra-Fernandez and Sabherwal (2006), who
state Over time, the emphasis of knowledge management shifted to groups and entire organiza-
tions, and now examples of interoranizational impacts of knowledge management are becoming
increasingly common. This trend in the impact of knowledge management is expected to continue
with its use across networks of organizations and across governments, enabling collaborations
across historical adversaries and integrating knowledge across highly diverse perspectives and
disciplines (p. 234).
Interpersonal Relationships
The use of CMC to foster, improve, or control interpersonal relationships has been dealt with on
a variety of levels. Kalbfeisch and Eckley (2003) studied how mentoring relationships change
following the introduction of CMC and fnd support for Walther and Burgoons (1992) earlier
work showing how online relationships develop over time. In more recent work, Duthler (2006)
looked at the growth of voicemail use and the impact of politeness in both e-mail and voice mail
on the development of interpersonal relationships. Jarvenpaa and Leidner (1999) showed that
trust is critical for virtual teams to succeed and that such trust may be fostered through supporting
information technologies; more recently, Sussman and Siegal (2003) studied the infuence CMC
has on establishing trust relationships leading to the adoption of advice (i.e., direct KT) received
via e-mail. van den Hooff, Elving, Meeuwsen, and Dumoulin (2003) reached similar conclusions
with respect to knowledge sharing in communities of practice.
Shared Understanding
We assume here that senders intend their communication to be comprehensible (Grice, 1957).
Shared understanding means not only that the receiver has understood the message but also that
the sender is aware that the receiver has understood it and the receiver knows that the sender
knows this. For this state to be achieved, the message has to be transmitted effectively and both
communicators must be aware that the message has been transmitted.
These studies all serve to explicate key aspects of the role of CMC in OF.
RETURNING TO KNOWLEDGE TRANSFER
With reference to the KT elements highlighted in our earlier discussion of Kings (2006)
defnition of KT, CMC has the potential to provide inherent information richness, though
at times can be complex to operate on the part of the user. So while elements of KT related
to interpersonal relationship, understanding, absorptive capacity, knowledge observability,
and credibility may all be addressed in some form by CMC, the specifc nature of KT that
we are supporting must be considered to ensure that the elements identifed above are in fact
taken into account.
In order for CMC to improve KT and make KT easier, it is not enough to indiscriminately in-
troduce CMC modalities to the KT process. The actors using each CMC modality and its expected
impact on the communication and key KT elements must be considered.
The elements described earlier contribute to our understanding of how CMC can affect KT.
When viewed together with research that identifes the challenges of KT, we can start to understand
the specifc impact CMC can have on KT and through KT to OF.
To summarize, we know that:
1. Organizational forms change due to the introduction of CMC.
2. The challenges of KT increase with the advent of virtual projects, teams, and organiza-
tions.
3. KT within organizations is infuenced and impacted by network structure and commu-
nications patterns.
4. The KT elements of interpersonal relationship, shared understanding, absorptive capac-
ity, knowledge observability, and credibility must be explicitly addressed by any model
that would effectively refect the relationship between KT and CMC.
5. Elements similar to these listed have been shown to be what ultimately drives changes
to OF.
In the section that follows, we take two of the KT elements that need to be addressed by CMC,
namely interpersonal relationships and shared understanding, and show how they are also primary
drivers to change in OF and can be viewed and understood through our model (Figure 8.3).
IMPLICATIONS OF THE CMC IMPACT CYCLE
Having presented a model explaining the interaction fows of the CMC Impact Cycle (Figure 8.3),
we demonstrate its implications on design by taking two of the organizational and technologi-
cal elements discussed above as being critical to the success of KTshared understanding and
interpersonal relationshipsand showing their importance to OF.
That shared understanding and interpersonal relationships are critical to KT success is not
surprising and is fortifed by our model. On the one hand, shared understanding and interpersonal
relationships are the objectives of organizational communication (Teeni, 2001), and it is therefore
clear that effective communication enhances KT, while ineffective communication hinders KT.
On the other hand, shared understanding and interpersonal relationships are closely tied to OF.
First, shared understanding bridges the knowledge gaps, which is a major obstacle in the face
of effective communication. Second, poor (or good) interpersonal relationships that result from
communication are directly refected as one of the dimensions of OF. So the impact of shared
understanding and interpersonal relationships is carried through both KT and OF (Figure 8.3).
What are the implications on design? The most profound conclusion, and probably most
extensively studied, is that communication support must consider OF, in particular, space, time,
knowledge, interpersonal relationships, and functional interdependencies. A greater distance in
space is one of the main determinants of the choice of medium (Caldwell, Uang, and Taha, 1995;
Webster and Treviono, 1995). Similarly, distance in time can be overcome effectively by asynchro-
nous CMC. The classic example is communication support for workers in different shifts (Huff,
Sproull, and Kiesler, 1989). Distance in time requires, at a minimum, greater use of OM and a
greater need for contextualization because context changes over time. In general, contextualization
requires richer media (e.g., Daft and Lengel, 1986).
There is a comparatively huge accumulation of research on the impact of knowledge gaps
on communication and KT. At the level of communication, knowledge gaps make it diffcult
to reach shared understanding and usually rely on contextualization to moderate their negative
impact (Stasser and Titus, 1987; Hightower and Sayeed, 1995). Here, too, the immediate impact
on CMC is the need for richer media. However, at the level of KT, knowledge gaps are actually
the reason for KT in the frst place. It is not clear, though, what are the upper limits on knowledge
gaps to make KT at all feasible. Without a minimum level of common ground, it will be diffcult
to exchange knowledge. One strategy to overcome such gaps is perspective taking (Boland and
Tenkasi, 1995), which requires advanced technologies such as support for cognitive mapping. In
their approach, the knowledge and communication layers of CMC we have described here are
explicitly and intentionally implemented as part of technology design.
The effects of functional interdependencies on KT and their design implications appear to
be somewhat different from the other effects. Strong functional interdependencies represent a
requirement of more intense information exchange (communication) that is not necessarily a
transfer of knowledge. Research indicates that richer exchange of knowledge is usually from
supervisor to subordinate (Lee and Heath, 1999), and this is presumably where we should require
richer media.
But more revealing are the complex interactions between actors. For instance, the greater
communications needed to ensure mutual understanding to overcome distance in space result in
lower communication of interpersonal (non-task-related) messages (Sarbaugh-Thompson, 1998),
which in turn may compromise interpersonal relations, and when interpersonal relations begin to
change, OF follows closely behind.
CONCLUSION
Investigations into the relationship between CMC and OF have been developing and evolving for
over 30 years, and the study of the relationship between CMC and KT has also developed during
that period since the frst studies on the impact of e-mail use on team behavior were conducted.
While the distinct pairwise relationships of CMC to OF and CMC to KT will no doubt continue
to receive much deserved attention in both research and practice, we have shown here how they can
be integrated into a single model and be studied as a continuum of activitya CMC Impact Cycle
in which the fow of organizational change moves from CMC to KT to OF and back to CMC.
In presenting an integrated model that explicates the relationship between CMC, KT, and OF,
we have focused on two elements central to both KT and OFinterpersonal relationships and
shared understandingwhich were examined in detail with respect to the model. We believe that
further study of these two elements is warranted and that the elements of absorptive capacity,
knowledge observability, and credibility should also be examined in terms of the model.
We hope that the model we have presented here, by explicitly representing the multiple levels
and interaction between KT, CMC, and OFs, will enable researchers to better understand the re-
lationships expressed in Figure 8.3 and drive the continued evolution of the CMC Impact Cycle.
When considering the impact of CMC on KT, consider how we could possibly envision KT in the
modern organization without computer-mediated communications. Would we even know how to
conceptualize such an environment? True, we could fall back on classic training modalities such
as videos, classroom and group meetings, manuals and guidelines, even lessons learned databases
with remote access. But CMC has changed the nature of our organizations forever, and with it has
enabled KT that is far more effective, precise, diverse, trusted, and personalized than we could
ever have achieved without it. Changes in OF have emerged and will continue to evolve from
the infuence of KT and will undoubtedly result in the development of new CMC modalities as
well. The continued growth, acceptance, development, and sophistication of CMC modalities will
continue to push the envelope and require us to develop even better ways to enhance KT, even as
our OFs continue to change.
We shape our environments, then our environments shape us is one of the many insightful
sayings attributed to Sir Winston Churchill. While undoubtedly true in the general societal context
to which Churchill referred, it is also true with respect to the societal macrocosms that we create
in our organizations. The introduction of CMCs into our organizational environments has shifted
the evolutionary course of KT, further affecting OF. We can expect CMC to continue driving the
cycle of change for many years to come.
REFERENCES
Argote, L. 1999. Organizational Learning: Creating, Retaining, and Transferring Knowledge. Boston:
Kluwer Academic Publishers.
Aunger, R. 2000. Darwinizing Culture: The Study of Memetics as a Science. Oxford: Oxford University
Press.
Basque, J.; Imbeault, C.; Pudelko, B.; and Lonard, M. 2004. Collaborative knowledge modeling between
experts and novices: A strategy to support transfer of expertise in an organization. In A.J. Caas, J.D.
Novak, and F.M. Gonzlez (eds.), Concept Maps: Theory, Methodology, Technology. Proceedings of the
First International Conference on Concept Mapping, Pamplona, Spain.
BBC News. E-mail tries out a sense of smell. BBC News, February 19, 2004. Available at http://news.bbc.
co.uk/2/hi/technology/3502821.stm. Accessed March 2006.
Becerra-Fernandez, I., and Sabherwal, R. 2006. ICT and knowledge management systems. In D.G. Schwartz
(ed.), Encyclopedia of Knowledge Management. Hershey, PA: Idea Group Reference, pp. 230236.
Berger C.R. 1998. Message plans, communication failure and mutual adaptation during social interaction. In
M.T. Palmer and G.A. Barnett (eds.), Progress in Communication Sciences, Volume XIV, Mutual Infuence
in Interpersonal Communication: Theory and Research in Cognition, Affect and Behavior. Stamford, CT,
Ablex Publishing, pp. 91111.
Bielema, C.L. 1997. How computer-mediated communication (CMC) can work to enhance distance delivery
of courses. Journal of Applied Communications, 8, 4, 317.
Birkinshaw, J.M.; Frost, T.S.; and Ensign, P.C. 2002. Centers of excellence in multinational corporations.
Strategic Management Journal, 23, 11, 9971019.
Burkhardt, M.E., and Brass, D.J. 1990. Changing patterns and patterns of changeThe effects of a change in
technology on social network structure and power. Administrative Science Quarterly, 35, 1, 104127.
Burstein, F., and Linger, H. 2006. Task-based knowledge management. In D.G. Schwartz (ed.), Encyclopedia
of Knowledge Management. Hershey, PA: Idea Group Reference, pp. 840847.
Caldwell, B.S.; Uang, S.T.; and Taha, L.H. 1995. Appropriateness of communication media use in organiza-
tions: Situation requirements and media characteristics. Behaviour and Information Technology, 14, 4,
199207.
Carley, K., and Wendt. K. 1991. Electronic mail and scientific communication. Knowledge, 12, 4,
406440.
Conrath, D. 1973. Communication environment and its relationship to organizational structure. Management
Science, 4, 586603.
ties for technical advice. Organization Science, 7, 119135.
Cross, R.; Johnson-Cramer, M.; and Parise, S. April 2005. Managing change through networks and values:
How a relational view of culture can facilitate large-scale change. New Roundtable at University of
Virginia. Available at https://webapp.comm.virginia.edu/NetworkRoundtable/Portals/0/Change_through_
Networks_and_Culture_Final.pdf. Accessed April 2006.
Daft, R.L. 1991. Management. Chicago: The Dryden Press.
Daft, R.L., and Lengel, R.H. 1986. Organizational information requirements, media richness and structural
design. Management Science, 32, 5, 554571.
Darr, E.D., and Kurtzberg, T.R. 2000. An investigation of partner similarity dimensions on knowledge transfer.
Organizational Behavior and Human Decision Processes, 82, 1, 2845.
Dennett, D.C., 2002. The new replicators. In M. Pagel (ed.), Encyclopedia of Evolution. Oxford, UK: Oxford
University Press, pp. E83E92.
DeSanctis, G., and Fulk, J. 1999. Shaping Organizational Form: Communication, Connection, and Com-
munity. Newbury Park, CA: Sage.
DeSanctis, G., and Monge, P. 1998. Communication processes for virtual organizations. Journal of Computer
Mediated Communications, 3, 4.
Eppler, M.J. 2006. Knowledge communication. In D.G. Schwartz (ed.), Encyclopedia of Knowledge Man-
agement. Hershey, PA: Idea Group Reference, pp. 317325.
Ewalt, D.M. E-mail time capsule. Forbes, December 1, 2005. Available at http://www.forbes.com/2005/10/18/
communication-networks-language-cx_mn_de_comm05land.html. Accessed February 10, 2006.
Fulk, J., and DeSanctis, G. 1995. Electronic communication and changing organizational forms. Organiza-
tional Science, 6, 338349.
Fulk, J., and Steinfeld, C.W. 1990. Organizations and Communication Technology. Newbury Park, CA:
Sage.
Galbraith, C.S. 1990. Transferring core manufacturing technologies in high technology frms. California
Management Review, 32,4, 5670.
Grice, H.P. 1957. Meaning. Philosophical Review, 66, 377388.
Hdrich, T., and Maier, R. 2006. Integrated modeling. In D.G. Schwartz (ed.), Encyclopedia of Knowledge
Management. Hershey, PA: Idea Group Reference, pp. 251258.
Harel, D.; Carmel, L.; and Lancet, D. 2003. Towards an odor communication systems. Computational Biol-
ogy and Chemistry, 27, 121133.
Herring, S.C. 2001. Computer-mediated discourse. In D. Tannen, D. Schiffn, and H. Hamilton (eds.), Hand-
book of Discourse Analysis. Oxford: Blackwell, pp. 612634.
Higgins, M., and Kram, K. 2001. Reconceptualizing mentoring at work: A developmental network perspec-
tive. Academy of Management Review, 26, 264288.
Hightower, R.T., and Sayeed, L. 1995. The impact of computer-mediated communication systems on biased
group discussion. Computers in Human Behavior, 11, 1, 3344.
Hinds, P., and Kiesler, S. 1995. Communication across boundaries: Work, structure, and use of communica-
tion technologies in a large organization, Organization Science, 6, 373393.
Huber, G.P. 2003. The effects of advanced information technologies on organizational design, intelligence,
and decision making. In R.D. Galliers and D.E. Leidner (eds.), Strategic Information Management: Chal-
lenges and Strategies in Managing Information Systems, 3rd ed. New York: Elsevier, pp. 460496.
Huff, C.; Sproull, L.; and Kiesler, S. 1989. Computer communication and organizational commitment: Tracing
the relationship in a city government. Journal of Applied Social Psychology, 19, 16, 13711391.
Huysman, M. and de Wit, D. 2002. Knowledge sharing in practice. Dordrecht: Kluwer Academic
Publishers.
Isaacs, E.; Walendowski, A.; Whittaker, S.; Schiano, D.; and Kamm, C. 2002. The character, functions,
and styles of instant messaging in the workplace. Proceedings of the CSCW02, New Orleans, pp.
126135.
Jacobson, C.M. 2006. Knowledge sharing between individuals. In D.G. Schwartz (ed.), Encyclopedia of
Knowledge Management. Hershey, PA: Idea Group Reference, pp. 507514.
Jarvenpaa, S., and Leidner, D. 1999. Communication and trust in global virtual teams. Organization Science,
10, 791815.
Kalbfeisch, P., and Eckley, V.K. 2003 Facilitating mentoring relationships: The case for new technology.
Informing Science (June), 15811590.
Katzy, B.; Evaristo, R.; and Zigurs, I. 2000. Knowledge management in virtual projects: A research agenda.
Proceedings of the 33rd Hawaii International Conference on System ScienceVolume 1, HICSS. IEEE
Computer Society.
Kilduff, M. 1992. The Friendship Network as a decision-making resource: Dispositional moderators of social
infuences on organizational choice. Journal of Personality and Social Psychology, 62, 168180.
King, W.R. 2006. Knowledge transfer. In D.G. Schwartz (ed.), Encyclopedia of Knowledge Management.
Hershey, PA: Idea Group Reference, pp. 538543.
Ko, D.G.; Kirsch, L.J.; and King, W.R. 2005. Antecedents of knowledge transfer from consultants to clients
in enterprise system implementations. MIS Quarterly, 20, 1, 5985.
Krackhardt, D. 1992. The strength of strong ties: The importance of philos in organizations. In N. Nohria
and R.G. Eccles (eds.), Networks and Organizations: Structure, Form and Action. Boston, MA: Harvard
Business School Press, pp. 216239.
Lee, J., and Heath R.L. 1999. Managerial media selection and information evaluation from the receivers
perspective in decision-making contexts. Management Communication Quarterly, 13, 1, 7699.
Lincoln, J.R., and Miller, J. 1979. Work and friendship ties in organizations: A comparative analysis of
relational networks, Administrative Science Quarterly, 24, 181199.
Majchrzak, A.; Rice, R.E.; King, N.; Malhotra, A.; and Ba, S. 2000a. Computer-mediated inter-organizational
knowledge-sharing: Insights from a virtual team innovating using a collaborative tool. Information Resources
Management Journal (JanuaryMarch), 4453.
. 2000b. Technology adaptation: The case of a computer-supported inter-organizational virtual team.
Nardi, B.; Whittaker, S.; and Bradner, E. 2000. Interaction and outeraction: Instant messaging in action.
Proceedings of CSCW 00, Philadelphia, PA, pp. 7988.
Newlands, A.; Anderson, A.H.; and Mullin, J. 2003. Adapting communicative strategies to computer-mediated
communication: An analysis of task performance and dialogue structure. Applied Cognitive Psychology,
17, 3, 325348.
New Scientist. 2005. Smell before you buyOver the Internet. New Scientist, 2502, June 4, p. 24. Available
at http://www.newscientist.com/channel/info-tech/mg18625025.200. Accessed March 2006.
Nissen, M.E. 2002. An extended model of knowledge-fow dynamics. Communications of the Association
for Information Systems, 8, 251266.
Nissen, M.E.; Kamel, M.N.; and Sengupta, K.C. 2000. Integrated analysis and design of knowledge systems
and processes. Information Resources Management Journal, 13, 1, 2443.
Orlikowski, W.; Yates, J.; Okamura, K.; and Fujimoto, M. 1995. Shaping electronic communication: The
metastructuring of technology in the context of use. Organization Science, 6, 4, 423443.
Pickering, J.M., and King, J.L. 1995. Hardwiring weak ties: Interorganizational computer-mediated commu-
nication, occupational communities, and organizational change. Organization Science, 6, 4, 479486.
Reagens, R., and McEvily, B. 2003. Network structure and knowledge transfer: The effects of cohesion and
range. Administrative Science Quarterly, 48, 240267.
Ribak, A., Jacovi, M., and Soraka, V. 2002. Ask before you search: Peer support and community building
with ReachOut. Proceedings of the CSCW02, New Orleans, 126135.
Rice, R.E. 1980. The impacts of computer-mediated organizational and interpersonal communication. In M.
Williams (ed.), Annual Review of Information Science and Technology. White Plains, NY: Knowledge
Industry Publications, pp. 221249.
. 1982. Communication networking in computer-conferencing systems: A longitudinal study of
group roles and system structure. In M. Burgoon (ed.), Communication Yearbook. Beverly Hills, CA:
Sage, pp. 925944.
. 1994. Relating electronic mail use and network structure to R&D work networks and performance.
Journal of Management Information Systems, 11, 1, 920.
Rice, R.E., and Gattiker, U. 2000. New media and organizational structuring. In F. Jablin and L. Putnam
(eds.), New Handbook of Organizational Communication. Newbury Park, CA: Sage, pp. 554581.
Sagie, A.; Zaidman N.; Amichai Hamburger, Y.; Teeni, D.; and Schwartz, D.G. 2002. An empirical assess-
ment of the loose-tight leadership model: Quantitative and qualitative analyses. Journal of Organizational
Behavior, 23, 303320.
Sarbaugh-Thompson, M., and Feldman, M.S. 1998. Electronic mail and organizational communication: Does
saying Hi really matter? Organization Science, 9, 6, 685698.
Schatz, B.R. 1991. Building an electronic community system. Journal of Management Information Systems,
8, 3, 87101.
Schwartz, D.G. 1998. Applying shared semantics to organizational email communications. Internet Research,
8, 2, 432441.
. 1999. When email meets organizational memories: Addressing threats to communication in a learn-
ing organization. International Journal of Human-Computer Studies, 51, 3, 599614.
Schwartz, D.G.; Divitini, M.; and Brasethvik, T. 2000. On knowledge management in the Internet age. In D.G.
Schwartz, M. Divitini, and T. Brasethvik (eds.), Internet-based Organizational Memory and Knowledge
Management. Hershey, PA: Idea Group Publishing, pp. 123.
Schwartz, D.G., and Sadan, Z. 2006. Strip-mining the Web with SAIM: A system for the analysis of instant
messages. International Journal of Business Data Communications and Networking, 2, 2, 120.
Schwartz, D.G., and Teeni, D. 2000. Tying knowledge to action with Kmail, IEEE Intelligent Systems, 15,
3, 3339.
Sproull, L., and Kiesler, S. 1992. Connections: New Ways of Working in the Networked Organization. Cam-
bridge, MA: MIT Press.
Stasser, G., and Titus, W. 1987. Effects of information load and percentage of shared information on the
dissemination of unshared information during group discussion. Journal of Personality and Social Psy-
chology, 53, 1, 8193.
Sussman, S.W., and Siegal, W.S. 2003. Informational infuence in organizations: An integrated approach to
knowledge adoption. Information Systems Research, 14, 1, 4765.
Szulanski, G. 1996. Exploring internal stickiness: Impediments to the transfer of best practice within the
frm, Strategic Management Journal, 17, 2743.
. 2000. The process of knowledge transfer: A diachronic analysis of stickiness. Organizational Be-
havior and Human Decision Processes, 82, 1, 927.
Teeni, D. 2001. Cognitive-affective model of organizational communication for designing IT. MISQ (Re-
view), 25, 2, 162.
. 2006. Organizational communication. In D.G. Schwartz (ed.), Encyclopedia of Knowledge Manage-
ment. Hershey, PA: Idea Group Reference, pp. 734740.
Teeni, D., and Schwartz, D.G. 2000. Contextualization in computer-mediated communicationTheory
informs design. New Review of Information Systems, 1, 1, 5975.
Torodan, H. 2005. Smelling objects for multimedia database applications. Studies in Informatics and Con-
trols, 14, 4, 273278.
Tsai, W., and Ghoshal, S. 1998. Social capital and value creation: The role of intrafrm networks. Academy
of Management Journal, 41, 464476.
Vallacher, R. R. and Wegner D. M. 1987. What do people think theyre doing? Action identifcation and
human behavior. Psychological Review, 94, 1, 315.
van den Hooff, B., and de Ridder, J.A. 2004. Knowledge sharing in context: The infuence of organizational
commitment, communication climate and CMC use on knowledge sharing. Journal of Knowledge Man-
agement, 8, 6, 117130.
van den Hooff, B.; Elving, W.; Meeuwsen, J. M.; and Dumoulin, C. 2003. Knowledge sharing in knowl-
edge communities. In M. Huysman, E. Wenger, and V. Wulf (eds.), Communities and Technologies. The
Netherlands: Kluwer, pp. 119141.
van Dijk, S.; Duysters, G.; and Beulens, A. 2003. Transparency dilemmas, information technology and alliances
in agriculture and food industry, Eindhoven Centre for Innovation Studies, Working Paper 03.23, Depart-
ment of Technology Management, Technische Universiteit Eindhoven, the Netherlands, November.
Walsh, J.P., and Ungson, G.R. 1991. Organizational memory. Academy of Management Review, 16, 1,
5791.
Walther, J.B.; Boos, M.; Jonas, K.; Prell, C.L.; DAddario, K.; and Bunz, U. 2001. Misattribution and at-
tributional redirection to facilitate effective virtual groups. Second Annual Conference of the Association
of Internet Researchers, Minneapolis, MN, October.
Walther, J.B., and Burgoon, J.K. 1992. Relational communication in computer-mediated interaction. Human
Communication Research, 19,1, 5088.
Webster, J., and Trevino, L.K. 1995. Rational and social theories as complementary explanations of com-
munication media choices: Two policy-capturing studies. Academy of Management Journal, 38, 6,
15441572.
Williams, F., and Rice, R.E. 1983. Communication research and the new media technologies. In R. Bostrom
(ed.), Communication Yearbook. Beverly Hills, CA: Sage Publications, pp. 200224.
163
ChaptEr 9
MOVING TOWARD A KNOWLEDGE
MANAGEMENT MATURITY MODEL (K3M) FOR
DEVELOPING KNOWLEDGE MANAGEMENT
STRATEGY AND IMPLEMENTATION PLANS
jay liEBowitZ and tom BECKman
Abstract: Most of the current research addressing knowledge management (KM) strategy and
implementation has focused on ad hoc approaches. Unfortunately, without the necessary rigor
behind these approaches, KM will become the management fad of the day and will fall into
demise (similar to the 70% failure rate estimated in business process reengineering projects).
To further advance the KM feld and to give senior management a stronger sense of trust in the
tangible advantages of KM in their organizations, a comprehensive KM maturity model (K3M)
is needed. K3M, as discussed in this paper, provides a model for KM development and strategy
formulation to ensure greater success of KM implementation efforts. K3M is unique as it is the
frst KM maturity model that is based on learning, competencies, and business strategy.
Keywords: Knowledge Management, Maturity Model, Learning, Business Strategy, Competencies,
Knowledge Management Strategy
INTRODUCTION
The focus of our research is to develop a knowledge management maturity model (KMMM, or
K3M for short) for knowledge management (KM) development and strategy formulation. In this
paper, we broaden the traditional notion of KM to include the management of intellectual and
other intangible assets. This framework will blend diverse schools of thought into a unifed busi-
ness strategy, better structure the assessment and formulation of business and KM strategies, and
provide a clear implementation path based on mapping stakeholder needs and values to necessary
competencies and capabilities. This model could potentially be as successful as the Capability
Maturity Model (CMM) and CMMI (Capability Maturity Model Integration), developed by the
Software Engineering Institute (SEI) at Carnegie Mellon University. The CMM/CMMI (http://
www.sei.cmu.edu/cmmi) has been adopted worldwide, including North America, Europe, India,
Australia, Asia Pacifc, and the Far East.
Following the business strategy formulation review and analysis, a new framework, K3M, will
be proposed and developed to implement KM strategies in the context of supporting and aligning
with the underlying business strategies. Existing K3Ms will be examined. Finally, we will describe
a unifed strategic model comprised of business, KM, competency, and learning components, and
envision directions for future investigation.
164 LIEBOWITZ AND BECKMAN
BUSINESS STRATEGY
During the past decade, the prevailing paradigm for formulating business strategy has shifted
away from the traditional Porterian model of value chains and competitive forces, moving through
a period of strategic design and contingency positioning, and presently is focused on intangible
resource- and intellectual assetbased approaches to strategy. Many strategic theorists and practi-
tioners, including Nonaka and Takeuchi (1995), Hamel and Prahalad (1996), Zack (1999), Kaplan
and Norton (2004), Ulrich and Smallwood (2004), and Saint Onge and Armstrong (2004), have
arrived at similar conclusions that intellectual and other intangible assets are the key to sustainable
competitive advantage and long-term growth.
During this same 10-year period, much has been written in the literature regarding business
strategy formulation and implementation, but less has been written about its counterpart, KM
strategy. In this paper, we frst defne the focus and scope of business strategy and then discuss
its evolution and direction by examining a number of alternative perspectives. Some of these ap-
proaches highlight intellectual asset management (see references given earlier), stakeholder analysis
and value propositions (Beckman, 2003), and design, learning, and cognitive schools (Mintzberg,
Ahlstrand, and Lampel, 1998). Next, we evaluate the relevance and synergy between business and
KM strategies and evaluate the potential value from formulating a KM strategy.
Several thought leaders have focused on value creation as the strategic imperative. In Competing for
the Future, Gary Hamel and C.K. Prahalad (1996) state, Firms need a strategic architecture consisting
of what benefts are to be delivered to customers and what mechanisms they should use to deliver those
benefts. More recently, Robert Kaplan and David Norton (2004) refned this idea: Strategy describes
how an organization intends to create sustainable value for its shareholders [and constituents].
Another school of thought has defned strategy primarily in terms of intangible and intellectual
assets. In their book The Knowledge Creating Company, Ikujiro Nonaka and Hirotaka Takeuchi
(1995) suggest, The essence of strategy lies in developing the organizational capability to acquire,
create, accumulate, and exploit knowledge. Michael Zack (1999) believes that every strategic
position is linked to some set of intellectual resources and capabilities. That is, given what the
frm believes it must do to compete, there are some things it must know and know how to do.
The strategic choices that companies makeregarding technologies, products, services, markets,
processeshave a profound infuence on the knowledge, skills, and core competencies required to
compete and excel in an industry. Tissen, Andriessen, and Deprez (2000) focus on the knowledge
needed to successfully develop a strategy: there must be a clear and positive link between the
business strategy of a company and the development and use of knowledge within the organiza-
tion. Strategic KM . . . gives a balance by linking the building of our companys knowledge to
our companys business strategy. Attention is given to the impact of IT and the need to design the
organizational structure accordingly. In a best practice study, APQC (2000) suggests, Managing
knowledge has to be part of the business model and embraced as offering competitive advantage,
or it is highly unlikely to become an enterprise-wide strategy.
Many authorsHamel and Prahalad (1996), Rothwell, Prescott, and Taylor (1998), and Ulrich
and Smallwood (2004)see the strategic value from developing and aligning core competencies
and capabilities with business strategy. For example, Hubert Saint Onge and Charles Armstrong
(2004) believe that strategy is the key to which all other capabilities align. However, despite the
body of work and thought around intellectual assets, key capabilities, and core competencies,
and their relationship to business strategy, this approach generally has not realized the advertised
benefts. This approach has not been integrated with older, more classic approaches to strategy
formulation, nor has it been well developed or aligned for the deployment of strategic plans.
MOVING TOWARD A K3M FOR DEVELOPING KM STRATEGY 165
Hamel and Prahalad (1996) suggest building a strategic architecture consisting of what benefts
are to be delivered to customers and what mechanisms they should use to deliver those benefts.
The two mechanisms described are stretch goals (strategic intent and vision) and core compe-
tencies. As Micklethwait and Wooldridge (1996) suggest, Competencies form the basis for the
modern equivalent of portfolio management: if a frm knows what it is good at, it will venture
only into new areas where its core skills can be applied to create new products, and it will buy
only the companies that can add substantially to its portfolio of skills. This view leads to a very
conservative approach to strategykeep doing only what you are already good at.
In order to achieve sustainable competitive advantage, high performance, and increased growth,
businesses must defne a strategic vision that improves satisfaction of selected stakeholders, focuses
on value delivered, and can be implemented. We must break free from planning incrementalism
if stretch goals are to be developed and achieved. We must adopt a mindset capable of creating
visions that can transform the organization. Table 9.1 summarizes alternative perspectives on and
approaches to business strategy. The next section looks at business strategy formulation.
Business Strategy Formulation
Until now, several key concepts essential for sustained organizational success have been miss-
inginnovative management, human resources, and information technology disciplines; a meth-
odology for strategy formulation; and a methodology for strategy implementation. We propose
adoption of several innovative disciplines and will create methodologies for strategy formulation
consisting of unifed disciplines and integrated components, as well as synchronized development
and implementation of these strategies, using the K3M approach. Adoption of these disciplines
and methodologies will ensure formulation of a robust, achievable business strategy, improved
execution of the vision and strategy using the latest enabling disciplines, and more effective uti-
lization of organizational resources.
Business strategy consists of two major phases: formulation/development and implementa-
tion/execution. In this chapter, we primarily focus on formulation of business strategy and on
implementation of KM strategy. Business strategy formulation starts with assessment and ends
with a strategic vision and plan. However, before we can discuss the characteristics of differing
strategic formulation methods and develop a more powerful formulation methodology, we must
understand the importance of frameworks, stakeholders, and value propositions.
Strategy Formulation Frameworks and Taxonomies
One hindrance to effective strategic assessment and formulation is the lack of robust business
system frameworks. Generic frameworks and models enable strategists to better describe their
mission, vision, and strategic plans, in terms of capabilities, outputs, and outcomes. Frameworks
also ensure that no important components or dimensions are overlooked during the strategic for-
mulation and implementation processes. Later in this paper, these frameworks are used to organize
and characterize knowledge within KM strategies.
Several authors, including Rothwell, Prescott, and Taylor (1998), Tissen (2000), and Kaplan
and Norton (2004), have proposed conceptual frameworks with components. Rothwell, Prescott,
and Taylor (1998) defned a generic framework to implement HR strategies with the following
components: goals, components, mechanisms and processes, outputs and outcomes, and feedback.
Kaplan and Norton (2004) have described a strategic framework with four perspectives: fnancial,
customer, internal (process), and learning/growth (asset). Within the fnancial perspective, they
Table 9.1
Alternative Approaches to Business Strategy Formulation
Competitive Strategy Porter (1979, 1996)
Valuechaincomponents:inbound-suppliers,research,marketing,operations,humanresources,
information technology, outbound-distribution, customer service, and management
Competitiveforces:
Attractiveness of industry
Competitive forces
Barriers to entry
Customer attractiveness
Availability of substitute products
Portersapproachaimsatanindustrylevelofanalysis,butcanbecustomizedtospecifccompanies
Combinedwithastrengths-weaknesses-opportunities-threats(SWOT)assessment,thiswasthe
gold standard for many years
The Strategy Safari Mintzberg et al. (1998)
Positsthatthereare10perspectivesorschoolsofthoughtonstrategyformulation:design,planning,
positioning, entrepreneurial, cognitive, learning, power, cultural, environmental, and confguration.
KM,inthisview,iscoveredunderstrategyformulationasanemergentprocessinthelearning
school. There are also aspects that fall into the design school.
Viewsstrategyasdecisionmaking,ratherthanaformulationprocess.
Creating the Knowledge Company Nonaka and Takeuchi (1996)
The essence of strategy lies in developing the organizational capability to acquire, create, accumu-
late, and exploit knowledge.
But,withouttheassumptionofapplicationtoacriticalbusinessorstakeholderneed,thisstrategy
will likely fail. This approach focuses more on knowledge rather than competency.
The Knowledge Dividend Tissen, Andriessen, and Deprez (2000)
The frst step in creating a strategy is to identify our core competencies: the combination of unique
knowledge, talent and organizational capability. . . . Do our people possess specifc talents? Does
the way our organization is structured result in us having unique capabilities in, say, R&D or logis-
tics or sales?
Believeinstrategyasfttothecustomerandtheenvironment.
Believeinsubjectivemethods,chaos,andexperimentation,ratherthanobjectivemethodssuch
as SWOT and decision analytics.
Believeincorecompetencies,andwarnagainstrandomknowledgeacquisition.
Recommendtheallbrains,nobodyorganizationwithnowell-defnedjobsroles.
Despiteendorsingcorecompetencies,donotlinkthemtostrategyformulation.
Failstodeveloptheconnectionbetweenproducts,concepts,andtheworkforce.
Failstodescribeanddeterminehowtotelliftheknowledgebeingmanagediscriticalandof
signifcant value to the business and customer.
Developing a Knowledge Strategy Zack (1999)
BelievesinobjectivemethodssuchasSWOTanddecisionanalytics.
Dividesknowledgestrategyintoobjectiveandsubjectiveoptions.
Failstounderstandthatmanyformsofknowledgearerequiredforsuccess.
Failstonotethattheknowledgeformislessimportantthanthedomainordiscipline.
Strategy Maps Kaplan and Norton (2004)
Developstheconceptofstrategymapstolinkandalignintangibleassetstothebusinessstrategy
and to strategic themes.
DefnesfourgenericvaluepropositionsverysimilartoPorterscompetitivestrategies.
(Table 9.1 continued)
Doesareasonableconceptualjoboflinkingcompetenciesandstrategies,butneverdoesgetinto
any details of how this might work.
Failstointegrateotherdisciplinesandconceptssuchasstrategicsourcing,KM,innovativeinforma-
tion technology, and innovative management and human resources practices into the framework.
Weakonprovidingtheguidanceandmethodologytoaccomplishandimprovethestrategicframework.
Failstounderstandhowcentralproductsandservicesareinlinkingtheselectionofkeystake-
holders to the value proposition, and to the conceptualization of the business strategy.
posit that there are two strategies: productivity and growth. Within the process perspective, they
list four strategic process types: operations management, customer management, innovation, and
regulatory and social processes. Under the learning and growth perspective, they also list three
asset categories, human capital, information capital, and organizational capital, that are further
broken down into culture, leadership, alignment, and teamwork.
Additionally, other authors such as Wiig (2004) and Saint Onge and Armstrong (2004) have
proposed taxonomies and abstraction hierarchies. Taxonomies and abstraction levels further defne
some frameworks. Wiigs models comprise a taxonomy from strategic to operationalprinciples,
structure, methods, and operational context.
Identifying Key Stakeholders
One of the initial activities in strategy formulation should be to identify the key stakeholders to
be served. The needs, interests, values, and motivations of selected stakeholders need to be con-
sidered in terms of the products and services that will satisfy their requirements and constraints.
Failure to identify the key stakeholders and their distinctive value propositions greatly reduces
the likelihood of organizational success.
Developing Value Propositions
After these key stakeholder segments and their needs have been analyzed, selected, and prioritized,
the next phase is to develop value propositions. Before presenting the proposed business strategy
formulation framework, the notion of stakeholders and contextual value needs to be better under-
stood. Value is the worth or utility assigned by recipients to something that is received. Tangible
things that contain or convey value include products, services, and capabilities. Intangible value
can also be present in beliefs, perceptions, and opinions of oneself or ones organization or held
by stakeholders. For example, we may value status or recognition. Value is often quite personal
and contextual and often varies from person to person and from organization to organization.
Values are often embedded in organizational cultures and practices. Differing types of orga-
nizations may value and reward different results and methods. For example, one company may
reward creativity and performance, while another one may value conformity and celebrity. The
value context describes the needs and requirements of a specifc person or organization at a par-
ticular time, the ftness for use of the product or service to be delivered, a measure of the perceived
benefts, and what compensation in what form would be offered in return.
Competency and Capability Management
Besides developing value propositions and identifying the key stakeholders, competency man-
agement should also be considered as an underpinning of the business and resulting KM strat-
egy. According to Zack (1999), the knowledge strategy, then, can be thought of as balancing
knowledge-based resources and capabilities to the knowledge required for providing products or
services in ways superior to those of competitors. Competency management (Beckman, 2000,
2001a, 2001b, 2003; Beckman, Novak, and Novak, 2004) determines workforce capabilities, core
competencies, and required profciency levels and identifes needed development and performance
opportunities.
Competency modeling is also critical to enable capability management (CM) which in turn is a
key discipline supporting the business strategy. One of the key problems facing existing business
and KM strategies is the failure of strategy formulation to adequately guide subsequent strategy
implementation and execution. One approach to mitigating this weakness is by applying capabil-
ity and competency assessment, modeling, and sourcing methods that would provide the missing
link between formulation and implementation. Thus, by applying intellectual and other intangible
assets through competency management and KM, we can greatly improve the success and value
from strategy formulation.
Several authors have suggested strategic capabilities that are necessary for success. Tissen,
Andriessen, and Deprez (2000) described six basic capabilities that companies must have to suc-
ceed: the ability to produce, respond, anticipate, create, learn, and last (persevere). Ulrich and
Smallwood (2004) have proposed 11 intangible strategic capabilities that are critical to success:
talent, speed, shared mindset and coherent brand identity, accountability, collaboration, learning,
leadership, customer connectivity, strategic unity, innovation, and effciency.
Unifed Business Strategy Formulation Methodology
Capability management specifes the capabilities required to accomplish the business strat-
egy. The business framework defnes the components and perspectives necessary to build the
required capabilities. Now that we have examined some of the enabling disciplines and have
laid the groundwork for a fundamentally improved approach to formulating business strategy,
the components must be integrated and sequenced into a methodology. The unifed business
strategy formulation methodology outlined in Table 9.2 is derived from Beckmans (2003)
intelligent organization concept. This framework describes business system components and
provides a robust integration of perspectives and natural fow of processes (a new value chain)
from intention to resources to processes to results to valueguided and optimized through
organization and management. In Table 9.2, a six-phase, multicomponent, multidiscipline,
and multiperspective unifed framework is described for formulating business, competency,
and KM strategies.
Even when business and KM strategies have been well formulated, implementation has proved
to be much more diffcult than expected. This has played a large part of the motivation for devel-
oping this framework, which integrates multiple perspectives to ensure a robust methodology for
strategy execution.
KM STRATEGY BACKGROUND
Now that we have an appreciation for business strategy, we need to create and align a KM
strategy to the business strategy. A number of researchers and practitioners have been study-
ing techniques and methodologies for developing KM strategies and implementation plans
(Liebowitz, 2003b, 2006a, 2006b). Rothwell, Prescott, and Taylor (1998) defned a methodol-
ogy to implement KM strategy as transforming strategy into action. According to Chourides,
Table 9.2
Unifed Business Strategy Formulation Methodology
1. Assess Theorganizationintermsof:
Strengths-weaknesses-opportunities-threats (SWOT), asset classes
(fnancial, intellectual, technical, physical), and business system
components (stakeholder, environment, resource, process, result,
organization, and management/administration)
Products,services,andcapabilities
Competitivelandscapeandexternalenvironment
2. Identify Targetedstakeholdersegments
Theirneeds,interests,perspectives,andmotivations
Valuepropositionsfortargetedstakeholdersegments
3. Research Industrybestpractices
Lessonslearned
Innovativeinformationtechnology,humanresources,andmanagement
practices
4. Analyze Valuepropositions,SWOT,andassessmentsofoutputs,assetclasses,
operational and organizational problems
Determinethefeasibilityofreachingtargetkeycapabilitiesandrelated
capacities with similar to existing workforce, systems, and resources, and
refne strategy accordingly
Employstrategicsourcingtocloseandmitigatecapability,capacity,and
knowledge gaps
Analyzeanddesigntheoverallbusinessandinformationtechnologysys-
tems including enterprise architecture
5. Defne and Describe Products,services,andsystems
Requirementsintermsoffeatures,functionality,performance,andre-
sources
Aspectsandconnectionsthatsatisfythevaluepropositions
Targetcapabilitiesandcapacitiesintermsofbusinesssystemcomponents
Assetclassesneededtodeliverresults
Knowledgedomainsanddisciplinesrequiredforsuccess,andrequired
forms of knowledge
Rolesandresponsibilities,corecompetencies,jobfamilies,andworkroles
Positiondescription,qualifcations,andprofciencylevelsneededtofulfll
the target capabilities
6. Design Integratetheassetclassesandunifythebusinesssystemcomponentsinto
key capabilities
Applyinnovativeinformationtechnologyandmanagement,humanre-
sources, and industry practices to enable the strategy
Designorganizationalformsandauthorityrelationships
Developprocessesneededtoproduceproductsanddeliverservices
Createdocumentsincludingprinciplesandvalues,mission,vision,targetcon-
cept of operations, conceptual enterprise architecture, and sourcing strategy
Longbottom, and Murphy (2003), to get anywhere with KM, you must have a strategy and
individuals must be persuaded to contribute to both formulation and implementation. The KM
strategic plan has greater focus on the knowledge needs of the organization and an evaluation
of capabilities. Apostolou and Mentzas (2003) developed the Know-Net KM approach, which
includes the interplay between strategy, assets, process, systems, structure, individuals, teams,
interorganizations, and the organization itself.
Hansen, Nohria, and Tierney (1999) suggest that there are two mutually exclusive KM strate-
gies: codifcation and personalization. Codifcation focuses on capturing and sharing explicit
knowledge in formal databases. Personalization relies on sharing knowledge through direct per-
son-to-person contacts and relationships. To choose the correct KM strategy, they suggest the
following questions:
Doyouofferstandardizedorcustomizedproducts?
Doyouhaveamatureorinnovativeproject?
Doyourpeoplerelyonexplicitortacitknowledgetosolveproblems?
ODell, Wiig, and Odem (1999) performed benchmarking studies on KM strategies. They
have found successful organizations using KM strategies as a matrix of: KM as a business
strategy, transfer of knowledge and best practices, customer-focused knowledge, personal
responsibility for knowledge, intellectual asset management, and innovation and knowledge
creation.
Even when business and KM strategies have been well formulated, implementation has proven
to be much more diffcult than expected. This is a large part of the motivation for developing the
KMM model, which integrates multiple perspectives to ensure a robust methodology for strategy
execution. Levett and Guenov (2000) have developed a methodology for KM implementation that
looks at a four-phase approach:
1. Defne case study.
2. Capture KM practice.
3. Build a KM strategy.
4. Implement and evaluate.
April (2002) developed guidelines for building a knowledge strategy looking at the in-
terlinking of assets or resources, complementary resource combinations, and the strategic
architecture of the company. Becerra-Fernandez and Sabherwal (2001) apply contingency
perspectives to KM strategy. Ulrich and Smallwood (2004) use a capabilities audit to assess
the strategic intent and feasibility and provide guidance into improvement and development
opportunities. Nickerson and Silverman (1998) examined intellectual capital management
strategies and proposed a strategy integration analysis methodology that involves competi-
tion, technology, and intellectual capital management. Other researchers and practitioners
like McElroy (2003), Alavi and Leidner (2001), Mertins, Heisig, and Vorbeck (2003), Hult
(2003), Davenport and Probst (2002), NASA KM Team (2001), and Zimmermann (2003)
have been performing research dealing with KM strategy and implementation. The American
Productivity and Quality Centers Knowledge Management Benchmarking Study (APQC,
2000) suggests that the key features of successful implementation of KM are a champion,
communities of practice, and measurement.
Seeley and Dietrick (2001) discuss building a KM strategy using the following components:
governance, culture, content management, technology, application, and measurement. Earl (2001)
discusses a knowledge mapping, cartographic approach to KM where knowledge networking
and incentives to share knowledge are critical success factors. AT&T and Bain and Company
use this approach. Chavel and Despres (2002) in their 19972001 review of survey research in
KM found that surveys are typically used in KM research. Beckman (2003) presented a KM
strategy with seven dimensions: intellectual assets, content management, collaboration/commu-
nity, learning, innovation, improvement, and information technology. Liebowitz (2004) discusses
the importance of KM as a key pillar in an organizations human capital strategy. Holsapple
(2003), Wexler (2001), Noll, Frohlich, and Schiebel (2002), Hylton (2003), and Grey (1999)
talk about the importance of performing a knowledge audit as a frst step in developing a KM
strategy for an organization.
KNOWLEDGE MANAGEMENT MATURITY MODELS: A REVIEW
Now that we have carefully examined business, KM, and CM strategies, we can put the pieces
together toward developing a comprehensive KM maturity model, which we call K3M. Before
doing so, however, we should take a brief look at what others have done toward developing similar
types of models.
Several organizations have worked on developing a K3M. One such organization is Siemens
AG (Ehms and Langen, 2002). Siemens KMMM comprises a development model and an analysis
model. The development model of KMMM, which provides valuable information for reaching
the next maturity level, was developed along the lines of the CMM of the SEI (see earlier) at
Carnegie Mellon University (Ehms and Langen, 2002) (please note that SEI uses CMMI). The
analysis model creates transparency in all key areas of KM and demonstrates the potential for
improvement. The analysis concept is derived on the basis of the EFQM (European Foundation
for Quality Management) Model for Business Excellence (Ehms and Langen, 2002). When the
key areas of the KMMM were defned, it was ensured that they were compatible with the EFQM
enablers (Ehms and Langen, 2002).
By 2000, Hewlett-Packard Consulting had developed a KMMM based on the fve-level CMM.
The key attributes found within each level were selected to measure the KM capability of HP
consultants:
Level 5Optimized
Organizationsknowledgepolicies,practices,andactivitiesareimprovedcontinuously.
Participationinimprovingknowledgeprocessesisorganization-wide.
Coachesworkwithteamstoimprovetheirknowledgecreationandsharingandtoleverage
competencies.
Individualscontinuouslyimprovetheircapabilitytoexecutivecoreprocesses.
Level 4Managed
Mentoringactivitiesarematchedtodefnedobjectives.
Teamassignmentsaremadetointegratecomplementaryknowledgeandskills.
Standardknowledgeprocessesareavailablefortailoringandusebyteammembersinper-
forming their work.
Theorganizationadjustsitspracticestomotivateandsupportthedevelopmentofteam-based
knowledge competencies within the organization; compensation and rewards are tailored to
motivate improved team performance.
Level 3Defned
Theorganizationdevelopsknowledgepracticesandstrategiescriticaltoitsbusiness.
Knowledgerequirementsincorebusinessprocessesaredefned;plansaredevelopedtoad-
dress gaps.
Individualsaremotivatedtoimprovetheirknowledgeandskillsinthecorecompetenciesof
the business.
Compensation and rewards are tailored to motivate growth in knowledge processes and
competencies.
Communications activities are enhanced to improve the fow of information within the
organization.
Somecommontoolsandprocessesareusedwithintheorganization.
Clientengagementmaterialisregularlysharedacrosstheorganization.
Level 2Repeatable
Anenvironmentthatsupportstheperformanceofknowledgeprocessesisestablishedand
maintained.
Resources needed by the work force to perform the knowledge processes are made
available.
Individualsdevelopskillstoshareinformationandcoordinatetheiractivities.
Informationissharedacrosslevelsoftheorganization.
Job performance is measured against knowledge-sharing criteria; desired behavior is
rewarded.
Trainingopportunitiesonknowledgeprocesses,products,andtoolsaremadeavailable.
Repository is populated with existing tools and materials, primarily output from client
engagements.
Level 1Ad Hoc
Sharingoccurswithinthecontextofpersonalnetworksandrelationshipsloosecommuni-
ties of practice.
Consultantsusetheirownprocessesandtools.
Knowledgeexistsunstructuredandscattered.
Therearenomeasuresorrewardsforsharingandleveragingknowledge.
The development model of KMMM is used to assess the current position of an organization
on fve maturity levels adopted from the CMM levels as mentioned above. According to Marco
(2002), CMM is intended to be an appraisal methodology for ranking a companys informa-
tion technology-related activities based on six levels of maturity. The CMM maturity levels
are ranked based on key processes in fve areas: goals, commitment, ability, measurement, and
verifcation (Kay, 2005). CMM is useful for several reasons such as easy management, insight-
ful judgment, and a reliable comparison mechanism between the company and others within
a particular industry and sequentially offers a proved order for improvement (Marco, 2002).
Siemens KMMM development model implements these characteristics of the CMM levels as
relatively robust states of an organization based on IT activities and processes practiced over
time (Ehms and Langen, 2002). The rigid requirements for a step-by-step progress of a CMM
model, however, make its applicability better suited to large organizations. Moreover, the actual
progress an organization chooses depends on a variety of factors, and CMM does not point out
specifcally what to implement for improvement and merely indicates where the improvement
is needed (Kay, 2005).
The other component of KMMM, the analysis model, mainly focuses on important aspects
of KM as well as arising development issues. The KMMM analysis model is an extended or
differentiated adaptation of the European Foundation for Quality Management model in eight
fundamental areas of KM (Ehms and Langen, 2002). The advantage of an analysis model is that
it digs deeper into organizational practices supporting KM and describes 64 KM topics arranged
in the areas: strategy/knowledge goals, environment/partnerships, people/competencies, col-
laboration/culture, leadership/support, knowledge structure/forms, technology/infrastructure, and
process/role/organization. Although the KMMM analysis model contributes necessary features
that an organization needs to consider, the organizational development cannot solely depend on
this checklist because it does not provide clear and concise action plans or prioritization of
topics (Ehms and Langen, 2002).
The effective implementation of KMMM (Mertins et al., 2003) is motivated through interviews,
workshops, and feedback sessions, which are structured for planning and data collections. This
process of acquiring essential information is often referred to as the auditing process of KMMM.
Six phases of a KMMM audit include orientation and planning, motivation and data collection,
consolidation and preparation, feedback and consensus, ideas for solutions and action proposals,
and report and presentation (Ehms and Langen, 2002).
Dalkir (2005) points out three other KM Maturity Models. Infosys has developed a KMMM
with fve levels: default, reactive, aware, convinced, and sharing. The Forrester Group also has
a KMMM with three phases: assisted, self-service, and organic. The knowledge process quality
model (KPQM) has fve maturity levels modeled after CMM, namely the initial maturity phase,
aware, established, quantitatively managed, and optimizing phases.
Somewhat related to these KM and people maturity models have been a number of KM
capability assessments that have been developed. Freeze and Kulkarni (2003) constructed the
Knowledge Management Capability Assessment to measure knowledge assets identifed as
knowledge capability areas. Their framework provides a method to assess the capability of an
organizations KM initiatives in lessons learned, knowledge documents, expertise, and data.
Harigopal and Satyadas (2001) developed the Cognizant Enterprise Maturity Model (CEMM)
to measure 15 key maturity areas within an organization to improve business value. Lee, Lee,
and Kang (2005) developed the Knowledge Management Performance Index to measure KM
performance in organizations. Montano et al. (2001) and Rubenstein-Montano et al. (2001)
developed the SMARTVision KM methodology based on four phases: conceptualize and
strategize, act, evaluate, and implement. The threaded tenets throughout the phases are project
management, technical/process reviews, change management, process training, and quality
assurance. Rubenstein-Montano et al. (2001) also stress the need for a double-loop learning
environment as part of the KM approach and show numerous KM frameworks that have been
developed over the years. Liebowitz (1999, 2003a) and Liebowitz and Megbolugbe (2003) echo
these fndings by discussing the need to put more rigor into KM to transform it into more of a
science than an art.
Gottschalk (2005) discusses four stages of maturity for KM technology in organizations: Stage 1
(person-to-technology systems), Stage 2 (person-to-person systems), Stage 3 (person-to-information
systems), and Stage 4 (person-to-system). The stages progress from the end-user tools to the who-
knows-what systems, to the what-they-know systems, to fnally the how-they-think systems.
DEVELOPING K3M: THE KNOWLEDGE MANAGEMENT
MATURITY MODEL
As organizations are delving into KM, there does not appear to be a KMMM that links KM,
learning, competencies, and business strategies. K3M is our model for trying to determine the
KM maturity level of an organization, based upon its KM, learning, competency, and business
strategies. K3M consists of the following maturity levels, developed by the authors as shown in
Table 9.3. The maturity levels range from being unaware of KM to understanding the full utility
and value of KM to the organization. The key indicators of each level are shown in Table 9.4.
Now we will demonstrate the maturity levels for each of the forms of knowledge, per Table
9.5. Table 9.5 shows the mapping between maturity levels and types of knowledge. Learning,
competency, and business strategies can also be linked to these KM maturity levels, as shown
in Table 9.6.
The learning states and business strategies mirror the KM maturity levels. For example, the
nonawareness maturity level (Level 0) is steady state and the business strategy is the status quo
(business as usual) because the organization is not cognizant of KM. As the awareness level
heightens for the organization (Maturity Level 1), the curiosity learning state sets in and the
business strategy might be somewhat conservative as KM isnt proven within the organization.
At Level 2 (Initiation Stage), the KM pilots foster an indoctrination learning style as people
are seeing how KM can be used in the organization. With this indoctrination period, a sense
of optimism is created if KM can show value-added benefts. Level 3, the Intrigue and Interest
Stage, creates an enthusiasm learning style as the KM pilots metamorphose into full-blown
projects with change management processes built in as part of the business strategy for the
organization. The Penetration stage, Level 4 in K3M, provides an enterprise-wide approach to
KM, which generates a replication and discovery learning state. This allows KM to be further
institutionalized (as the business strategy) within the organization. Level 5, the Utility Stage,
fosters a renewal and creation learning style for generating new knowledge and replenishing the
knowledge base of the organization. This should ultimately lead to organizational transforma-
tion as a business strategy.
The competencies associated with these levels could be analogized with those in the
professorial ranks in academe. Level 0 is the instructor level, where there is a basic level of
competency and profciency. Level 1 might be similar to the assistant professor, where there
is tremendous curiosity and a thriving search for knowledge. As the assistant professor ma-
tures into the associate professor, after gaining promotion and tenure, the assistant professor
becomes indoctrinated (Level 2) by going through the tenure process and, once achieving
tenure, is now part of the permanent establishment. The associate professor typically wants to
strive to become a professor and beyond, and shows the enthusiasm (Level 3) toward moving
to that rank. Once a professor, new discoveries and increased international visibilities may
result in becoming an endowed professor (Level 4), with the dream of aspiring to hold an
endowed chair (Level 5).
As components of Levels 4 and 5, knowledge sharing profciencies should be part of the
employees daily work life. Some profciencies can be used as shown in Table 9.7.
Table 9.3
K3M Levels
Maturity
Level Description
0 Nonawareness: Organization has not heard of knowledge management
1 Awareness: Organization is introduced to KM and knowledge sharing concepts
2 Initiation: Small KM pilots are developed and sprinkled throughout the organization
3 Intrigue and interest: KM pilots spark enthusiasm and interest and give way to full-blown
KM projects
4 Penetration: The discovery phase leads to mass appeal and employees are embracing
knowledge sharing tenets and are embedding KM activities into their daily work lives
5 Utility: The organization (and its people) realizes the collective value of knowledge sharing
and can assess the utility and resulting impact of KM on the organization
Table 9.4
Key Indicators for the K3M Levels
Maturity
Level Key Indicators
0 Ineffcienciesinsearchingforneededinformationandknowledge
Notknowingwhotocontactwithintheorganizationforansweringquestions
Lackofinnovation
Silo/stovepipingeffects
1 RealizationthatyourcompetitionisapplyingKMconceptsandthatyoumaybelagging
behind your competition
Realizationthattheremaybeabetterwayofdoingbusinessandreachingouttothe
customers/stakeholders by using KM techniques
2 TestingthewatersthroughselectedKMpilots
KMtechnologyinfrastructurestartstobebuilt
Metricsaredevelopedtomeasureinitialsuccess
3 Employeemoraleincreasesduetoanimprovedsenseofbelonging/communityfromKM
efforts
KMorganizationalinfrastructurestartstobeputinplace
KMpilotsmigrateintofull-blownprojects
4 Enterprise-wideknowledgemanagement
Learningandknowledgesharingprofcienciesarebuiltintotherecognitionandreward
system of the organization
KMprocessesareputinplacetoensurethecapture,sharing,application,andcreation
of knowledge
Newproductsorservicesarecreated
Functionalsilosbegintocrumble
5 NirvanaiscreatedthroughtheKMactivitiesbyestablishinghighlevelsofemployee
morale, increased worker productivity, improved institutional memory building, improved
customer relations, increased innovation, a thriving continuous learning culture exists,
improved access to people and associated information and knowledge, and a working
smarter, not harder environment permeates
Table 9.6
Mapping Learning, Knowledge, and Business Strategy to K3M Levels
Maturity
Level Learning Knowledge Business Strategy
0 Steady state Nonawareness Absenceno strategy
1 Curiosity Awareness Conservatism
2 Indoctrination Initiation Incrementalism
3 Enthusiasm Intrigue and interest Improvement and institutionalization
4 Replication and discovery Penetration Modernization
5 Renewal and creation Utility Transformation
Table 9.7
Individual Knowledge Sharing Profciencies/Competencies
Communicateswellwiththosewithinhisorherdepartment(intradepartmentcommunications)
Communicateseffectivelywiththoseinotherdepartments(interdepartmentcommunications)
Sharesknowledgethroughvariousknowledgemanagementmechanisms,suchasmentoring,
conference trip report discussions via brown bag lunches, storytelling (organizational narratives),
lessons learned/best practice content contribution, online communities/threaded discussions,
newsletter contributions, etc.) (knowledge contribution)
Activelyparticipatesincross-functionalteams(collaboration)
Regularlydistributesarticlesofinteresttootheremployees(knowledgedissemination)
Showsvalue-addedbeneftsfromknowledgereceivedfromothersandknowledgegainedby
others (knowledge value)
Willingtobeinnovative,takerisks,andtrynewideas(knowledgecreation)
Table 9.5
KM Maturity Model Defned in Terms of Knowledge Forms
Maturity
Level
Objective
Knowledge
Subjective
Knowledge
Interactive
Knowledge
Organizational
Knowledge
0 Little documented
data, information,
and knowledge
Ad hocwork
not supported by
competencies
Minimal, stifed,
and random
communications;
Rumor reigns
Chaos, little
leadership,
dysfunctional
bureaucracy
1 Some documents Identify core
competencies
One-way
communication
Structure and lines of
authority
2 Indoctrination Assess profciencies Conversation Assessment and
metrics
3 Enthusiasm Develop workforce Collaboration Processes and
practices
4 Replication and
discovery
Allocatestrategic
sourcing
Communities of
practice
Strong management
and human resources
practices
5 Expert systems;
data mining and
knowledge
discovery
Leverage
organizational
capabilities key to
success
High degree of
knowledge sharing
Governance, policies,
and culture support
and enable strategic
intent
SUMMARY
Anecdotal and qualitative studies reveal that most organizations are somewhere between Levels
0 and 3 in K3M. Levels 4 and 5 typically require organizations to have senior leadership in har-
nessing the true value that KM and strategic human capital have to offer. Moving to the highest
KM maturity levels requires a senior management commitment to KM and requires mechanisms,
with associated recognitions and rewards, to instill knowledge sharing within the fabric of the
organization. This will ultimately contribute toward creating a continuous learning culture in the
organization and should foster an innovative climate among the employees.
This paper provides a basis for linking business and knowledge strategies toward developing
a KMMM for an organization. Integrating learning, competency, and business strategies within a
KM context is needed for organizations to succeed in KM. A few KMMMs have been developed,
but most of them fail to make the connections between these areas. The K3M approach has been
proposed as a possible solution to this defciency. Future work needs to be conducted to apply
this framework to organizations, and to fesh out the necessary metrics at each K3M level. This
paper provides the underlying literature foundation and roadmap to move in this direction.
REFERENCES
American Productivity and Quality Center (APQC). 2000. Successfully implementing knowledge manage-
ment. Best Practice Report.
Apostolou, D., and Mentzas, G. 2003. Experiences from knowledge management implementations in com-
panies of the software sector. Business Process Management Journal, 9, 3.
April, K. 2002. Guidelines for developing a k-strategy. Journal of Knowledge Management, 6, 5.
Becerra-Fernandez, I., and Sabherwal, R. 2001. Organizational knowledge management: A contingency
Becerra-Fernandez, I.; A. Gonzalez, A.; and Sabherwal, R. 2004. Knowledge Management: Challenges,
Solutions, and Technologies. Upper Saddle River, NJ: Prentice-Hall.
Beckman, T. 2000. A framework for measuring and valuing knowledge management, intellectual assets, and
organizational learning. KM World 2000 Conference Proceedings, Santa Clara, CA.
. 2001a. Intelligent Organizations: A Unifying Framework for Knowledge Management, Organi-
zational Learning, and Intellectual Capital Development. Presentation at Society for the Advancement
of Socio-Economics International Conference on Knowledge: The New Wealth of Nations. Amsterdam,
Netherlands: Society for the Advancement of Socio-Economics.
. 2001b. Creating business value from knowledge management: The fusion of knowledge and
technology. In R.Barquin, A. Bennett, and S. Remez (eds.), Knowledge Management: The Catalyst
for Electronic Government. Vienna, VA: Management Concepts.
. 2003. KM strategy: Integrating seven dimensions of KM and aligning KM with the business strategy.
Presented at KM West Executive Summit, Bellevue, WA.
Beckman, T.; Novak, M.; and Novak, M., Jr. 2004. Competency-based socio-economic development. Society
for the Advancement of Socio-Economics. Available at www.sase.org.
Chourides, P.; Longbottom, D.; and Murphy, W. 2003. Excellence in knowledge management: An empirical
study to identify critical factors and performance measures. Measuring Business Excellence, 7, 2. West
Yorkshire, England: MCB University Press.
Dalkir, K. 2005. Knowledge Management in Theory and Practice. Burlington, MA: Elsevier Butterworth-
Heinemann.
Davenport, T., and Probst, G. (eds.). 2002. Knowledge Management Case Book: Siemens Best Practices, 2nd
ed. Berlin: John Wiley & Sons/Publicis Corporate Publishing.
Systems, 18, 1, 225233.
Ehms, K., and Langen, M. 2002. Holistic development of knowledge management with KMMM.
Available at http://www.knowledgeboard.com/doclibrary/knowledgeboard/kmmm_article_siemens_
2002.pdf.
Freeze, R., and Kulkarni, U. 2003. Knowledge management capability assessment: Validating a knowledge
assets measurement instrument [white paper]. Tempe, AZ: W.P. Carey School of Business, Arizona State
University.
Gottschalk, P. 2005. Strategic Knowledge Management Technology. Hershey, PA: Idea Group Publishing.
Grey, D. 1999. Knowledge mapping: A practical overview. Available at http://www.it-consultancy.com/ex-
tern/sws/knowmap.html.
Hamel, G., and Prahalad, C. 1996. Competing for the Future. Boston: Harvard Business School Press.
Hansen, M.; Nohria, N. and Tierney, T. 1999. Whats your strategy for managing knowledge? Harvard Busi-
ness Review, MarchApril, 106116.
Harigopal, U., and Satyadas, A. 2001. Cognizant Enterprise Maturity Model (CEMM). IEEE Transactions
in Systems, Man, and Cybernetics Part CApplications and Reviews, 31.
Holsapple, C. (ed.). 2003. The Handbook on Knowledge Management (2 vols.). London: Springer-Verlag.
Hult, G.T. 2003. An integration of thoughts on knowledge management. Decision Sciences, 34, 2,
189195.
Hylton, A. 2003. The knowledge audit. In J. Edwards (ed.), Proceedings of the Knowledge Manage-
ment Aston Conference 2003. Birmingham, UK: The Operational Research Society/Aston Business
School.
Kaplan, R., and Norton, D. 2004. Strategy Maps: Converting Intangible Assets into Tangible Outcomes.
Boston: Harvard Business School Press.
Kay, R. 2005. CMMI. Available at http://www.computerworld.com/managementtopics/outsourcing/
story/0,10801,99159,00.html.
Lee, K.; Lee, S.; and Kang, I. 2005. KMPI: Measuring knowledge management performance. Information
& Management, 42, 3, 469482.
Levett, G., and Guenov, M. 2000. A methodology for knowledge management implementation. Journal of
Knowledge Management, 4, 3, 258269.
Liebowitz, J. (ed.). 1999. The Knowledge Management Handbook. Boca Raton, FL: CRC Press.
Liebowitz, J. 2003a. Putting more rigor into knowledge management. In J. Edwards (ed.), Proceedings
of the Knowledge Management Aston Conference 2003. Birmingham, UK: The Operational Research
Society/Aston Business School.
. 2003b. Aggressively pursuing knowledge management over two years: A case study at a US govern-
ment organization. Knowledge Management Research & Practice Journal, 1, 2.
. 2004. Addressing the Human Capital Crisis in the Federal Government: A Knowledge Management
Perspective. London: Butterworth-Heinemann/Elsevier.
. 2006a. Strategic Intelligence: Business Intelligence, Competitive Intelligence, and Knowledge
Management. New York: Auerbach Publishing/Taylor & Francis.
. 2006b. What They Didnt Tell You About Knowledge Management. Lanham, MD: Scarecrow Press/
Rowman & Littlefeld.
Liebowitz, J., and Beckman, T. 1998. Knowledge Organizations: What Every Manager Should Know. Boca
Raton, FL: CRC Press.
Liebowitz, J., and Megbolugbe, I. 2003. A set of frameworks to aid the project manager in conceptualizing
and implementing knowledge management initiatives. International Journal of Project Management,
21, 3, 189198.
Marco, D. 2002. Meta data & knowledge management: Capability maturity model: An introduction. Avail-
able at http://www.dmreview.com/article_sub.cfm?articleId=5567.
Marquardt, M. 1996 Building the Learning Organization. New York: McGraw-Hill.
McElroy, M. 2003. The New Knowledge Management. Burlington, MA: Butterworth-Heinemann/Elsevier.
Mertins, K.; Heisig, P.; and Vorbeck, J. (eds.). 2003. Knowledge Management: Concepts and Best Practices.
London: Springer-Verlag.
Mintzberg, H.; Ahlstrand, B.; and Lampel, J. 1998. Strategy Safari: A Guided Tour through the Wilds of
Strategic Management. New York: Free Press.
Montano, B.; Liebowitz, J.; Buchwalter, J.; McCaw, D.; Newman, B.; and Rebeck, K. 2001. SMARTVision:
A knowledge management methodology. Journal of Knowledge Management, 5, 4, 300310.
NASA Knowledge Management Team. 2001. Strategic Plan for Knowledge Management. Washington, DC,
March 18.
Nickerson, J., and Silverman, B. 1998. Intellectual capital management strategy: The foundation of successful
new business generation. Journal of Knowledge Management, 1, 4, 320331.
Noll, M.; Frohlich, D.; and Schiebel, E. 2002. Knowledge maps of knowledge management tools: Informa-
tion visualization with BibTechMon. In D. Karagiannis and U. Reimer (eds.), Practical Applications of
Knowledge Management 2002 Conference Proceedings. London: Springer-Verlag.
ODell, C.; Wiig, K.; and Odem, P. 1999. Benchmarking unveils emerging knowledge management strate-
gies, Benchmarking Journal, 6, 3, 202211.
Porter, M. 1979. How competitive forces shape strategy. Harvard Business Review, March/April,
137145.
. 1996. What is strategy? Harvard Business Review, November/December, 6178.
Rothwell, W.; Prescott, R.; and Taylor, M. 1998. Strategic Human Resource Leader. Mountain View, CA:
Davies-Black Publishing.
Rubenstein-Montano, B.; Liebowitz, J.; Buchwalter, J.; McCaw, D.; Newman, B.; and Rebeck, K. 2001. A
systems thinking framework for knowledge management. Decision Support Systems, December, 516.
Saint Onge, H., and Armstrong, C. 2004. The Conductive Organization: Building Beyond Sustainability.
London: Elsevier Butterworth-Heinemann.
Sabherwal, R., and Becerra-Fernandez, I. 2003. An empirical study of the effect of knowledge management
processes at individual, group, and organizational levels. Decision Sciences, 34, 2, 225260.
Seeley, C., and Dietrick, W. 2001. Crafting a knowledge management strategy [KM Review Special Report].
London: Melcrum Publishing.
SEI (Software Engineering Institute). 2001. People Capability Maturity Model (P-CMM) Version 2. Pitts-
burgh, PA: Carnegie Mellon University. Available at http://www.sei.cmu.edu/publications/documents/01.
reports/01mm001.html. Accessed October 2007.
Sveiby, K.E. 2001. A knowledge-based theory of the frm to guide in strategy formulation. Journal of Intel-
lectual Capital, 2, 4, 344358.
Tissen, R.; Andriessen, D.; and Deprez, F. 2000. The Knowledge Dividend: Creating High-Performance
Companies through Value-Based Knowledge Management. Upper Saddle River, NJ: Prentice-Hall.
Treacy, M., and Wiersema, F. 1995. The Discipline of Market Leaders: Choose Your Customers, Narrow Your
Focus, Dominate Your Market. New York: HarperCollins.
Ulrich, D., and Smallwood, N. 2004. Capitalizing on capabilities. Harvard Business Review, June,
119127.
Wexler, M. 2001. The who, what, and why of knowledge mapping. Journal of Knowledge Management, 5,
3, 249263.
Zack, M. 1999. Developing a knowledge strategy. California Management Review, Spring, 41, 3,
125145.
Zimmermann, K. 2003. Learning from success and failure. KM World, October.
180
ChaptEr 10
BUILDING KNOWLEDGE MANAGEMENT
SYSTEMS TO IMPROVE PROFITS AND
CREATE LOYAL USERS
Lessons from the Pharmaceutical Industry
alan r. dEnnis, donG-Gil Ko, and paUl f. Clay
Abstract: Although many organizations are implementing knowledge management systems (KMS),
there is little empirical evidence about the impacts of KMS use on performance, and what infuences
employees to use a KMS. This chapter describes the experiences of a multinational pharmaceutical
frm that implemented a KMS to support its sales representatives. We frst describe how the frm
developed and implemented its KMS, including the failure of the frst version of the system, and
the transition to the ultimately successful second and third versions. We then examine the impact
that this KMS has had on sales performance. Next, we investigate the factors that infuenced the
target users choice to use or not use it. We conclude with four lessons for other frms.
Keywords: Knowledge Management Systems, Lessons Learned, Multi-Method, Longitudinal KMS
Implementations, Loyal Use, KM Strategy.
INTRODUCTION
Many corporations are implementing knowledge management systems (KMS), a class of informa-
tion systems applied to managing organizational knowledge (Alavi and Leidner, 2001; Kaplan,
2002). Many government and non-for-proft organizations are also beginning to implement KMS
(Motsenigos and Young, 2002). The goal of KMS is to provide knowledge to its users so that they
perform their work more effectively and effciently (Markus, 2001).
Popular press accounts suggest that KMS use can have benefcial impacts on frm performance
(e.g., Derr, 2006). In this chapter, we examine the efforts of a large multinational pharmaceutical
frm in the United States (herein called Farmaco) to develop and deploy a KMS to support their
feld sales representatives. The system is widely used and generally considered to be successful by
both its users and senior management at Farmaco. We begin by describing how Farmaco designed,
developed, and implemented its KMS, including its initial failure and redesign into a successful
system. In the next section, we examine the impact of the use of the system of sales performance
to evaluate whether KMS use in this context actually produced bottom line impacts as believed
by its users and managers. In the third section, we examine the factors infuencing the use of the
systemwhat factors induce users to actually use the system to change the way they work. We
conclude with four general lessons learned and implications for other organizations.
BUILDING KM SYSTEMS IN THE PHARMACEUTICAL INDUSTRY 181
DEVELOPING THE KNOWLEDGE MANAGEMENT SYSTEM
Farmaco
Farmaco is one of the top 10 global pharmaceutical frms. It has offces, research centers, and
manufacturing sites in more than a dozen countries, but its administrative heart lies in its headquar-
ters in the United States. Farmaco is organized by both functional business units and geographic
business units. The key research and development function is located in a separate business unit
that is managed throughout the world under one executive. The same is true of manufacturing.
However, sales is organized by geographic region. By far the largest individual sales unit is the
U.S. business unit, which is responsible for all sales within the United States. Other business
units are also organized by country or geographic region within the world (e.g., Canada, United
Kingdom, Africa). The U.S. business unit, in turn, is subdivided into eight principal operating
divisions, each focusing on a different set of disease states for which the frm offers drugs (e.g.,
asthma, gastrointestinal disorders) or on types of physicians (e.g., primary care physicians, spe-
cialist physicians, HMO administrators).
Because Farmaco focuses on prescription drugs, almost all of its revenues come from a very
small set of productsusually a few dozen individual drugs. The pressure to manage this select
set of brands, each of which has only a limited life before generics arrive, is extremely important
to the frms success.
In the market for prescription drugs, the end consumer has historically played only a small role in
the purchase decision. Drugs are prescribed by physicians and nurse practitioners, limited in some
cases by the rules imposed by hospitals, health maintenance organizations (HMOs), and government
health programs (e.g., Medicare). Like most pharmaceutical frms, Farmaco has feld sales repre-
sentatives who visit physicians and clinics to provide information about its products to physicians.
The role of the feld sales representatives is constrained by government regulation, so there are many
rules that representatives must be careful to follow when talking with physicians.
Sales representatives often fnd it diffcult to meet with physicians, and when they do meet, they
often have only a few minutes to present their information. For this reason, many frms, includ-
ing Farmaco, employ full-time physicians and also hire physicians as external contractors to run
workshops and present papers at medical conferences to present evidence on the effectiveness of
their products.
In recent years, marketing strategies have expanded. Pharmaceutical frms have begun to ad-
vertise directly to the end consumer in an attempt to induce individuals with chronic disease states
who have not sought medical treatment to visit their physician, because once they see a physician
there is a good probability that the physician will prescribe the frms product. Such advertisements
have also been shown to create brand awareness, so that some patients request specifc drugs when
they visit their physician, who may prescribe what the patient requests.
The pharmaceutical industry is a knowledge intensive industry; medical research is inherently
knowledge work, and the relationship between sales representatives and the physicians is intensely
knowledge based. It is not surprising, then, that most pharmaceutical frms have implemented
knowledge management programs. Farmaco is no exception.
The First Field Sales KMS
The development of the feld sales KMS began in 1998. The senior vice president responsible
for the U.S. business unit recognized the need to improve knowledge sharing among the several
182 DENNIS, KO, AND CLAY
thousand members of the companys feld sales force. Because of the knowledge intensive nature
of the drug selling process and the fact that many of the feld sales representatives were new col-
lege hires with no prior experience beyond a university degree in pharmaceuticals or business,
he decided it was important to provide key knowledge that sales representatives needed through
the KMS.
By 1998, all feld sales representatives had been equipped with laptops with Lotus Notes as
the primary e-mail platform. Lotus Notes was well suited to knowledge management and so the
initial system was developed using Notes. The team that developed the initial system was led by an
information technology (IT) manager with specialists assigned to the team from the IT group.
In the summer of 1998, with help from a leading consulting frm with an expert knowledge
management practice, the team began developing a knowledge management strategy for the feld
sales force. They immediately focused on the communities of practice strategy (Brown and Duguid,
1998), based on extensive popular press articles and conversations with feld sales representatives.
The team worked closely with consultants from the same leading multinational consulting frm who
were working on the corporate KMS to design the initial KM strategy and the system itself.
The frst version of the KMS was deployed during the spring of 1999. Because of the size of
the system, it could not be easily downloaded over the Internet through the traditional dial-up
modems used by the sales representatives laptops. Therefore, the system was deployed division by
division as the sales representatives attended the annual sales meetings held by each division over
that spring. The roll-out to all sales representatives was completed before the end of the spring,
and sales representatives began using the KMS.
The results were disappointing. By the fall of 1999, the several thousand feld sales represen-
tatives had contributed a total of only 35 items across the entire KMS. The KMS had been built
with a belief that if you build it, they will come, but, unfortunately, after it had been built, the
sales representatives did not come. It was clearly a failure; there was no community of practice.
Something had to change.
The Second Field Sales KMS
In February 2000, a new KM manager was appointed to the U.S. business unit. The manager was
a district sales manager with many years of feld experience. He began by hiring several experi-
enced feld sales representatives and brand marketing associates from both inside and outside the
frm into his KM group.
The KM manager adopted a very different approach to knowledge management from the
previous system. To him, the success of a KMS was 75% behavior change among the target
users, 20% business process change, and 5% technology. He and the rest of the KM team went
throughout the U.S. business unit talking with sales representatives, managers, and key opinion
leaders to learn what was needed from the KMS and to sell the idea of knowledge sharing. Based
on these discussions, they came to understand that the communities of practice strategy failed
not because of implementation but because it was not an appropriate strategy for an organization
where communities of practice did not formally exist, although it was intuitively appealing given
their multiple divisions and thousands of sales representatives.
Instead, the KM manager came to see knowledge as product and the sales representatives as
potential consumers. Once in this mindset, the KM team set about to develop and sell the KMS
product as they would any other product; after all, they were experienced sales managers and
brand managers with many years of experience, so designing and selling consumer products were
their expertise. They changed the KMS mission from one of promoting the sharing of knowledge
among sales representatives to one of designing and building both the knowledge content and
the KMS structure to deliver it. In other words, they shifted from the communities of practice
strategy to a codifcation strategy (Hansen, Nohria, and Tierney. 1999). They branded KM in a
new way, and although they encouraged the sharing of knowledge among sales representatives,
the key focus was on identifying knowledge useful to the sales representatives and developing a
formal, sustainable, means to develop and maintain that knowledge over the long term. A second-
ary focus was on the development of a KMS that would make it simple for sales representatives
to consume that knowledge.
The KM team worked closely with the original IT team and key managers in the headquar-
ters and sales districts to develop a taxonomy of important knowledge needed by the feld sales
representatives. It became apparent that while there was a core set of knowledge needed by all
feld sales representatives, the knowledge needed about individual drugs differed depending upon
which division the sales representative worked for. Sales representatives selling asthma drugs, for
example, had no use for knowledge pertaining to drugs intended to treat gastrointestinal disorders.
Therefore, a new knowledge taxonomy was developed so that knowledge about each of the frms
drugs was organized separately. Sales representatives would only have access to knowledge about
the drugs they sold.
The next step was to create the knowledge itself needed to populate these taxonomies. In this
case, the KM team turned to the managers and associates within sales operations and the brand
management team responsible for each drug. These non-KM experts would develop initial drafts
of the knowledge content, which they would provide to the KM team. The KM team would format
the documents and put them in the proper locations in the KMS.
The system was intended to be the primary knowledge repository used by the feld sales repre-
sentatives and the sales managers. All knowledge communication with the feld sales representatives
was expected to be conducted through the feld sales system. Instead of mailing paper marketing
materials and advisories, for example, managers would now create them in Word and PowerPoint
and post them into the feld sales system.
The KM team also realized that it was important to enable the feld sales representatives them-
selves to contribute sales tips and practical advice for use by other sales representatives. However,
because of strict government regulatory control over communication with the physicians, all such
tips needed to be frst approved by the frms legal department. A formal four-step process was
therefore developed for validating all content sent in from the feld sales representatives. Tips
were frst vetted by the KM team itself to ensure the content was coherent and complete. Next,
the tip was submitted to the legal group to ensure it was consistent with all rules, regulations, and
good promotional practice guidelines. Then, the tip was sent to the brand management team to
ensure that it was consistent with the marketing strategy for the drug. Next, the tip was sent to
the sales operations group for peer review by a panel of fve sales representatives to ensure the
contribution had real value. Finally, once the tip had been approved, it was entered into the feld
sales system by the KM team. While this sounds like a lengthy process, most tips were approved
and posted (or rejected) within 4 weeks of receipt. Field sales representatives were rewarded by
receiving sales points for each tip that was ultimately accepted (these points were part of the
usual commission structure received by all sales representatives; the points received for each tip
was equivalent to approximately $60).
At the same time, an entirely new user interface was designed for Lotus Notes. The new interface
was intended to better match the structure of the knowledge and to be much more user-friendly
than the original Notes interface. The interface was very graphic, using images and icons for key
tasks, rather than the text-based interface of the old system.
Although the graphic design was important, it was not the key technical difference between the
frst and second KMS. The key difference lay in rethinking the users desktops. In the frst version,
the KMS was available on the desktop through Lotus Notes, but it required several mouse-clicks to
reach. In the new design, the KMS was the users desktops. The frst screen the users saw after turning
on their computers was the KMS. All other tools (e-mail, expense accounts, status reports, Microsoft
Offce) were links available from within the KMS. Thus, the KMS was always present and easy to
access. It became a part of the users work environment the moment they used their computers.
The new system, and the frst release of second version, was deployed in the spring of 2001,
with some promising initial success. Field sales representatives began to use the knowledge man-
agement system, and, very slowly, knowledge contributions started to be received from the feld.
By late summer of 2001, the feld sales system was frmly established within the U.S. business
unit, with most sales representatives using it on a regular basis.
The Third Field Sales KMS
By late summer of 2001, it also had become clear that the initial knowledge taxonomies were
not ideal, and the easy-to-use user interface was not as easy to use as expected. The technical
architecture also proved to be too infexible to easily accommodate changes. Therefore, in late
summer of 2001, the U.S. business unit began redesigning the feld sales system. The goals of
this version were to streamline and improve the fundamental knowledge taxonomies and to use
a more standard Internet-like user interface rather than the Notes-specifc graphic user interface.
This time, the taxonomies were changed so that each division had its own taxonomy.
The KM team met with numerous feld sales representatives, including all KM leads; each
district sales manager had appointed one sales representative as the KM lead or primary point
of contact for KMS for his or her district. The team also met with many district sales managers
and brand management teams to design a new knowledge structure and a new interface. The team
started by focusing on the largest division within the U.S. business unit and a pilot version of the
system was deployed in November 2001.
The knowledge structure for the system was designed in what the team called a T-structure,
which had two distinct parts. Across the top of the T (and presented horizontally near the top
of the Lotus Notes screen) was the general sales knowledge designed to be pertinent to all sales
divisions. This contained knowledge on topics such as rules and guidelines for sales promotions,
templates for sales processes, forms for sales functions, as well as directories with phone numbers
of key experts within the U.S. business unit.
Down the middle of the T (and presented vertically near the left edge of the Lotus Notes
screen) was the division-specifc knowledge that typically pertained to drugs sold by that division.
This contained knowledge such as fundamental sales knowledge about the drugs sold by the sales
representatives, competitive analysis, results in recent drug trials, and letters from expert physicians.
Tips and best practices submitted by the feld sales representatives would either ft across the top
or down the side of the screen depending on whether they focused on general sales knowledge or
on product-specifc knowledge.
The knowledge structure and lessons learned from designing this pilot system were quickly
adjusted for the next division, and the system began its regular deployment in January 2002. Once
again, the system was deployed division by division through the remainder of the winter and spring
as the sales representatives attended each divisions annual sales meeting. By May 2002, all divi-
sions had received the new system. By June 2002, system use among the feld sales representatives
was approximately 98%, and the system was widely considered to be a success.
The KMS is heavily used today. Approximately 40,000 to 60,000 knowledge documents are
read per month. Approximately 250 to 300 new knowledge documents are created each month. A
typical knowledge document has a life span of 6 months and a half-life (the length of time it takes
for the knowledge to receive half of its reads) of 1 to 2 months. More fundamental knowledge
documents (e.g., clinical trials, drug side effects) have much longer life spans and half-lives. This
shows the knowledge-intensive nature of the pharmaceutical sales process.
KNOWLEDGE MANAGEMENT SYSTEM USE AND PERFORMANCE
The Impact of Knowledge on Performance
While user adoption and user satisfaction represent one key measure of success (DeLone and
Mclean, 1992), organizations usually invest in KMS for the same reason they invest in other
information systems: they believe that the value derived by the individuals using the system will
exceed its cost (Santhanam and Hartono, 2003). Does this KMS improve the performance of sales
representatives and lead to more revenue for Farmaco? With a codifcation-based KMS such as
this one, the knowledge in the KMS has been explicitly designed and validated to be correct and
useful within the target users work context, and such context-specifc knowledge should improve
individual performance and consequently organizational performance (Davenport and Glasser,
2002). Thus, the knowledge should enable the sales representative to perform his or her job more
effciently or effectivelyafter all, this is why the knowledge was added to the KMS.
However, prior research shows that using a KMS impairs performance (Haas and Hansen,
2005). Haas and Hansen speculate using a KMS hurts performance because it takes users away
from more productive activities. That is, the time spent using a KMS likely adds no value and at
best is a waste of time, and wasting time using a KMS reduces performance.
One key to this apparent counterintuitive fnding may lie in learning curves (Lapre and van
Wassenhove, 2001; Pisano, Bohmer, and Edmonson, 2001; Wright, 1936). When individuals
acquire new knowledge, it takes some time for the benefts of the new knowledge to appear. As
knowledge is transferred, the recipient of the knowledge must change his or her work processes
to refect the new knowledge. Initial performance often drops, as workers must learn to apply the
new knowledge; performance will gradually improve as practice makes perfect (Pisano, Bohmer,
and Edmondon, 2001, p. 753). Performance improves as acquired knowledge is refned by practice
(Ackerman, 1988; Farrell and McDaniel, 2001) and applied continuously under different yet related
circumstances (Schilling, Vidal, Ployhart, and Marangoni, 2003). Thus measuring performance the
frst time new knowledge is used (as Haas and Hansen did) may lead one to conclude that KMS
use impairs performance, when in fact the value from KMS use may only emerge after the user
has used the knowledge several times and moved down the learning curve. We believe that KMS
use will improve individual performance but that performance benefts are more likely after the
user has had some time to use the knowledge and refne it.
Another possible key to the apparent counterintuitive fndings of Haas and Hansen (2005) may
lie in the prior knowledge and skills of the potential knowledge user. A given knowledge asset in
the KMS may or may not lead to performance improvement for everyone. Some individuals will
fnd little value in the knowledge contained in the KMS because they either already know the
knowledge or fnd it irrelevant (McDermott, 1999). Others will fnd it relevant to their situation
and use it to improve their performance.
One factor that may affect performance benefts (or impairments) from KMS use is the amount
of experience one has, because experience is a reasonable proxy for job-related knowledge (Haas
and Hansen, 2005; McDaniel, Schmidt, and Hunter, 1988; Schmidt, Hunter, and Outerbridge,
1986). Veterans are better able to understand and internalize new knowledge (Arnold, Collier,
Leech, and Sutton, 2004; Bell and Kozlowski, 2002; Kanfer and Ackerman, 1989). Their more
developed mental models better enable them to engage in deep processing, which involves elabo-
ration, critical thinking, and the integration of new information with their existing mental models
(Bell and Kozlowski, 2002). Veterans are more likely to successfully adapt new knowledge to
their individual contexts because of the greater organizational, contextual, domain-relevant, and
job-related tacit and explicit knowledge in their mental models (Choudhury and Sampler, 1997;
Cohen and Levinthal, 1990; Glaser and Chi, 1988; Markus, 2001; Van Overschelde and Healy,
2001). Thus, veterans are better equipped to act on the new knowledge they receive (Jacoby, Mor-
rin, Johar, Gurhan, Kuss, and Mazursky, 2001).
In contrast, lack of experience and underdeveloped mental models constrain the amount of
job-relevant knowledge that is available to a novice (Kolodner, 1983). To a novice, such limited
knowledge, even if available in their mental model, is not nearly as well organized or useable
relative to that in veterans mental models (McKeithen, Reitman, Rueter, and Hirtle, 1981). Thus,
veteran knowledge workers fnd it easier to absorb and apply new knowledge (Chi, Glaser, and
Farr, 1988; Zack, 1999; Markus, 2001).
In summary, we believed that the benefts from KMS use may only appear over time. It may
take several weeks before sales representatives are able to effectively apply the new knowledge
they gain from the KMS and reap performance benefts from it. Likewise, veterans are better able
to contextualize, absorb, and apply new knowledge they do fnd; thus, veterans are more likely to
derive more beneft from each use of the KMS than are novices. In order to test these arguments,
we conducted a study of KMS use and performance, with the help of Farmaco.
Method
Data
We used three independent sources of data: KMS usage data managed by Farmacos knowledge
management group, experience data from Famacos personnel management system, and sales
performance data managed by a third-party outside contractor. We received 24 months of data for
a 2-year period (January 2002 through December 2003), plus the prior years performance, which
we used as a control. During the spring of 2002, Farmaco consolidated some sales divisions and
deployed new information systems to others. To avoid any potential impacts, we eliminated these
divisions from our study, resulting in a total of six divisions studied.
We eliminated any sales representative with less than 6 months of experience at the start of
our study, because sales representatives spend most of their frst 6 months in a training program.
We also eliminated all posttransfer data from sales representatives who transferred between ter-
ritories; their pretransfer data are included. The resulting data set includes data for 2,154 sales
representatives with 31,418 monthly observations.
Consistent with prior studies of sales representatives (e.g., Engle and Barnes, 2000; VandeWalle,
Brown, Cron, and Slocum, 1999), our dependent variable was the monthly sales performance,
defned as an individuals percent of sales quota achievedthat is, how an individuals actual
monthly sales compared to the sales quota for their sales territory set for that month. A sales
representative who exactly met his or her quota would receive a 100, while a representative under
quota by 1% would receive a 99. Because different divisions sell different products, the mean
percentages of sales quota achieved for all the sales representatives in a division were different
across the divisions studied. For this reason, we include the sales representatives division as a
control factor in our model.
The frst set of independent variables is KMS use. Computer-recorded measures of system use
are generally preferred to subjective measures such as self-report measures (LeBlanc and Kozar,
1990; Straub, Limayem, and Karahanna-Evaristo, 1995), so we used the number of knowledge as-
sets displayed on an individual sales representatives screen in the current month (current use) and
in the prior 3 months (prior use), because there may be some lag in the effects of use (a document
read in 1 month may have an important impact on sales performance in the following months).
These data exclude all navigation information and simply count the number of knowledge docu-
ments in the Notes repository that each person opened. Because use was non-normally distributed
and because we expected declining marginal benefts to use, we used a log transformation on the
data prior to analysis.
The use data suffer from the usual problems associated with computer-recorded metrics (e.g.,
undercounting of use for documents printed and referred many times on paper rather than in the
system, and overcounting of documents that are displayed on the screen but not read or used). Such
issues are generally believed to be outweighed by the greater accuracy that computer recording
brings over perceptions (LeBlanc and Kozar, 1990; Straub, Limayem, and Karahanna-Evaristo,
1995) because any imperfections in the measure are likely to be less than those in a perceptual
measure and because such imperfections would tend to increase error variance and render insig-
nifcant any relationship between use and performance. While counting the number of knowledge
documents displayed is an imperfect measure of knowledge use, it is a conservative and reliable
measure.
Job tenure, the length of time in the current job in the current organization, has commonly been
used as a measure of experience in sales contexts (Naumann, Widmier, and Jackson, 2000; OHara,
Boles, and Johnston, 1991; Saxe and Weitz, 1982). Therefore, we defned experience as the number
of months an individual had been employed by Farmaco as a feld sales representative. Recall, the
vast majority of sales representatives are hired directly out of college. Because experience was
non-normally distributed, we used a log transformation on the data prior to analysis.
Individual sales performance can be affected by many contextual variables. Farmacos knowl-
edge management group was quick to identify division and prior sales performance as major
factors in performance. Therefore, we included these two as control variables.
Analysis Technique
Similar to Ang, Slaughter, and Ng (2002), we used Hierarchical Linear Modeling (HLM) (Rauden-
bush and Bryk, 2002) for our analysis because traditional regression techniques were not appro-
priate for our matched set of performance and use nested within individuals. HLM is designed to
analyze this type of multilevel research design
1
(Hoffman, 1997; Raundenbush and Bryk, 2002;
Snijders and Bosker, 1999). We have a two-level model: the lowest level (level 1) is the month
(with the matched set of performance, current use, and 3 months of prior use); the second level
(level 2) is the individual sales representative and his or her characteristics (experience, prior
performance, and division).
Because our sample size was large, we used grand mean centering to reduce the potential im-
pacts of multicolinearity (Hoffman, 1997; Raundenbush and Bryk, 2002). We used the formulas of
Snijders and Bosker (1999) to calculate level 1 R
2
(within-person variance from month to month)
and level 2 R
2
(between-person variance from one sales representative to another). Following the
procedures of Snijders and Bosker (1999), we treated the 19 indicator variables for the time period
effects (
5
through
24
) as fxed effects, which means that they are treated exactly like standard
regression coeffcientsone value for each month coeffcient calculated for the entire sample
which remains constant across all data. We treated the intercept (
0
) and the value derived from
current and prior use (
1
through
4
) as random effects factors, which mean that they are different
for each individual sales representative. The level 2 models for current and prior use (
1
through
4
) enable us to determine whether more experienced sales representatives derive different value
from KMS use than do less experienced representatives. If the coeffcients on experience in the
level 2 models for current and prior use (
11
through
41
) are different from zero, then experience
has an impact on the value a sales representative derives from use. If the coeffcients on experi-
ence are not signifcant, then experience has no impact on the value from use; everyone derives
the same value as the intercept coeffcients (
10
to
40
).
Results
We followed the HLM analysis process recommended by Hoffman (1997) and Snijders and Bosker
(1999). Table 10.1 summarizes the results of each step in this process.
2
The frst step shows the
results of an unconditional model. The second step adds the level 1 control variables for month
as a series of 20 fxed effect indicator variables, resulting in a signifcant decrease in deviation,
indicating that this model better fts the data than the step 1 model. The third step builds a random
coeffcient model by adding current and three months prior use into the level 1 model, resulting in
a signifcant decrease in deviation. The fourth step builds a level 2 model that explains the vari-
ance in the level 1 intercept (
0
)that is, a model to explain mean individual performance. Once
again, there is signifcant decrease in deviation.
Our focus is on the fnal model (Step 5 in Table 10.1), in which level 2 models are added to
explain the slope terms on current and prior months use resulting in the full intercept and slope
model. There is a signifcant decrease in deviation, indicating that this model better fts the data
than the models in the prior steps. The equation for the impact of current months use does not
have a signifcant intercept but the coeffcient on experience is positive and signifcant. In contrast,
the equations for the impact of the 3 prior months use all have signifcant intercepts. The equa-
tions for the prior months use and 2 months prior use both have signifcant positive coeffcients
on experience, but the equation for 3 months prior does not.
Therefore, we conclude that use of a codifcation-based KMS improves individual sales representa-
tive performance, but not in the current month; use only affects performance in subsequent months.
Veteran sales representatives gained greater performance benefts than did novice sales representa-
tives for three of the four months (current month, prior month, 2 months prior). Veterans were able
to more quickly capitalize on new knowledge, gaining immediate benefts from KMS use.
FACTORS INFLUENCING KNOWLEDGE MANAGEMENT SYSTEM USE
The results show that use of this codifcation-based KMS had signifcant positive effects on
individual performance and that these performance benefts grew over time. The more a sales
representative used the KMS, the more likely it was that they would meet or beat their sales quota.
Only veterans derived immediate value from KMS use in the same month they used the KMS,
but all users derived signifcant value in the second, third, and fourth months after use. Veteran
sales representatives improved their performance by a greater extent each time they used the KMS
than did novice sales representatives, although there were no differences in performance due to
experience in the fourth month after use.
BUILDING KM SYSTEMS IN ThE PharMacEUTIcaL INDUSTrY 189
T
a
b
l
e

1
0
.
1
R
e
s
u
l
t
s

o
f

H
L
M
M
o
d
e
l
i
n
g

a
t

L
e
v
e
l

1
M
o
d
e
l
i
n
g

a
t

L
e
v
e
l

2
S
t
e
p

1
:

U
n
c
o
n
d
i
t
i
o
n
a
l

M
o
d
e
l
S
t
e
p

2
:

M
o
d
e
l

w
i
t
h

M
o
n
t
h

I
n
d
i
c
a
t
o
r
s
S
t
e
p

3
:

R
a
n
d
o
m

C
o
e
f
f
c
i
e
n
t

M
o
d
e
l
S
t
e
p

4
:

I
n
t
e
r
c
e
p
t

M
o
d
e
l
S
t
e
p

5
:

I
n
t
e
r
c
e
p
t

a
n
d

S
l
o
p
e

M
o
d
e
l
M
e
a
n
p
M
e
a
n
p
M
e
a
n
p
M
e
a
n
p
M
e
a
n
p
M
o
d
e
l

f
o
r

0

(
i
n
t
e
r
c
e
p
t
)
I
n
t
e
r
c
e
p
t

(
0
0
)
9
2
.
0
8
0
.
0
0
1
9
1
.
6
9
0
.
0
0
1
9
2
.
7
6
0
.
0
0
1
9
1
.
5
7
0
.
0
0
1
9
1
.
5
9
0
.
0
0
1
P
r
i
o
r

s
a
l
e
s

(
0
1
)
0
.
0
5
2
0
.
0
2
7
0
.
0
5
2
0
.
0
2
7
D
i
v
i
s
i
o
n

1

(
0
2
)
6
.
2
8
0
0
.
0
0
1
6
.
3
9
5
0
.
0
0
1
D
i
v
i
s
i
o
n

2

(
0
3
)
9
.
2
8
6
0
.
0
0
1
9
.
1
6
0
0
.
0
0
1
D
i
v
i
s
i
o
n

3

(
0
4
)
0
.
2
8
1
N
S
0
.
1
7
5
N
S
D
i
v
i
s
i
o
n

4

(
0
5
)
4
.
3
7
1
0
.
0
0
1
4
.
4
4
2
0
.
0
0
1
D
i
v
i
s
i
o
n

5

(
0
6
)
3
6
.
9
0
1
0
.
0
0
1
3
6
.
8
6
6
0
.
0
0
1
2
0
-
M
o
n
t
h

i
n
d
i
c
a
t
o
r
s
*
*
*
0
.
0
0
1
*
*
*
0
.
0
0
1
*
*
*
0
.
0
0
1
*
*
*
0
.
0
0
1
M
o
d
e
l

f
o
r

1

(
u
s
e

t
h
i
s

m
o
n
t
h
)
I
n
t
e
r
c
e
p
t

(
1
0
)
0
.
0
3
1
N
S
0
.
0
0
9
N
S
0
.
0
1
8
N
S
E
x
p
e
r
i
e
n
c
e

(
1
1
)
0
.
6
3
1
0
.
0
0
1
M
o
d
e
l

f
o
r

2

(
u
s
e

l
a
s
t

m
o
n
t
h
)
I
n
t
e
r
c
e
p
t

(
2
0
)
1
.
2
9
7
0
.
0
0
1
1
.
3
1
0
0
.
0
0
1
1
.
3
0
4
0
.
0
0
1
E
x
p
e
r
i
e
n
c
e

(
2
1
)
0
.
2
9
1
0
.
0
1
6
M
o
d
e
l

f
o
r

3

(
u
s
e

2

m
o
n
t
h
s

a
g
o
)
I
n
t
e
r
c
e
p
t

(
3
0
)
1
.
8
3
2
0
.
0
0
1
1
.
8
5
5
0
.
0
0
1
1
.
8
5
5
0
.
0
0
1
E
x
p
e
r
i
e
n
c
e

(
3
1
)
0
.
2
7
7
0
.
0
1
3
M
o
d
e
l

f
o
r

2
4

(
u
s
e

3

m
o
n
t
h
s

a
g
o
)
I
n
t
e
r
c
e
p
t

(
4
0
)
2
.
2
3
1
0
.
0
0
1
2
.
3
2
2
0
.
0
0
1
2
.
3
0
7
0
.
0
0
1
E
x
p
e
r
i
e
n
c
e

(
4
1
)
0
.
2
0
5
N
S
D
e
v
i
a
n
c
e
3
0
1
,
7
3
1
3
0
0
,
0
2
3
0
.
0
0
1
2
9
8
,
1
5
5
0
.
0
0
1
2
9
7
,
2
0
9
0
.
0
0
1
2
9
7
,
1
8
5
0
.
0
0
1
O
v
e
r
a
l
l

L
e
v
e
l

1

R
2

(
w
i
t
h
i
n

p
e
r
s
o
n
)
4
.
3
%
1
4
.
7
%
2
6
.
4
%
2
6
.
4
%
O
v
e
r
a
l
l

L
e
v
e
l

2

R
2

(
b
e
t
w
e
e
n

p
e
r
s
o
n
)
1
.
9
%
8
.
7
%
4
5
.
4
%
4
5
.
5
%
Given the signifcant benefts from KMS use, it appears that Farmaco should strive to incul-
cate the value of the KMS so its use becomes embedded into the habitual routines of the target
users. Unlike passively volitional systems that supplement existing modes of communication
(Malhotra and Galletta, 2005), the KMS at Farmaco was designed in part to replace the existing
paper-based knowledge artifacts. Use is widespread and the KMS is integrated into the daily
lives of many sales representatives. However, some users are more active than others. What fac-
tors infuence a sales representative to use the KMS to a greater or lesser extent? We conducted
a follow-up study to examine the factors that infuence the ongoing repetitive-use behavior of
the sales representatives.
Factors Infuencing Use
Much research has examined use, and especially the factors leading to the intent to use (e.g.,
Davis, Bagozzi, and Warshaw, 1989; Venkatesh and Davis, 2000). In this chapter, we propose a
model that describes the important factors that infuence the loyal use of a KMS (Figure 10.1).
The model is based upon the Information Systems (IS) Success Model (DeLone and McLean,
1992) and the Technology Acceptance Model (TAM) (Davis, 1989; Venkatesh and Davis, 2000).
This section describes the seven constructs in the model (loyal use, perceived usefulness, volun-
tariness, extrinsic motivation, perceived ease of use, system quality, and knowledge quality) and
how they are related.
Loyal Use
The primary dependent variable in TAM is the intent to use. In this case, we are less interested in
the intent to use, given that 98% of the target users use the KMS on a regular basis. We are instead
interested in the extent to which a user has become a routine userwhere the KMS has been fully
appropriated into the individuals standard work practices.
The marketing literature has the concept of consumer loyalty, one type of consistent use be-
havior, where the customer makes consistent repetitive purchases (Oliver 1999). These consistent
repetitive purchases demonstrate a prolonged consistent pattern of behaviors that become a part of
ongoing routines. Building on Olivers defnition of loyalty, we defne loyal use in the context of
KMS as a commitment to repeatedly and consistently consume knowledge content from the KMS
in the future. In the context of KMS, loyal use can be viewed as the prolonged appropriation of
the system that fundamentally changes behavior to incorporate KMS use into the users ongoing
routine. Hence, loyal use refers to a set of consistent, ongoing, and routine behaviors. Loyal use
is an appropriate dependent variable, because it captures the prolonged systematic appropriation
of the KMS.
Perceived Usefulness
Perceived usefulness is the degree to which a user believes that use of the system will result in
benefts being accrued to the user or the users organization and often includes increases to job
performance and productivity (Davis, 1989; Seddon, 1997). Previous research has established
that perceived usefulness is positively related to use (Davis, Bagozzi, and Warshaw, 1989; Ven-
katesh, 2000; Venkatesh and Davis, 2000). In the context of KM, perceived usefulness refers to
the assessment that loyal use will result in benefts to the user or the users organization over a
period of time. Repeated and consistent use will result in benefts beyond those from the initial
use of the system. Therefore, a positive relationship between perceived usefulness and loyal
use is expected.
Perceived Ease of Use
Perceived ease of use has been defned as the degree to which an individual believes that using
a particular system would be free of physical or mental effort (Moore and Benbasat, 1991). This
conceptualization of perceived ease of use focuses upon the ease of using a system separate from
quality of the system itself. Perceived ease of use has been shown to infuence use in a variety of
contexts, such as e-commerce (Gefen, Straub, and Boudreau, 2000), manufacturing and fnancial
services (Venkatesh and Davis, 2000), and offce work (Rai, Lang, and Welker, 2002). Behavioral
inertia suggests that one will continue with a course of action unless there is a compelling reason
to do otherwise. KMS that are easier to use will reduce the impetus to deviate from ones existing
use behavior, resulting in increased loyal use.
KMS that are easier to use require less mental effort to use. Lower levels of mental effort to
use a system mean that the content of the system can be accessed with less effort, increasing the
relative value of the use of the system. Therefore, the productivity gains from systems that are
easier to use enable increased perceived usefulness.
Voluntariness
Voluntariness refers to how much control a user believes that he or she has over his or her use
of a system. If a user perceives that he or she is obligated to use a system, then he or she would
exhibit a low level of voluntariness. In the case of low perceived voluntariness, where a user feels
that use of the system is mandatory, loyal use of the system will be higher. Without a perceived
choice to use, respondents will feel a need to comply with the perceived mandatory level of use
or risk disciplinary action.
Extrinsic Motivation
Extrinsic motivation is defned as the motivation to work primarily in response to something apart
from the work itself, such as reward or recognition (Amabile, Hill, and Hennessey, 1994). The infu-
ence of extrinsic motivation over ones actions is based outside of the individual and is the result of
external incentives such as fnancial compensation and promotion. Losier, Bourque, and Vallerand
Figure 10.1 Factors Infuencing KMS Use
(1993) indicate that people engage in activities for reasons other than the activity itself. These reasons
comprise extrinsic motivation. External reasons such as fnancial compensation and promotion are
organizational rewards based upon patterns of consistent behavior. Tyagi (1985) found that extrinsic
motivation was positively related to the job performance of sales representatives.
Within the feld of human resources, Wiley (1997) identifed professional accomplishment,
career advancement, and professional recognition as important extrinsic factors infuencing people
in their effort to gain professional certifcation through the completion of coursework and work
experience. The tasks required to complete certifcation were ongoing, repetitive and consistent,
requiring a commitment to continued use of the training materials. The use of the training materi-
als is very similar to the use of a KMS, where benefts are derived from ongoing repetitive and
consistent use.
Extrinsic motivation provides the reason(s) to use a KMS other than use itself. Use of a KMS
itself provides little in the way of benefts; it is the subsequent application of knowledge gained
from use of the KMS that generates benefts. Extrinsic motivation may help bridge the temporal
discontinuity between use and value from use of a KMS.
System Quality
In their initial paper on IS success, DeLone and McLean (1992) classifed 12 previous stud-
ies that empirically measured system quality. They referred to system quality as the quality of
the processing system itself. Seddon (1997) defned system quality as being concerned with
whether or not there are bugs in the system, the consistency of the user interface, ease of use,
quality of documentation, and sometimes, quality and maintainability of the system code. Both
of these characterizations of system quality capture in part the hardware and software dimensions
of systems.
Perceived ease of use is distinct from system quality, in that it refers to aspects of the system
such as the humancomputer interface design. A system may be supremely reliable in that it
performs the requested operations on time, every time; however, the way in which the user
interfaces with the system may be awkward resulting in low perceived ease of use while being
very reliable and having a high system quality (e.g., command line UNIX). Systems that tend
to exhibit ease of use are those that are clear and understandable and those that require little
mental effort to use.
We defne system quality as the reliability and predictability of the system independent of the
knowledge it contains and the ease of using the system. The degree to which a system embodies
reliability and predictability within its hardware and software functionality affects the amount of
cognitive effort required to use the system. Systems that are reliable and predictable create less
ambiguity that users must process prior to use. Therefore, higher-quality systems impose a lower
cognitive load on users than do low-quality systems. We posit that system quality is positively
related to perceived ease of use.
Knowledge Quality
One component of every information system is the quality of the information, knowledge, or other
content it provides. Information quality has been used in models of IS success for traditional information
systems whose primary content is information (DeLone and McLean, 1992; Rai, Lang, and Welker,
2002; Seddon, 1997). The higher construct is content quality that incorporates data quality for data
processing systems, information quality for information systems, and knowledge quality of KMS.
In the context of KMS success, knowledge quality is substituted for information quality as
the type of content contained within the system. Knowledge quality is defned as the degree to
which the knowledge contained in a KMS is useful in assisting the user in accomplishing tasks,
independent of the KMS in which it is contained. The relationship between knowledge quality
and perceived usefulness is expected to be positive, refecting increased perceived benefts derived
from using a system that contains high-quality knowledge.
Method
To test the model in Figure 10.1, we conducted a survey of the feld sales representatives after the
third version of the KMS had been operational for more than 1 year.
Item Development
Prior to testing the research model, we evaluated existing scales and, where necessary, developed
new scale items for system quality and loyal use. The initial pool of items was based upon the
IS and marketing literature. The model in Figure 10.1 is an extension and reinterpretation of the
literature on system success. We developed a pool of items to capture the full scope of system
quality and loyal use.
New question items were developed collaboratively through the interaction of the three authors
and two members of the Farmacos KM team. The initial set of items for each construct was vetted
for content validity in an iterative process. The question items from existing scales were similarly
vetted to ensure that they continued to meet content validity.
Existing scales (refer to Table 10.2) were used as a basis for the fnal scales used to capture
some of the constructs. However, minor changes were made to ensure contextual consistency and
to enhance question clarity. All of the question items were measured on a 7-point Likert type scale,
anchored with strongly disagree and strongly agree.
Four academics unrelated to the research project were recruited to participate in a sorting pro-
cess to help establish the construct validity of the scales. Each academic sorted the question items
into groups of similar questions, without an indication of the number or nature of the constructs
of interest. The results of the sorting process indicated that the question items satisfactorily cor-
responded to the a priori classifcations from the model in Figure 10.1. Interviews with each of the
four academics included discussions regarding the length of time required to complete the survey,
clarity of question items, formatting, and instructions. Several items were reworded or dropped,
and minor changes were made to the layout of the survey.
Instrument Pretesting and Refnement
Sixty-fve senior members of the feld sales force were invited to participate in the pilot test of
the survey. Thirty-one completed the pilot test survey, resulting in a response rate of 45% for the
pilot test. One outcome of the pilot test was the modifcation of a few questions and the elimina-
tion of questions that did not demonstrate construct validity, content validity, or reliability. Poorly
worded, obscure, or ambiguous questions were either dropped from the survey or reworded to
increase the reliability and validity of the scales. The changes to the survey were consistent with
theory and the model in Figure 10.1. The fnal items are provided in Table 10.2. The pilot was
conducted online to test both the psychometric characteristics of the constructs of interest and the
Web-based data collection site.
Table 10.2
Survey Items
Item Question Mean SD
Loyal Use ( = 0.76; Sirohi, McLaughlin, and Wittink, 1998 and newly created)
LU1 Use of the Field KM has become a part of my normal routine. 4.88 1.51
LU2 I will continue using the Field KM. 5.58 1.22
Perceived Usefulness ( = 0.90; Davis, 1989)
PU1 Using the Field KM enables me to accomplish job related tasks more
quickly.
4.57 1.54
PU2 Using the Field KM in my job increases my productivity. 4.53 1.50
PU3 Using the Field KM makes it easier to do my job. 4.58 1.47
Voluntariness ( = 0.73; Karahanna, Straub, and Chervany, 1999)
VOL1 My manager does not expect me to use the Field KM. 2.07 1.37
VOL2 Although it might be helpful, using the Field KM is not compulsory in
my job.
2.74 1.55
Motivation ( = 0.72; Amabile, Hill, Hennessey, and Tighe, 1994)
MOT1 I am motivated by recognition I can earn from other people for being a
better sales representative.
5.59 1.53
MOT2 My income goals are dependent upon being a better sales
representative.
6.01 1.32
MOT3 To become a better sales representative, I have to feel that I am
personally benefting.
5.60 1.44
Perceived Ease of Use ( = 0.92; Davis, 1989)
PEOU1 The Field KM is easy to use. 4.44 1.68
PEOU2 The Field KM is clear and understandable. 4.60 1.59
PEOU3 I fnd it easy to get the Field KM to do what I want it to do. 3.93 1.72
System Quality ( = 0.82; newly created)
SQ1 The Field KM is reliable. 5.25 1.46
SQ2 The Field KM rarely crashes. 5.44 1.57
SQ3 The Field KM contains few software errors (e.g., bugs, glitches). 5.10 1.52
Knowledge Quality ( = 0.92;Doll and Torkzadeh, 1988; Seddon and Kiew, 1994;
Rai et al., 2002)
KQ1 The knowledge content meets my needs. 4.93 1.47
KQ2 Knowledge content provided by the Field KM is helpful in resolving my
questions and problems.
4.90 1.43
KQ3 The knowledge content in the Field KM is relevant to what I need/do. 5.11 1.38
Procedure
An invitation to participate in the survey was e-mailed to 4,500 feld sales representatives by the
manager of the feld sales KMS. Within the e-mail invitation, a link connected the respondents to
the Web-based survey. From the 4,500 invited, we received 1,013 useable responses, resulting in a
response rate of about 22.5%. Of those who responded, 47% were female (n = 826), with 64% between
the ages of 25 and 35 years (n = 687). The average tenure with the frm was 2.99 years (n = 992).
The model was tested using LISREL 8.54, where the covariance matrix was analyzed with
maximum likelihood estimation. Each of the seven latent constructs was captured with multiple
refective indicators. Confrmatory factor analysis was used to assess convergent and discriminant
validity. Subsequently, the structural model was tested (see Figure 10.2).
Assessment of Measurement Properties
To test the measurement properties, we followed previously established procedures (Basselier,
Benbasat, and Reich, 2003). First, the ft between the indicators and the constructs was tested.
Second, the convergent validity of each set of indicators and constructs was assessed. Third, dis-
criminant validity was assessed between each of the constructs. Fourth, the relationships between
each of the constructs were assessed through a test of the structural model. Confrmatory factor
analysis was used to assess the measurement properties of the theoretical model in Figure 10.1. A
priori relationships were specifed between the constructs and their indicators. The relationships
between the indicators and the constructs were then tested against the data.
Figure 10.2 Results of Model Testing
The goodness of ft indices include both absolute measures and relative measures (see Table
10.3). Absolute measures include
2
, goodness of ft index (GFI), and root mean square residual
(RMR) and are measures of the ability of the model to reproduce the covariance matrix from the
data. The
2
(578.18, 0.00) statistic is large and signifcant, indicating that the null hypothesis of
covariance equality between the data and the model should be rejected. However, a large
2
is
sensitive to the sample size and number of model parameters (Bentler and Bonett, 1980). Therefore,
reliance upon
2
as the sole indicator of ft is inappropriate. GFI (0.94), a measure of the overall
goodness of ft, is strong, surpassing the recommendation of 0.90, indicating good ft. The standard-
ized RMR is used instead of RMR due to diffculties in the interpretation of RMR. Standardized
RMR (0.044) is below the threshold of 0.05 for a well-ftting model (Byrne, 1998).
Relative measures of ft include the adjusted goodness of ft (AGFI), root mean square of ap-
proximation (RMSEA), normed ft index (NFI), and the comparative ft index (CFI). Both the
adjusted goodness of ft index (AGFI) and the normed ft index (NFI) are statistics that range
from 0 to 1 inclusively. After adjusting the GFI for the number of degrees of freedom (df) of the
proposed model, the AGFI (0.92) is greater than the recommendation of 0.90 indicating good
ft. The normed ft index (0.98) compares the proposed model to the null model, where a value
of 1.0 indicates a perfect ft. Values greater than 0.90 indicate a good ft; values approaching 1.0
indicate a perfect ft. Both the RMSEA and the
2
/df are used to assess the ft of the model to the
data compared to models with additional parameters. The RMSEA (0.058) is within the range for
good ft (0.050.08). The
2
/df (4.41) conforms to a recommendation that the
2
/df should be less
than 5 (Hair, Anderson, and Tatham, 1998).
Convergent validity is concerned with demonstrating the degree to which the measures for each
construct capture a single underlying construct. Convergent validity was assessed by comparing
the average of the squared multiple correlations for each construct, which should be greater than
0.50 (Fornell and Larcker, 1981). All of the average squared multiple correlations were greater
than 0.50, establishing convergent validity. After calculating the average of the squared multiple
correlations for each construct if the average was less than 0.50, poorly ftting items were dropped,
when their omission did not adversely affect the content validity of the scales. The result was a
parsimonious set of scales that demonstrated convergent validity.
Discriminant validity is concerned with demonstrating the degree to which the measures for
each construct are distinct from the other measures. It can be assessed where the average shared
variance for each construct is greater than the square of the correlation between the constructs
(Anderson and Gerbing, 1988). For each construct, the average shared variance was greater than
Table 10.3
Confrmatory Factor Analysis-Based Goodness of Fit Indices
Model Desired Level
2
578.18 Smaller
df 131
2
/df 4.41 <5
GFI 0.94 >0.90
AGFI 0.92 >0.80
Standardized RMR 0.044 <0.05
RMSEA 0.058 <0.08
NFI 0.98 >0.90
CFI 0.98 >0.90
BUILDING KM SYSTEMS IN ThE PharMacEUTIcaL INDUSTrY 197
the square of the correlations between constructs, indicating that each of the measures of the latent
constructs demonstrates discriminant validity.
The fnal set of question items demonstrates suffcient reliability, with cronbachs for each
construct exceeding 0.70 for each set of indicators. Table 10.2 includes the cronbachs for each
construct.
Results
The results of a test of the structural model are presented in Figure 10.2. The model explains 68%
of the variance in loyal use of the KMS, 38% of the variance in respondents perception of the
usefulness of the KMS, and 38% of the variance in perceived ease of use of the KMS. The model
accounts for a signifcant portion of the variance in the perceptions of the respondents about loyal
use of the KMS.
Perceived usefulness was found to have a signifcant positive (0.58, p < 0.001) relationship with
loyal use. Perceived usefulness was the single most infuential factor in explaining the variance
in loyal use. The perception of system voluntariness was found to be negatively and signifcantly
related to (0.31, p < 0.001) loyal use. an interpretation of the negative path coeffcient for vol-
untariness to loyal use is that as people perceive use to be nonvoluntary (i.e., mandatory), they
choose to use the KMS to a greater extent. Extrinsic motivation had a positive and signifcant
(0.26, p < 0.001) infuence on loyal use. The fnal direct factor infuencing loyal use is perceived
ease of use (0.08, p < 0.05). although signifcant, the path coeffcient indicates that the ease of
using a KMS has only minimal infuence on loyal use.
Perceived usefulness was hypothesized to be infuenced by two factors, perceived ease of use
and the quality of the knowledge contained in the system. Perceived ease of use was positively
and signifcantly (0.45, p < 0.001) related to perceived usefulness. Knowledge quality was signif-
cantly positively (0.25, p < 0.001) related to perceived usefulness. Both factors exhibit a signifcant
capacity to explain the variance in perceived usefulness (R
2
= 38%).
Perceived ease of use was hypothesized to be a function of system quality. System quality was
positively and signifcantly related (0.61, p < 0.001) to perceived ease of use.
LESSONS LEARNED FROM FARMACOS EXPERIENCE
We believe there are four key lessons to be learned from Farmacos experiences.
System Quality and Knowledge Quality Are the Root Causes of Loyal KMS Use
We found that, as in prior studies, perceived usefulness and perceived ease of use signifcantly
infuenced loyal use (Davis, Bagozzi, and Warshaw, 1989; Venkatesh, 2000; Venkatesh and Davis,
2000). Of all the factors infuencing loyal use, perceived usefulness was found to be the most
important factor. although there are many factors infuencing loyal use, it is an individuals own
perception of the relative costs and benefts from use that are the primary driving force affecting
routine use over the long term.
System quality signifcantly affects perceived ease of use and, through it, perceived usefulness.
Similarly, the quality of the knowledge contained in the KMS is a key driver of positive perceptions
of usefulness and loyal use. Knowledge quality can be manipulated through the alignment between
the relevance and appropriateness of the knowledge content and the needs of the users.
These two factors, system quality and knowledge quality, lie at the heart of use. They are the
root causes of the choice to use or not use a KMS. At one level, this is an obvious conclusion:
more than a decade ago, Delone and McLean (1992) proposed system quality and information
quality as key factors in information system success. However, we have seen few frms focus on
both system quality and knowledge quality with the same intensity as Farmaco. In our experience,
most frms focus on system quality rather than knowledge quality.
In implementing the frst failed KMS, Farmaco chose to focus primarily on system quality,
leaving knowledge quality in the hands of the intended KMS users. After the failure of this frst
version of the KMS, Farmaco stepped back and carefully reconsidered what to do. The new
KMS manager, a former sales manager, choose to focus frst on knowledge quality and second
on system quality. The strategy paid off and the second version of the KMS was successful. The
key lesson here is to deliberately and forcefully focus on knowledge quality (and system quality)
when implementing a new KMS.
We do not intend our arguments to be interpreted as an implied criticism of communities of
practice, in which knowledge quality is left in the hands of the community members, not the KMS
leaders (who in this case were former users, not technologists). We agree that communities of
practice may be viable for some KMS applications. In this case, however, with the KMS targeted
at such a large-scale audience (thousands of users) and with knowledge being added, updated, and
removed at such an intensive pace, the community of practice strategy did not work. A strategy
that deliberately focused on knowledge quality was key to success.
Lesson 1. Managing both KMS system quality and the quality of the knowledge in the KMS
increases the chance users will integrate the KMS into their routine work processes.
Ease of Use Is Key to Loyal KMS Use
As in prior research, perceived ease of use, although signifcantly related to loyal use, exerted a
minimal direct infuence on long-term routine, use behavior (about 15% of the impact of perceived
usefulness). Thus, we conclude that although perceived ease of use may be important in the initial
acceptance of a technology, its direct impact fades away quite sharply after the technology has been
in use for an extended period of time. This result is consistent with previous fndings exploring
perceived ease of use and use, where use is the anticipated near-term use of a technology (Gefen,
Straub, and Boudreau, 2000; Rai, Lang, and Welker, 2002; Venkatesh and Davis, 2000).
Nonetheless, despite the small direct infuence on loyal use, perceived ease of use exerted a
signifcant infuence on perceived usefulness. Therefore, perceived ease of use exerted primarily
an indirect infuence on loyal use as mediated by perceived usefulness, where perceived usefulness
is the integrated assessment of the costs and benefts of use.
Ease of use is routinely cited as a key factor infuencing use, so this conclusion is again obvi-
ous. But take a careful look at how Farmaco acted to improve ease of use. After the frst KMS
failed, it totally reconsidered the look and feel of the KMS to make it more graphic and, supposedly,
easier to use. The second version of the KMS was a success, but in implementing the third version,
Farmaco moved away from the easy-to-use graphical interface to a more traditional Web-based
look and feel. The easy-to-use interface of the second version clearly was not instrumental in
the success of the second KMS.
Instead, we believe one critical technical difference between the frst failed KMS and the sec-
ond successful one was the decision to place the KMS at the center of the desktop. By building
the desktop around the KMS, it greatly simplifed users access to the KMS and made the KMS
much easier to use. They no longer had to navigate to the KMS; it was immediately available as a
part of their desktop. This had a great impact on the users choice to integrate the KMS into their
ongoing work routines and become loyal users of the KMS.
Lesson 2. Integrating the KMS into the users desktop increases the chance users will integrate
the KMS into their routine work processes.
Perceived Usefulness Is Central to Loyal KMS Use
Similar to prior research, we found that perceived usefulness signifcantly infuenced loyal use (Ven-
katesh, 2000; Venkatesh and Davis, 2000). In fact, perceived usefulness was found to be the most
important factor infuencing loyal use. In other words, a commitment to repeatedly and consistently
consume knowledge content from the KMS in the future arises from the degree to which a user
believes that use of the KMS will result in benefts accrued to him or her or the organization.
The fndings in this study extend prior research and provide empirical evidence that expected
benefts from the use of a system (e.g., KMS) not only affect initial use (Venkatesh, 2000; Ven-
katesh and Davis, 2000) but prolonged routine use as well. This is particularly important in a
knowledge-intensive industry, such as the pharmaceutical industry, where tasks are inherently
knowledge based and hundreds of new knowledge documents are created each month. Although
it may be easy to conclude that use of the KMS is inevitable given the volume of new knowledge
and the nature of sales representatives tasks, we notice in Farmaco, especially the frst KMS, that
lack of routine use was largely due to greater costs and minimal benefts for the users in using the
KMS. Upon closer inspection, we fnd that Farmacos efforts to improve their knowledge quality
and perceived ease of use, which positively impact perceived usefulness, were critical for sales
representatives to assess the costs and benefts of using the KMS.
Thus, the quality of knowledge contained in the KMS and the ease of using the KMS positively
infuenced the extent to which the users believed they will derive benefts from using the KMS.
Farmaco took great care in ensuring their sales representatives maximize the beneft derived from
using the KMS. The failure of the frst KMS lead to interviews with relevant stakeholders to gain a
better understanding of what was needed. The focus was on providing useful knowledge that was
easy to obtain and consume. Their fndings resulted in designing and building both the knowledge
content and the KMS structure.
A successful deployment of the second KMS included a new knowledge taxonomy and a for-
mal four-step knowledge validation process for handling knowledge content while the third KMS
streamlined and improved the fundamental knowledge taxonomies and structure. Similarly, an
entirely new graphic user interface took center stage as the second KMS became the users desk-
tops where all other tools became available through the KMS. Although much improved, the third
KMS moved toward a more standard Internet-like user interface and provided greater fexibility
in accommodating changes. Thus, Farmacos focus on providing useful knowledge that was easy
to obtain and consume lead to the improvement of knowledge quality and perceived ease of use,
which in turn infuenced perceived usefulness.
It is also interesting to note that Farmaco made concerted efforts to align knowledge content and
structure with KMS design and user interface for both the second and the third KMS. We believe
that aligning knowledge content with the KMS design improved the expected benefts users will
derive, which affected the users choice to become a routine KMS user.
Lesson 3. The expected benefts derived from use of the KMS increases the chance users will
integrate the KMS into their routine work processes.
Voluntariness and Motivation Affect Loyal KMS Use
Voluntariness and extrinsic motivation are individual characteristics that affect ones willingness
or incentives to engage in an activity. In this study, voluntariness referred to whether an individual
believed he or she was obligated to use the KMS. Perceived obligations can come in the form of
use related expectations. Persons who believe that they are expected to use the KMS (low volun-
tariness) are more likely to do so. As stated earlier, the overarching goal of the KMS at Farmaco
is to increase sales performance through the reuse of knowledge assets. Some knowledge assets
are required by legal oversight to be used during sales presentations, while others are optional.
Without mandating the use of the KMS, there is no certainty that sales representatives acquire
the necessary knowledge to comply with legal oversight. By mandating use for some knowledge
assets and developing expectations of use for other knowledge assets, the sales representatives
are more likely to develop patterns of consistent and ongoing use behavior
Extrinsic motivation is largely driven by external incentives unrelated to the work activity
itself. Financial and professional incentives for sales representatives are not uncommon, and
Farmaco is no exception. We found that sales representatives use the KMS and participate regu-
larly on submitting best practices where an approved tip results in sales points equivalent
to $60. We also found that dissemination of time-sensitive knowledge (e.g., fashes on how
to respond to breaking news or competitors actions, new marketing campaigns) potentially
impacts ones individual performance, and occurs frequently enough for sales representatives
to continually use the KMS.
Lesson 4. Providing fnancial/professional incentives and/or mandating the use of the KMS
increases the chance users will integrate the KMS into their routine work processes.
CONCLUSION
We believe that the knowledge management strategy adopted by Farmacoof focusing both
on the system quality and the knowledge qualitywas a key factor infuencing the widespread
adoption of its KMS by the target users and the signifcant impacts that its use has had on sales
performance. We have drawn four lessons from Farmacos experience that we believe are likely
to apply to other frms that attempt to build and deploy similar large-scale KMS in other environ-
ments with rapidly changing knowledge needs.
NOTES
1. With traditional regression, there is a problem with the unit of analysis. If the data are analyzed at the
lowest level (time period in our case), then the impact of the individual must be omitted. Because there is
likely to be signifcant correlation among the performance and use sets for a specifc individual, this approach
can erroneously infate the signifcance and cause type 1 errors. If the data are analyzed at the second level
(individual in our case), then we cannot include the level 1 variables (i.e., use and performance) in the model,
except in aggregate, which removes precision.
2. For simplicity, we do not report the coeffcients or signifcance of the 20 indicator variables for month
(19 of which were signifcant in the fnal model).
REFERENCES
Ackerman, P.L. 1988. Determinants of individual differences during skill acquisition: Cognitive abilities and
information processing. Journal of Experimental Psychology, 117, 288313.
Amabile, T.M.; Hill, K.G.; Hennessey, B.A.; and Tighe, E.M. 1994. The Work Preference Inventory: As-
sessing intrinsic and extrinsic motivational orientations. Journal of Personality and Social Psychology,
66, 5, 950967.
Anderson, J.C., and Gerbing, D.W. 1988. Structural equation modeling in practice: A review and recom-
mended two-step approach. Psychological Bulletin, 103, 3, 411423.
Ang, S.; Slaughter, S.; and Ng, K.Y. 2002. Human capital and institutional determinants of information
technology compensation: Modeling multilevel and cross-level interactions. Management Science, 48,
11, 14271445.
Arnold, V.; Collier, P. A.; Leech, S.A.; and Sutton, S.G. 2004. Impact of intelligent decision aids on expert
and novice decision-makers judgments. Accounting and Finance, 44, 126.
Basselier, G.; Banbasat, I.; and Reich, B.H. 2003. The infuence of business managers IT competence on
championing IT. Information Systems Research, 14, 4, 317336.
Bell, B.S., and Kozlowski, S.W.J. 2002. Goal orientation and ability: Interactive effects on self-effcacy,
performance, and knowledge. Journal of Applied Psychology, 87, 3, 497505.
Bentler, P.M., and Bonett, D.G. 1980. Signifcance tests and goodness of ft in the analysis of covariance
structures. Psychological Bulletin, 88, 588606.
Blessing, S.B., and Anderson, J.R. 1996. How people learn to skip steps. Journal of Experimental Psychol-
ogy: Learning, Memory, and Cognition, 22, 3, 576598.
Brown, J.S., and Duguid, P. 1998. Organizing knowledge. California Management Review, 40, 3, 90111.
Byrne, B.M. 1998. Structural Equation Modeling with LISREL, PRELIS, and SIMPLIS: Basic Concepts,
Applications, and Programming. Mahwah, NJ: Lawrence Erlbaum.
Chi, M.T.H.; Glaser, R.; and Farr, M.J., 1988. The Nature of Expertise. London: Lawrence Erlbaum.
Choudhury, V., and Sampler, J.L. 1997. Information specifcity and environmental scanning: An economic
perspective. MIS Quarterly, 21, 1, 2553.
Cohen, J. 1988. Statistical Power Analysis for the Behavioral Sciences, 2nd ed. Mahwah, NJ: Lawrence
Erlbaum.
Cohen, W., and Levinthal, D. 1990. Absorptive capacity: A new perspective on learning and innovation.
Davenport, T.H., and Glasser, J. 2002. Just-in-time delivery comes to knowledge management. Harvard
Business Review, July, 59.
Davis, F.D. 1989. Perceived usefulness, perceived ease of use, and user acceptance of information technol-
ogy. MIS Quarterly, 13, 319339.
Davis, F.D.; Bagozzi, R.P.; and Warshaw, P.R. 1989. User acceptance of computer technology: A comparison
of two theoretical models. Management Science, 35, 8, 9821003.
DeLone, W.H., and McLean, E.R. 1992. Information systems success: The quest for the dependent variable.
Information Systems Research, 3,1, 6095.
Derr, K.T. 2006. Managing knowledge the Chevron way. In Laurence Prusak and Eric Matson (eds.), Knowl-
edge Management and Organizational Learning: A Reader. Oxford, UK: Oxford University Press.
Doll, W.J., and Torkzadeh, G. 1988. The measurement of end-user computing satisfaction. MIS Quarterly,
12, 2, 259274.
Engle, R.L., and Barnes, M.L. 2000. Sales force automation usage, effectiveness, and cost-beneft in Germany,
England and the United States. The Journal of Business & Industrial Marketing, 15, 4, 216241.
Farrell, J.N., and McDaniel, M.A. 2001. The stability of validity coeffcients over time: Ackermans (1988)
model and the General Aptitude Test Battery. Journal of Applied Psychology, 86, 6079.
Fedorowicz, J.; Oz, E.; and Berger, P.D. 1992. A learning curve analysis of expert system use. Decision
Sciences, 23, 4, 797818.
Fornell, C., and Larcker, D.F. 1981. Evaluating structural equation models with observable variables and
measurement error. Journal of Marketing Research, 18, 3950.
Gefen, D.; Straub, D.W.; and Boudreau, M.-C. 2000. Structural equation modeling and regression: Guidelines
for research practice. Communications of the Association for Information Systems, 4, 7, October.
Glaser, R., and Chi, M.T.H. 1988. Overview. In M.T.H. Chi, R. Glaser, and M.J. Farr (eds.), The Nature of
Expertise. Hillsdale, NJ: Lawrence Erlbaum, pp. xvxxviii.
Haas, M.R., and Hansen, M.T. 2005. When using knowledge can hurt performance: The value of organizational
capabilities in a management consulting company. Strategic Management Journal, 26, 124.
Hair, J.F.; Anderson, R.E.; Tatham, R.L.; and Black, W.C. 1998. Multivariate Data Analysis, 5th ed. Engle-
wood Cliffs, NJ: Prentice-Hall.
Business Review, MarchApril, 106116.
Hoffman, D.A. 1997. An overview of the logic and rationale of hierarchical linear models. Journal of Man-
agement, 23, 6, 723744.
Jacoby, J.; Morrin, M.; Johar, G.; Gurhan, Z.; Kuss, A.; and Mazursky, D. 2001. Training novices to become
more expert: The role of information accessing strategy. The Journal of Psychology and Financial Mar-
kets, 2, 2, 6979.
Kanfer, R., and Ackerman, P.L. 1989. Motivation and cognitive abilities: An integrative/aptitude-treatment
interaction approach to skill acquisition. Journal of Applied Psychology, 74, 657690.
Kaplan, S. 2002. KM the right way: A step-by-step approach that will ensure that your expensive KM system
actually gets used. CIO, July 15, 7481.
Karahanna, E.; Straub, D.W.; and Chervany, N.L. 1999. Information technology adoption across time: A
cross-sectional comparison of pre-adoption and post-adoption beliefs. MIS Quarterly, 23, 2, 183213.
Kolodner, J.L. 1983. Towards an understanding of the role of experience in the evolution from novice to
expert. International Journal of Man-Machine Studies, 19, 497518.
Lapre, M.A., and van Wassenhove, L.N. 2001. Creating and transferring knowledge for productivity in
factories. Management Science, 47, 10, 13111325.
LeBlanc, L., and Kozar, K. 1990. An empirical investigation of the relationship between DSS usage and
system performance: A case study of a navigation support system. MIS Quarterly, 263277.
Leigh, T.W. 1987. Cognitive selling scripts and sales training. Journal of Personal Selling & Sales Manage-
ment, 7, 3948.
Losier, G.F.; Bourque, P.E.; and Vallerand, R.J. 1993. A motivational model of leisure participation in the
elderly. The Journal of Psychology, 127, 2, 153170.
Malhotra, Y., and Galletta, D. 2005. A multidimensional commitment model of volitional systems adoption
and usage behavior. Journal of Management Information Systems, 22, 1, 117151.
McDaniel, M.A.; Schmidt, F.; and Hunter, J. 1988. Job experience correlates of job performance. Journal
of Applied Psychology, 73, 2, 327330.
McDermott, R. 1999. Why information technology inspired but cannot deliver knowledge management.
California Management Review, 41, 4, 103117.
McKeithen, K.; Reitman, J.; Rueter, H.; and Hirtle, S., 1981. Knowledge organization and skill differences
in computer programmers. Cognitive Psychology, 13, 3, 307325.
Moore, G.C., and Benbasat, I. 1991. Development of an instrument to measure the perceptions of adopting
an information technology innovation. Information Systems Research, 2, 3 (September), pp. 192222.
Motsenigos, A., and Young, J. 2002. KM in the US government sector. KM World, October, 1617.
Naumann, E.; Widmier, S.M.; and Jackson, D.W., Jr. 2000. Examining the relationship between work attitudes
and propensity to leave among expatriate salespeople. Journal of Personal Selling & Sales Management,
20, 4, 227241.
OHara, B.S.; Boles, J.S.; and Johnston, M.W. 1991. The infuence of personal variables on salesperson
selling orientation. Journal of Personal Selling & Sales Management, 11, 1, 6167.
Oliver, R.L. 1999. Whence consumer loyalty? Journal of Marketing, 63, 3344.
Pisano, G.P.; Bohmer, R.M.J.; and Edmonson, A.C. 2001. Organizational differences in the rates of learning:
Evidence from the adoption of minimally invasive cardiac surgery. Management Science, 47, 6, 752768.
Rai, A.; Lang, S.S.; and Welker, R.B. 2002. Assessing the validity of IS success models: An empirical test
and theoretical analysis. Information Systems Research, 13, 1, 5069.
Raundenbush, S.W., and Bryk, A.S. 2002. Hierachical Linear Models: Application and Data Analysis Meth-
ods, 2nd ed. Thousand Oaks, CA: Sage Publications.
Rubash, A.R.; Sullivan, R.R.; and Herzog, P.H. 1987. The use of an expert to train salespeople. Journal of
Personal Selling & Sales Management, 7, 4955.
Santhanam, R., and Hartono, E. 2003. Issues in linking information technology capability to frm performance.
Saxe, R., and Weitz, B.A. 1982. The SOCO scale: A measure of the customer orientation of salespeople.
Journal of Marketing Research, 19, 3, 343351.
Schilling, M.; Vidal, P.; Ployhart, R.; and Marangoni, A. 2003. Learning by doing something else: Variation,
relatedness, and the learning curve. Management Science, 49, 1, 3956.
Schmidt, F.; Hunter, J.; and Outerbridge, A. 1986. Impact of job experience and ability on job knowledge,
work sample performance, and supervisory ratings of performance. Journal of Applied Psychology, 71,
432439.
Seddon, P., and Kiew, M.-Y. 1994. A partial test and development of the DeLone and McLean model of IS
success. Presented at the International Conference on Information Systems, Vancouver, Canada, p. 12.
Seddon, P.B. 1997. A response location respecifcation and extension of the DeLone and McLean model of
IS success. Information Systems Research, 8, 3, 240253.
Sirohi, N.; McLaughlin, E.W.; and Wittink, D.R. 1998. A model of consumer perceptions and store loyalty
intentions for a supermarket retailer. Journal of Retailing, 74, 2, 223245.
Snijders, T.A.B., and Bosker, R.J. 1999. Multilevel Analysis: An Introduction to Basic and Advanced Multi-
level Modeling. Thousand Oaks, CA: Sage Publications.
Sonnentag, S., and Lange, I. 2002. The relationship between high performance and knowledge about how
to master cooperation situations. Applied Cognitive Psychology, 16, 491508.
Spiro, R.L., and Weitz, B.A. 1990. Adaptive selling: Conceptualization, measurement, and nomological
validity. Journal of Marketing Research, 27, 1, 6169.
Straub, D.; Limayem, M.; and Karahanna-Evaristo, E. 1995. Measuring system usage: Implications for IS
theory testing. Management Science, August, 13281342.
Tyagi, P.K. 1985. Relative importance of key job dimensions and leadership behaviors in motivating sales-
person work performance. Journal of Marketing, 49, 7686.
Van Overschelde, J.P., and Healy, A.F. 2001. Learning of nondomain facts in high- and low-knowledge do-
mains. Journal of Experimental Psychology: Learning, Memory, and Cognition, 27, 5, 11601171.
VandeWalle, D.; Brown, S.P.; Cron, W.L.; and Slocum, J.W., Jr. 1999. The infuence of goal orientation and
self-regulation tactics on sales performance: A longitudinal feld test. Journal of Applied Psychology,
84, 2, 249259.
Venkatesh, V. 2000. Determinants of perceived ease of use: Integrating control, intrinsic motivation, and
emotion into the technology acceptance model. Information Systems Research, 11, 4, 342365.
Venkatesh, V., and Davis, F.D. 2000. A theoretical extension of the technology acceptance model: Four
longitudinal feld studies. Management Science, 46, 2, 186204.
Wiley, C. 1997. What motivates employees according to over 40 years of motivation surveys. International
Journal of Manpower, 18, 3, 263280.
Wright, T.P. 1936. Factors affecting the costs of airplanes. Journal of the Aeronautical Science, 3,
122128.
Zack, M.H. 1999. Developing a knowledge strategy. California Management Review, 41, 3, 125145.
204
ChaptEr 11
CAN WE LEARN FROM OUR PAST?
Managing Knowledge and Learning Within and Across Projects
patriCK s.w. fonG
Abstract: Management by projects is becoming common. Organizations see projects embracing
peculiar features that provide fexibility and adaptable solutions to the challenges, changes, and
problems they face. However, it is also these peculiar features of projects that make knowledge
management even more important in projects. Without such deliberate management efforts, knowl-
edge assets will be lost once the project is completed. This results in organizational knowledge
fragmentation and the loss of organizational learning. Successful project management is based,
on the one hand, on accumulated knowledge and, on the other hand, on individual and collective
competences. Because of the uniqueness of the product produced or service rendered, project
organizations need to create new knowledge and learn throughout the project life cycle so that
they can meet the project objectives as well as complete them successfully. The amount of new
knowledge generated in a project depends on the novelty and uniqueness of the expected product.
However, it is often argued that the processes involved are more or less similar even though the
projects are unique. As a result, most projects do not need to start anew; those involved can learn
from what has been done in previous projects. In other words, historical data, information, knowl-
edge, and experience are valuable. Project deliveries can be enhanced through the integration of
lessons learned from previous projects. Moreover, the tight schedules of projects make the reuse of
knowledge necessary. Projects are often required to be completed within a relatively short period
of time. Without the reuse of existing knowledge, project organizations have to create solutions to
every problem they face. Projects are further characterized by the temporary assemblage of experts
with diverse expertise and from various backgrounds. The project organizations will be disbanded
once the project is completed, and these experts will be absorbed by their own organizations and
engaged in other projects. Sometimes, personnel may be relocated before a project is fnished.
This high fuctuation of experts produces a need to keep the important part of their knowledge
within the company before they leave, so that the lessons learned are not dispersed along with the
personnel themselves. Finally, knowledge is seen as a strategic asset and a primary resource for
competitive advantage; thus, the possibility to gain from earlier projects becomes central, not just
to save time and money but also to avoid reinventing the wheel, which results from the failure
to capture and transfer project knowledge. However, unlike permanent organizations, projects
do not have any organizational memory: they are only temporary organizations. In comparison
with organizations that are supported by structure and routines to absorb knowledge, projects do
not support any natural transfer mechanism. Deliberate management efforts and incentives are
crucial to the creation, capture and transfer of knowledge. For instance, lessons learned have to be
socialized consciously among individuals before they leave the project. The absence of knowledge
CAN WE LEARN FROM OUR PAST? 205
management will make projects unable to contribute to any improvement of the organizational
business processes. Lack of improvement will eventually lead to an obsolete business process.
Keywords: Knowledge Management, Project Management, Projects, Learning, Knowledge
INTRODUCTION
The objective of a project is to deliver a benefcial change or outcome. That change can take the
form of tangible or intangible deliverables or output. The former can be in the form of a new physi-
cal facility, object, or asset. The latter can be represented in social form, such as by organizational
change or intervention. As such, projects involve temporary organization and efforts in terms of
project input and output. They are unique, with complex and interdependent activities, and are
usually undertaken on a one-off basis. Projects are often constrained by resources such as time,
budget, manpower, and quality. Due to the complex nature of projects, they require cross-functional
or multidisciplinary team input. More and more organizations are adopting the project-organizing
approach to subdivide tasks among their employees, which causes projects to become dispersed
throughout organizations. Projects can be initiated internally within an organization for developing
internal activities or externally for delivering business solutions (Artto, 2001).
One central feature of this postindustrial era is that knowledge is being treated as one of the
most important resources of an organization (Nonaka and Takeuchi, 1995) and may be the only
meaningful resource (Drucker, 1993, p. 42). This has also led to the development of ways to
involve all project stakeholders for their unique knowledge inputs. A project, as a knowledge-in-
tensive activity, often involves cross-functional linkages, where different participants join a team
with differing viewpoints (see Figure 11.1). Duwairi, Fiddian, and Gray (1996) suggest that this
team interaction brings about a need to organize, integrate, flter, condense, and annotate the col-
laborative data and other relevant information that these team members contribute.
Managing new knowledge and perspectives is fundamental to project development. For ex-
ample, the development of a new product, treated as a project, can be considered as a package of
features and benefts, each of which must be conceived, articulated, designed and operationalized,
or brought into existence (Dougherty, 1996, p. 425). Another example is that the development
of a constructed facility can be viewed as a new project, with customers or end-users purchasing
or using the facility. They assess their own needs and the affordability of the facility before they
purchase it. The development of a new product entails the application of knowledge to new prob-
lem-oriented situations, thus requiring uncertainty reduction (Cyert and March, 1963). The same
applies to projects, with each project unique in terms of design and construction/manufacturing.
With the many constraints faced by projects (due to limited space, increasing project complexity,
limited budgets, tight programs, and the constant demand for innovation), project teams are faced
with the challenges of utilizing diverse knowledge and creating new knowledge in order to meet
stringent requirements and fulfll ever-changing needs. Project team members have to incorporate
new information into their understanding in order to solve the technical challenges they face.
The objective of this chapter is to investigate how projects can generate knowledge and
provide learning in a project environment, which, in turn, contributes to establishing effective
knowledge-managing and learning mechanisms within and across projects. The focus is on
why it is crucial to manage knowledge in projects that involve a small number of staff, as well
as the learning process on two levels: intraproject and interproject learning. At the same time,
the implications of the learning processes in the project environment will provide management
teams with a clearer picture of their role in enhancing learning within and across projects. In
206 FONG
addition, this chapter will form a basis for researchers who would like to pursue studies into
these interconnected issues.
The next section examines projects and their organization, covering aspects such as what constitutes
a project, features of project-based organizations, and different perspectives of projects (i.e., resources
and knowledge perspectives). The state-of-the-art in project knowledge management is examined
next, together with developments in practices for project knowledge management. A major section
on learning and project capability building is included, with the aim of examining organizational and
project-based learning. Project-based learning is further subdivided into intraproject and interproject
learning, together with their interdependency. The contribution of this chapter to research and practice
follows the previous section, and conclusions are drawn based on the whole chapter.
PROJECTS AND THEIR ORGANIZATION
What Is a Project?
The defnition of project most often cited in the project management literature (Gaddis, 1959;
Butler, 1973; Pinto and Prescott, 1988; Sabbag, 1999; Prencipe and Tell, 2001; Gould, 2002) is
the one provided by the Project Management Institute (PMI, 2000), which defnes a project as a
temporary endeavor undertaken to create a unique product or service. According to this defni-
tion, two peculiar features of a project are identifedtemporary and unique. Temporary means
that projects are limited in time (Lundin and Steinthrsson, 2003): they have a defned beginning
and end (Gould, 2002; Turner and Mller, 2003). However, temporary does not necessarily imply
short in duration: as long as the project is fnite in life span, it is fundamentally temporary. Further,
the project team is assembled on a temporary basisit disbands once the project is completed
(Prencipe and Tell, 2001). On the other hand, unique means that the project is a once-in-a-lifetime
task (Packendorff, 1995) in the sense that no two projects are exactly the same (Turner and Mller,
2003). Projects differ from each other in several critical aspects (Prencipe and Tell, 2001). They
Figure 11.1 Different Stakeholders Involved in a Project
Other Stakeholders
Project Organizations
Subcontractors
Suppliers Project
Organizations
Statutory Authorities
Construction Companys
Consultants
Clients/Owners
involve heterogeneous activities that may not be repeated in successive projects. In addition to
these two essential features, Koskinen, Pihlanto, and Vanharanta (2003) suggest that projects usu-
ally entail large, expensive, and high-risk undertakings that have to be completed within certain
time, cost, and quality constraints. In order to carry out the required tasks, all projects need to
have well-defned goals or objectives (Packendorff, 1995; Gould, 2002; Koskinen et al., 2003;
Lundin and Steinthrsson, 2003).
Notwithstanding the widespread acceptance for the PMIs defnition, Turner and Mller (2003,
p. 2) question its completeness by saying that it is adequate for some purposes, but does not
defne fully the roles, functions and limits of projects. They readdress the nature of projects in
terms of a production function, a temporary organization, an agency for change, an agency for
resource utilization, and an agency for uncertainty management. After reviewing the nature of
projects in the aforementioned perspectives, Turner and Mller (2003, p. 7) propose a revised
defnition of a project:
A project is a temporary organization to which resources are assigned to undertake a unique,
novel and transient endeavor managing the inherent uncertainty and need for integration in
order to deliver benefcial objectives of change.
Although there is no universal consensus on the defnition of a project, considering most general
defnitions, the common understanding is nonetheless that it is a unique, specifc, and nonrecurrent
task that has to be completed by a certain date, for a certain amount of money, and within some
expected level of performance and that it is complex in implementation and subject to evaluation
(Packendorff, 1995).
Features of Project-Based Organizations
In terms of the execution of projects, Eskerod (1996) defned project organizations as ones with
the following organizational characteristics: (1) the tasks of the frms are carried out as projects;
(2) the projects share common resources including human resources; (3) the projects can be of
different natures (e.g., marketing, product development); (4) the projects are managed by project
managers; (5) all project team members report and refer to the project manager; and (6) employ-
ees can work on several projects at the same time. Consequently, project organizations manage
the execution of multiple projects concurrently or sequentially (Nobeoka, 1995; Cusumano and
Nobeoka, 1998), which creates a multiproject environment.
Because of the uniqueness of the product produced or service rendered, project organizations
need to create new knowledge and learn throughout the project so that they can meet their objec-
tives. The amount of new knowledge needed to be generated in a project depends on the novelty
and uniqueness of the expected product (Pohjola, 2003); compare, for example, designing a
spacecraft for a mission to Saturn and building a series of houses using the same design. However,
it is often argued that the processes involved are more or less similar even though the projects
are unique. As a result, most projects do not need to start anew. The workers can learn from what
has been done in previous projects. In other words, historical data, information, and knowledge
are valuable. Project delivery can be enhanced with the integration of lessons learned and/or
project reviews from previous projects. Like other organizations, project organizations are also
required to improve the quality of their products and services in order to succeed in a competitive
environment. The competitive environment includes shortened product development life cycles
and consumer demands for new designs and better functionality of products. These all contribute
208 FONG
to an increased need for knowledge. Examples include designing a new drug to cure asthma or
enhancing the functions of mobile phones by incorporating digital cameras, MP3, and movies.
In addition, the time restrictions of projects make the reuse of knowledge necessary. Projects are
often required to be completed within a relatively short period of time. Without the reuse of ex-
isting knowledge, project organizations have to create their own solutions to every problem they
face. However, these problems may have been dealt with previously in other projects. With the
reuse of knowledge, project organizations can operate more effciently, and thus the time require-
ment can be met more easily. Projects are further characterized by the temporary assemblage of
experts with diverse expertise and from various backgrounds. These experts involved in projects
are not only organizationally but can also be geographically dispersed (Kasvi, Vartiainen, and
Hailikari, 2003), such as virtual project teams where members are located in different countries
or time zones. Project teams will be disbanded once the project is completed and these experts
will be absorbed by their own organizations and engaged in other projects. Sometimes, personnel
change during a project. This high fuctuation of experts produces a need to keep at least part of
their knowledge within the project before they depart, so that the lessons learned are not dispersed
along with them.
Different Perspectives of Projects
In the following sections, two different views of looking at projects are developed. They are not
meant to supersede one another: in fact, they provide a new perspective on viewing projects in
organizations or project management, as many industries, such as construction, engineering,
information technology (IT), health services, business, arts, and media, are classifed as project
based. The frst project perspective is the more widely known, normative view of managing projects
through the deployment of various resources, including plant, machinery, people, materials, and
so on. The second view looks at projects from the knowledge perspective as, from time to time,
knowledge acquired or created in previous projects can be reused or transferred to current ones.
This prevents reinventing the wheel. Examples include the development of a new car model or
digital camera, conducting a surgical operation, or relocating the head offce. It is very unusual to
start a project from scratch if there are precedent cases to follow. The business benefts brought
about by properly managing knowledge in projects are increased effciency, enhanced leverage
of past knowledge, fewer mistakes or errors, and increased profts for the organization. The chal-
lenge for project team members is the timely application of real-time emerging knowledge and
knowledge from past projects in order to realize such benefts.
Projects from a Resources Perspective
The dominant traditional view of project management emphasizes normative techniques and meth-
ods for planning, controlling, and evaluating the use of resources to accomplish a project where the
owner or clients objective(s) needs to be met. These tools and techniques are developed primarily
by consultants, project managers, and engineers and are usually prescriptive in nature. Examples
of these project management tools include critical path analysis, work breakdown structure, Monte
Carlo simulation, resource histogram, and others. Project objectives can be defned in terms of
quality, time, and cost, but they very much depend on the focus of each project (see Figure 11.2).
The underlying assumption of the primary function of project management is to get something
done on time, within budget, and to a specifed quality or function level. This performance-based
approach is used to evaluate competencies in projects (Crawford, 1999). This is especially true
in the construction and IT industries, where the timely delivery of projects is key while learning,
capturing, and sharing knowledge are totally ignored. To illustrate the concept of meeting speci-
fed project goals by a couple of examples, the design and construction of the HSBC (formerly
the Hong Kong and Shanghai Banking Corporation) headquarters were intended to project an
image of prestige to the general public. Unusually, the project objective did not include a defnite
cost budget. On the other hand, the project to build the Hong Kong headquarters of the Bank of
China just 100 m away concentrated more on the cost aspect by using combinations of different
geometrical shapes in the external wall design in order to maximize design effciency and ease
of construction.
The proper planning of a project is crucial in measuring whether a project is successful in terms
of the objectives already laid down. The controlling process is mainly to run a project according
to the predetermined plan. Of course, due to the many uncertainties and risks inherent in projects,
the original plan will have to be revised from time to time to refect the actual progress of the
project. The evaluating process is undertaken by comparing outcomes with targets prepared during
the planning stage. A quick glance at some of the popular project management journals, such as
the International Journal of Project Management and the Project Management Journal, reveals
that a large number of the articles published are related to tools and techniques used in planning,
controlling and evaluating projects. The Project Management Body of Knowledge (PMBOK),
published by the American-based PMI, also emphasizes integration, scope, time, cost, quality,
human resources, communications, risk, and procurement management (PMI, 2000). Such lit-
erature belongs to the traditional view of project management, as classifed by Engwall (1998). It
provides practical tools, procedures, and checklists of an applicable nature, rather than manage-
ment practice. An organization can easily adopt these tools for its own internal consumption. In
addition, the success of project management is measured by effciency in the project processes,
accomplished through rationalistic procedures. In fact, there is evidence that two of the major
project management associations, the Association for Project Management in the UK (APM, 2003)
and the Project Management Institute in the United States (PMI, 2003), have increased activities
related to linking project management with knowledge management.
Figure 11.2 Projects from a Resources Perspective: Intraproject Focus
Internal Focus
Time
Project C
Project D
Project B
Project A
Project E
210 FONG
The traditional project management approach views projects as separate entities that are de-
tached from other projects and activities. This short-term focus on tasks may result in the loss of
localized learning (Wenger, 2000) or a lack of systematic and institutionalized learning (Pinto,
1999). Compared with a functional manager in an organization, who manages a particular function
within the organization, a project manager only manages that individual project. Turner and Keegan
(1999) called for a different approach in management practices to these differing organizational
structures. In fact, it is not uncommon for project managers within an organization to compete
for resources. Because there is little connection between projects, project managers may not even
know what projects others are managing.
The widely accepted and traditional view of project management is focused on the systematic ap-
plications of the tools and techniques for project management practitioners and has been questioned
in the past decade by Engwall (1998) for ignoring the many different nuances of the project concept.
Packendorff (1995) also contended that a project can be seen as fulflling both external prearranged
objectives and objectives of its own. He further adds a distinction between projects as plans and
projects as temporary organizations. The former seems to be the approach widely used in many
countries, where the focus is on planning, controlling and evaluating. This is classifed as the traditional
view of project management. The latter approach is becoming more prominent in Scandinavia, where
they focus more on the temporary nature of project organizations, the process dynamics within project
organizations, the relationship between projects and their parent organizations, and so on.
Projects from a Knowledge Perspective
This chapter takes the perspective of the frm as a knowledge creation and application system and
as such falls within the domain of the knowledge-based view of the frm. A core issue that has been
raised in the literature is the tension between exploration for new organizational knowledge and the
exploitation of current organizational knowledge. This tension has become the focus of consider-
able theoretical and empirical research (e.g., March, 1991; Levinthal and March, 1993).
Within the existing literature, exploration can be defned as the pursuit of new knowledge
of things that might come to be known and exploitation as the use and development of things
already known (Levinthal and March, 1993, p. 95). March (1991, p. 71) observes that the main-
tenance of a balance between exploration and exploitation is a primary factor in system survival
and prosperity. In this respect, a central component of success is the maintenance of a balance
of exploration and exploitation within projects, as both knowledge development and knowledge
use are crucial in fast-changing environments.
Knowledge Exploitation. Althoff, Bomarius, and Tautz (2000) typify knowledge as experience,
as it is thought that what contributes to the quality of a decision is the experience of the decision
maker. According to the PMBOK (PMI, 2000, p. 24), decision making is a key management
skill that includes analyzing the problem to identify viable solutions, and then making a choice
from among them. However, the topic was not developed further. During project progress,
project teams must make many decisions in multidimensional contexts with different degrees of
uncertaintyeven for repetitive projects. Decision making is therefore an essential ongoing feature
of project management. Moreover, decision making depends on knowledge input, and the qual-
ity of the decisions is directly related to the accessibility, structure, and quality of the knowledge
available. The PMBOK (PMI, 2000) defnes project management as the application of knowledge,
skills, tools, and techniques to project activities in order to meet or exceed stakeholder needs and
expectations from a project.
Through deliberating projects, new knowledge is created and both positive and negative experi-
ence is gained by project team members. Such experience is made available not only from projects
but also through the social interactions among project stakeholders and networks of individual
team members. Bowen, Clark, Holloway, and Wheelwright (1994) emphasized that the knowledge
gained from a project has to be recognized by employees in two dimensionsnot only as obvious
products or processes resulting from the project but also in terms of how the results are achieved,
which stresses the importance of learning in and between projects.
Knowledge Exploration. In Leonard-Bartons (1992, p. 23) paper The Factory as a Learning
Laboratory, she described a learning laboratory as an organization dedicated to knowledge
creation, collection, and control. A project is assembled very much like a learning laboratory,
where the contribution of knowledge, skills, and experiences is a key principle for all activities in
order to optimize the use of resources and meet the owners objectives. In addition, as suggested
by Leonard-Barton (1992, p. 23), tremendous amounts of knowledge and skill are embedded in
physical equipment and processes and embodied in people. More important, activities evolve
around the project, so that it is often dynamic, and renewing and changing according to the activities
is inevitable. It is almost true to say that a project can symbolically be classifed as an experiment
or laboratory where some of the knowledge and skills required for the project may come from
past experiences or the memories of individual project participants. On the other hand, there are
occasions where newly created knowledge is required during the execution of the project (Fong,
2003). It is not uncommon for the same situation to arise in two different projects but for changes
in circumstances or technology to warrant new knowledge input to the situation. For example,
the adoption of the same house design for different locations may require new knowledge for the
foundation, as the bearing capacity of the soil underneath may differ and so may the climate. These
projects are often uniqueone-offsin their design, location, end user, and budget.
During their participation in the project, team members engage in a learning process, gaining
new knowledge and experience that could be applicable to other projects within their organiza-
tions (see Figure 11.3). This means that a project can be viewed as a learning laboratory for the
entire organization if a proper channel for transferring knowledge is established. Some researchers
(Hedberg, 1981; McGill and Slocum, 1993) have suggested that other than acquiring new knowl-
edge, organizations have to deal with knowledge no longer requiredorganizational unlearning.
Hedberg (1981, p. 18) defnes unlearning as the process through which learners discard knowl-
edge to make way for new responses and mental maps. In a similar vein, project personnel need
to recognize that knowledge deteriorates, becomes obsolete, and results in bad decisions during
project organizing.
To use the perspective of the project as a learning laboratory for the company means that proj-
ects are not seen as separate entities to be managed as in the past. Rather, it should be recognized
that knowledge gained from one project can be used in other projects, whether those projects
are carried out concurrently or sequentially. This knowledge-based view of projects is not meant
to substitute for the traditional project management view: instead, it complements the latter by
recognizing knowledge as a key ingredient during the execution of the project.
KNOWLEDGE MANAGEMENT IN PROJECTS
As management by projects is becoming common, organizations see projects as embracing peculiar
features that provide fexible and adaptable solutions to the challenges, changes, and problems they
face. However, it is also these peculiar features of projects that make knowledge management even
212 FONG
more important in projects. This concept of project knowledge management is lacking, although
there is a considerable literature on knowledge management as well as on project management
(Kasvi et al., 2003). Kasvi et al. (2003) argue that systematic project knowledge management is
needed if a project organization wants to evolve into a learning organization and apply the solutions
and lessons from one project to another. Lundin and Midler (1998) shared a similar view, as the
project in itself cannot and has not been created to memorize and store what has been learnt.
Crawford (2000) identifed knowledge management as one of the seven major areas of project
competences seen as critical success factors for projects. However, Kasvi et al. (2003) adopted those
seven factors proposed by Crawford (2000) in their study and found that knowledge management
does not appear to be a critical project competence area among the competence areas of commu-
nication and interaction, processes and procedures, technology, leadership, project management,
and interest group connections, as there was an intrinsic confict among their interviewees between
perceived and observed competences.
Knowledge management in projects is important because of the increasingly complex situation,
together with clients or owners ever-evolving requirements, where additional information and
knowledge are needed to produce excellent solutions for products or services. On the other hand,
more knowledge is available with new technologies (Korppi-Tommola, 2003). The management
of knowledge is thus made crucial because without such management efforts, knowledge assets
will be lost once the project is completed. This results in organizational knowledge fragmentation
and loss of organizational learning (Kotnour, 2000). This makes more relevant the knowledge
capture aspect. As Kasvi et al. (2003) state, the identifcation of critical knowledge and the ability
to utilize it are a challenge for any project organization. Successful project management is based,
on the one hand, on accumulated knowledge and, on the other hand, on individual and collective
competences. Deliberate management efforts and incentives are crucial to the creation, capture, and
transfer of knowledge. For instance, lessons learned have to be made explicit among individuals
before they leave the project. The absence of knowledge management will make projects unable
to contribute to any improvement of the organizational business processes. Lack of improvement
will eventually lead to obsolete business processes.
In the United States, lessons learned systems (LLS) supported by knowledge management
Figure 11.3 Projects from a Knowledge Perspective: Intra- and Interproject Focus
Time
Project C
Project D
Project B
Project A
Project E
and organizational learning achieve great success in real-life practice for many organizations and
government agencies, such as Lockheed Martin, NASA, and the U.S. Army. The goal of LLS is
to capture and provide lessons that can beneft employees who encounter situations that closely
resemble a previous experience in a similar situation (Weber, Aha, and Becerra-Fernandez, 2001).
In addition, LLS are probably one of the most frequently used knowledge management systems
associated with projects. Lessons learned systems include the processes of acquiring, handling,
and verifying good or bad practices from projects in different stages, disseminating the verifed
and approved lessons learned to related parties, and recording such practices in the preferred man-
ner. Through LLS, practitioners can learn from previous recorded lessons to improve their work-
ing performance. It is a formal system that allows individual learning to become organizational
learning through systematic records in different stages of projects. Project professionals can learn
from previous lessons learned as well as share their knowledge or experience by submitting new
lessons learned. In this manner, individual knowledge can become organizational knowledge. We-
ber, Aha, and Becerra-Fernandez (2001) suggested fve lessons learned processes: collect, verify,
store, disseminate, and reuse. They highlighted six lesson collection methods: passive, reactive,
after action, proactive, active, and interactive collection. Lesson verifcation should be performed
by a team of experts to ensure correctness, consistency, and relevance, and their storage should
be represented and indexed in a way that can be easily located by users (Weber et al., 2001). The
dissemination subprocess is the most important step regarding the promotion of lessons reuse,
and the various dissemination methods that can be adopted include passive dissemination, active
casting, broadcasting, and proactive and reactive dissemination (Weber et al., 2001). Although an
LLS can provide valuable organizational knowledge to users, it is at the entire discretion of users
how they want to make use of the lessons.
Various means have been used for managing knowledge in project environments: they include
after action review (AAR) (e.g., the U.S. Army [Dixon, 2000]), LLS (Weber et al., 2001; e.g.,
NASA [Liebowitz, 2005]), communities of practice (e.g., IBM [DeFillippi and Arthur, 1998]),
Peer Assist (e.g., British Petroleum [Dixon, 2000]), Best Practice (e.g., Lockheed-Martin [Dixon,
1999]), Project Reviews/Project Audits (e.g., British Petroleum [Gulliver, 1987]), Project De-
briefng (e.g., UBS [Schindler and Eppler, 2003]), and storytelling (Denning, 2001). Weber et al.
(2001) and Becerra-Fernandez et al. (2004) provide a comparison of various types of knowledge
repositories based on their attributes. Table 11.1 is an expanded version of the comparison table
to cover other practices not specifed in these two sources but included above.
Table 11.1
Types of Knowledge Practices
Knowledge Practices
Originates
from
Experiences
Describes
a Complete
Process
Describes
Failures
Describes
Successes Orientation
After action review 3 3 3 3 Organization
Lessons learned systems 3 7 3 3 Organization
Communities of practice 3 7 3 3 Inter-/intraorganization
Peer assist 3 7 7 3 Organization
Best practices 3 3 7 3 Industry/organization
Project reviews/project
audits
3 7 3 7 Organization
Project debriefng 3 7 3 3 Organization
Storytelling 3 3 3 3 Organization
214 FONG
LEARNING AND PROJECT CAPABILITY BUILDING
Kotnour (1999) argued that without having the right capabilities, the organization cannot deliver
a successful project, still less a series of projects. Among the various core capabilities of a project
environment, the most important capability that helps the project team to accomplish its goals is
the project learning process. Researchers (e.g., Nobeoka, 1995; Kotnour, 1999, 2000) have dem-
onstrated that the generative beneft of learning in projects is the increase in the performance and
capabilities of multiproject organizations. Wheelwright and Clark (1992) stated that (1) complex
interaction results, (2) time separation of cause and effect in projects, and (3) incentives, which typi-
cally press forward to the next project, present considerable obstacles to learning from projects.
Kotnour (1999, p. 32) defned learning in projects as the process by which knowledge is cre-
ated from experience and the path by which improvement takes place. The knowledge created
through the learning process is of long-term beneft to the organization. Yet, as noted by Ayas
and Zeniuk (2001), project knowledge is not always disseminated within the same project, and it
can hardly be transferred across projects accordingly. A number of researchers studying project
management have stressed that an increased risk of reinventing the wheel, wasted activity, and
impaired project performance would result if project organizations failed to capture and transfer
knowledge from projects (Siemieniuch and Sinclair, 1999; Kamara et al., 2002). The problem
of reinventing the wheel occurs when existing knowledge and experiences are not available for
access and use because they are not stored and disseminated (Disterer, 2002). The larger the in-
vestment at stake, the greater is the potential cost of failure to learn (Barker and Neailey, 1999).
Notwithstanding the importance of project learning in avoiding such detrimental effects, in the
studies conducted by Kotnour (1999) and Prencipe and Tell (2001), all the researchers arrive at
the interesting conclusion that project managers are aware of the reinventing the wheel problem
and agree that project performance can be improved through not repeating the mistakes of the
past. They nevertheless accepted that this was inevitable and that they could do nothing about it
due to their projects-as-islands mentality and the prime focus of completing projects according
to the predetermined objectives. As a result of this misperception of project managers, Williams
(2003) comes to the similar fnding that business displays signifcant weakness in project learning,
as reasons for project success or failure are rarely explored and management behavior is rarely
adapted to the lessons learned, owing to the fact that facilitated learning or refection in a concerted
fashion is rarely carried out.
It is not diffcult to observe that many projects are not successful. This may partly be explained
by the aforementioned phenomenon. It is therefore essential for project organizations to learn from
both successes and failures, as well as problem resolutions. According to Antoni (2000), these
are often problems or mistakes similar to the ones that cause project uncertainties. If the project
team is able to learn from past mistakes and prevent them from happening again, the uncertainties
that might lead to ineffciencies can be greatly reduced. Furthermore, as suggested by a number
of writers, project learning also enables organizations to develop project competencies that lead
to a sustainable competitive advantage (Barker and Neailey, 1999; Huber, 1999; Gnyawali and
Stewart, 2003; Schindler and Eppler, 2003; Sense, 2003). Competitive advantage is a determinant
of the survival of organizations in todays hypercompetitive environment, thus it is necessary to
consider project learning seriously.
Fiol and Lyles (1985) identifed four key factors that are critical to organizations capacity to learn
but can equally apply to the project management context: (1) a culture that encourages learning,
(2) a strategy that permits learning, (3) an organizational structure that promotes innovation, and
(4) an appropriate environment. The APQC (2005) reports that by integrating a learning-centric
organization together with a systems view of knowledge, these outcomes will be expected, which
have been modifed to suit project organizations: (1) increasing project team members capabilities
and/or project-organizing effciencies or building new capabilities for organizing by projects, (2)
leveraging and using stakeholder/project resources to better serve stakeholders; not solely to save
time and cost but to avoid reinventing the wheel and/or repeating mistakes in every new project,
(3) capturing and transferring project knowledge and best practices to improve concurrent/future
projects, and (4) excelling beyond previous/concurrent projects, leading to improved performance
in future. Because projects connect to permanent organizations as well as acting as individual
entities, it is essential to study learning at both the organizational and project levels.
Organizational Learning
The distinction between individual and organizational learning is a vital issue in the organiza-
tional learning literature. As Dodgson (1993) pointed out, focusing on organizational learning
alone undervalues some of the characteristics of learning. For example, routines in processes are
argued to be independent of the individuals who operate and use them. In addition, behind the
theory of organizational learning, there is an assumption of uniformity in learning processes within
frms, while in complex organizations different learning processes can occur at the same time,
in different directions or at different speeds. Miller (1996) defned organizational learning as the
acquisition of new knowledge by employees who are able and willing to apply that knowledge
in making decisions or infuencing others in the organization. He differentiated between learning
and decision making: while the former increases ones knowledge, the latter involves a sequence
of steps in problem solving. By contrast, Simon (1991) maintained that an organization learns in
two ways: by the learning of its staff or by introducing new members who possess unique knowl-
edge that the organization did not previously own. He found that what an individual learns in an
organization is very much dependent on what is already known (or believed) by other members
of the organization and what kind of information is present in the organizational environment
(Simon, 1991, p. 125). With this defnition, one important element is highlighted that forms the
basis of the knowledge management discipline (i.e., the role of memory).
Kim (1993) portrayed the relationship between individual and organizational learning through
the concept of memory, where individual learning is depicted as a Deming (1986) cycleplan, do,
check, act (PDCA) (Figure 11.4). It represents the creation of knowledge and memory is located
in individual mental models. According to Senge (1990), mental models are collections of ideas,
experiences, and schemas through which individuals interpret their own reality. On the other hand,
organizational learning is interpreted as a loop of individual beliefsindividual actionorgani-
zational actionenvironmental response (March and Olsen, 1976). This loop is complete if the
organizational response infuences individual beliefs. If the project environment changes, individual
beliefs can also change, also creating different individual and organizational actions. This model of
organizational learning is coherent with the interpretation of the organization as an interpretation
system (Daft and Weick, 1984), where organizational learning is a process of exploration (col-
lection of data), interpretation (meaning to data), and learning (action undertaken). According to
Kims model, individual learning does not necessarily imply organizational learning. In order to
fully understand organizational learning, it is necessary to introduce the concept of shared mental
models (theories in action) that represent the memory of the organization. Shared mental models
constitute the way an organization interprets its own world and assumes that individuals improve
their mental models. This model of individual and organizational learning fts with the levels of
learning introduced by Argyris and Schn (1978). Moreover, the distinction between individual
216 FONG
and organizational learning has been very useful in representing situations where learning does not
occur, as barriers to learning arise due to incomplete loops at either individual or organizational
or between the two levels (March and Olsen, 1976; Kim, 1993).
Organizational learning is a multilevel phenomenon that occurs at several levels in an organiza-
tion (Levitt and March, 1988; Huber, 1991). A number of empirical studies on learning from the
management studies perspective have differentiated the various types and levels of learning. Argyris
and Schn (1978) developed a three-fold typology of learning which they described as single-loop,
double-loop, and deutero-learning. Single-loop learning occurs whenever an error is detected and
corrected without questioning or altering the underlying values of the system (Argyris and Schn,
1978). It is found adequate in business units. The concept of double-loop learning is derived from the
earlier research of Bateson (1972), who found that dolphins have a second layer scanning system that
monitors how they translate signals into actions and check for defects. This type of learning occurs
when an error is detected and corrected in ways that involve the modifcation of an organizations
underlying norms, policies and objectives (Argyris and Schn, 1978, p. 3). Project teams are more
likely to be motivated to develop this type of learning orientation. Deutero-learning is learning how
to learn that serves to evaluate and improve the learning process itself.
Huber (1991) classifed learning into fve types: congenital learning, experiential learning, vicari-
ous learning, grafting, and searching and noticing. Congenital learning refers to the acquisition of
knowledge prior to the founding of organizations. By contrast, experiential learning is learning by
doing or learning from experience from experimenting and testing. This occurs when project team
members are assigned to projects in which they do not have prior experience. Vicarious learning
is second-hand learning or learning by imitating others. Grafting is bringing new members who
possess new knowledge which they share with the organization or acquiring new technologies.
Searching and noticing refers to scanning the environment for information and changes.
Given that both projects and learning are recognized and used as vehicles for competitive
business survival, the next section examines the interfaces between projects and learning.
Project-Based Learning
Learning and projects are not a natural combination (Bartezzaghi, Coros, and Verganti, 1997).
This view is shared by Ayas and Zeniuk (2001), who stated that learning is not a natural outcome
Figure 11.4 The PDCA Learning Loop
Source: Adapted from Deming, 1986.
P
D
C
A
LEARNING
PROJECTS
of projects and a project-based organization is not necessarily conducive to learning. Antoni
(2000) explains this phenomenon by suggesting that projects and learning involve a number
of contradictory interests. These conficts usually arise from the peculiar characteristics of
projects. Unlike permanent organizations, projects are temporary in nature, with predefned
commencement and completion dates. With their foreseeable short life span compared with
permanent organizations, there is little incentive to learn from them, because they are not a go-
ing concern. Another peculiar characteristic of projects, uniqueness, also furthers the conficts
between projects and learning. Project teams are often prejudiced against the applicability of
the knowledge generated in a project, because they think that every project is unique and there
is no point learning from something that may well not be repeated again. Cooper (1994, p. 15)
emphasized that if an organization thinks it has a unique situation it is unlikely to gain from
the experiences of past works or others. True systemic causes and transferable project manage-
ment lessons are there to be learnt.
Due to the aforementioned conficts, project teams are reluctant to engage in learning activities
deliberately, notwithstanding the signifcant personal learning opportunities provided by projects.
In that context, there is a built-in assumption that learning occurs randomly and will appear un-
inhibitedly during the project. However, learning in projects requires some deliberate attention,
commitment and resources (Sense and Antoni, 2003). Smith and Dodds (1997, p. 2) considered
a project as a vehicle enabling a manager to undertake a journey resulting in both learning and
practical beneft to the business. The opportunity for learning is an inherent part of the project
management process: the learning process can go alongside the project management process. From
his study, Kotnour (1999) revealed that project managers often found themselves lacking time to
undertake the learning process. They in fact viewed learning as a separate activity. Kotnour (1999),
however, explained from the learning perspective that the project management process can serve
as a basis for creation and transfer of knowledge for the project team.
The learning generated by project teams provides an opportunity for organizational learning
(Keegan and Turner, 2001). Adapting the lessons learned generated by the project team into or-
ganizational routines, information and process, the experience and insights of a particular project
team can be shared throughout the organization. In other words, it will be of great beneft to the
organization to manage the learning process as well. The PMI (2000) suggests a way for project
learning through capturing knowledge at regular review points during the life cycle of a project.
Schindler and Eppler (2003) have discovered that continuous project learning through regular
reviews can be a critical success factor for projects.
Depending on the relationships between the project in which experience is acquired and the
project in which it is applied, two different forms of learning can be distinguishedintraproject
and interproject learningand they will be discussed in the following sections.
Intraproject Learning
Intraproject learning refers to a learning process in which the acquisition and the use of experience
both occur within the same project (Antoni, 2000). Intraproject learning (Figure 11.5) emphasizes
improving project performance by identifying problems and solving them during the project. Antoni
(2000) considered intraproject learning as able to align and develop the capacity of a team to create
the result its members truly desire. Intraproject learning thus supports successful project delivery
by creating and sharing the experience during the project life cycle. Through this process, the
uncertainty involved in a project can be reduced by applying the experience obtained during it.
Learning occurs whenever project team members discuss approaches for completing a task
218 FONG
or overcoming problems. For example, sharing successes/failures through project reviews at
key decision-making stages of projects can facilitate this type of learning. New knowledge
is created accordingly, and it has to be disseminated and applied in order to accomplish the
desired results. In this context, new knowledge is already shared during its lifetime or after
the project is completed. A survey of project managers conducted by Kotnour (2000) reveals
that the majority of project managers produced lessons learned at the end of the project cycle.
Only a few of them produced the lessons learned throughout the project. As emphasized by
Kotnour (2000) and Schindler and Eppler (2003), intraproject learning is not simply postproject
review. Instead, it should be regarded as a continuous improvement process. An obvious prob-
lem with producing lessons learned at the end of the project is that there are biases infuencing
the subject matter about which the project team will create lessons learned (Kotnour, 2000).
In addition, key project personnel may depart or be reassigned to other projects, or lessons
learned may be forgotten with the passing of time. Besides generating and gathering lessons
learned at the end of a project, this can be done throughout the project life cycle. Each stage of
a project can be regarded as a learning cycle in itself. With the learning cycle, lessons learned
can be produced for each cycle in a project, in order to carry the learning on to the next cycle
and to the next project.
It is often argued that the lifetime of a project might be too short to learn or make improve-
ments. Time is an important consideration in measuring project success. As a result, completing
projects on time is given high priority by the project team. Project learning activities, in the
majority of project managers understanding, have long-term characteristics that may extend the
project duration as well as incur additional expense for the project at hand. Williams (2003) fnds
that insuffcient resources are allocated by management to refecting and learning from projects,
and that many times rework consumes enormous amounts of time and money that go beyond the
resources available to the project. Love, Irani, and Edwards (2003) suggest that project learning
enables the accumulation and sharing of experience and that the possibility of rework could be
progressively reduced through this process.
Another diffculty in project learning, mentioned by Pinto (2000), is that in some cases the
scope of the project is ever-changing during its life. These changes may involve a change of mind
by the customer or changes related to organizational politics, the environment, or stakeholders
preferences. Even if the project has allowed its members to learn something by the end of its
lifetime, the possibility of applying this experience to other projects may be limited by the fact
that the experience could be obsolete by then.
Figure 11.5 Intraproject Learning
Source: Adapted from Bartezzaghi et al., 1997.
MODELS
PROJECT
A C
D P
Interproject Learning
Interproject learning is the combining and sharing of lessons learned across projects to apply
and develop new knowledge (Kotnour, 1999). Interproject learning takes place by transferring
the experience acquired on different projects over a period of time. It is the accumulation
of knowledge and its transfer to subsequent projects. Interproject learning is certainly not a
natural outcome of any project activities; it requires a high degree of awareness as well as
some sort of knowledge-transfer mechanism (Figure 11.6). From an interproject learning
standpoint, learning should be seen as a means of enhancing organizational performance in
the long term.
Learning in projects should not be studied only within projects, according to Packendorff (1995),
but also between projects; learning in one project has an impact on the learning in another, even
though the project activities and stakeholders are no longer the same. By continuously building
an organizations capability to execute the project management, product and learning processes,
interproject learning aims at delivering a series of successful projects (Kotnour, 1999). Similar to
intraproject learning, interproject learning achieves a reduction in future uncertainty by exploiting
the experience gained in previous or concurrent projects. Good practices and procedures can be
derived from the experience of other project organizations. These good practices can contribute
to enhancing organizational capability so that it can perform better in current and future projects
(Antoni, 2000).
Knowledge can be transferred directly or indirectly to other projects through two different
interproject linkages proposed by Nobeoka (1995): the concurrent and sequential transfer modes.
The concurrent transfer mode involves the transfer of knowledge before the projects have been
completed. Direct communication and interactions between the project teams are made possible
in this kind of project learning. The sequential transfer mode, on the other hand, means that
knowledge and experience are transferred after the initial project has come to an end. It may not
be effective and effcient compared to concurrent design transfer, because direct interactions and
mutual adjustments are not possible between two projects.
Prencipe and Tell (2001) studied the problems associated with interproject learning and con-
cluded that as projects involve heterogeneous activities that are only applicable to a particular
project, project organizations may face the diffcult task of learning from samples of one or
fewer. Furthermore, the temporary constellation of people implies that new human encounters
Figure 11.6 Interproject Learning
MODELS
PROJECT
META
MODELS
220 FONG
and relationships take place whenever a new project is started, which may increase the barriers to
learning from the previous experience of others.
Finally, a central prerequisite for interproject learning is a certain degree of repetitiveness be-
tween projects, because similarities allow construction and refnement of procedures in projects,
whereas total uniqueness in a project hinders learning (Lundin and Sderholm, 1995; Partington,
1996). Newell (2004) fnds that the most effective sharing of learning across projects is done
through sharing the right content at the right time to the receiver project team. This is done through
the strong networks established between team members of the source and receiver project teams.
Wenger (1998) found that multiple memberships contribute to informal webs of relationships,
which may enhance knowledge sharing within and across projects.
Interdependency of Intraproject and Interproject Learning
Intraproject learning and interproject learning are by no means independent. As suggested by
Smeds, Olivari, and Corso (2001), they are interdependent as the success of an individual project
depends on how the lessons learned are applied during the execution of the project and how
feedback is generated and gathered at the end of the project. Bartezzaghi, Coros, and Verganti
(1997) found intraproject learning easier as the feedback or learning loops are faster. Inter-
project learning is more diffcult as it is more context specifc and contrary to inherent project
characteristics. Figure 11.7 highlights the processes of and relationship between intraproject
and interproject learning.
Kotnour (1999) suggested that the intraproject learning cycle supports the interproject learning
Figure 11.7 Processes of and Relationship between Intra- and Interproject Learning
past project
decision
(plan)
implementation
(do)
analysis of variances
(check)
past project
decision
(plan)
implementation
(do)
(check)
abstraction and
generalization abstraction and
generalization
meta models
embodiment embodiment
models
dissemination
application
decision
(plan)
implementation
(do)
(check)
new projects
Intraproject Learning
Interproject
Learning
cycle by providing a routine, ongoing store of data, information, and knowledge that is integrated
for interproject lessons learned. The intraproject learning produces a living lessons learned
journal for interproject learning. Bartezzaghi et al. (1997), however, recognized that despite the
importance of intraproject learning in improving the decisions and performance of a project, it is
not suffcient to ensure the exploitation of experience from previous projects. They considered the
lessons learned by intraproject learning too context specifc and thus too diffcult, if not impos-
sible, to transfer to subsequent projects. Interproject learning is needed to abstract and generalize
the experience so that it can be used in other projects.
No matter how useful project learning is, it cannot be implemented successfully without man-
agement support. Previous research reveals that numerous barriers deter project organizations from
achieving project learning. Among those cited are poor project management, internal politics,
organizational inertia, expense cutbacks, communication diffculties within and between project
teams, and the sticky nature of organizational learning outcomes from projects (Keegan and
Turner, 2001). These authors add that time pressures, centralization, and deferral of decisions are
also threats to project learning. Williams (2003) concludes that managements role in facilitating and
encouraging learning from projects is vital. Lessons needed to be considered during the project, and
when the project is completed, interpersonal structures need to be in place to distribute that learning
throughout the organization.
CONTRIBUTIONS TO RESEARCH AND PRACTICE
This chapter not only has made a contribution to knowledge in the immediate felds of knowledge
management, theories of learning, theories of knowledge, and project management but also
has implications for the parent disciplines as well as other related areas. After all, organizations
use a project-organizing approach in managing their assignments.
From a knowledge management point of view, a project can be seen as an occasion for the
generation of new knowledge, whether it is product-related, process-related, or project-organizing
knowledge. By sharing knowledge explored in one project, this knowledge can be utilized and
exploited in later projects or in other parts of the organization. Knowledge reuse does not mean
knowledge copying; instead, knowledge is translated and reconstructed by the project team mem-
bers involved. In organizations where projects are performed in different departments or divisions,
there is a risk of partiality toward inventing ones own solutions and hence exploring instead of
exploiting corporate experiences. Not to take advantage of any earlier project experiences within
the organization is in fact learning myopia. In addition, a project learning culture is not solely
for the current project but also for future ones.
What is proposed in this chapter is an alternative approach to project management, that is, to
focus not only on the project per se, such as the new product produced or service rendered, but also
on the learning possibilities that a particular project provides. In short, to take a dual approach to
project management, such an approach bringing forth both the traditional view and the knowledge
management view regarding project management.
Adopting the contemporary knowledge perspective of projects, project managers, team mem-
bers and senior management should have a better understanding of the importance of enhancing
learning and knowledge capability within and across projects, rather than just focusing on the cur-
rent narrow view of project performance. In this way, projects will not become islands but rather
learning laboratories/experiments and repositories of knowledge and experience where both
the short term and the long term are taken into account. Moreover, project teams can enhance their
performance through communities of practice or social networks established over projects.
222 FONG
FUTURE DIRECTIONS
This chapter has provided a basis for understanding knowledge transfer within and between proj-
ects and between a project and the permanent organization; on this basis, further research can
be developed. One example of the latter would be to examine organizations where all or some
operations are organized in project form, in order to study enablers and barriers for learning and
knowledge transfer within a multiproject environment.
Another interesting area would be to study in more depth the role of project team members
as carriers of knowledge within projects. To penetrate deeper into the theoretical feld of strategy
or to further investigate different strategies for managing projects within an organization so as
to facilitate knowledge transfer between projects would also be fruitful. Further studies into the
role of information and communication technologies in transferring knowledge between projects
could make a valuable contribution. Finally, it would be worthwhile to study knowledge on a more
operational level and perhaps even measure it.
CONCLUSIONS
Managing knowledge gained from projects and learning from them are important assets for
project organizations. An organization needs to exploit its knowledge in order to achieve short-
term benefts, such as reducing overheads and enhancing operations. However, frms also need to
explore new knowledge or combine existing knowledge in order to fnd ways of achieving more
sustainable growth.
This chapter shows that project organizations face many diffculties in recognizing and ensuring
learning and knowledge managing within and between projects. Projects do not share knowledge
and experiences with other projects naturally and routinely. Some of the impediments to the learning
and knowledge aspects in project organizations include the diversity of knowledge and experience
in project personnel, which makes overcoming professional or knowledge boundaries more dif-
fcult as each member is so used to his or her own mental models and neglects the felds beyond
his or her expertise. From the team dynamic perspective, the diverse experiences of project team
members enable exposure to other practices or experience sets, which will create opportunities to
learn and share knowledge with other team members.
The lack of project memory or infrastructure (including search and retrieval functions) in some
project organizations infuences the collection of essential project knowledge and experience.
Without such an infrastructure, the repository will end up not being used, individuals will not
know where to fnd the knowledge, and there will be no mechanism for distributing knowledge to
new projects. In addition, there may be an overreliance on the transactive memory of knowing
who within project organizations affects individuals connection to the right people who have the
right experience. Many organizations are not fully aware of the knowledge they possess (Daven-
port and Prusak, 1998) and they cannot fully utilize their expertise as they either have only a few
people who realize that particular types of expertise exist or only a few who know how to locate
them (Moreland and Argote, 2003). Hansen (1999, 2002) fnds that a limited social network in the
receiver project (i.e., a lack of connections with past or concurrent projects) reduces the receiver
projects capacity for identifying sources of relevant knowledge and expertise. In addition, a greater
awareness of who knows what via, for example, expertise locator systems (Becerra-Fernandez,
2006) will greatly improve how project teams are staffed, as people with relevant knowledge can
be assigned to projects that require such expertise.
Time constraints in projects also undermine project learning, as short-term goals that focus
solely on remedying or overcoming immediate problems or fnishing tasks often prevent project
team members from suggesting or making changes to project routines. This fnding is similar to
those of Tucker, Edmondson, and Spear (2002), who carried out observations of hospital nurses.
This emphasis on short-term effciency for projects makes it hard to shift the focus from action to
refection, which is needed in learning (Ayas and Zeniuk, 2001).
Executing a project from a knowledge management point of view means not only focusing on
the project per se, such as the end product produced or service rendered, but also on the learning
possibilities that a particular project can provide. It is recommended that a dual approach to project
management be adopted. Such an approach has a focus on both the resources perspective and the
learning and knowledge perspective regarding project management. This approach signifes the
importance of having a strategic perspective on project organizing, where both the short-term goal
of meeting the project objectives and the long-term goal of learning within and between projects
are taken into consideration. This will enable us to learn from successful as well as failed les-
sons and to reuse the lessons in the same or other projects. This will create an awareness of the
knowledge synergies and the productive use of a frms resources.
REFERENCES
Althoff, K.-D.; Bomarius, F.; and Tautz, C. 2000. Knowledge management for building learning software
organizations. Information Systems Frontiers, 2,3, 349367.
Antoni, M. 2000. Inter-Project LearningA Quality Perspective. Linkping Studies in Science and Technol-
ogy, Thesis No. 839, Linkping, Sweden: Linkping University.
APM Special Interest Group Forum: Knowledge Management. 2003. Association for Project Management,
Hong Kong Branch.
APQC. 2005. Integrating Knowledge Management and Organizational Learning. Consortium Benchmarking
Study: Best-Practice Report. Houston, TX: APQC.
Argyris, C., and Schn, D.A. 1978. Organizational Learning: A Theory of Action Perspective. Reading,
MA: Addison-Wesley.
Artto, K.A. 2001. Management of project-oriented organization: Conceptual analysis. In K.A. Artto, M.
Martinsuo, and T. Aalto (eds.), Project Portfolio Management. Strategic Management through Projects.
Helsinki, Finland: Project Management Association, pp. 522.
Ayas, K., and Zeniuk, N. 2001. Project-based learning: Building communities of refective practitioners.
Management Learning, 32, 1, 6176.
Barker, M., and Neailey, K. 1999. From individual learning to project team learning and innovation: A
structural approach. Journal of Workplace Learning, 11, 2, 6067.
Bartezzaghi, E.; Coros, M.; and Verganti, R. 1997. Continuous improvement and inter-project learning in
new product development. International Journal of Technology Management, 14, 1, 116138.
Bateson, G. 1972. Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution,
and Epistemology. San Francisco: Chandler Publishing.
Becerra-Fernandez, I. 2000. The role of artifcial intelligence technologies in the implementation of people-
fnder knowledge management systems. Knowledge Based Systems, 13, 5, 315320.
. 2006. Searching for experts on the web: A review of contemporary expertise locator systems. ACM
Transactions on Internet Technology, 6, 4, 333355.
tions, and Technologies. Upper Saddle River, NJ: Pearson/Prentice-Hall.
Bowen, H.K.; Clark, K.B.; Holloway, C.A.; and Wheelwright, S.C. 1994. Development projects: The engine
of renewal. Harvard Business Review, SeptemberOctober, 110120.
Butler, A.G. 1973. Project management: A study in organizational confict. Academy of Management Journal,
16, 1, 84101.
Cooper, K. 1994. The $2000 hour: How managers infuence project performance through the rework cycle.
Project Management Journal, 25, 1, 1124.
Crawford, L. 1999. PM competence: People and organizations. In K.A. Artto, K. Khknen, and K. Koskinen (eds.),
Managing Business by Projects (vol. 2). Helsinki, Finland: Project Management Association, pp. 672689.
224 FONG
. 2000. Profling the competent project manager. In Proceedings of PMI Research Conference. Sylva,
NC: Project Management Institute, pp. 315.
Cusumano, M.A., and Nobeoka, K. 1998. Thinking beyond Lean: How Multi-Project Management Is Trans-
forming Product Development at Toyota and Other Companies. New York: Free Press.
Cyert, R., and March, J. 1963. A Behavioral Theory of the Firm. Englewood Cliffs, NJ: Prentice-Hall.
Daft, R.L., and Weick, K.E. 1984. Towards a model of organizations as interpretation systems. Academy of
Management Review, 5, 2, 281290.
Davenport, T.H., and Prusak, L. 1998. Working Knowledge: How Organizations Manage What They Know.
Boston, MA: Harvard Business School Press.
DeFillippi, R.J., and Arthur, M.B. 1998. Paradox in project-based enterprise: The case of flm making. Cali-
Deming, W.E. 1986. Out of the Crisis. Cambridge, MA: Massachusetts Institute of Technology, Center for
Advanced Engineering Study.
Denning, S. 2001. The Springboard: How Storytelling Ignites Action in Knowledge-Era Organizations.
Boston, MA: Butterworth-Heinemann.
Disterer, G. 2002. Management of project knowledge and experiences. Journal of Knowledge Management,
6, 5, 512520.
Dixon, N.M. 1999. The changing face of knowledge. The Learning Organization, 6, 5, 212216.
. 2000. Common Knowledge: How Companies Thrive by Sharing What They Know. Boston, MA:
Harvard Business School Press.
Dodgson, M. 1993. Organizational learning: A review of some literatures. Organization Studies, 14, 3,
375394.
Dougherty, D. 1996. Organizing for innovation. In S.R. Clegg, C. Hardy, and W.R. Nord (eds.), Handbook
of Organization Studies. London: Sage, pp. 424439.
Drucker, P.F. 1993. Post-Capitalist Society. Oxford: Butterworth-Heinemann.
Duwairi, R.M.; Fiddian, N.J.; and Gray, W.A. 1996. Schema integration meta-knowledge classifcation and
reuse. In Proceedings of the 14th British National Conference on Databases: Advances in Databases.
London: Springer-Verlag, pp. 117.
Engwall, M. 1998. The project concept(s): On the unit of analysis in the study of project management. In
R.A. Lundin and C. Midler (eds.), Projects as Arenas for Renewal and Learning Processes. Boston:
Kluwer, pp. 2536.
Eskerod, P. 1996. Meaning and action in a multi-project environment. International Journal of Project
Management, 14, 2, 6165.
Fiol, C.M., and Lyles, M.A. 1985. Organizational learning. Academy of Management Review, 10, 4,
803813.
Fong, P.S.W. 2003. Knowledge creation in multidisciplinary project teams: An empirical study of the
processes and their dynamic interrelationships. International Journal of Project Management, 21,
7, 479486.
Gaddis, P.O. 1959. The project manager. Harvard Business Review, 37, 3, 8997.
Gnyawali, D.R., and Stewart, A.C. 2003. A contingency perspective on organizational learning: Integrating
environmental context, organizational learning processes and types of learning. Management Learning,
34, 1, 6389.
Gould, F.E. 2002. Managing the Construction Process: Estimating, Scheduling, and Project Control (2nd
ed). Columbus, OH: Prentice-Hall, Merrill College Publishing Company.
Gulliver F.R. 1987. Post-project appraisals pay. Harvard Business Review, 65, 2, 12832.
Hansen, M.T. 1999. The search-transfer problem: The role of weak ties in sharing knowledge across orga-
nizational subunits. Administrative Science Quarterly, 44, 1, 82111.
. 2002. Knowledge networks: Explaining effective knowledge sharing in multiunit companies. Or-
ganizational Science, 13, 3, 232248.
Hedberg, B. 1981. How organizations learn and unlearn. In P.C. Nystrom and W.H. Starbuck (eds.), Handbook
of Organizational Design: Volume 1, Adapting Organizations to their Environments. Oxford: Oxford
University Press, pp. 327.
Huber, G.P. 1991. Organizational learning: The contributing processes and the literatures. Organization
Science, 2, 1, 88115.
. 1999. Facilitating project team learning and contributions to organizational knowledge. Creativity
and Innovation Management, 8, 2, 7076.
Kamara, J.M.; Augenbroe, G.; Anumba, C.J.; and Carrillo, P.M. 2002. Knowledge management in the archi-
tecture, engineering and construction industry. Construction Innovation, 2, 1, 5367.
Kasvi, J.J.J.; Vartiainen, M.; and Hailikari, M. 2003. Managing knowledge and knowledge competencies in
projects and project organizations. International Journal of Project Management, 21, 8, 571582.
Keegan, A., and Turner, J.R. 2001. Quantity versus quality in project-based learning practices. Management
Learning, 32, 1, 7798.
Kim, D.H. 1993. The link between individual and organizational learning. Sloan Management Review, 35,
1, 3750.
Korppi-Tommola, S. 2003. Knowledge management in projects. TU-22.165 Seminar in Industrial Manage-
ment, Department of Industrial Engineering and Management, Helsinki University of Technology.
Koskinen, K.U.; Pihlanto, P.; and Vanharanta, H. 2003. Tacit knowledge acquisition and sharing in a project
work context. International Journal of Project Management, 21, 4, 281290.
Kotnour, T. 1999. A learning framework for project management. Project Management Journal, 30, 2,
3238.
. 2000. Organizational learning practices in the project management environment. International
Journal of Quality & Reliability Management, 17, 4/5, 393406.
Leonard-Barton, D. 1992. The factory as a learning laboratory. Sloan Management Review, 34, 1, 2338.
Levinthal, D.A., and March, J.G. 1993. The myopia of learning. Strategic Management Journal, 14 (Special
Issue), 95112.
Levitt, B., and March, J.G. 1988. Organizational learning. Annual Review of Sociology, 14, 319340.
Liebowitz, J. 2005. Conceptualizing and implementing knowledge management. In P.E.D. Love, P.S.W. Fong,
and Z. Irani (eds.), Management of Knowledge in Project Environments. Oxford: Elsevier/Butterworth-
Heinemann, pp. 118.
Love, P.E.D., Irani, Z.; and Edwards, D.J. 2003. Learning to reduce rework in projects: Analysis of frms
organizational learning and quality practices. Project Management Journal, 34, 3, 1325.
Lundin, R.A., and Midler, C. 1998. Emerging convergences or debates. In R.A. Lundin and C. Midler (eds.),
Projects as Arenas for Renewal and Learning Processes. Boston: Kluwer, pp. 231241.
Lundin, R.A., and Sderholm, A. 1995. A theory of the temporary organization. Scandinavian Journal of
Management, 11, 4, 437455.
Lundin, R.A., and Steinthrsson, R.S. 2003. Studying organizations as temporary. Scandinavian Journal of
Management, 19, 2, 233250.
March, J.G. 1991. Exploration and exploitation in organizational learning. Organization Science, 2, 1,
7187.
March, J.G., and Olsen, J.P. 1976. Ambiguity and Choice in Organizations. Bergen, Norway: Universitets-
forlaget.
McGill, M.E., and Slocum, J.W., Jr. 1993. Unlearning the organization. Organizational Dynamics, 22, 2,
6779.
Miller, D. 1996. Confgurations revisited. Strategic Management Journal, 17, 7, 505512.
Moreland, R.L., and Argote, L. 2003. Transactive memory in dynamic organizations. In R. Peterson and E.
Mannix (eds.), Understanding the Dynamic Organization. Mahwah, NJ: Erlbaum, pp. 135162.
Newell, S. 2004. Enhancing cross-project learning. Engineering Management Journal, 16, 1, 1220.
Nobeoka, K. 1995. Inter-project learning in new product development. In Academy of Management Best
Papers Proceedings. Briarcliff Manor, NY: Academy of Management, pp. 432436.
the Dynamics of Innovation. Oxford: Oxford University Press.
Packendorff, J. 1995. Inquiring into the temporary organization: New directions for project management
research. Scandinavian Journal of Management, 11, 4, 319333.
Partington, D. 1996. The project management of organizational change. International Journal of Project
Management, 14, 1, 1321.
Pinto, J.K. 1999. Managing information systems project: regaining control of a runaway train. In K.A. Artto,
K. Khknen, and K. Koskinen (eds.), Managing Business by Projects (vol. 1). Helsinki, Finland: Project
Management Association, pp. 3143.
. 2000. Understanding the role of politics in successful project management. International Journal
of Project Management, 18, 2, 8591.
Pinto, J.K., and Prescott, J.E. 1988. Variations in critical success factors over the stages in the project life
cycle. Journal of Management, 14, 1, 518.
226 FONG
PMI. 2000. A Guide to the Project Management Body of Knowledge (PMBOK guide). Newtown Square,
PA: Project Management Institute.
. 2003. Research program open working session-group discussion notes. Newtown Square, PA:
Project Management Institute.
Pohjola, M. 2003. Knowledge management in projects. TU-55.165 Seminar in Industrial Management,
Department of Industrial Engineering and Management, Helsinki University of Technology.
Prencipe, A., and Tell, F. 2001. Inter-project learning: Processes and outcomes of knowledge codifcation in
project-based frms. Research Policy, 30, 9, 13731394.
Sabbag, P.Y. 1999. The nature of projects: A tool for improving management. In 30th Annual Project Man-
agement Institute Seminars and Symposium. Philadelphia, PA: Project Management Institute.
Schindler, M., and Eppler, M.J. 2003. Harvesting project knowledge: A review of project learning methods
and success factors. International Journal of Project Management, 21, 3, 219228.
Senge, P.M. 1990. The Fifth Discipline: The Art and Practice of Learning Organization. New York: Double-
day.
Sense, A.J. 2003. A model of the politics of project leader learning. International Journal of Project Man-
agement, 21, 2, 107114.
Sense, A.J., and Antoni, M. 2003. Exploring the politics of project learning. International Journal of Project
Management, 21, 7, 487494.
Siemieniuch, C.E., and Sinclair, M.A. 1999. Organizational aspects of knowledge life cycle management in
manufacturing. International Journal of Human-Computer Studies, 51, 3, 517547.
Simon, A.S. 1991. Bounded rationality and organizational learning. Organization Science, 2, 1, 125134.
Smeds, R.; Olivari, P.; and Corso, M. 2001. Continuous learning in global product development: A cross-
cultural comparison. International Journal of Technology Management, 22, 4, 373392.
Smith, B., and Dodds, B. 1997. Developing Managers Through Project-based Learning. Aldershot, Hamp-
shire, UK: Gower.
Tucker, A.L.; Edmondson, A.C.; and Spear, S. 2002. When problem solving prevents organizational learning.
Journal of Organizational Change Management, 15, 2, 122137.
Turner, J., and Keegan, A. 1999. The management of operations in the project-based organization. In K.A.
Artto, K. Khknen, and K. Koskinen (eds.), Managing Business by Projects (vol. 1). Helsinki, Finland:
Project Management Association, pp. 5785.
Turner, J.R., and Mller, R. 2003. On the nature of the project as a temporary organization. International
Journal of Project Management, 21, 1, 18.
Wenger, E. 1998. Communities of Practice: Learning, Meaning, and Identity. Cambridge, UK: Cambridge
University Press.
. 2000. Communities of practice: The key to knowledge strategy. In E.L. Lesser, M.A. Fontaine, and
J.A. Slusher (eds.), Knowledge and Communities. Boston: Butterworth-Heinemann, pp. 320.
Wheelwright, S.C., and Clark, K.B. 1992. Revolutionizing Product Development: Quantum Leaps in Speed,
Effciency, and Quality. New York: Free Press.
Williams, T. 2003. Learning from projects. Journal of the Operational Research Society, 54, 443451.
PART IV
KNOWLEDGE MANAGEMENT AND TEAMS
WITHIN AND ACROSS ORGANIZATIONS
229
ChaptEr 12
MANAGING KNOWLEDGE IN VIRTUAL
COMMUNITIES WITHIN ORGANIZATIONS
narEn B. pEddiBhotla and mani r. sUBramani
Abstract: The intellectual capital of the frman intangible asset embedded in the knowledge and
expertise of a frms employeesis increasingly recognized as a source of sustainable competi-
tive advantage. The establishment of virtual communities is among the most common approaches
toward knowledge management adopted within frms. Such communities provide contexts where
individuals within frms can come together, bound by shared roles, bases of expertise, or passion
for specifc topics, and interact. Virtual communities are recognized as important contributors to
both the development of social networks among individuals but also individual performance and
frm performance. This chapter provides a synthesis of prior work on the topic and highlights the
key contributions of different streams of research on virtual communities. It also suggests a variety
of directions for future research.
Keywords: Knowledge Management, Virtual Community, Knowledge Creation, Knowledge Stor-
age, Knowledge Sharing and Transfer, Knowledge Application
INTRODUCTION
Intangible assets embedded in the knowledge and expertise of a frms employees that are created,
transferred, and adapted through patterns of interaction among individuals and groups within frms
are increasingly recognized as a source of sustainable competitive advantage. Firms have imple-
mented a variety of information technology (IT)enabled systems aimed at leveraging these assets
that are often broadly described as knowledge management systems (KMS). Technologies to create
and manage virtual communities are commonly components of knowledge management (KM)
initiatives and intended to facilitate collaboration and social interactions among employees.
Information systems supporting virtual communities provide a variety of featuresthey allow
participants to examine details of community members such as their personal profles, their exper-
tise, and follow links to documents submitted by members. Such systems also provide discussion
forums where individuals can post questions to members of the group or respond to questions or
comments of other members. Such IT-enabled virtual communities are often organized around topic
arease.g., Server Administration and Network Monitoring in a technology frm and Ocean Marine
Underwriting and Reinsurance in an insurance companyand provide a shared communal space
for individuals with similar roles, bases of expertise, or passion for specifc topics to come together,
share experiences, and help each other. Studies of such communities within organizations suggest
that they are important forums where individuals disseminate knowledge by sharing personal expe-
riences with other members and creating new knowledge through discussions about topics. Virtual
230 PEDDIBHOTLA AND SUBRAMANI
communities within frms are often important resources for members to request help and for problem
solving through interactions between community members. Instances of virtual communities that
are viewed as exemplars include ShareNet within Siemens that, in the year 2002, reportedly had
19,000 members located in over 73 countries and contained over 20,000 knowledge objects (reports,
proposals, drawings, etc.) contributed by community members (McCormack, 2002).
From one perspective, the creation of virtual communities represents the next step by organiza-
tions that have been providing e-mail and teleconferencing facilities for electronic communication
as well as creating distributed databases with information to be accessed by employees in the frm.
The variety of technologies underlying virtual communities is highlighted in Table 12.1. While
each of these technologies such as e-mail and document repositories have individually helped
organizations to overcome geographic and temporal barriers to coordination and sharing of in-
formation, the technologies underlying virtual communities integrate complementary technology
components to provide the facility for individuals to interact electronically, to create commonly
accessible content, and to self-organize and mobilize the resources of the group. The juxtaposition
of these complex capabilities in virtual communities represents a qualitative shift in the nature of
IT-enabled support that in turn creates several challenges for establishment and maintenance. First,
the success of virtual communities is linked to bottom-up efforts of individualsquite different
from the top-down processes linked to enabling access to databases and data manipulation tools.
Second, the roles of individuals as contributors and users of content and as creators and monitors
of social norms of system use in virtual communities are fuid in comparison to the structured
and clearly delineated roles (e.g. providers of specifc sets of inputs, users of specifc outputs,
database administrators ) with respect to traditional database systems. Third, participation in vir-
tual communities is often voluntary (although encouraged). Virtual communities thus represent
novel contexts of action that rely on mutual engagement by members, the development of shared
norms, and the perception of accountability to the joint enterprise among members to remain
vital and useful. The fragility of virtual communities is highlighted by reports suggesting that a
large proportion of KM initiatives involving virtual communities often fail to meet expectations
(Hahn and Subramani, 2001). Overall, virtual communities in organizations present a variety of
intriguing challenges. Surprisingly, given the importance of virtual communities to frms, prior
research on virtual communities has been fragmented and scattered.
The goal of this chapter is to provide a synthesis of prior research on virtual communities in
organizations, discuss the key fndings, and highlight avenues for future research. We focus our
attention in this paper on virtual communities within organizations. Virtual communities in public
spaces (such as Usenet groups, Internet blogs, etc.) and those that span multiple organizational
boundaries (such as supplier exchanges) are outside the scope of our review.
Table 12.1
Technologies Supporting Virtual Communities
Features Applications
Text-basedasynchronousmessaging
(e.g., 11, 1-many and many-many)
E-mail, e-mail list, Usenet, bulletin boards, Web-
based collaboration tools (e.g., Groove, Webex,
Lotus Notes), Web-based repositories, databases
accessed through the Internet
Text-basedsynchronousmessaging
(11, 1-many and many-many)
Text,videoposts(e.g.,onblogs)
Videoconferencing
MANAGING KNOWLEDGE IN VIRTUAL COMMUNITIES 231
BACKGROUND
Virtual communities within organizations comprise groups of individuals who draw on the resources
of the community to accomplish their tasks and also participate in interactions within the community.
Membership in these communities is generally open to all organizational members. The larger the
member base, the more is the knowledge potentially accessible by members of the community. While
participation through posting is open to all members, the legitimacy of participation often derives from
the organizational role or function of the individual (Lave and Wenger, 1991, p. 35). This is an important
feature of organizational communities as contextual features of frms such as hierarchical position and
organizational reputation are transferred into the community from the organizational context.
In compiling the body of prior work on virtual communities in organizations, we examined the
work published in major MIS journals (Information Systems Research, Journal of Management
Information Systems, MIS Quarterly) as well as in other management journals (Administrative
Science Quarterly, Academy of Management Journal, Organization Science). We also included
papers published after 1995 in proceedings of the International Conference on Information Systems
(ICIS) and the Americas Conference on Information Systems (AMCIS). In addition, we performed
a keyword search for papers on Business Source Premier for relevant work. Our search yielded a
total of 58 papers from a variety of sources (Table 12.2).
Table 12.2
List of Sources
S. No. Journal Number of papers
1 MIS Quarterly 9
2 Organization Science 6
3 AMCIS proceedings 4
4 ICIS proceedings 4
5 Information Systems Research 4
6 IEEE Transactions on Software Engineering 3
7 Journal of Management Information Systems 3
8 Decision Support Systems 2
9 Information Systems Journal 2
10 Journal of Organizational Computing and Electronic Commerce 2
11 Academy of Management Best Paper Proceedings 1
12 ACM Transactions on Information Systems 1
13 Administrative Science Quarterly 1
14 Behavior and Information Technology 1
15 British Journal of Management 1
16 Communication Research 1
17 Decision Sciences 1
18 Human Communication Research 1
19 IEEE Transactions on Engineering Management 1
20 IEEE Transactions on Professional Communication 1
21 Industrial and Corporate Change 1
22 Information and Management 1
23 Journal of Business Communication 1
24 Journal of Management Studies 1
25 Journal of Organizational Change Management 1
26 Journal of Strategic Information Systems 1
27 Management Communication Quarterly 1
28 Management Science 1
29 Production Planning and Control 1
Methodological Approaches in Prior Research
The methodological approaches used in prior work on virtual communities are presented in
Table 12.3.
As an emerging topic, it is not surprising to fnd a signifcant focus on qualitative approaches
case studies, ethnography, and action-research. For instance, Newell, Scarbrough, and Swan (2001)
draw on a case study of the KM initiatives of a large global bank to highlight the important role of
social factors such as organizational norms for information sharing and the shared understanding
among employees in infuencing knowledge sharing. In addition, studies have drawn on prior work
on electronic communication to highlight the role of factors infuencing helping behaviors within
organizations (Nidumolu, Subramani, and Aldrich, 2001; Massey, Montoya-Weiss, and ODriscoll,
2002; Sabherwal and Becerra-Fernandez, 2005). We see few studies attempting to draw precise
conclusions regarding interconstruct relationships and controlling for the role of extraneous fac-
tors through lab experiments. The work of Poston and Speier (2005) on the infuence of content
ratings and credibility indicators on knowledge search and evaluation in a document repository
represents one of the few instances of this approach.
Level of Analysis
The different levels of analysis used in prior research on virtual communities refect varying
perspectives on the phenomenon (see Table 12.4).
Research at the individual level has largely viewed virtual communities as forums for com-
munication and resource leverage. Such a view focuses on the help and resources obtained
through interactions among individuals in virtual communities. For instance, Constant, Sproull,
and Kiesler (1996), Jarvenpaa and Staples (2000), and Sussman and Siegal (2003) highlight the
factors infuencing knowledge sharing and the dynamics of interaction in virtual communities.
Adopting this view highlights a variety of interesting questions such as the determinants of the
usefulness of knowledge obtained through virtual communities, the relative hierarchical positions
of helpers, and the factors infuencing the willingness of individuals to assist unknown others. Prior
research has also addressed questions such as, Why do individuals supply knowledge to stock the
repositories? Why do people draw upon the stored knowledge available in a repository? (Fulk,
Table 12.3
Methodological Approaches to Examine Virtual Communities
S. No. Nature of data Method of data collection Number and percentage of papers*
1 Quantitative Experiment
Cross-sectional 5 (8.62%)
Longitudinal 1 (1.72%)
Survey 20 (34.48%)
Archival data 4 (6.90%)
2 Qualitative Case Study 36 (62.97%)
Ethnography 2 (3.45%)
Action-research-based ethnography 3 (5.17%)
Archival data 8 (13.79%)
3 Modeling Analytical 1 (1.72%)
*Percentages do not add up to 100% since some papers use multiple methods.
Heino, Flanagin, Monge, and Bar, 2004; Gray and Meister, 2004; Yuan, Fulk, Shumate, Monge,
Bryant, and Matsaganis, 2005) The theoretical lenses applied to address these questions include
cognitive psychology and communication theories.
The dominant perspective underlying work at the group level is of virtual communities as
collaborative systems. The work at the group level of analysis in Table 12.4 views virtual com-
munities as a means to pool expertise and knowledge to accomplish common goals. For example,
Chidambaram and Tung (2005), Miranda and Saunders (2003), and Yates and Orlikowski (2002)
focus on decision-making teams and knowledge sharing using collaboration software. Research
adopting this view of virtual communities highlights the applicability of group phenomena observed
in face-to-face contexts to virtual communities. For instance, researchers have focused on group
characteristics and examined their interactions with features of technology to infuence outcomes.
The primary lenses underlying this stream of research are theories from social psychology, social
cognition, and group decision making.
Research examining virtual communities at the organizational level of analysis views them
as organizational resources and part of the social infrastructure of the frm. Adopting this
Table 12.4
Level of Analysis
S. No. Level of analysis
Number and
percentage of papers Papers
1 Individual 24 (41.38%) Ackerman (1998), Barreto and Heckman (2004),
Bock et al. (2005), Constant et al. (1996), Faniel
and Majchrzak (2002), Firth (2004), Fulk et al.
(2004), Goodman and Darr (1998), Gray and
Meister (2004), Irmer et al. (2002), Jarvenpaa and
Staples (2000), Jarvenpaa and Staples (2001),
Kankanhalli et al. (2001), Kankanhalli et al. (2005),
Linger et al. (1999), Majchrzak et al. (2005),
Olivera (2000), Poston and Speier (2005), Staples
and Jarvenpaa (2000), Stenmark (2001), Sussman
and Siegal (2003), Teigland and Wasko (2003),
Thorn and Connolly (1987), Yuan et al. (2005)
2 Group 10 (17.24%) Chidambaram and Tung (2005), Favela (1997),
Kock and McQueen (1998), Miranda and Saun-
ders (2003), Orlikowski et al. (1995), Purvis et
al. (2001), Rafaeli and Ravid (2003), Weiser and
Morrison (1998), Yates and Orlikowski (2002),
Yates et al. (1999)
3 Organization 24 (41.38%) Banker and Kauffman (1991), Banker et al.
(1993), DAdderio (2003), Dube and Bourhuis
(2005), Garud and Kumaraswamy (2005),
Hansen and Haas (2001), Hsiao et al. (2003),
Ji and Salvendy (2002), Kwan and Balasubra-
manian (2003), Lee and Cole (2003), Lee and
Litecky (1997), Levina and Vaast (2005), Mann
et al. (1991), Massey et al. (2002), Morisio et al.
(2002), Newell et al. (2001), Purvis et al. (2000),
Ruppel and Harrington (2001), Sherif (1999),
Sherif and Vinze (2002), Skok and Kalmanovitch
(2004), Smeds and Alvesalo (2003), Stenmark
and Lindgren (2003), Vaast (2004)
perspective, studies have examined issues such as the support for software projects through
virtual communities through the availability of expert help and easy access to prewritten soft-
ware modules (Banker and Kauffman, 1991). Virtual communities are viewed as being internal
markets for knowledge that are ordered by the unseen hand of competition for attention among
suppliers and users of knowledge (Hansen and Haas, 2001). From this perspective, prior research
has highlighted the role of virtual communities as supporting the creation and maintenance of
boundary objects to ease the sharing of work practices among geographically dispersed units
(Levina and Vaast, 2005; Newell et al., 2001). Theoretical approaches used include practice
theory and structuration theory.
KNOWLEDGE PROCESSES IN VIRTUAL COMMUNITIES
Virtual communities in organizations are generally linked to KM initiatives to enable the social
collectives to serve as the context to nurture and support knowledge processes in the frm. We
focus on four key processes related to knowledge: creation, storage, transfer, and application of
knowledge and examine the role of virtual communities in each of these processes.
Virtual Communities as Forums for Knowledge Creation
Virtual communities are forums that, by enabling interaction among individuals, can help create
new knowledge (Nonaka, 1994). For instance, Siemenss ShareNet enables feld support engineers
in remote locations to post customer problems that they have been unable to resolve. These posts
set up cycles of interaction among domain experts and others who may have solved similar prob-
lems elsewhere that often lead to novel solutions that subsequently become part of the knowledge
base of the frm (McCormack, 2002). Recent work by Sabherwal and Becerra-Fernandez (2005)
suggests that virtual communities within organizations provide a fertile ground for each of the
four modes of knowledge creation outlined by Nonaka (1994): externalization, internalization,
combination, and socialization (see Table 12.5).
Externalization
Externalization involves the conversion of the tacit knowledge within an individual to a form that
is explicit and that can be stored in an artifact that is accessible to others in a virtual community.
However, it is well recognized that even if individuals are motivated to disclose such knowledge,
they often fnd it diffcult because they do not clearly know what they know. In order for a virtual
community to serve as an effective context where individuals can document knowledge that may
Table 12.5
Creation of Knowledge in Virtual Communities
Level of analysis Dependent variable (knowledge creation) Papers
Individual By externalization Stenmark (2001)
By internalization Gray and Meister (2004)
Group
Organization By socialization Lee and Cole (2003)
By combination Sherif (1999)
be useful to the community, this issue needs to be addressed. This is often an intractable problem
because this arises from cognitive limitations of individuals (Hahn and Subramani, 2001). Another
issue pertaining to knowledge creation is that individuals are unable to effectively disclose their
possession of professional expertise to others. If they were able to do so, this would enable new
knowledge creation as they could be targeted by other members with requests for help. Stenmarks
(2001) interpretive case study within the context of an agent-based recommender system suggests
that tacit knowledge in the form of ones professional interests can be communicated to others
through the trail of documents that one fnds interesting. Such knowledge on individual interests
is then used by the system to retrieve information from the knowledge repository of documents.
This is a case of virtual communities supported by a novel pull-technology to support knowledge
creation through the identifcation of individuals with expertise to potentially involve in problem
solving.
Internalization
Knowledge creation through internalization involves the conversion of explicit knowledge avail-
able in a virtual community to tacit knowledge within a person, that is, individual learning. This
is the inverse of the externalization process and is achieved by the access and use of knowledge
provided by others. Gray and Meister (2004) examine how knowledge that is made explicit in a
companys intranet and in printed documents is internalized by employees. An important insight
from their study is that such internalization is infuenced by the levels of knowledge seeking and
learning orientation of individuals.
Combination
Virtual communities also enable multiple individuals to pool their expertise to create new knowl-
edge. Knowledge creation by combination occurs in virtual communities when components of
explicit knowledge contributed by different individuals are put together, often in the course of
solving a problem. Such combination often poses signifcant challenges since the different pieces
of explicit knowledge often need to be classifed, sorted, and recontextualized for combination to
occur. Sherif (1999) conducted case studies in three software-developing organizations that had
adopted the Lotus Notes collaborative tool for its developer teams. A major insight from her study
is that even if people record their knowledge in such a system, it is not useful to others unless it is
synthesized and aggregated for wider applicability. Individuals would submit content about their
particular experience, but unless it was generalized or abstracted at higher level, others (especially
those in other teams) would not fnd it immediately relevant.
Socialization
Virtual communities support a variety of processes involved in socialization. Novices can observe
other knowledgeable individuals and learn by observation, imitation and practice (Nonaka, 1994)
and by legitimate peripheral participation (Wenger, 1988). The work of Lee and Cole (2003) on
the Linux community suggests that the mutual criticism and evaluation of software code written by
participants in the virtual communities provide a context for socialization and the creation of new
knowledge within individuals. Discussions of software code in which a few members participate,
and which all members can observe, lead to knowledge creation both at the individual level and
at the level of the community.
Virtual Communities as Contexts for Knowledge Storage
Virtual communities are natural contexts to store knowledge that can be accessed by all members of
the community. A major challenge is to design and implement a system that is suitable for the stor-
age and retrieval of knowledge. Appropriately structuring knowledge (e.g., by indexing documents,
creating FAQs, etc.) is necessary to enable the individuals to access and use knowledge. However,
this poses a signifcant challenge for designers since the appropriate structure is often a function of
the type of conceptualization of storage and retrieval needs and is an issue that has been examined
in prior research (see Table 12.6). One stream of research on the issue deals with specifc virtual
communities contexts, and offers designs of systems and evaluates them against the needs of the
context. Another stream focuses more on overall storage features in virtual communities and how
they can address broad requirements across many types of virtual community contexts.
The frst stream has followed the design science paradigm of building and evaluating knowledge
storage models and prototype systems based on those models (March and Smith, 1995). Such
models are usually based on a theoretical framework claimed to be relevant to knowledge storage
in the particular virtual community context. In this line of work, a common theme has been that
knowledge can best be captured and stored in a memory system when that system is designed to
be closely aligned with the tasks or business processes that individuals engage in as part of their
organizational roles. Ackerman (1998) argued that since organizations are goal-driven entities that
have limited resources, a system that is tied to an ongoing activity is likely to be more useful to
individuals. He presented a system called Answer Garden that focused on a specifc topic (e.g.,
X Window GUI system, Department courses) and showed how information can be retrieved
in question-and-answer sessions from the storage or obtained from experts. Unlike the Answer
Garden system, however, Linger, Burstein, Zaslavsky, and Crofts (1999) adopted a broader view
of tasks in terms of their pragmatic, process and conceptual layers. They developed a system
within the task context of an epidemiological research community in a research organization. Weiser
and Morrison (1998) focused on a project team context. Their prototype system named Project
Memory was aimed at a general representation of project contexts and focused on structured search
and retrieval of information. Favela (1997) also studied the general project context and developed
a model to aid knowledge retrieval based on guided exploration of information using an adaptive
resonance theory neural network. Finally, continuing the theme of task-system alignment, Kwan
and Balasubramanian (2003) proposed and tested the KnowledgeScope system that serves as
a store of knowledge for a workfow context and extends systems described in earlier studies by
providing contextual information about the stored knowledge.
In the second stream of research, scholars have examined the impact of knowledge storage
Table 12.6
Storage of Knowledge in Virtual Communities
Level of analysis Dependent variable Papers
Individual Effectiveness of general storage Olivera (2000)
Effectiveness of specifc system Ackerman (1998), Linger et al. (1999)
Group Effectiveness of specifc system Favela (1997), Weiser and Morrison (1998)
Organization Effectiveness of specifc system Kwan and Balasubramanian (2003)
Effectiveness of general storage Ji and Salvendy (2002), DAdderio (2003)
systems on the people using them. Unlike the other stream where the focus is on a specifc model
or prototype of a virtual community, these studies examine virtual communities at the overall level.
Olivera (2000) conceptualized knowledge storage in virtual communities as being comprised of
content, structure, and operating processes. His fndings from a qualitative study of both
IT-based and nonIT-based systems in a consulting frm show how features addressing the three
components respectively infuence individual perceptions about the effectiveness of the system.
Employees expected relevance and generality in the system content. Ease of retrieval was the key
factor with respect to the structure of knowledge storage. Finally, people preferred formal evalua-
tion of knowledge stored in a system, updated information, and person-to-person interaction with
respect to the operating processes of a system. The paper by Sherif (1999) also touches upon the
issue of organizing knowledge appropriately in a repository in order for it to be useful to knowledge
seekers. Ji and Salvendy (2002) focused on the effectiveness of two tools for retrieving knowledge
from an intranet-based system. Derived from two basic information-seeking strategies (analytical
and browsing), they show that their memory structure and history map tools improve an
individuals retrieval of knowledge from the system. Finally, DAdderio (2003) uses structuration
theory concepts to highlight the challenges of codifying knowledge in a system. They show how
knowledge often needs to be restructured when stored in computer-embedded representations,
specifcally in the Product Data Manager Software system at an automotive organization.
Virtual Communities as Contexts for Knowledge Sharing and Transfer
Virtual communities can play a crucial role in knowledge sharing and transfer by facilitating interaction
among dispersed individuals and connect individuals with problems to those who may help provide
a solution. This aspect has received considerable attention by researchers (see Table 12.7).
The most important issues regarding knowledge transfer in virtual communities are the com-
ponent processes of contribution by knowledge providers and acquisition by knowledge seekers.
As a result, prior work has focused mostly on these two aspects of knowledge transfer in virtual
communities at the individual, group, and organizational levels of analyses. Some of these papers
combine the two factors into a more general usage construct (Jarvenpaa and Staples, 2000; Staples
and Jarvenpaa, 2000; Yuan et al., 2005) or consider them together (Goodman and Darr, 1998).
Most studies, however, have focused on one of the two aspects. Apart from knowledge contribu-
tion and acquisition levels, other important aspects of knowledge transfer in virtual communities
include the quality of knowledge transferred, knowledge ownership, and knowledge search, but
the examination of these issues has been limited.
Researchers who have examined antecedents of contribution include those who focused on
contribution level (Fulk, Kankanhalli, Tan, and Wei, 2004; Heino, Flanagin, Monge, and Bar, 2004;
Thorn and Connolly, 1987; Yuan et al., 2005) and decision to contribute (Barreto and Heckman,
2004; Bock, Zmud, Kim, and Lee, 2005; Firth, 2004; Goodman and Darr, 1998; Irmer, Bordia,
and Abusah, 2002). A dominant theme in studies examining contribution level is that individuals
incur costs in making available their knowledge to others and that they derive benefts only from
the knowledge contributions of others. Grounded in public goods and collective action theory, the
key fnding of these studies is that individual costs from making ones own contribution and gains
to oneself from contributions of others are respectively negatively and positively associated with
ones contributions (Fulk et al., 2004; Thorn and Connolly, 1987; Yuan et al., 2005). On the other
hand, studies that have looked at decision to contribute have tended to rely on a framework based
on individual attitudes and organizational norms (Barreto and Heckman, 2004; Bock, Zmud, Kim,
and Lee, 2005; Irmer, Bordia, and Abusah, 2002). These attitudes usually pertain to the individual
knowledge contributors propensity to help others, confdence in ones expertise, personal beneft
from knowledge sharing, and attitude to ones job.
Those who have looked at acquisition of content have studied the antecedents of knowledge
adoption (Sussman and Siegal, 2003) and knowledge seeking (Faniel and Majchrzak, 2002; Gray
and Meister, 2004; Kankanhalli, Tan, and Wei, 2001). Characteristics of the content (quality,
tacitness, availability of meta-knowledge, generality), content contributor (expertise), content
recipient (expertise, involvement), technology (ease of use), and organizational factors (incentives,
task interdependence) have been found to be important factors. These studies have mostly framed
knowledge use in virtual communities within well-known models of technology use, namely, the
technology acceptance model, theory of planned behavior, and task-technology ft theory. Based
on an examination of the unique characteristics of knowledge seeking, scholars have also begun
enriching these frameworks with theories of information infuence (in Sussman and Siegal, 2003)
and adult education (in Gray and Meister, 2004). In their exploratory study, Faniel and Majchrzak
(2002) outline the importance of certain knowledge-contentrelated factors.
Prior research has also studied antecedents of three other aspects of knowledge transfer:
the quality of knowledge transferred (Constant et al., 1996), perception of knowledge owner-
ship (Jarvenpaa and Staples, 2001), and knowledge search (Poston and Speier, 2005) in virtual
Table 12.7
Sharing of Knowledge in Virtual Communities
Level of
analysis Dependent variable Papers
Individual Contribution level Staples and Jarvenpaa (2000), Yuan et al.
(2005), Fulk et al. (2004), Thorn and Connolly
(1987), Kankanhalli et al. (2005), Jarvenpaa
and Staples (2000),
Idea quality/usefulness Constant et al. (1996)
Contribution decision Goodman and Darr (1998), Firth (2004),
Barreto and Heckman (2004), Bock et al.
(2005), Irmer et al. (2002)
Perception of organization
ownership of knowledge
Jarvenpaa and Staples (2001)
Information/knowledge adoption Sussman and Siegal (2003)
Decision to adopt knowledge Kankanhalli et al. (2001)
Knowledge sourcing/seeking
acquisition
Gray and Meister (2004), Faniel and Majchrzak
(2002)
Knowledge search and evaluation Poston and Speier (2005)
Group Overall knowledge sharing Orlikowski et al. (1995), Yates and Orlikowski
(2002), and Yates et al. (1999)
Idea quantity, breadth/depth of
knowledge sharing, contribution level
Chidambaram and Tung (2005), Miranda and
Saunders (2003), Kock and McQueen (1998),
Knowledge adoption Kock and McQueen (1998), Purvis et al. (2001)
Organization Contribution level Purvis et al. (2000), Levina and Vaast (2005)
Attention to ideas Hansen and Haas (2001)
Ease of knowledge transfer Hsiao et al. (2003)
Software reuse Lee and Litecky (1997), Banker et al. (1993)
communities. These studies, although few and far between, extend the traditional focus on
overall contribution and acquisition of knowledge and highlight factors that in turn infuence
knowledge transfer.
Group-level aspects of knowledge transfer that scholars have studied are overall knowledge
sharing (Orlikowski, Yates, Okamura, and Fujimoto, 1995; Yates and Orlikowski, 2002; Yates,
Orlikowski, and Okamura, 1999); level of knowledge contribution (Chidambaram and Tung,
2005; Kock and McQueen, 1998; Miranda and Saunders, 2003), and knowledge adoption (Kock
and McQueen, 1998; Purvis, Sambamurthy, and Zmud, 2001). Unlike research that has examined
individual-level knowledge contribution and acquisition on a variety of topics among a diffuse set
of people, these studies have been within the context of well-defned groups that have to accomplish
specifc tasks. Orlikowski et al. (1995), Yates and Orlikowski (2002), and Yates et al. (1999) draw
upon the genre-based theory of communication to show how members of virtual communities
share knowledge. Their arguments suggest that members either explicitly or implicitly shape the
genres used within in the group to interact with each other using collaborative tools such as Lotus
Notes, e-mail, and Usenet postings. Anonymity of individual contributors and the asynchronous
nature of electronic interaction also infuence knowledge contribution in groups (Chidambaram
and Tung, 2005; Kock and McQueen, 1998; Miranda and Saunders, 2003). The paper by Purvis,
Sambamurthy, and Zmud (2001) examines the special case of adoption of knowledge embedded
in CASE tool databases and draws upon the broader technology assimilation literature. Thus, re-
search within a group-level context, although sparse, has been focused on specifc contexts such
as decision making and software technology reuse.
At the organization level, prior research has examined the following factors refecting knowledge
transfer: Contribution level (Levina and Vaast, 2005; Purvis et al., 2000), software use (Banker,
Kauffman, and Zweig, 1993; Lee and Litecky, 1997), ease of knowledge transfer (Hsiao, Tsai,
and Lee, 2003), and attention to ideas (Hansen and Haas, 2001). Purvis et al. (2000) conceptual-
ize software and software objects embedded in CASE tool repositories as particular examples of
knowledge and draw upon the theory of technology assimilation to identify factors (e.g., manage-
ment support, methodology compatibility), enabling higher levels of knowledge contribution to
such repositories. Levina and Vaast (2005), on the other hand, focus more on work practices of
knowledge workers compared to the technology used for knowledge sharing. They study the broad
usage of IT artifacts for knowledge sharing across boundaries of communities by drawing upon
and extending the theory of boundary spanning. Their key insight is the need to support a com-
mon or joint community across existing communities by people who take on the role of boundary
spanners in practice. Hsiao et al. (2003) focus on the characteristics of knowledge in their context
of a chip manufacturing plant and how they affect knowledge transfer among engineers. They
identify the complexity of knowledge, the nature of collaboration among the knowledge workers,
and dynamics of the work context as three dimensions that infuence the ease of knowledge transfer
using a repository. Hansen and Haas (2001) adopt a novel knowledge-market view of document
supply and demand in a knowledge repository in a consulting frm. In contrast to the demand-side
view of the study by Poston and Speier (2005), their study suggests supplier-side tactics that draw
higher attention to documents that people contribute to the repository. Both the organization-level
studies on acquisition of knowledge in virtual communities (Banker et al., 1993; Lee and Litecky,
1997) deal with software reuse. They identify organizational, technical, and software-architectural
factors as antecedents of the rate or level of software reuse.
In all, research to date has covered extensive ground on three broad aspects of knowledge
transfer in virtual communities: the provider (e.g., contribution level), seeker (e.g., acquisition),
and the transfer process (e.g., search).
Application of Knowledge Gained from Virtual Communities
The knowledge that is created, stored, and transferred in virtual communities creates value when it
is applied to a specifc task depending on the particular context. The gains could be to an individual
personally or collectively to a group or organization. Research on KM in virtual communities has
examined various outcomes of knowledge application (see Table 12.8).
At the individual level, scholars have examined outcomes such as development of collabora-
tion know-how (Majchrzak, Malhotra, and John, 2005), decision quality (Poston and Speier,
2005), and creativity (Teigland and Wasko, 2003). Majchrzak et al. (2005) show that providing
context information in the IT systems used by distributed teams can lead to either better or poor
collaboration know-how development depending on whether the task is nonroutine or routine,
respectively. Poston and Speier (2005) focus on evaluation of content in a repository by individu-
als based on content ratings and their own experience with it. Their study shows that those who
anchor on low-quality content but later adjust away end up making higher quality decisions based
on the content than those who do not adjust away. In their study of IT consultants, Teigland and
Wasko (2003) fnd that individual creativity improves with increased level of knowledge access
from noncollocated individuals.
Two papers have examined group-level outcomes of knowledge application in virtual com-
munities: Chidambaram and Tung (2005) and Miranda and Saunders (2003). In their study on
decision making within a simulated managerial context, Chidambaram and Tung (2005) found
that decision quality decreased as the size of the group increased owing to increased complexity in
integrating inputs from more individuals and also because of social loafng. In a similar decision-
making context, Miranda and Saunders (2003) found that the breadth and depth of information
sharing via electronic media led to better decision quality.
At the organization level, two papers that examined knowledge application deal with software
development: Banker and Kauffman (1991) and Lee and Cole (2003). Application of knowledge
embedded in software objects in a CASE repository and their reuse in other projects was found to
improve software development productivity (Banker and Kauffman, 1991). Lee and Coles (2003)
study of the development of the Linux operating system highlights the important role of criticism
and the selection from different variants of software code modules in the innovation process. In this
exemplar of the community-based model of new product development, knowledge possessed by
individual software developers is applied to code written by others through a peer-review process.
Table 12.8
Application of Knowledge Gained from Virtual Communities
Individual Collaboration know-how Majchrzak, Malhotra and John (2005)
Decision quality/speed Poston and Speier (2005)
Individual creativity Teigland and Wasko (2003)
Group Decision quality Chidambaram and Tung (2005), Miranda
and Saunders (2003)
Organization Coordination Smeds and Alvesalo (2003)
Software productivity Banker and Kauffman (1991)
Innovation Lee and Cole (2003)
Boundary spanning Levina and Vaast (2005)
Community development Vaast (2004)
A study by Smeds and Alvesalo (2003) on coordination of new product development between
two sites of a company focused on the facilitation of a joint community of practice. The use of a
process simulation via IT artifacts enabled the two sites to better integrate their knowledge in the
development of a global process design. Studies by Levina and Vaast (2005) and Vaast (2004) in
similar contexts of cross-community knowledge application also highlight the importance of the
development of a joint community or network of practice.
Outcomes Linked to Virtual Communities
Knowledge management using virtual communities can have broader outcomes than those dis-
cussed earlier. Unlike the immediate objectives discussed in the previous section, these outcomes
are emergent and are an indirect result of applying knowledge (see Table 12.9).
The only individual-level study so far that examines KM success via virtual communities is
by Teigland and Wasko (2003). They fnd that the level of knowledge sharing (contribution and
acquisition) by individual employees via electronic discussion networks in a consulting frm
enhanced their individual performance. At the group level, the lone study by Rafaeli and Ravid
(2003) examines knowledge sharing by business executives in a simulated supply chain. The key
fnding is that knowledge sharing via e-mail enhanced the performance of the group.
Most studies of broader outcomes as a result of using virtual communities have been at the
level of the organization. Of these, Mann, Rudman, Jenckes, and McNurlin (1991), Morisio,
Ezran, and Tully (2002), Sherif and Vinze (2002), and Ruppel and Harrington (2001) adopt a
project-implementation view of virtual communities. These papers indicate that in order for an
organization to introduce a new KM project, factors such as management commitment, resource
availability, modifcation of work practices, marketing of the initiative, training, and facilitation
of appropriate organization culture are important. In contrast to this top-managers view of goal-
directed KM implementation, other scholars have taken a view of KM as being situated within
individual work practices. Skok and Kalmanovitch (2005) identify three views of knowledge among
employeescognitive, connectionist, and autopoeticand discuss their differential impacts on
implementation of an intranet for KM. Stenmark and Lindgren (2003), Massey, Montoya-Weiss,
and ODriscoll (2002), and Dube and Bourhuis (2005) stress the importance of alignment between
the work activities and knowledge that a KM is designed to support in order for the effort to be
successful.
Scholars have also focused on the sustenance of KM efforts involving virtual communities.
Stenmark and Lindgren (2003) highlight the need to develop a critical mass of individuals who
Table 12.9
Outcomes of Knowledge Management Using Virtual Communities
Individual Performance Teigland and Wasko (2003)
Group Group performance Rafaeli and Ravid (2003)
Organization KM system implementation
success/effectiveness
Mann et al. (1991), Stenmark and Lindgren
(2003), Newell et al. (2001), Ruppel and Har-
rington (2001), Dube and Bourhuis (2005),
Massey et al. (2002), Skok and Kalmanovitch
(2004), Garud and Kumaraswamy (2005), Sherif
and Vinze (2002), Morisio et al. (2002)
use the virtual communities, whose use of the system in turn sustains the KM effort. Garud and
Kumaraswamy (2005) point out that initial management support is not enough to sustain virtual
communities. Conscious effort is also required to manage unintended consequences of these initia-
tives on a continuous basis. Newell et al. (2001) discuss the need to manage tensions among dif-
ferent aspects of intranet usage for KMthe infrastructure, infostructure, and infoculture.
DIRECTIONS FOR FUTURE RESEARCH
In general, prior research has viewed virtual communities as arenas for participation by individu-
als. What is missing is recognition of the infuence of interpersonal networks on participation. For
instance, the case of ShareNet in Siemens suggests that as the system was rolled out, individuals
from different geographies, departments, and functional groups joined the virtual community.
Clearly, group-level variables, such as the norms of cooperation and sharing among members of
the engineering groups in India who joined ShareNet at the same time, are likely to have infuenced
the dynamics of participation. There has been little attention paid to the infuence of group level
and this represents a very fruitful direction for future work.
Further research on the lines of the work by Sabherwal and Becerra-Fernandez (2005) can also
not only examine the four processes of knowledge creation in virtual communities but also provide
a fner-grained view of the nature of knowledge bring shared. More research is also needed to probe
the nature of the internalization process that leads to individual learning in virtual communities.
Applying complementary perspectives in examining interactions in virtual communities can also
be useful. For instance, viewed from the perspective of behaviorist theory (Wenger 1998), learning
occurs through stimulus-response pairs and selective reinforcement. Knowledge is imparted by
means of appropriate stimuli and training of an individual by reinforcing one response behavior
as opposed to other behaviors. This view underlies the work of Poston and Speier (2005), where
they examine the relationships among ratings of content in a repository, indicators of credibility
of the raters, and search for content by seekers of knowledge. Such constructivist theories of
learning emphasize the building of an individuals mental structures vis--vis different aspects of
the environment. From this view, individuals are seen as gaining knowledge when they actively
engage in tasks through self-discovery. Based on the paper by Constant et al. (1996), it is likely
that the providers of information might themselves be learning by applying their minds to techni-
cal customer-support problems faced and posed by others. By trying out different approaches and
stepping through failures, they could potentially gain knowledge. A similar learning is potentially
experienced by contributors of ideas to new product development projects such as the ones studied
by Smeds and Alvesalo (2003) and Massey et al. (2002).
Social learning theory focuses on individual learning occurring through imitation. Observation
of others is thus a critical part of the process. The contribution of visual inputs, such as through
videoconferencing in communities involving repeated interaction (e.g., online discussion groups),
is an interesting issue that can be studied. Learning can also occur when individuals participate in
an activity according to the community-of-practice view (Wenger 1998). In a context of a reposi-
tory where content ratings provide important signals of quality, social learning theory suggests a
focus on the behavior of the content raters. From this perspective, it can be interesting to examine
how individuals develop an understanding of standards by examining the ratings provided to
content by others.
We suggest that researchers can broaden their focus and build theory by recognizing the
presence of multiple modes of learning. While some types of learning may be more important
than others, each may be relevant for different people (e.g., contributors versus users of knowl-
edge). An explicit focus on the specifc types of learning and the types of knowledge shared in
virtual communities can lead to a better understanding of the KM processes and highlight how
information technologies can facilitate such processes. For instance, Griffth, Sawyer, and Neale
(2003), who consider knowledge to be of two types (individual versus social), and Sabherwal and
Becerra-Fernandez (2005) view knowledge shared in communities as being of three types: context
specifc, technology specifc, and context-and-technology specifc. Building on these approaches,
examining the dynamics and outcomes of the process linked to different types of knowledge and
how different models of learning are involved can offer useful insights on knowledge fows in
virtual communities.
The limited research so far on knowledge storage in virtual community contexts calls for pio-
neering work to address this important aspect of the KM process. We need a greater understand-
ing of how knowledge artifacts are best organized in virtual communities for access by others to
encourage further knowledge creation, sharing, and application.
Potentially fruitful ways to extend prior work are (1) building and systematically testing
virtual communities to examine the effects of different features and (2) assessing the impact of
storage systems on individuals and organizations. To this end, the design perspective presented by
Wijnhoven (1998) and the theory of knowledge reuse by Markus (2001) provide useful guidance.
Wijnhoven (1998) suggests a framework for designing knowledge storage with the twin goals of
effectiveness and effciency in mind. He defnes storage effectiveness as the match between avail-
able organizational memory contents with organizational needs and effciency based on optimal
memory-handling means (p. 31). Within his framework, two important aspects of storage are
content and means. Content comprises know-how (factual knowledge), know-why (con-
ceptual understanding), meta-memory (information about the content), and memory-informa-
tion (information about retrieving content). Means consist of process and media. Wijnhoven
presents techniques for memory content analysis and means analysis to build a knowledge storage
system. Complementary to his framework is Markuss (2001) theory of knowledge reuse. Based on
the source and recipient of knowledge, she identifes four types of knowledge reuse: shared work
producers, shared work practitioners, expertise-seeking novices, and secondary knowledge
miners. She points out that knowledge storage designers need to keep in mind the different types of
knowledge users with respect to how they are related to the knowledge producers: the same people,
similar others, or dissimilar others. To achieve successful knowledge reuse, Markus suggests that
designers also consider the costs and incentives to producers, and use of appropriate human and
technical intermediaries who bridge the distance between producers and users.
With the exception of the work by Jarvenpaa and Staples (2001), there has been little attention
paid to the role of contextual variables involved in knowledge transfer in virtual communities. It is
widely recognized that knowledge sharing depends not only on the source and recipient involved
in the transfer but also on the relationship between the two and the nature of the knowledge itself
(Argote, McEvily, and Reagans, 2003; Markus, 2001). Examining the factors facilitating and
hindering knowledge transfer in virtual communities would be useful. Prior research has also
overlooked potential variations of factors within populations, such as individuals who are not
driven by cost-beneft calculations as postulated in public goods theory versus individuals who
also consider benefts from their own contributions to themselves.
Knowledge Application and KM Success
Building on work that suggests the utility of virtual communities to frms, future research might
consider contingent effects of virtual communities, such as what features of the context, nature of
use, dynamics of participation, or profle of interpersonal relationships might bring about nega-
tive outcomes rather than positive outcomes. Prior research in organizational memory has shown,
for example, that reliance on outdated knowledge may be detrimental to performance (Alavi and
Leidner, 2001) and that KM technologies can lead to the reinforcement of existing functional and
national boundaries within frms rather than to greater organization-wide information fows (Newell
et al., 2001). Scholars can provide a more balanced view by exploring both pros and cons of KM
in virtual communities. Longitudinal studies examining how communities evolve and factors that
infuence different potential trajectories are needed.
CONCLUSION
In this paper, we have presented our view of how research to date on KM in virtual community
contexts within organizations has evolved. We organized the literature by drawing upon the tra-
ditional view of a KM process as comprising knowledge creation, storage, transfer, and applica-
tion. We also identifed the broad dependent variables that scholars have examined, their research
methods, and unit of analysis.
We noted the rich stream of work on the knowledge transfer and application aspects of KM in
virtual community contexts and relatively sparse research on knowledge creation and storage. The
gaps in prior work point to obvious avenues for future research. Based on this review and our own
understanding of the possibilities of the feld, we have made suggestions for future research in the areas
of knowledge creation (explicating in depth the nature of knowledge creation, e.g., by internalization),
knowledge storage (adoption of a unifed framework for building and evaluating knowledge storage
systems), knowledge transfer (recognition of the three dimensions of knowledge transfer context;
examination of subpopulations of virtual communities participants), and knowledge application/suc-
cess (contingent effects of KM via virtual communities; evolution of virtual communities).
REFERENCES
Ackerman, M.S. 1998. Augmenting organizational memory: A feld study of Answer Garden. ACM Transac-
tions on Information Systems, 16, 3, 203224.
Argote, L.; Beckman, S.; and Epple, D. 1990. The persistence and transfer of learning in industrial settings.
Management Science, 36, 17501763.
Argote, L.; McEvily, B.; and Reagans, R. 2003. Managing knowledge in organizations: An integrative
framework and review of emerging themes. Management Science, 49, 4, 571582.
Banker, R.D., and Kauffman, R.J. 1991. Reuse and productivity in integrated computer-aided software
engineering: An empirical study. MIS Quarterly, 15, 3, 375401.
Banker, R.D.; Kauffman, R.J.; and Zweig, D. 1993. Repository evaluation of software reuse. IEEE Transac-
tions on Software Engineering, 19, 4, 379389.
Barreto, C., and Heckman, R. 2004, Understanding knowledge sharing motivators within knowledge man-
agement initiatives. In Christine Bullen and Edward Stohr (eds.), Proceedings of the Tenth Americas
Conference on Information Systems. Association for Information Systems. Available at www.aisnet.org,
pp. 22732283.
Bock, G.; Zmud, R.W.; Kim, Y.; and Lee, J. 2005. Behavioral intention formation in knowledge sharing:
Examining the roles of extrinsic motivators, social psychological forces, and organizational climate. MIS
Quarterly, 29, 1, 87111.
Chidambaram, L., and Tung, L.L. 2005. Is out of sight, out of mind? An empirical study of social loafng in
technology-supported groups. Information Systems Research, 16, 2, 149168.
DAdderio, L. 2003. Confguring software, reconfguring memories: The infuence of integrated systems on
the reproduction of knowledge and routines. Industrial and Corporate Change, 12, 2, 321350.
Darr, E.D.; Argote, L; and Epple, D. 1995. The acquisition, transfer and depreciation of knowledge in service
organizations: Productivity in franchises. Management Science, 41, 11, 17501762.
Davenport, T.H., and Prusak, L. 2000. Working Knowledge: How Organizations Manage What They Know.
Boston, MA: Harvard Business School Press.
Dube, L.; Bourhis, A.; and Jacob, R. 2005. The impact of structuring characteristics on the launching of virtual
communities of practice. Journal of Organizational Change Management, 18, 2, 145166.
Faniel, I. M., and Majchrzak, A. 2002. An exploratory study of the factors associated with successful reuse
of others knowledge. In Rajiv D. Banker, Hsihui Chang, and Yi-Ching Kao (eds.), Proceedings of the
Eight Americas Conference on Information Systems. Association for Information Systems. Available at
www.aisnet.org, pp. 20492058.
Favela, J. 1997. Capture and dissemination of specialized knowledge in network organizations. Journal of
Organizational Computing and Electronic Commerce, 7, 2 & 3, 201226.
Firth, D.R. 2004. Is free information really free? Information supply into an IT-based organizational memory
system. In Ritu Agarwal, Laurie Kirsch and Janice DeGross (eds.), Proceedings of the Twenty-Fifth
International Conference on Information Systems. Association for Information Systems. Available at
www.aisnet.org, pp. 583595.
Fulk, J.; Heino, R.; Flanagin, A.J.; Monge, P.R.; and Bar, F. 2004. A test of the individual action model for
organizational information commons. Organization Science, 15, 5, 569585.
Garud, R., and Kumaraswamy, A. 2005. Vicious and virtuous circles in the management of knowledge: The
case of Infosys Technologies. MIS Quarterly, 29, 1, 933.
Goodman, P.S., and Darr, E.D. 1998. Computer-aided systems and communities: Mechanisms for organiza-
tional learning in distributed environments. MIS Quarterly, 22, 4, 417440.
Gray, P.H., and Meister, D.B. 2004. Knowledge sourcing effectiveness. Management Science, 50, 6,
821834.
Griffth, T.L.; Sawyer, J.E.; and Neale, M.A. 2003. Virtualness and knowledge in teams: Managing the love
triangle of organizations, individuals, and information technology. MIS Quarterly, 27, 2, 265287.
Hahn, J., and Subramani, M. 2000. A framework of knowledge management systems: Issues and challenges
for theory and practice. In Soon Ang, Helmut Krcmar, Wanda Orilikowski, Peter Weill and Janice DeGross
(eds.), Proceedings of the Twenty-First International Conference of Information Systems. Association for
Information Systems. Available at www.aisnet.org.
Hansen, M.T., and Haas, M.R. 2001. Competing for attention in knowledge markets: Electronic document
dissemination in a management consulting company. Administrative Science Quarterly, 46, 1, 128.
Hsiao, R.; Tsai, S.; and Lee, C. 2003. The problem of embeddedness: Knowledge transfer, situated prac-
tice, and the role of information systems. In Salvatore March, Anne Massey and Janice DeGross (eds.),
Proceedings of the Twenty-Fourth International Conference on Information Systems. Association for
Information Systems. Available at www.aisnet.org, pp. 3646.
Huber, G. P. 1991. Organizational learning: The contributing processes and the literatures. Organization
Science, 2, 1, 88115.
Irmer, B. E.; Bordia, P.; and Abusah, D. 2002. Evaluation apprehension and perceived benefts in interpersonal
and database knowledge sharing. In Academy of Management OCIS Division Best Paper Proceedings.
Academy of Management. Available at www.aomonline.org, pp. B1B6.
Jarvenpaa, S.L., and Staples, D.S. 2000. The use of collaborative electronic media for information sharing:
An exploratory study of determinants. Journal of Strategic Information Systems, 9, 129154.
. 2001. Exploring perceptions of organizational ownership of information and expertise. Journal of
Management Information Systems, 18, 1, 151183.
Ji, Y.G., and Salvendy, G. 2002. Development and validation of user-adaptive navigation and information
retrieval tools for an intranet portal organizational memory information system. Behavior and Informa-
tion Technology, 21, 2, 145154.
Kankanhalli, A.; Tan, B.C.Y.; and Wei, K. 2001. Seeking knowledge in electronic knowledge repositories: An
exploratory study. In Veda Storey, Sumit Sarkar and Janice DeGross (eds.), Proceedings of the Twenty-
Second International Conference on Information Systems. Association for Information Systems. Available
at www.aisnet.org, pp. 123133.
. 2005. Contributing knowledge to electronic knowledge repositories: An empirical investigation.
Kock, N., and McQueen, R.J. 1998. Groupware support as a moderator of interdepartmental knowledge
communication in process improvement groups: An action research study. Information Systems Journal,
8, 183198.
Kwan, M.M., and Balasubramanian, P. 2003. KnowledgeScope: Managing knowledge in context. Decision
Support Systems, 35, 467486.
Lee, G.K., and Cole, R.E. 2003. From a frm-based to a community-based model of knowledge creation: The
case of the Linux kernel development. Organization Science, 14, 6, 633649.
Lee, N., and Litecky, C.R. 1997. An empirical study of software reuse with special attention to Ada. IEEE
Transactions on Software Engineering, 23, 9, 537549.
Levina, N., and Vaast, E. 2005. The emergence of boundary spanning competence in practice: Implications
for implementation and use of information systems. MIS Quarterly, 29, 2, 335363.
Linger, H; Burstein, F; Zaslavsky, A.; and Crofts, N. 1999. A framework for a dynamic organizational
memory information system. Journal of Organizational Computing and Electronic Commerce, 9, 2 & 3,
189203.
Majchrzak, A.; Malhotra, A.; and John, R. 2005. Perceived individual collaboration know-how development
through information technologyenabled contextualization: Evidence from distributed teams. Information
Mann, M.M.; Rudman, R.L.; Jenckes, T.A.; and McNurlin, B.C. 1991. Eprinet: Leveraging knowledge in
the electric utility industry. MIS Quarterly, 15, 3, 403421.
March, S.T., and Smith, G.F. 1995. Design and natural science research on information technology. Decision
Support Systems, 15, 251266.
Markus, L. 2001. Toward a theory of knowledge reuse: Types of knowledge reuse situations and factors in
reuse success. Journal of Management Information Systems, 18, 1, 5793.
Massey, A.P.; Montoya-Weiss, M.M.; and ODriscoll, T.M. 2002. Knowledge management in pursuit of
performance: Insights from Nortel Networks. MIS Quarterly, 26, 3, 269289.
McCormack, A. 2002. Siemens Share-Net: Building a knowledge network. Boston: Harvard Business School
Case 9603036.
Miranda, S.M., and Saunders, C.S. 2003. The social construction of meaning: An alternative perspective on
Morisio, M.; Ezran, M.; and Tully, C. 2002. Success and failure factors in software reuse. IEEE Transactions
on Software Engineering, 28, 4, 340357.
Newell, S.; Scarbrough, H.; and Swan, J. 2001. From global knowledge management to internal electronic
fences: Contradictory outcomes of intranet development. British Journal of Management, 12, 97111.
Nidumolu, S.; Subramani, M.R.; and Aldrich, A. 2001. Situated learning and the situated knowledge web:
Exploring the ground beneath knowledge management. Journal of Management Information Systems,
18 (Special Issue on Knowledge Management), 115151.
Nonaka, I. 1994. A dynamic theory of knowledge creation. Organization Science, 5, 1, 1437.
Oh, W., and Jeon, S. 2004. Membership dynamics and network stability in the open-source community: The
Ising perspective. In Ritu Agarwal, Laurie Kirsch and Janice DeGross (eds.), Proceedings of the Twenty-
Fifth International Conference on Information Systems. Association for Information Systems. Available
at www.aisnet.org, pp. 413425.
Olfman, L. 1999. Introduction to the special issue on organizational memory systems. Journal of Organiza-
tional Computing and Electronic Commerce, 9, 2 & 3, 101103.
Olivera, F. 2000. Memory systems in organizations: An empirical investigation of mechanisms for knowledge
collection, storage, and access. Journal of Management Studies, 37, 6, 811832.
Orlikowski, W.J.; Yates, J.; Okamura, K.; and Fujimoto, M. 1995. Shaping electronic communication: The
metastructuring of technology in the context of use. Organization Science, 6, 4, 423444.
Poston, R.S., and Speier, C. 2005. Effective use of knowledge management systems: A process model of
content ratings and credibility indicators. MIS Quarterly, 29, 2, 221244.
Purvis, R.L.; Sambamurthy, V.; and Zmud, R.W. 2000. The development of knowledge embeddedness in CASE
technologies within organizations. IEEE Transactions on Engineering Management, 47, 2, 245257.
. 2001. The assimilation of knowledge platforms in organizations: An empirical investigation. Orga-
nization Science, 12, 2, 117135.
Rafaeli, S., and Ravid, G. 2003. Information sharing as enabler for the virtual team: An experimental ap-
proach to assessing the role of electronic mail in disintermediation. Information Systems Journal, 13,
191206.
Ruppel, C.P., and Harrington, S.J. 2001. Sharing knowledge through intranets: A study of organizational
culture and intranet implementation. IEEE Transactions on Professional Communication, 44, 1, 3752.
Sabherwal, R., and Becerra-Fernandez, I. 2005. Integrating specifc knowledge: Insights from the Kennedy
Space Center. IEEE Transactions on Engineering Management, 52, 3, 301315.
Schultze, U., and Leidner, D.E. 2002. Studying knowledge management in information systems research:
Discourses and theoretical assumptions. MIS Quarterly, 26, 3, 213242.
Sherif, K. 1999. The prerequisites for learning from collaborative technologies. In Proceedings of the Americas
Conference on Information Systems, pp. 151153.
Sherif, K., and Vinze, A. 2002. Domain engineering for developing software repositories: A case study.
Decision Support Systems, 33, 5569.
Skok, W., and Kalmanovitch, C. 2005. Evaluating the role and effectiveness of an intranet in facilitating
knowledge management: A case study at Surrey County Council. Information and Management, 42,
731744.
Smeds, R., and Alvesalo, J. 2003. Global business process development in a virtual community of practice.
Production planning and control, 14, 4, 361371.
Staples, D.S., and Jarvenpaa, S.L. 2000. Using electronic media for information sharing activities: A replica-
tion and extension. In Soon Ang, Helmut Krcmar, Wanda Orilikowski, Peter Weill and Janice DeGross
(eds.), Proceedings of the Twenty-First International Conference on Information Systems. Association
for Information Systems. Available at www.aisnet.org, pp. 117133.
Stenmark, D. 2001. Leveraging tacit organizational knowledge. Journal of Management Information Systems,
17, 3), 924.
Stenmark, D., and Lindgren, R. 2003. Intranets for knowledge management: Applications affording user
participation. In Dennis Galletta and Jeanne Ross (eds.), Proceedings of the Ninth Americas Confer-
ence on Information Systems. Association for Information Systems. Available at www.aisnet.org, pp.
26002606.
Sussman, S. W., and Siegal, W. S. 2003. Informational infuence in organizations: An integrated approach to
knowledge adoption. Information Systems Research, 14, 1, 4765.
Teigland, R., and Wasko, M.M. 2003. Integrating knowledge through information trading: Examining the
relationship between boundary spanning communication and individual performance. Decision Sciences,
34, 2, 261286.
Thorn, B.K., and Connolly, T. 1987. Discretionary data bases: A theory and some experimental fndings.
Communication Research, 14, 5, 512528.
Vaast, E. 2004. O Brother, where are thou? Management Communication Quarterly, 18, 1, 544.
Von Krogh, G.; Spaeth, S.; and Lakhani, K.R. 2003. Community, joining, and specialization in open source
software innovation: A case study. Research Policy, 32, 12171241.
Webster, J., and Watson, R. T. 2002. Analyzing the past to prepare for the future: Writing a literature review.
MIS Quarterly, 26, 2, xiiixxiii.
Weiser, M., and Morrison, J. 1998. Project Memory: Information Management for project teams. Journal of
Management Information Systems, 14, 4, 149166.
Wenger, E. 1998. Communities of Practice: Learning, Meaning and Identity. Cambridge: Cambridge Uni-
versity Press.
Wijnhoven, F. 1998. Designing organizational memories: Concept and method. Journal of Organizational
Computing and Electronic Commerce, 8, 1, 2955.
Yates, J., and Orlikowski, W. 2002. Genre systems: Structuring interaction through communicative norms.
Journal of Business Communication, 39, 1, 1335.
Yates, J.; Orlikowski, W. J.; and Okamura, K. 1999. Explicit and implicit structuring of genres in electronic
communication: Reinforcement and change of social interaction. Organization Science, 10, 1, 83103.
Yuan, Y.; Fulk, J.; Shumate, M.; Monge, P. R.; Bryant, J. A.; and Matsaganis, M. 2005. Individual participation
in organizational information commons: The impact of team-level social infuence and technology-specifc
competence. Human Communication Research, 31, 2, 212240.
248
ChaptEr 13
DYNAMIC TEAM MEMORY SYSTEMS
Enabling Knowledge Sharing Effectiveness in
Structurally Diverse Distributed Teams
ann majChrZaK, arVind malhotra, and riChard john
Abstract: Firms are increasingly leveraging their globally distributed knowledge resources
through the deployment of structurally diverse distributed teams. Because face-to-face meetings
are becoming less common among distributed teams, team members are more frequently sharing
their knowledge through the use of virtual workspacesan integrated set of tools that offer a
variety of communication support capabilities including a common team repository organized
for easy search and retrieval, application sharing, electronic whiteboards, and group discussion
forums. Despite this increased use, there is much we do not understand about how to effectively
use these virtual workspaces. Using an organizational memory information system framework,
we propose that teams that use their virtual workspace technologies to provide functions of
what we call a dynamic team memory system will report higher levels of knowledge sharing
effectiveness. We suggest, however, that this relationship is mediated by increases in perceived
task analyzability. A test of this model on a sample of 54 structurally diverse distributed teams
supports our hypotheses.
Keywords: Virtual Teams, Distributed Teams, IT Support, Knowledge Sharing, Organizational
Memory, Task Analyzability
Increasingly, globally distributed expertise is being brought to bear on innovative and challeng-
ing tasks. Such expertise hails not only from diverse functions within an organization, but also
across multiple organizations. Such expertise is brought together in what Cummings (2004) calls
structurally diverse distributed teams, virtual teams that are diverse in terms of geographic loca-
tion, functional assignment, reporting managers, and business units. Co-locating such expertise is
often neither economically feasible, nor appropriate to deriving benefts from globally distributed
expertise (Malhotra, Majchrzak, Carmen, and Lot, 2001). The fact that the expertise is not co-
located requires that most of the work by team members be conducted electronically.
Research has found that some distributed teams may need face-to-face interactions to succeed
(Cramton, 2001; Maznevski and Chudoba, 2000). However, other researchers have described that
even the most structurally diverse distributed teams can be successful beyond expectations without
any face-to-face meetings (Malhotra et al., 2001). In light of this somewhat conficting research,
one has to ask the question: how can distributed teams succeed without the luxury of face-to-face
interactions among sources of expertise?
Some researchers have recommended the use of virtual workspaces as collaborative technolo-
DYNAMIC TEAM MEMORY SYSTEMS 249
gies in such environments (e.g., DeSanctis and Jackson, 1994; Malhotra and Majchrzak, 2005).
However, Hinds and Bailey (2003) suggest that the use of collaborative technologies in diverse
virtual teams can lead to greater task confict than experienced in face-to-face interactions because
the technology mediation leads to uneven information, disrupted sequences of responses, and
members being out of sync with the larger conversation.
Complicating the matter is the fact that many of the distributed teams that are being deployed
in practice are being entrusted with innovative and complex tasks (Malhotra and Majchrak, 2006).
The nature of these tasks affects the amount and nature of information that needs to be shared
between team members distributed across the globe. One approach that can be taken by leaders
of structurally diverse distributed teams is to make the tasks more analyzable and thereby reduce
the need of information sharing between team members. However, Daft and MacIntosh (1981)
(in a nonvirtual setting) reported surprise at their fndings that task analyzability was positively
correlated with amount of information processing. Benbasat and Lim (1993) reported surprise in
their results in which groups with low-complexity tasks benefted more, not less as hypothesized,
from group support systems. Jarvenpaa and Staples (2000) noted as surprising their result that a
more structured culture was positively correlated with greater (not less) use of electronic media
for sharing information, reasoning that these results suggest that users may want more structured
information fow in place, possibly due to their need to have reliable access to other individuals
knowledge and information (p. 129). Thus, even the highly structured and low-complexity tasks
performing distributed teams may require appropriate technology in place to reduce task conficts
while optimizing the fow of information between team members.
However, there has been only limited research on identifying characteristics of virtual work-
spaces that facilitate knowledge sharing effectiveness in structurally diverse distributed teams.
For example, while recent research by Majchrzak, Malhotra, and John (2005) examined specifc
characteristics of virtual workspaces, their research was focused on how those characteristics
facilitated contextualization, rather than on knowledge sharing in general.
In this study, we build on the organizational memory information system theoretical framework
of Stein and Zwass (1995) to suggest hypotheses describing how virtual workspaces might be
structured to facilitate knowledge sharing in globally distributed and structurally diverse teams
whose members rarely, if at all, see each other face-to-face. We then empirically test the hypotheses
using a sample of 54 structurally diverse distributed teams.
CONCEPTUAL DEVELOPMENT
Knowledge Sharing Effectiveness in Structurally Diverse Distributed Teams
As individuals work within a structurally diverse distributed team, they must be able to integrate
and synthesize knowledge across diverse members (Grant, 1996). Synthesizing knowledge across
different areas of expertise requires that teams have members with the absorptive capacity to
use the knowledge shared by others (Szulanski, 2000) as well as the practices in place to make
knowledge available to the team (Becerra-Fernandez and Sabherwal, 2001). Additionally, Daft
and MacIntosh (1981) combine volume of information gathered and the capacity of participants
to absorb the gathered information in characterizing how well individuals process information.
Thus, we defne knowledge sharing effectiveness in such teams as member satisfaction with the
way team members share and absorb each others knowledge. Our research is intended to test
theoretically derived hypotheses for explaining the role of virtual workspaces in knowledge shar-
ing effectiveness within structurally diverse distributed teams.
250 MAJCHRZAK, MALHOTRA, AND JOHN
An Organizational Memory Information System Framework for Studying
Virtual Workspaces
Virtual workspaces are defned by Dourish and Belloti (1992) as shared electronic workspaces that
provide team members with awareness information that can be exploited for knowledge creation,
allowing them to move smoothly between close and loose collaboration and to coordinate their
work dynamically. These technologies may consist of a whole host of features, ranging from in-
stant messaging to confguration control, from keyword searches to automatic notifcation, from
group brainstorming aids to simple email (DeSanctis and Jackson, 1994; Greenberg and Roseman,
2003; Zigurs and Buckland, 1998). Studying how virtual workspaces might be used to facilitate
knowledge sharing in structurally diverse distributed teams requires a framework for understand-
ing the role these various features serve.
Following the earlier work of Ackerman and Malone (1990) and Walsh and Ungson (1991),
Stein and Zwass (1995) proposed an organizational memory information systems framework for
studying collaborative technology. They argued that organizational memory refers to the social
process of articulating, exchanging, and sharing information from the past to bear on creating
shared interpretations and decisions in the present and that information systems can be designed
to support this process. They defne an organizational memory information system as a system
that functions to provide a means by which knowledge from the past is brought to bear on present
activities, thus resulting in increased levels of effectiveness for the organization (p. 95). Recently,
the concept of an organizational memory information system was extended to memory systems
supporting small groups (Linger, Burstein, Zaslavsky, and Crofts, 1999).
Using an information-processing model of memory, Stein and Zwass describe how an infor-
mation system can support organizational memory. They argue that knowledge must be acquired
from producers, then retained and maintained in distributed and ever-changing repositories, and
then searched and retrieved by information consumers. These mnemonic functions are dynamic
processes requiring the active engagement of members of the organizational unit. As an example,
annotations of documents by members help others in the organizational unit to relate their per-
sonal experiences to the problem. Stein and Zwass propose three requirements for supporting the
mnemonic needs of a team:
Design Requirement #1: Intelligently, Continuously, and Easily Acquire and Encode Knowl-
edge from Information Producers. While Stein and Zwass refer to this design requirement as
simply knowledge acquisition and retention, we use the more descriptive label to indicate the
intent that knowledge acquisition is ongoing, and it is the continuous encoding that refects
currently retained knowledge that infuences the acquisition process. Features meeting this
requirement dynamically and regularly transfer knowledge from the information producer
to the memory system through encoding to make the knowledge understandable to others.
Examples of technology features suggested by Stein and Zwass include intelligent sum-
marizing, selectively targeting knowledge acquisitions, linking to information already in the
memory system, and indexing structures for encoding information as it comes in.
Design Requirement #2: Maintaining Integrity of Knowledge Over Time. Stein and Zwass
refer to this design requirement simply as knowledge maintenance. Instead, we label it
more descriptively because the critical element of their concern is not simply that knowl-
edge be retained but that the integrity of the retained knowledge is maintained, especially as
new emerging knowledge is integrated with existing knowledge and obsolete knowledge is
deleted. Stein and Zwass suggest that this requirement can be met with technology features
that provide revision histories, maintain appropriate versions, and reorganize indexing cat-
egories as new information is added.
Design Requirement #3: Pattern Matching During Search & Retrieval. According to Stein
and Zwass, an organizational memory information system must not only provide the means
to allow for knowledge retrieval but must also facilitate bringing the right knowledge to bear
on a problem, going considerably beyond traditional Boolean retrieval of unstructured text.
Technology features that meet this requirement include those that offer pattern matching and
recognition, multiple forms of knowledge representation to facilitate making comparisons,
and comparison of earlier to current problem contexts.
There are parallels between Stein and Zwass design requirements for effective organizational
memory information systems and the literature on virtual workspaces (Greenberg and Roseman,
2003; Malhotra and Majchrzak, 2004). In support of the frst design requirement for ongoing intel-
ligent knowledge acquisition, Bordetsky and Mark (2000) describe the use of virtual workspace
features that provide intelligent encoding of knowledge being placed in the repository. They argue
that without such encoding, distributed team members face great diffculty in maintaining the
integrity of the knowledge and searching and retrieving this knowledge at a later time.
Supporting the second design requirement, Jang, Steinfeld, and Pfaff (2000) describe how
virtual workspaces need to help maintain the integrity of knowledge over time by enabling team
members to add comments to any fle or folder that allows others to understand their perspective.
In addition, such workspaces must also maintain a history of who reads and modifes documents
and messages posted on the electronic discussion boards and link the different objects in the team
repository so as to enable easy access.
Finally, in support of the third requirement, Dourish and Belloti (1992) point out the role of
virtual workspaces in enabling pattern matching during search and retrieval. Such pattern match-
ing is supported by features that store details of past activities within documents in the repository,
which also allows presentation of information at various levels of specifcity as needed. Features
can also allow for pattern matching by enabling the storage of information that evolves over time
to refect the progress of activities. Further, this information can be presented differently to the
various team members depending on the relevance to their evolving activities.
When these features are integrated into a virtual workspace, teams are better able to grow their
organizational memory through extensive acquisition, maintenance, and search/retrieval capabili-
ties (Ackerman, 1994; Ackerman and Malone, 1990; Conklin, 1993). In doing so, these features
provide team members with perspectives on the ongoing activities of other team members, insight
into the rationale and thoughts of other team members, and an understanding of where the team
is in terms of its project life cycle (Bordetsky and Mark, 2000; Dourish and Belloti, 1992; Jang
et al., 2000).
Integrating the Stein and Zwass general organizational memory information systems framework
with the specifc features offered for structurally diverse distributed teams by virtual workspaces,
we refer to a virtual workspace that is able to meet all three design requirements for a mnemonic
organizational memory information system as a dynamic team memory system (DTMS). A DTMS,
then, is an important precursor to the development of knowledge sharing effectiveness. Following
Stein and Zwasss argument, a memory system will only have the intended effect if it is designed
properlymeeting the three design requirementsand used in a way that encourages integrating
the past information with the present information.
Therefore, we propose that a virtual workspace is only a precursor to knowledge sharing ef-
fectiveness when it is used as a DTMS. Use as a dynamic team memory system implies that the
virtual workspace features are (1) intended to provide a DTMS for the team (what DeSanctis and
Poole [1994] refer to as the technologys spirit), (2) implemented as part of the tools to support the
accomplishment of group tasks (Zigurs and Buckland, 1998), and (3) judged by team members as
assisting them in performing their portfolio of tasks (Goodhue and Thompson, 1995). In essence,
for a virtual workspace to be a precursor of knowledge sharing effectiveness, it must be designed,
implemented, and found useful as a DTMS.
Effect of Using a DTMS on Knowledge Sharing Effectiveness
Stein and Zwass complete their framework by indicating that an information system-enabled or-
ganizational memory will have its intended impacts only if used in a way that creates a routine of
making present decisions in light of a critically appraised past (p. 108). These routines increase
the perceptions of analyzability of the process of converting past knowledge into knowledge
pertinent for the current context (Perrow, 1967). According to Perrow (1967), analyzable tasks
are those for which there are more information cues available and more procedures established
and for which individuals can identify a correct response. The more the conversion process is
deemed analyzable, the less time will the organizational actors have to spend thinking about how
to use the past knowledge to perform their current work (Rice, 2001; Withey, Daft, and Cooper,
1983). Task analyzability has been found to lead to increased information processing capacity
(Daft and MacIntosh, 1981; Zeffane and Gul, 1993), which in turn increases knowledge sharing
effectiveness (Malhotra, Gosain, and El Sawy, 2005; Zahra and George, 2002) as people are able
to absorb new knowledge more effectively.
A DTMS can lead to increased perceptions of task analyzability by establishing these routines.
For example, by establishing routines for encoding knowledge from information producers, the
DTMS can make it easier for other team members to fnd the knowledge (Bordestky and Mark,
2000; Griffth and Neale, 2001), increasing perceptions of task analyzability. By using routines
for maintaining the integrity of the knowledge over time, the DTMS can enhance the analyzability
of the task of responding to the new knowledge that is added to the workspace. By using routines
for pattern matching during search and retrieval, the DTMS can make the search and retrieval
process more analyzable. In essence, following Stein and Zwasss arguments, DTMSs that fail
to increase task analyzability are unlikely to enhance knowledge sharing effectiveness. This is
particularly the case for structurally diverse distributed teams where there is signifcant diversity
in past knowledge and evolving knowledge bases (Malhotra et al., 2001).
In sum, we hypothesize:
H1: The effect of the use of a DTMS on knowledge sharing effectiveness within a structurally
diverse distributed team will be mediated by task analyzability.
Clarity of expertise location has been shown to be related to knowledge sharing effectiveness
(Faraj and Sproull, 200), particularly in groups with members having complementary skills (Hol-
lingshead, 1998b). Expertise location allows members to assess accuracy of the different solutions
proposed by the members to avoid the effort of searching for their own solutions. In this way,
enhanced expertise location tends to reduce the information processing needs of individuals on
structurally diverse distributed teams. Expertise location helps in cueing one anothers expertise to
retrieve already existing solutions without having to generate ones own solution (Wegner, 1987).
When groups have this shared conceptualization of one anothers expertise, they pool information
more effectively (Lewis, 2003). Stein and Zwass argue that successful task-specifc memory in-
formation systems are expressly designed to augment the interaction between a knowledge seeker
and human organizational experts (p. 91). Augmenting such interactions through the use of the
DTMS should lead to more clarity about where the expertise is located and how that expertise is
related to the work at hand (Griffth and Neale, 2001). However, for the same reasons as before,
this increased clarity is likely only when the use of the DTMS leads to increased task analyzability.
Therefore, we hypothesize:
H2: The effect of the use of a DTMS on expertise location within a structurally diverse distributed
team is mediated by task analyzability, with increases in expertise location related to higher
knowledge sharing effectiveness.
MODEL SUMMARY
In summary, our model is illustrated in Figure 13.1. In proposing organizational memory infor-
mation system theory as a framework for examining virtual workspaces in the model, we have
responded to the call by Dennis and Kinney (1998) for alternative, more foundational, technology
constructs than media richness. In addition, the model suggests that perceptions of task analyz-
ability is a critical intervening variable in explaining the relationship of virtual workspaces to
team outcomes. Finally, by introducing expertise location in the model, we suggest that not only
is knowledge sharing facilitated by a DTMS but also the diffcult task of locating and maintaining
an understanding of the evolving expertise on the team.
RESEARCH METHODS
Sample
A cross-sectional survey of 54 structurally diverse distributed teams was conducted. The sample
was identifed through solicitation by virtual team consultants, professional organizations, col-
laboration tool providers, and personal contacts. The solicitation promised a benchmarking report
Figure 13.1 Proposed Research Model
USE OF
DYNAMIC TEAM
MEMORY
SYSTEM
TASK
ANALYZABILITY
EXPERTISE
LOCATION
KNOWLEDGE
SHARING
EFFECTIVENESS
+
+
+
+
Clarity of
task context
Reduced information
processing needs
Enhanced
information
processing
capacity
in exchange for participation. Email exchanges or interviews were conducted to confrm that the
team was structurally diverse and virtual. Of the 90 contacts, 54 teams met the criteria.
Once a team was identifed, a 1-hour semistructured interview with the team leader was con-
ducted asking about the teams objective and background, the practices used in the team to manage
its distributed nature and the features of the virtual workspace utilized by the team to collaborate.
Analysis of the interviews indicated that 75% of the teams included members from more than
one national culture. Further, 60% of the teams also included members three or more time zones
apart or with different native languages. Half of the teams included members from more than one
company. Sixty percent of the teams included members from more than one organization or func-
tion within a company. This confrmed that teams in our sample were indeed structurally diverse
(crossing geographically and either corporate, organizational, or functional boundaries).
Thirty-three companies in 15 different industries were represented in the sample (e.g., telecom-
munications, consumer products, engineering design, medical device manufacturing). The teams
were performing a variety of tasks (e.g., consulting, corporate mergers, help desk support, new
product development). Team sizes ranged from 3 to 50, with a median of 10 members and all but
seven teams having 20 or fewer members. The teams had been in operation, on the average, 18
months, and thus were beyond the start-up phase of their task performance.
The team leaders reported that the teams used a wide variety of virtual workspace platforms,
including Lotus Notes, Groove, Livelink, Microsoft products, NetMeeting, Webex, and E-Room.
Team leaders confrmed that they had selected these platforms and the specifc features used by
the team at the outset of the start of the team. All but three of the teams supplemented their virtual
workspace technologies with audioconferencing.
The team leader was asked to encourage team members to complete a 35-minute online survey.
A total of 263 individuals completed the survey representing, on average, 50% of the members of
each team. Surveyed members reported spending only 2.5% of their time on the team in face-to-face
meetings with all other members, and only 15% of their time in face-to-face meetings with any other
member. Thus, the teams could not rely on face-to-face interactions for their communication. Team
members reported little to no previous experience working with other members on the team. This
suggests that their assessments of knowledge sharing effectiveness were focused on the actions of team
members during the teams life cycles, with minimal carryover from previous working relationships
with the members. Finally, team members reported being highly dependent on each other to perform
their task, substantiating the need for team members to interact regularly to perform their tasks.
Measures
Knowledge Sharing Effectiveness
As defned earlier, knowledge sharing effectiveness (KSE) is the satisfaction of team members
with the way team members shared and absorbed each others knowledge. A seven-item scale
combined the Becerra-Fernandez and Sabherwal (2001) scale measuring satisfaction with knowl-
edge sharing practices of organizations with Szulanskis (2000) measure of absorptive capacity,
as recommended by Griffth et al. (2003) for virtual teams. The items, measured using a 5-point
agree/disagree response scale, are listed in Table 13.3. To ensure that we could aggregate to the
team level, the James index of interrater agreement was used (James, Demaree, and Wolf, 1984).
Janz, Colquitt, and Noe (1997) suggested a median of 0.70 or above as a guideline justifying
aggregation of individuals responses into a group. We obtained an interrater agreement with a
median of 0.89 across teams in our sample.
Use of a DTMS
A three-pronged defnition of the use of a virtual workspace was proposed earlier, that is, as a
set of features that (1) are designed to provide a DTMS for the team (i.e., spirit), (2) have been
implemented as part of the tools to support the accomplishment of group tasks, and (3) are judged
by team members as assisting them in performing their portfolio of tasks. To measure use we frst
listed the set of features depicting a DTMS as described in the Conceptual Development section.
These features included indexing, version history, links, keyword searching, annotations, and
status/progress indicators to organize and retrieve task information. We expanded on this initial
list by examining and including features used by the major workspace platform providers (Groove,
Eroom, Livelink, Webex, Lotus Notes, and netMeeting). We then identifed an expert panel of
16 practitioners using or selling workspace technologies and researchers studying workspace
technologies. We asked the expert panelists to confrm that our list comprehensively described
various features in use in virtual workspaces at the time of the study. This yielded a list of 18
features (Table 13.1).
To assess the spirit of each virtual workspace feature (i.e., the extent to which each feature has
been designed to support a DTMS), we asked the 16 experts to assess the extent on a 7-point scale
(from not enabled at all to enabled to a great extent) to which each of these features enabled
the teams acquisition, maintenance, and search and retrieval needs as specifed for a DTMS. As
a control, we included three virtual workspace features that we expected would not be related to
these needs: instant messaging, electronic whiteboard, and synchronous application sharing. As
reported in Table 13.1, the expert ratings for the DTMS features were higher (from 3.6 to 5.7)
than the ratings for the other features (1.4 to 1.9). The average interreliability among experts was
0.74, indicating a general agreement between diverse experts in their assessment of the features
of virtual workspaces as enabling the spirit of a DTMS.
To assess which features had been implemented by the team, we discussed each feature with
the team leader to determine if the team leader had implemented it as part of the virtual work-
space used by the team. We only marked features as having been implemented if the team leader
understood the feature we were discussing, if she or he could demonstrate it verbally while on the
telephone, and if it had been in use for some time prior to the interview (so that team members
had had some experience in using it). The percentages of team leaders implementing each feature
are presented in Table 13.1.
Finally, to measure the degree to which the confguration of features assisted an individual in
performing his or her portfolio of tasks, we asked team members to report on the degree to which
their virtual workspaces (defned for them as the central electronic locatione.g., Web site,
repository, discussion room, etc.for the teams knowledge) helped them to contribute, organize,
and retrieve knowledge from the team. We developed six items specifcally tailored to measure
knowledge contribution, organization, and retrieval by asking the degree (15 agree/disagree) to
which the team members felt that their virtual workspaces enabled them to (1) easily know who
contributed a piece of knowledge to the team repository, (2) easily allow different people to fnd
summaries as well as details in the repository, (3) easily fnd specifc entries in repositories that
have been contributed by specifc individuals, (4) easily label an entry with multiple keywords it
pertains to, (5) easily know rationale behind the decisions made by team members so that deci-
sion and rationale can be revisited later, and (6) easily identify historical connections between
entries. We piloted these six items on a sample of eight individuals who were members of differ-
ent structurally diverse distributed teams to ensure that these items represented important aspects
of their work. To ensure that we could aggregate to the team level, the James index of interrater
agreement was used. We obtained an interrater agreement with a median of 0.91 across teams in
our sample suggesting that the items could be aggregated at the team level. After aggregating the
items, we obtained a reliability coeffcient of 0.89 at the team level for the six items.
To create the fnal index of the use of a DTMS, we summed the extent of enabling scores in Table
13.1 for only the virtual workspace features that had been implemented (possible range from 0 to
86.88), and multiplied by the aggregated mean score across the six items of the degree to which the
features assisted team members in performing their work. The fnal index ranged from 0 to 175.95,
with a mean of 66.74. Therefore, the index is a composite measure of the implementation of features
that enable a DTMS weighted by the degree to which the features support DTMS in spirit, and
team members perceptions of degree to which the features assisted team members work. The high
Table 13.1
Virtual Workspace Features Extent of Enabling Ratings and Implementation (in percent)
Virtual Workspace Features
Enables Teams
Acquisition, Maintenance,
and Search and
Retrieval Needs
Percent of Teams
Implementing Each
Feature (of 54)
1. Repository query for patterns via indexed
and summarized multimedia data
5.73 13.0%
2. Text mining tools 5.73 7.4%
3. Entries have subjects, keywords, and
dates felds to make them searchable
5.70 44.4%
4. Threaded discussion database 5.58 38.9%
5. Online capture of teams progress
towards milestone
5.50 61.1%
6. Intelligent search capabilities for team
repository
5.35 37.0%
7. Automatically generated revision history
for entries
5.33 40.7%
8. View multiple entries simultaneously for
comparison and contrast
5.13 37.0%
9. Teams knowledge repository integrated
with corporate databases
4.65 33.3%
10. Automatic suggestion of keywords for
entries
4.60 11.1%
11. Visualization tools to exhibit links between
entries
4.50 5.6%
12. Links associated with entries categorized
by types
4.50 18.5%
13. Field identifying authorship of all entries
in team repository
4.48 72.2%
14. Identifcation of authors of electronic
annotations to entries
4.48 51.9%
15. Electronic annotation capabilities for
entries
4.43 55.6%
16. Online team virtual meeting scheduler 3.83 64.8%
17. Automatic notifcation to members of any
addition to team repository
3.73 33.3%
18. Easy changing of keywords associated
with entries
3.63 33.3%
Electronic whiteboard capabilities 1.93 NA
Synchronous application sharing 1.85 NA
Instant messaging without fle sharing 1.48 NA
interrater reliability among experts regarding the spirit of the features, and also the high James
index (at the individual level) and high alpha reliability (at the team level) related to team members
perceptions of the assistance of the features for task performance suggests that the overall index is
reliable and consistent. Of the teams in the sample, 20% had a zero score on the index as their lead-
ers reported that none of the features in Table 13.1 were present in their virtual workspaces (or, in
other words, they were not supplementing their e-mail and audioconferencing).
Expertise Location Within the Team
Expertise location was measured using the Faraj and Sproull (2000) scale of the degree to which
members know how knowledge is distributed among the members. The items, using a 5-point
agree/disagree response scale, are listed in Table 13.3. Because expertise location is a team-level
measure, we aggregated to the team level. A James index of 0.93 confrmed that this aggregation
was appropriate.
Task Analyzability
As have others studying technology use in electronically mediated settings (e.g., Rice, 1992), we
used the measure of Withey et al. (1983) of task analyzability, listed in Table 13.3. A James index
of 0.88 was obtained confrming the aggregation to the team level.
Control Variables
We included three variables to control for the effect of group structure on all three endogenous
constructs: goal similarity among team members, open team membership, and team task type. Goal
similarity was measured at the individual-level using Jehns (1995) scale (see Table 13.3), and
aggregated to the team-level after obtaining a James index of 0.95. Open team membership refers
to the predictability of participation in the team and was measured by team leaders response to:
Does the team have members who are not formally assigned to the team? Team task type refers
to the innovativeness of the team and was measured as its inverse, by team leaders responses to:
Is your team characterized as operational (defned as problem-solving tasks carried out repeat-
edly for customers such as customer support) or project/innovation based (defned as one-off
problems a team is attempting to solve such as a merger or a new product development). All three
controls were treated as covariates in the model, and the hypothesized model paths were estimated
in a hierarchical fashion, after controlling for these three variables.
Assessing Discriminant Validity Amongst Salient Latent Constructs
Due to sample size restrictions, we followed the work of Kirsch et al. (2002) and Malhotra et al.
(2005) and submitted only a subset of the three salient latent endogenous to an exploratory factor
analysis. Table 13.2 presents the results of the principal component factor analysis, resulting in a
three-factor solution that accounted for almost three-quarters of the variance in the 15 indicators.
The factor loadings, following an oblique rotation, suggest good discriminant validity among the
three key constructs. The minimum loading for an indicator on its hypothesized construct was
0.59, with all but two above 0.70. The maximum cross-loading was only 0.25, with the other
14 below 0.20. The correlation among the factors that resulted from the oblique rotation ranged
between 0.41 and 0.58.
RESULTS
Assessing the Measurement Model
We conducted a PLS analysis to test the proposed model. We followed Hullands (1999) prescrip-
tion for PLS analysis to assess the test for reliability of constructs used in the measurement model.
The results of the assessment, along with all manifest variable means and standard deviations, are
shown in Table 13.3. Individual item reliability was measured as the degree to which an item loaded
on its intended construct (Hulland, 1999). All the items in our study loaded on their hypothesized
constructs with loadings greater than 0.7 (Hulland, 1999).
The composite reliability of constructs was used as an indicator of internal consistency (Grey
and Meister, 2004; Werts, Linn, and Joreskog, 1974). While Cronbachs coeffcient is generally
used as a measure of construct reliability, it tends to make an assumption that items are identically
correlated with their construct (Bollen, 1989). Chin (1998) suggests that therefore represents a
lower boundary of internal consistency. Instead, it has been advocated that item loadings must be
taken into account to assess internal consistency (Chin; 1998; Fornell and Larcker, 1981; Grey and
Meister, 2004). The internal consistency associated with various constructs (shown in Table 13.4)
is measured following this prescription and therefore is a superior estimate of construct reliability.
The results are all above the 0.7 threshold (minimum = 0.89) that is suggested by Nunnally and
Bernstein (1994). Finally, AVE (average variance extracted) for each of the constructs exceeds
(minimum = 0.67) the thresholds of 0.5, i.e., 50% of variance in indicators of a construct has been
accounted for (Chin, 1998).
The discriminant validity among the constructs in the model can be assessed by comparing the
square root of average variance extracted (AVE) for each of the constructs with the intercorrelations
between constructs (Hulland, 1999). The square root of AVE for each of the constructs is shown in
Table 13.2
Exploratory Factor Analysis of Key Latent Constructs
Factors
Item
Knowledge Sharing
Effectiveness Expertise Location Task Analyzability
KSE1 0.81 0.04 0.15
KSE2 0.76 0.07 0.19
KSE3 0.84 0.09 0.06
KSE4 0.99 0.05 0.14
KSE5 0.78 0.01 0.18
KSE6 0.76 0.10 0.05
KSE7 0.82 0.03 0.07
EXPL01 0.06 0.95 0.12
EXPL02 0.06 0.65 0.17
EXPL03 0.25 0.59 0.08
EXPL04 0.10 0.75 0.09
TSKANAL1 0.08 0.05 0.87
TSKANAL2 0.12 0.12 0.82
TSKANAL3 0.01 0.18 0.84
TSKANAL4 0.06 0.00 0.89
Rotation method: Oblimin (75.08% variance extracted).
259
T
a
b
l
e

1
3
.
3
M
e
a
n
s
,

S
t
a
n
d
a
r
d

D
e
v
i
a
t
i
o
n
s
,

a
n
d

I
t
e
m

L
o
a
d
i
n
g
s

o
f

I
n
d
i
c
a
t
o
r
s
C
o
n
s
t
r
u
c
t
I
n
d
i
c
a
t
o
r
M
e
a
n
S
D
I
t
e
m

L
o
a
d
i
n
g
s
*
K
n
o
w
l
e
d
g
e

S
h
a
r
i
n
g

E
f
f
e
c
t
i
v
e
n
e
s
s

(
K
S
E
)
I
n

t
e
r
m
s

o
f

k
n
o
w
l
e
d
g
e

s
h
a
r
i
n
g

p
r
a
c
t
i
c
e
s

o
f

m
y

t
e
a
m
,

I

a
m

s
a
t
i
s
f
e
d

w
i
t
h
:
1
t
h
e

k
n
o
w
l
e
d
g
e

t
h
a
t

i
s

a
v
a
i
l
a
b
l
e

t
o

m
e

f
r
o
m

o
t
h
e
r

m
e
m
b
e
r
s

t
o

h
e
l
p

m
e

p
e
r
f
o
r
m

m
y

a
c
t
i
v
i
t
i
e
s
3
.
8
7
0
.
5
5
0
.
8
8
2
t
h
e

w
a
y

t
h
e

k
n
o
w
l
e
d
g
e

i
s

m
a
n
a
g
e
d

w
i
t
h
i
n

t
h
e

t
e
a
m
4
.
2
1
0
.
4
2
0
.
9
2
3
t
h
e

d
e
g
r
e
e

o
f

k
n
o
w
l
e
d
g
e

s
h
a
r
i
n
g

a
m
o
n
g

t
e
a
m

m
e
m
b
e
r
s
3
.
9
6
0
.
4
8
0
.
9
3
4
t
h
e

k
n
o
w
l
e
d
g
e

t
h
a
t

t
e
a
m

m
e
m
b
e
r
s

a
r
e

a
b
l
e

t
o

p
r
o
v
i
d
e

f
r
o
m

t
h
e
i
r

e
x
t
e
r
n
a
l

c
o
n
t
a
c
t
s
4
.
0
2
0
.
4
5
0
.
8
6
I

a
m

s
a
t
i
s
f
e
d

t
h
a
t

t
h
e

t
e
a
m

m
e
m
b
e
r
s

i
n

m
y

t
e
a
m
:
5
c
l
e
a
r
l
y

u
n
d
e
r
s
t
a
n
d

t
h
e

p
r
o
c
e
s
s
e
s

u
s
e
d

t
o

a
c
c
o
m
p
l
i
s
h

t
h
e

t
e
a
m
s

a
c
t
i
v
i
t
i
e
s
3
.
9
2
0
.
4
9
0
.
8
9
6
h
a
v
e

e
n
o
u
g
h

c
o
m
p
e
t
e
n
c
e

i
n

k
n
o
w
i
n
g

h
o
w

t
h
e

o
r
g
a
n
i
z
a
t
i
o
n

w
o
r
k
s

t
o

u
n
d
e
r
s
t
a
n
d

a
n
d

u
s
e

t
e
a
m

k
n
o
w
l
e
d
g
e
3
.
7
8
0
.
6
3
0
.
7
8
7
a
r
e

g
e
n
e
r
a
l
l
y

a
b
l
e

t
o

e
x
p
l
o
i
t

n
e
w

k
n
o
w
l
e
d
g
e

c
r
e
a
t
e
d

w
i
t
h
i
n

t
h
e

t
e
a
m
3
.
7
6
0
.
5
0
0
.
7
8
E
x
p
e
r
t
i
s
e

L
o
c
a
t
i
o
n
T
e
a
m

m
e
m
b
e
r
s
:
1
h
a
v
e

a

g
o
o
d

m
a
p

o
f

e
a
c
h

t
e
a
m

m
e
m
b
e
r
s

t
a
l
e
n
t
s

a
n
d

s
k
i
l
l
s
3
.
5
4
0
.
4
9
0
.
7
7
2
w
o
r
k

o
n

t
a
s
k
s

a
p
p
r
o
p
r
i
a
t
e

w
i
t
h

t
h
e
i
r

t
a
s
k
-
r
e
l
e
v
a
n
t

k
n
o
w
l
e
d
g
e

a
n
d

s
k
i
l
l
s
3
.
9
6
0
.
3
8
0
.
8
5
3
k
n
o
w

t
h
e
i
r

o
w
n

s
k
i
l
l
s

a
n
d

h
o
w

t
h
e
y

r
e
l
a
t
e

t
o

t
h
e

t
e
a
m
s

w
o
r
k
3
.
8
9
0
.
3
1
0
.
8
4
4
k
n
o
w

w
h
o

o
n

t
h
e

t
e
a
m

h
a
s

s
p
e
c
i
a
l
i
z
e
d

s
k
i
l
l
s

a
n
d

k
n
o
w
l
e
d
g
e

t
h
a
t

i
s

r
e
l
e
v
a
n
t

t
o

t
h
e

t
e
a
m
s

w
o
r
k
3
.
9
1
0
.
4
7
0
.
8
3
T
a
s
k

A
n
a
l
y
z
a
b
i
l
i
t
y
1
T
h
e
r
e

i
s

a

c
l
e
a
r
l
y

k
n
o
w
n

m
e
t
h
o
d

t
o

d
o

t
h
e

t
a
s
k

o
f

t
h
e

t
e
a
m
3
.
1
0
0
.
5
2
0
.
8
4
2
T
h
e
r
e

i
s

a

c
l
e
a
r
l
y

d
e
f
n
e
d

s
e
t

o
f

p
r
i
n
c
i
p
l
e
s
/
b
e
s
t

p
r
a
c
t
i
c
e
s

t
h
a
t

c
a
n

g
u
i
d
e

t
h
e

t
e
a
m

i
n

d
o
i
n
g

i
t
s

w
o
r
k
3
.
3
9
0
.
5
4
0
.
8
4
3
T
h
e
r
e

i
s

a
n

u
n
d
e
r
s
t
a
n
d
a
b
l
e

s
e
q
u
e
n
c
e

o
f

s
t
e
p
s

t
h
a
t

t
h
e

t
e
a
m

c
a
n

f
o
l
l
o
w

i
n

d
o
i
n
g

i
t
s

w
o
r
k
3
.
4
4
0
.
6
0
0
.
9
3
4
T
h
e
r
e

a
r
e

e
s
t
a
b
l
i
s
h
e
d

p
r
o
c
e
d
u
r
e
s

a
n
d

p
r
a
c
t
i
c
e
s

t
h
a
t

t
h
e

t
e
a
m

c
a
n

u
s
e

i
n

d
o
i
n
g

i
t
s

w
o
r
k
3
.
4
1
0
.
6
9
0
.
9
2
D
y
n
a
m
i
c

T
e
a
m

M
e
m
o
r
y

S
y
s
t
e
m

U
s
e

(
D
M
S

U
s
e
)
S
u
m

(
e
x
p
e
r
t

e
n
a
b
l
i
n
g

s
c
o
r
e
s

f
o
r

f
e
a
t
u
r
e
s

i
m
p
l
e
m
e
n
t
e
d
)

X
6
6
.
7
4
5
3
.
4
4
1
.
0
0
M
e
a
n

(
t
e
a
m

m
e
m
b
e
r

r
e
p
o
r
t
s

o
n

w
o
r
k
s
p
a
c
e

e
n
a
b
l
i
n
g
6

i
t
e
m
s
)
G
o
a
l

S
i
m
i
l
a
r
i
t
y

(
c
o
n
t
r
o
l

v
a
r
i
a
b
l
e
)
1
A
s

a

t
e
a
m
,

w
e

h
a
v
e

s
i
m
i
l
a
r

g
o
a
l
s
3
.
8
6
0
.
4
9
0
.
9
3
2
T
h
e

m
a
i
n

g
o
a
l
s

o
f

o
u
r

t
e
a
m

a
r
e

t
h
e

s
a
m
e

f
o
r

a
l
l

m
e
m
b
e
r
s

i
n

t
h
e

t
e
a
m
3
.
7
0
0
.
5
6
0
.
9
0
3
W
e

o
n

t
h
e

t
e
a
m

a
l
l

a
g
r
e
e

o
n

w
h
a
t

i
s

i
m
p
o
r
t
a
n
t

t
o

t
h
e

t
e
a
m
3
.
6
8
0
.
6
1
0
.
9
2
O
p
e
n

T
e
a
m

M
e
m
b
e
r
s
h
i
p

(
c
o
n
t
r
o
l

v
a
r
i
a
b
l
e
)
D
o
e
s

t
h
e

t
e
a
m

h
a
v
e

m
e
m
b
e
r
s

w
h
o

a
r
e

n
o
t

f
o
r
m
a
l
l
y

a
s
s
i
g
n
e
d

t
o

t
h
e

t
e
a
m
?

[
1

=

y
e
s
,

0

=

n
o
]
0
.
3
1
0
.
4
7
1
.
0
0
T
e
a
m

T
a
s
k

T
y
p
e

(
c
o
n
t
r
o
l

v
a
r
i
a
b
l
e
)
I
s

y
o
u
r

t
e
a
m

c
h
a
r
a
c
t
e
r
i
z
e
d

a
s

o
p
e
r
a
t
i
o
n
a
l

o
r

p
r
o
j
e
c
t
/
i
n
n
o
v
a
t
i
o
n

b
a
s
e
d
?

[
1

=

o
p
e
r
a
t
i
o
n
a
l
,

0

=

p
r
o
j
e
c
t
]
0
.
3
3
0
.
4
8
1
.
0
0
*
I
t
e
m

l
o
a
d
i
n
g
s

a
r
e

f
r
o
m

f
n
a
l

f
u
l
l

P
L
S

m
o
d
e
l

(
M
o
d
e
l

4
)
.
T
a
b
l
e

1
3
.
4
I
n
t
e
r
n
a
l

C
o
n
s
i
s
t
e
n
c
y
,

C
o
n
s
t
r
u
c
t

I
n
t
e
r
c
o
r
r
e
l
a
t
i
o
n
s
,

a
n
d

A
v
e
r
a
g
e

V
a
r
i
a
n
c
e

E
x
t
r
a
c
t
e
d
N
o
.

o
f

I
n
d
i
c
a
t
o
r
s
I
n
t
e
r
n
a
l

C
o
n
s
i
s
t
e
n
c
y
C
o
n
s
t
r
u
c
t

N
o
.
N
o
.
C
o
n
s
t
r
u
c
t
1
2
3
4
5
6
7
1
K
n
o
w
l
e
d
g
e

s
h
a
r
i
n
g

e
f
f
e
c
t
i
v
e
n
e
s
s

(
K
S
E
)
7
0
.
9
5
0
.
8
7
2
E
x
p
e
r
t
i
s
e

l
o
c
a
t
i
o
n
4
0
.
8
9
0
.
7
0
0
.
8
2
3
T
a
s
k

a
n
a
l
y
z
a
b
i
l
i
t
y
4
0
.
9
4
0
.
6
0
0
.
5
4
0
.
8
9
4
D
y
n
a
m
i
c

t
e
a
m

m
e
m
o
r
y

s
y
s
t
e
m

u
s
e

(
D
T
M
S

u
s
e
)
1
1
.
0
0
0
.
2
2
0
.
0
7
0
.
2
5
1
.
0
0
C
o
n
t
r
o
l

V
a
r
i
a
b
l
e
s
5
G
o
a
l

s
i
m
i
l
a
r
i
t
y
3
0
.
9
4
0
.
6
2
0
.
5
6
0
.
4
3
0
.
0
4
0
.
9
1
6
O
p
e
n

t
e
a
m

m
e
m
b
e
r
s
h
i
p
1
1
.
0
0
0
.
0
1
0
.
0
6
0
.
2
6
0
.
0
8
0
.
1
2
1
.
0
0
7
T
e
a
m

t
a
s
k

t
y
p
e
1
1
.
0
0
0
.
0
1
0
.
0
4
0
.
2
5
0
.
0
4
0
.
1
3
0
.
0
6
1
.
0
0
N
o
t
e
:

A
l
l

e
n
t
r
i
e
s

f
r
o
m

f
n
a
l

f
u
l
l

P
L
S

m
o
d
e
l

(
M
o
d
e
l

4
)
;

d
i
a
g
o
n
a
l

e
n
t
r
i
e
s

(
b
o
l
d
)

a
r
e

t
h
e

s
q
u
a
r
e

r
o
o
t

o
f

t
h
e

a
v
e
r
a
g
e

v
a
r
i
a
n
c
e

e
x
t
r
a
c
t
e
d

(
A
V
E
)
.
Table 13.4
Internal Consistency, Construct Intercorrelations, and Average Variance Extracted
No. of
Indicators
Internal
Consistency
Construct No.
No. Construct 1 2 3 4 5 6 7
1 Knowledge sharing
effectiveness (KSE)
7 0.95 0.87
2 Expertise location 4 0.89 0.70 0.82
3 Task analyzability 4 0.94 0.60 0.54 0.89
4 Dynamic team
memory system use
(DTMS use)
1 1.00 0.22 0.07 0.25 1.00
Control Variables
5 Goal similarity 3 0.94 0.62 0.56 0.43 0.04 0.91
6 Open team
membership
1 1.00 0.01 0.06 0.26 0.08 0.12 1.00
7 Team task type 1 1.00 0.01 0.04 0.25 0.04 0.13 0.06 1.00
Note: All entries from fnal full PLS model (Model 4); diagonal entries (bold) are the square root of the
average variance extracted (AVE).
bold along the diagonal in Table 13.4. As is evident, the square root of average variance extracted
for each of the constructs is higher than the intercorrelations among constructs in the model.
Structural Model
The research model shown in Figure 13.1 was tested by conducting partial least squares (PLS)
using the PLS-Graph Version 3.0. A series of four PLS models were estimated in order to clearly
delineate the effects of the control variables, and to systematically test alternative models includ-
ing both direct and indirect (mediated) paths.
Control Variables Only (Model 1)
A model with only the nine paths from the three control variables to the three endogenous vari-
ables was estimated as a baseline and the results are presented in the top panel of Table 13.5.
Goal similarity signifcantly predicted task analyzability ( = 0.45, p < 0.01), expertise location
( = 0.57, p < 0.01), and knowledge sharing effectiveness (KSE) ( = 0.64, p < 0.01). In addition,
team task type was a signifcant predictor of task analyzability ( = 0.29, p < 0.01). The other fve
paths between the three control variables and the three endogenous variables were not signifcantly
different from zero. The control variables alone accounted for 32% of the variance in task analyz-
ability, 33% in expertise location, and 40% in KSE.
All nine control variable paths, including the fve not signifcantly different from zero, are
estimated in the following three models in order to provide a more conservative test of the hy-
pothesized path coeffcients in Figure 13.1. While the pattern of control variable path coeffcients
does not vary across the three following models (the same four paths are always signifcantly
different from zero and the other fve paths not different from zero), we include the estimates of
these nine paths for completeness.
Table 13.5
Path Coeffcients and Explained Variance for Nested PLS Models
Dependent Variables
Predictors Task Analyzability Expertise Location
Knowledge Sharing
Effectiveness
Model 1. Control variables only
Team task type 0.29** 0.03 0.09
Goal similarity 0.45** 0.57** 0.64**
Team membership 0.20 0.01 0.08
Total R2 0.32 0.33 0.40
Model 2. Dynamic Team Memory System (DTMS) use direct effects + control variables
Team task type 0.28** 0.03 0.08
Goal similarity 0.45** 0.57** 0.65**
Team membership 0.22* 0.00 0.07
DTMS use 0.27** 0.08 0.23*
Total R2 0.39 0.33 0.45
R2 from Model 1 0.07 0.00 0.05
F ratio R2 5.62* < 1.00 4.45*
Model 3. DTMS use mediated by task analyzability + control variables
Team task type 0.28** 0.09 0.05
Goal similarity 0.45** 0.38** 0.45**
DMS use 0.27** 0.00 0.12
Task analyzability 0.42** 0.43**
Total R2 0.39 0.44 0.56
R2 from Model 2 0.11 0.11
F ratio R2 9.43** 12.00**
Model 4. DTMS use mediated by task analyzability and expertise location + control variables
Team task type 0.28* 0.09 0.02
Goal similarity 0.45* 0.37** 0.30**
DTMS use 0.27** 0.03 0.13
Task analyzability 0.43** 0.26*
Expertise location 0.39**
Total R2 0.39 0.44 0.64
R2 change from Model 3 0.08
F ratio R2 10.44**
*p < 0.05; **p < 0.01.
DTMS Use Direct Effects + Control Variables (Model 2)
Direct effects of DTMS use on the three endogenous variables, over and above the three control
variables, were estimated and tested in Model 2 (Table 13.5). DTMS use predicts KSE ( = 0.23,
p < 0.05), suggesting a direct effect of DMS use on KSE when no other predictors (other than
controls) are entered. Likewise, the incremental KSE variance accounted for (R
2
= 0.450.40)
is signifcantly greater than zero [F(1,50) = 4.45, p < 0.05].
Model 2 is also used to test part of the hypothesis that task analyzability and/or expertise loca-
tion mediates the relationship between DTMS use and KSE. In order for such a mediated model
to hold, there must be a signifcant path coeffcient, over and above the control variables, between
DTMS use and either or both of the potential mediators. As can be observed from Model 2 results
(Table 13.5), DTMS use does predict task analyzability ( = 0.27, p < 0.01), and the increment
in task analyzability variance accounted for (R
2
= 0.390.32) is signifcantly greater than zero
[F(1,50) = 5.62, p < 0.05]. In contrast, the path coeffcient between DMS and expertise location
is not signifcantly different from zero. We can conclude from Model 2 that task analyzability is
a potential mediating variable between DTMS use and KSE but that expertise location does not
directly mediate this relationship.
DTMS Use Mediated by Task Analyzability + Control Variables (Model 3)
Direct effects of task analyzability on both expertise location and KSE are tested in Model 3.
If task analyzability mediates the relationship between DTMS use and KSE, then there should
be a signifcant path from task analyzability to KSE, and when this path is included, the path
coeffcient between DTMS use and KSE should not be different from zero. As can be seen from
Model 3 results (Table 13.5), task analyzability signifcantly predicts KSE ( = 0.43, p < 0.05),
and the increase in KSE variance accounted for (R
2
= 0.560.45) is signifcantly greater than
zero [F(1,48) = 12.00, p < 0.01]. This suggests that task analyzability is a signifcant predictor of
KSE, over and above DTMS use and the three control variables.
It is important that once DTMS use is entered into the PLS model, the path between DTMS
use and KSE is no longer signifcantly different from zero ( = 0.12, not signifcant). This result,
coupled with the signifcant paths between task analyzability from DTMS use and between KSE
from task analyzability, strongly suggest that consistent with our Hypothesis 1, task analyzability
mediates the relationship between DTMS use and KSE.
As it can be observed from Model 3 results (Table 13.5), consistent with Hypothesis 2, task
analyzability signifcantly predicts expertise location ( = 0.42, p < 0.01), over and above the
three control variables and DTMS use, and the increase in expertise location variance accounted
for (R
2
= 0.11) is signifcantly greater than zero [F(1,49) = 9.43, p < 0.01]. This signifcant path
tests one element of the hypothesis of an indirect path between task analyzability and KSE.
DTMS Use Mediated by Task Analyzability and Expertise Location + Control Variables (Model 4)
The fnal model adds a single additional path between expertise location and KSE. This path al-
lows us to test the direct effect of expertise location on KSE, and to establish whether there is an
indirect path between task analyzability and KSE through expertise location. As presented in the
Model 4 results (Table 13.5), expertise location is a signifcant predictor of KSE ( = 0.39, p <
0.01), and the increment in KSE variance accounted for (R
2
= 0.08) is also signifcantly greater
than zero [F(1, 48) = 10.44, p < 0.01]. This suggests that, consistent with our Hypothesis 2, there
is a direct effect of expertise location on KSE, over and above the combined effects of task analyz-
ability, DTMS use, and the three controls. It is also important to note that the direct path from task
analyzability to KSE remains signifcantly different from zero in Model 4 ( = 0.26, p < 0.05).
These two results for Model 4, coupled with the signifcant paths predicting expertise location
from task analyzability, indicate an indirect path from task analyzability to KSE, through expertise
location. This suggests that expertise location plays the role of a partial mediator, allowing task
analyzability to affect KSE both directly and indirectly. Like Model 3, there is no signifcant direct
path between DTMS use and KSE ( = 0.13, not signifcant).
Figure 13.2 provides a summary of the direct paths and R
2
for the fnal PLS model presented
in Model 4 in Table13.5. This model is completely consistent with the hypothesized model in
Figure 13.1. As it can be observed from Figure 13.2, the model explains a substantial proportion
of variance for the three endogenous constructs: knowledge sharing effectiveness (R
2
= 0.64), task
analyzability (R
2
= 0.39), and expertise location (R
2
= 0.44). All the R
2
values were greater than
the recommended 0.10 (Falk and Miller, 1992).
Overall, the results suggest that use of a DTMS has a mediated effect on knowledge sharing
effectiveness in structurally diverse distributed teams. The mediated effect is manifest through
increases in the perceived analyzability of the teams task. Likewise, task analyzability has both
direct and indirect (mediated) effects on knowledge sharing effectiveness through the extent to
which team members can locate expertise within their teams.
DISCUSSION
Our results supported the hypothesized associations among the constructs in our research model.
While common method variance can never be ruled out completely in a survey design, our dis-
criminant analysis (exploratory factor analysis in Table 13.2 and AVEs versus interconstructs
correlations in Table 13.4) indicated little common method variance. The inclusion of measures at
both the individual and team level, as well as incorporation of independent experts ratings in the
measure of DTMS use, may have also helped to avoid common method variance. In addition, our
testing for alternative paths in the PLS modelnone of which proved to be signifcantreduces
the likelihood that our results are due to common method variance.
Our fndings are based on a cross-sectional survey, making causal ordering diffcult to justify.
Figure 13.2 Final PLS (Model 4) Signifcant Path Coeffcients and R2
Note: Control variables and paths omitted.
USE OF
DYNAMIC TEAM
MEMORY SYSTEM
TASK
ANALYZABILITY
EXPERTISE
LOCATION
KNOWLEDGE
SHARING
EFFECTIVENESS
.27** .43**
.26*
.39**
R
2
=.39
R
2
=.44
R
2
=.64
Yet, we believe there are several reasons to infer that a causal ordering may be appropriate. First,
the fact that earlier research has already anecdotally demonstrated a causal link between virtual
workspace use and knowledge sharing effectiveness (Malhotra et al., 2001; Mark, 2002) sug-
gests that our causal ordering of salient constructs may be theoretically viable. Second, while
the selection of the technology features was made by team leaders at the inception of the team,
we measured the perceptions of task analyzability by specifcally asking team members to think
about how they were feeling at the time of our survey (on the average was 18 months after team
inception). This suggests that we have taken into account the time lag between technology imple-
mentation/use and task analyzability perceptions. Nevertheless, further longitudinal research is
needed to confrm fndings.
In summary, our results suggest that the more that the virtual workspace technology deployed
by the structurally diverse distributed team enables development and use of a DTMS (a form of
an organizational memory information system) and increases team members task analyzability
perceptions, the greater knowledge sharing experienced by team members. Further, the greater is
the clarity in expertise location experienced by team members due to the use of DTMS, but only if
the DTMS increases task analyzability perceptions. Last, the greater the expertise location reported
by team members, the greater is the knowledge sharing effectiveness in the team.
We believe our research has several implications. First, we have introduced a richer way to
measure technology use that integrates three related perspectives: task-technology ft (Goodhue
and Thompson, 1995), technology spirit (DeSanctis and Poole, 1994), and implementation of
specifc features (Zigurs and Buckland, 1998). We believe this richer measure not only integrates
the three perspectives but also ensures that no single perspective is given preference or ignored
when understanding the role of technology in organizational performance. Our hope is that our
measure will inspire similarly rich assessments of technology use in future research on informa-
tion technologies in general.
Second, our fndings demonstrate that the Stein and Zwass (1995) organizational memory in-
formation system theoretical framework is helpful in predicting why and how virtual workspaces
are helpful to members of structurally diverse distributed teams. Their framework suggested that
knowledge sharing is more likely to be effective when the information system used by the team
is designed to meet three requirements:
Intelligently, continuously, and easily acquire and encode knowledge from information
producers.
Maintainintegrityofknowledgeovertime.
Patternmatchduringsearchandretrieval.
Our application of this framework to virtual workspace technologies, labeled DTMSs, sug-
gests the promise of this framework for research on virtual team technologies. This framework
has two signifcant implications for virtual team research literature. First, this framework may
help to explain why some structurally diverse distributed teams need face-to-face interaction
and others are able to succeed without face-to-face meetings. We argue that published studies
describing teams needing face-to-face meetings typically use technologies that fail to meet these
three design requirements. Mere use of e-mail coupled with audioconferences does not allow for
maintenance of knowledge integrity over time, encoding of the acquired knowledge, nor pattern
matching during retrieval. Further, the use of simple team repositories (e.g., fle folders) rarely
allows for continuous acquisition of knowledge, or provides a means to maintain integrity over
time, or pattern matching during retrieval. Our fndings of structurally diverse teams who rarely
met face-to-faceyet are able to share knowledge effectivelyindicates that the presence of these
three design requirements facilitates knowledge sharing effectiveness in the absence of face-to-face
meetings. Therefore, our fndings suggest that virtual team researchers should assess the degree
to which the technologies used by virtual team members conform to these design requirements
before concluding that face-to-face interaction is needed.
The organizational memory framework as applied to virtual team research also suggests how
to use collaborative technologies to help distributed team members resolve task confict. Based
on their review of the literature on electronically mediated confict in virtual teams, Hinds and
Bailey (2003) conclude; Research to date reveals that problems arise from nearly every available
technology. . . . On these grounds, we propose that technology mediation engenders confict of
all types in distributed teams (p. 622). We believe that the organizational memory framework as
applied in this chapter identifes design requirements that need to be met to engender less confict,
rather than assume, as do Hinds and Bailey, that all collaborative technologies result in confict.
The organizational memory framework suggests that confict in distributed teams can be avoided
(or minimized) if team memory systems are constructed to foster ongoing knowledge acquisition
that is encoded as it is gathered, in which integrity of the knowledge is monitored as it evolves,
and that, when searching to retrieve knowledge, identifes patterns in the knowledge that arise.
Moreover, our research suggests that such systems are in place today; we do not need to wait for
future technology evolutions. Thus, rather than assuming that technology mediation will create
task confict, our framework offers guidance on how the technology can evolve to identify ways
to reduce task confict among distributed teams. Since we did not explicitly examine task confict,
we hope that our research will encourage future research to move in this direction.
A third implication of our fndings is to the task analyzability literature. Researchers continue
to report that they are surprised that task analyzability is related to increased rather than decreased
information processing (Daft and MacIntosh, 1981). We believe the surprise of researchers stems
from the continuing assumption in this literature that task analyzability is a causally prior in-
put variable (Rice, 1992, p. 479). In the spirit of adaptive structuration theory (DeSanctis and
Poole, 1994), an implication of our fndings is for researchers to reconsider if task analyzability
is indeed causally prior and independent of technology or is more properly treated as a variable
characterizing both the technology and task environment in which the team performs. As such,
task analyzability may be an outcome of how technology is used, not simply a moderator or con-
textual variable for technology.
We theorized that the mediating role of task analyzability is due to the presence of routines about
how to bring past knowledge to bear on current workroutines that are enabled by DTMS use.
We did not specifcally measure the use of routines. Instead we relied on a more general assess-
ment of perceived task analyzability. Future research should explicitly examine the form of these
routines. The routines may be norms for keywords, knowledge access, and retrieval, as suggested
by Bordetsky and Mark (2000), or they may be common artifacts, as suggested by Carlile (2002).
In any case, it is also possible that alternative theories may offer equally plausible explanations for
the mediating role of task analyzability. One such theory suggests that information sharing will
occur when members are interdependent and it is an accepted cultural norm to share (Jarvenpaa
and Staples, 2000). Task analyzability may then be measuring not the use of routines, but a more
general attitude that sharing is important and all members of the team should therefore exchange
information. If such an explanation was found to be supported by follow-on research, this would
suggest that, as long as a collaborative tool was used to create a culture of sharing, the precise
functionalities of that tool are not important.
Another plausible theory explaining the role of task analyzability is information load theory,
in which people are unable to process information when they are unable to control the selecting,
sorting, and organizing of the information (Schultze and Vandenbosch, 1998). Task analyzability,
then, may be a measure of control over information load. If true, then, as long as a collaborative
technology was used in a way that fostered user control, the precise functionalities of that technol-
ogy may not be important.
Yet another theory explaining the role of task analyzability is that of presence awareness (Dour-
ish and Belotti, 1992; Jang et al., 2000). This theory suggests that task analyzability provided by
these tools indicate that users have a greater understanding of the situational context of others on
the team (Cramton, 2001). Such a theory then suggests that any tool that provides presence aware-
ness (e.g., instant messaging) may be suffcient for knowledge sharing effectiveness.
Each of these theories has been proposed in the literature independent of each other. Future
research should attempt to test alternative theories in the same study. Research is needed that ex-
amines different psychological states of users (e.g., control over information load, awareness of
others presence, general information sharing culture, expectations of specifc routines for bringing
past knowledge to bear) as they interact with the technology. Such research may conclude that
all of these states are needed for people to share knowledge effectively through electronic media,
resulting in a richer and more comprehensive theory.
Our application of the organizational memory information systems framework to virtual work-
spaces is primarily focused on the information processing and communication support features
of group support systems (Zigurs and Buckland, 1998). Process-structuring support from such
features as agenda setting, agenda enforcement, decision making, voting, etc., were not covered
in our framework, in large part because our sample of teams made little use of these features.
However, we suggest that the organizational memory information systems framework might be
extendable to include process-structuring features. For example, can the voting feature be conceived
not simply as process structuring, but as one feature meeting the design requirement of intelligent
ongoing, easily encoded knowledge acquisition from information producers? We believe that vot-
ing provides more than the simple capability to structure participation about an issue and allow
anonymity; following the organizational memory information systems framework, we believe that
electronic voting may provide an ongoing means of capturing, maintaining, and then observing the
evolution of the teams memory over time. That is, votes are not simply taken and then discarded
in a DTMS. Votes can be stored and trends over time in the ability of the team to quickly reach
unanimous agreement can be observed with such a feature. As process structuring is increasingly
used in DTMSs, future research should address this issue.
For practitioners, this study suggests principles for designing and using virtual workspaces.
As distributed teams become increasingly structurally diverse, these fndings suggest that such
teams will beneft from technology features that meet the three requirements of organizational
memory information systems. E-mail and occasional audioconferences are just not suffcient,
nor is the occasional post to the teams Web site. However, DTMSs will help only to the degree
that they are perceived by users to increase analyzability of the teams task. If poor training, in-
consistent management direction, or poorly followed norms of use (Bordetsky and Mark, 2000;
Fulk, Schmitz, and Steinfeld, 1990) occur, the virtual workspace will be of no greater value than
a poorly conceived and executed face-to-face meeting.
In conclusion, although one of the oft-cited benefts of the virtual team is the ability of its
members to contribute diverse knowledge and expertise, the question of how virtual team mem-
bers can bring their respective knowledge and expertise to solve the problems they face has been
largely ignored in past research on virtual teams (Yoo and Kanawattanachai, 2001, p. 187). We
have sought to address this issue in the hopes of encouraging other research to follow.
ACKNOWLEDGMENTS
We would like to thank the Society for Information Management, Netage, Kathy Chudoba, and
Groove for helping us to identify the sample. We would also like to thank Samer Faraj, Sanjay
Gosain, Arjan Raven, Huseiyn Tardivat, Sabine Hirt, and Claudia Kubowicz Malhotra for provid-
ing comments that were helpful in revising this manuscript, as well as the reviewers for inspiring
us to push harder in our thinking. We would also like to thank the members of our expert panel
for providing us their invaluable insights regarding workspace technologies that helped guide our
research.
REFERENCES
Ackerman, M.S. 1994. Augmenting the organizational memory: A feld study of answer garden. In Proceed-
ings of the ACM Conference on Computer-Supported Cooperative Work (CSCW 94), pp. 243252.
Ackerman, M.S., and Malone, T.W. 1990. Answer Garden: A tool for growing organizational memory. In
Proceedings of the ACM Conference on Offce Information Systems, pp. 3139.
Becerra-Fernandez, I., and Sabherwal, R. 2001. Organizational knowledge management: A contingency
Benbasat, I., and, Lim, L. 1991. The effects of group, task, context, and technology variables on the useful-
ness of group support systems: A meta-analysis of experimental studies. Small Group Research, 24, 4,
430462.
Bollen, K.A. 1989. Structural Equations with Latent Variables. New York: Wiley.
Bordestky, A., and Mark, G. 2000. Memory-based feedback controls to support groupware coordination.
Information Systems Research, 11, 4, 366385.
Carlile, P.R. 2002. A pragmatic view of knowledge and boundary objects in new product development.
Chin, W.W. 1998. The partial least squares approach for structural equation modeling. In G.A. Marcoulides
(ed.), Modern Methods for Business Research. Mahwah, NJ: Lawrence Erlbaum, pp. 295336.
Conklin, J. 1993. Capturing organizational memory. In R.M. Baecker (ed.), Readings in Groupware and
Computer-Supported Cooperative Work: Assisting Human-Human Collaboration. San Francisco, CA:
Morgan Kauffmann, pp. 133137.
Cramton, C. 2001. The mutual knowledge problem and its consequences for dispersed collaboration. Orga-
Cummings, J.N. 2004. Work groups, structural diversity, and knowledge sharing in a global organization.
Daft, R.L., and Macintosh, N.B. 1981. A tentative exploration in the amount and equivocality of information
processing in organizational work unit. Administrative Science Quarterly, 26, 207224.
Dennis, A.R., and Kinney, S.T. 1998. Testing media richness theory in new media: The effects of cues,
feedbacks, and task equivocality. Information Systems Research, 9, 3, 256274.
DeSanctis, G., and Jackson, B. 1994. Coordination of information technology management: Team-based
structures and computer-based communication systems. Journal of Management Information Systems,
10, 4, 85111.
DeSanctis, G., and Poole, M. 1994. Capturing the complexity in advanced structuration theory. Organization
Science, 5, 2,121147.
Dourish, P., and Bellotti, V. 1992. Awareness and coordination in shared workspaces. In Proceedings of the
ACM Conference on Computer Supported Cooperative Work (CSCW 1992), pp. 107114.
Falk, R.F., and Miller, N.B. 1992. A Primer for Soft Modeling. Akron, OH: University of Akron Press.
Faraj, S., and Sproull, L. 2000. Coordinating expertise in software development teams. Management Sci-
ence, 46, 15541568.
Fornell, C., and Larcker, D. 1981. Evaluating structural equation models with unobservable variables and
measurement error. Journal of Marketing Research, 18, 1, 3950.
Fulk, J.; Schmitz, J.A.; and Steinfeld, C. 1990. A social infuence model of technology use. In J. Fulk
and C. Steinfeld (eds.), Organizations and Communication Technology. Newbury Park, CA: Sage, pp.
117140.
Goodhue, D.L., and Thompson, R.L. 1995. Task-technology ft and individual performance. MIS Quarterly,
213236.
Grant, R.M. 1996. Prospering in dynamically competitive environments: Organization capability as knowledge
integration, Organization Science, 7, 4, 375387.
Greenberg, S., and Roseman, S. 2003. Using a room metaphor to ease transition in groupware. In M. Acker-
man, V. Pipek, and V. Wulf (eds.), Sharing Expertise: Beyond Knowledge Management. Cambridge, MA:
MIT Press, pp. 203256.
Grey, P.H., and Meister, D.B. 2004. Knowledge sourcing effectiveness. Management Science, 50, 6,
821834.
Griffth, T.L., and Neale, M.A. 2001. Information processing in traditional, hybrid, and virtual teams: From
nascent knowledge to transactive memory. Research in Organizational Behavior, 23, 379421.
Griffth, T.L.; Sawyer, J.E.; and Neale, M.A. 2003. Virtualness and knowledge in teams: Managing the love
triangle of organizations, individuals, and information technology. MIS Quarterly, 27, 2, 265287.
Hinds, P.J., and Bailey, D.E. 2003. Out of sight, out of sync: Understanding confict in distributed teams.
Hollingshead, A.B. 1998. Retrieval processes in transactive memory systems. Journal of Personality and
Social Psychology, 74, 659671.
Hulland, J. 1999. Use of partial least squares (PLS) in strategic management research: A review of four
studies. Strategic Management Journal, 20, 195204.
James, L.R.; Demaree, R.G.; and Wolf, G. 1984. Estimating within-group inter-rater reliability with and
without response bias. Journal of Applied Psychology, 69, 85.
Jang, C.Y.; Steinfeld, C.; and Pfaff, B. 2000. Supporting awareness among virtual teams in a web-based
collaborative system: The TeamSCOPE system. SIGGROUP Bulletin, 21, 3, 2834.
Janz, B.D.; Colquitt, J.A.; and Noe, R.A. 1997. Knowledge worker team effectiveness: The role of au-
tonomy, interdependence, team development and contextual support variables. Personnel Psychology,
50, 877904.
Jarvenpaa, S.L., and Staples, D.S. 2000. The use of collaborative electronic media for information sharing:
An exploratory study of determinants. Journal of Strategic Information Systems, 9, 129154.
Jehn, K.A. 1995. A multimethod examination of the benefts and detriments of intragroup confict. Admin-
istrative Science Quarterly, 40, 256282.
Kirsch, L.J., Sambarmurthy, V., and Ko, D.G. 2002. Controlling information systems development projects:
The view from the client. Management Science, 48, 4, 484498.
Lewis, K. 2003. Measuring transactive memory systems in the feld: Scale development validation. Journal
of Applied Psychology, 88, 4, 587604.
Linger, H.; Burstein, F.; Zaslavsky, A.; and Crofts, N. 1999. A framework for a dynamic organizational
memory information system. Journal of Organizational Computing and Electronic Commerce, 9, 2 &
3, 189203.
through IT-enabled contextualization: Evidence from distributed teams. Information Systems Research,
16, 1, 927.
Majchrzak, A.; Malhotra, A.; Stamps, J.; and Lipnack, J. 2004. Can absence make a team grow stronger?
Harvard Business Review, 82, 5, 131137.
Malhotra, A.; Gosain, S.; and El Sawy, O.A. 2005. Absorptive capacity confgurations in supply-chains:
Gearing for partner enabled market knowledge creation. MIS Quarterly, 29, 1, 145187.
Malhotra, A., and Majchrzak, A. 2004. Enabling knowledge creation in far-fung teams: Best practices for
IT support and knowledge sharing. Journal of Knowledge Management, 8, 4, 7588.
. 2006. Teams across borders. Financial Times: Mastering Uncertainty Series, March 21, 11.
Malhotra, A.; Majchrzak, A.; Carmen, R.; and Lott, V. 2001. Radical innovation without collocation: A case
study of Boeing-Rocketdyne. MIS Quarterly, 25, 2, 229249.
Mark, G. 2002. Conventions for coordinating electronic distributed work: A longitudinal study of groupware
use. In P.J. Hinds and S. Kiesler (eds.), Distributed Work. Cambridge, MA: MIT Press.
Maznevski, M.L., and Chudoba, K.M. 2000. Bridging space over time: Global virtual team dynamics and
effectiveness. Organization Science, 11, 473492.
Nunally, J.C., and Bernstein, I.H. 1994. Psychometric Theory, 3rd ed. New York: McGraw-Hill.
Perrow, C.A. 1967. Framework for the comparative analysis of organizations. American Sociological Review,
32, 194208.
Rice, R.E. 1992. Task analyzability, use of new media, and effectiveness: A multi-site exploration of media
richness. Organization Science, 3, 4, 475500.
Schultze, U., and Vandenbosch, B. 1998. Information overload in a groupware environment: Now you see it,
now you dont. Journal of Organizational Computing and Electronic Commerce, 8, 2, 127148.
Szulanski, G. 2000. The process of knowledge transfer: A diachronic analysis of stickiness. Organizational
Behavior and Human Decision Processes, 82, 1, 927.
Walsh, J.P., and Ungson, G.R. 1991. Organizational memory. The Academy of Management Review, 16, 1,
5791.
Wegner, D.M. 1987. Transactive memory: A contemporary analysis of the group mind. In B. Mullen and
G.R. Goethals (eds.), Theories of Group Behavior. New York: Springer-Verlag, pp. 185208.
Werts, C.E.; Linn, R.L.; and Joreskog, K.G. 1974. Interclass reliability estimates: Testing structural assump-
tions. Educational and Psychological Measurement, 34, 2533.
Withey, M.; Daft, R.L.; and Cooper, W.H. 1983. Measures of Perrows work unit technology: An empirical
assessment and a new scale. Academy of Management Journal, 26, 1, 4563.
Yoo, Y., and Kanawattanchai, P. 2001. Development of transactive memory systems and collective mind in
virtual teams. International Journal of Organizational Analysis, 9, 2, 87208.
Zahra, S.A., and George, G. 2002. Absorptive capacity: A review, reconceptualization, and extension, Academy
of Management Review, 27, 2, 185203.
Zeffane, R.M., and Gul, F.A. 1993. The effects of task characteristics and sub-unit structure on dimensions
of information processing. Information Processing & Management, 29, 6, 703720.
Zigurs, I., and Buckland, B.K. 1998. A theory of task/technology ft and group support systems effectiveness.
MIS Quarterly, September, 313334.
270
ChaptEr 14
ELECTRONIC KNOWLEDGE NETWORKS
Processes and Structure
samEr faraj, molly mClUrE wasKo, and stEVEn l. johnson
Abstract: Advances in information and communication technologies have enabled the creation of
online communities, where individuals congregate via a shared technology to engage in a variety
of social interactions. In addition to purely social communities (such as Facebook or MySpace),
some communities develop for the primary purpose of knowledge exchange, creating electronic
communities of practice akin to an online help desk. We refer to these communities as electronic
knowledge networks (EKNs). Given the success and proliferation of EKNs openly available on
the Internet, organizations are investigating the creation and management of EKNs as a crucial
knowledge management technology to support intraorganizational knowledge exchange. This
chapter reviews the literature on two key issues that are essential for understanding knowledge
exchange in EKNs: why people participate and how these networks are structured to ensure sus-
tainability over time. Findings suggest that individuals participate in EKNs due to self-interest,
such as gaining access to information and advice, and enhancing ones reputation or infuence.
In addition to self-interest, research also indicates that participation in EKNs is also driven by
high levels of social capital, such as feelings of trust and obligation, that stem from the social
interactions in the network. In terms of network sustainability, research indicates the EKNs are
predominately structured as scale-free networks, where the vast majority of effort is contributed
by a small minority of individuals. Participation in these networks also exhibits certain underlying
patterns of exchange, such as patterns of reciprocity, generalized exchange, and, fnally, preferential
attachment. This chapter ends by suggesting the implications of this research for organizations,
and areas in need of future research.
Key Words: Electronic Knowledge Networks, Electronic Networks of Practice, Social Networks,
Network Analysis, Online Communities, Virtual Communities
INTRODUCTION
Practitioners and researchers in the area of knowledge management have focused almost exclu-
sively on how knowledge is stored, transferred, and used within organizations. However, much
knowledge is exchanged in open online forums that cross organizational boundaries. We defne an
electronic knowledge network (EKN) as a self-organizing, open activity system that focuses on a
shared interest or practice and exists primarily through computer-mediated communication. These
networks are sustained through a range of technology including e-mail (Finholt and Sproull, 1990;
Wu, Huberman, Adamic, and Tyler, 2004), USENET newsgroups (Butler, 2001; Jones, Ravid, and
ELECTRONIC KNOWLEDGE NETWORKS 271
Rafaeli, 2004), and organizational discussion groups (Constant, Sproull, and Kiesler, 1996; Ravid
and Rafaeli, 2004; Wasko and Faraj, 2005). In EKNs, individuals congregate based on shared in-
terest and use the asynchronous nature of the technology to overcome the same-place/same-time
limitation inherent in face-to-face settings. Activities such as the exchange of technical advice now
link thousands of people who lack previous familiarity and physical proximity and are unaware
of each others demographic characteristics, organizational setting, or even national culture (Fulk
and DeSanctis, 1995; Preece and Maloney-Krichmar, 2003; Sproull and Faraj, 1997; Sproull and
Kiesler, 1991; Wellman and Gulia, 1999).
EKNs are important new ways of organizing that have not been well studied so far (Butler,
2001; Fulk and Desanctis, 1995). As with any new phenomenon, a number of researchers have
offered different defnitions, each based on a specifc emphasis. For example, EKNs focused on
practice have been called electronic networks of practice (Wasko, Faraj, and Teigland, 2004). Those
focused on a task have been called virtual groups (Ahuja, Galletta, and Carley, 2003), and those
focused on shared interests and social support have been called virtual community (Rheingold,
1993). We use the EKN defnition to emphasize the communality of these collectives in terms of
electronic mediation, knowledge exchange orientation, and social network structure. This chapter
is based on the premise that knowledge exchange in EKNs is sustained by both individual and
network infuences, and that the networks structure of ties infuences individual participation and
network dynamics, which results in the exchange of messages that add value to the network and
create a public good of collective knowledge.
EKNs overcome the limitations of knowledge repositories by supporting knowledge exchange
and new knowledge creation through conversation rather than static documentation. The advice
accessed through EKNs is useful because EKNs provide access to knowledge from more diverse
resources and from better resources (Constant et al., 1996). EKNs are critical information resources
for knowledge workers because they bridge the gap between knowledge available in codifed
form (including knowledge stored in knowledge management systems [KMS]) and the actual
requirements of knowledge work (Brown and Duguid, 1991; Wenger, 1998). EKNs overcome
the limitations of knowledge maps by promoting self-identifcation of expertise, encouraging the
voluntary participation of individuals willing and able to assist, and creating a codifed record of
the knowledge as it is exchanged. These EKNs are also having an impact on a number of organi-
zational activities. For example, electronic networks support distributed R&D efforts (Orlikowski,
Yates, Okamura, and Fujimoto, 1995) and enable open source software development (von Hippel
and von Krogh, 2003) and extraorganizational work practices (Wasko and Faraj, 2005).
The purpose of this chapter is to summarize how EKNs sustain themselves by promoting vol-
untary participation and knowledge exchange. We focus on two major aspects. First, we examine
the factors that explain individual participation and knowledge exchange behaviors with an em-
phasis on individual motivations, social ties, and critical mass theories. Understanding why and
how individuals exchange knowledge in EKNs is an important factor for the success of KMS, as
EKNs help fll the knowledge gaps left by knowledge repositories and knowledge maps. Next, we
investigate studies that explore structural factors associated with knowledge exchange in EKNs.
Recent advances in social network analysis have enabled the study of large-scale networks based
on network properties and are useful for deepening our understanding of interaction patterns on
EKNs. We offer a summary and an evaluation of recent empirical and simulation work that focuses
on whether interaction patterns follow competing theories of reciprocal exchange, generalized
exchange, or critical mass. In summary, this chapter contributes to theories of new organizational
forms by answering open questions as to the structure of EKNs (DeSanctis and Monge, 1999) and
offers an early evaluation of the social and structural factors that both drive and limit individual
272 FARAJ, WASKO, AND JOHNSON
participation in EKNs. Finally, from this emergent body of research fndings, some evidence-based
recommendations can be made for practitioners and organizations concerned with the building
of effective EKNs.
THE ROLE OF TECHNOLOGY IN ELECTRONIC
KNOWLEDGE NETWORKS
The purpose of this chapter is to examine how and why individuals use technology to voluntarily
exchange advice and ideas in EKNs. It should be noted that one of the most prevalent theories shap-
ing our understanding of technology-supported communication to date is the media richness theory
(Daft and Lengel, 1986). This theory proposes that communication media can be arrayed along a
continuum of media richness based on differing objective characteristics of the communication
channel. Media richness has been referred to as a cues fltered out approach because it predicts
that the richness of the communication media will result in predictable changes in intrapersonal
and interpersonal variables (Culnan and Markus, 1987). Media richness has been interpreted
to suggest that because media forms differ in their ability to convey information, technology
imposes constraints that restrict and ultimately determine the outcomes of actors. This objective,
deterministic view of technology is not generally appropriate and is particularly inadequate in the
case of information technologies (Orlikowski, 1992). According to Dennis and Kinney (1998), the
limiting factor may not be the communications media but rather our preconceived perceptions
of their limitations based on media richness theory.
In reality, despite the pessimistic characterization of computer-mediated communication as a
lean channel of exchange, there is ample evidence that people are using technology to support a
wide variety of social interactions. Communication media have objective characteristics that deter-
mine the capacity of the media to carry information but in using technology, individuals interpret,
manipulate, and appropriate technology in different ways. Technology is physically constructed
but is also socially constructed by users through the various features they emphasize and use, as
well as the meanings that individuals attach to the technology (Orlikowski, 1992; Panteli, 2004).
Media richness has become a popular straw man theory, set up to be easily refuted. There are
currently such a wide variety of electronically enabled collaborative networks, contrary to the early
prediction of media richness theory, that media richness now has little validity in predicting how
media are used today. Electronically enabled collaborative network activities range from primarily
social purposes, such as social software (e.g., facebook.com) and personal blogs (e.g., livejournal.
com) to informational collaboration and repository sites such as Wikipedia. There are sites that
started primarily for the purpose of e-commerce but evolved into EKNs, such as the BTX Web
site where Bruce Springsteen fans started, using the ticket exchange functionality to send mes-
sages to each other (Culnan, 2005). Technology is used to coordinate the efforts of a worldwide
network of volunteers devoted to the creation of complex software through open source software
development sites like sourceforge.com. Additionally, collaborative technologies are no longer
limited to text-based asynchronous exchange. The technology supports both synchronous (i.e.,
chat and instant messaging) and asynchronous exchanges consisting of text, voice, graphics, or
a combination of these.
Rather than compare and contrast how these different characteristics of the technology infu-
ence knowledge exchange across the different forms, we limit the scope of this chapter to focus
solely on EKNs that are text based and asynchronous and support the exchange of special interest
or practice-related knowledge through message posting and response.
Individuals in EKNs limited to text have found ways to adapt this technology to convey a variety
of signals and symbols. Individuals create faces using colon and parentheses keys (emoticons)
to convey emotion. Individuals SHOUT AT EACH OTHER using capital letters to convey anger
or frustration or to gain attention. Individuals translate common sayings through the use of the
symbols available in text, such as let me add my $1/50 worth, or by creating shorthand symbols
(FWIW, BTW, IMHO). Individuals build complex pictures using text characters to imitate the
drawing of lines. Thus, despite the objective aspect of the technology that limits communication
to text, the use of the technology is adapted, based on social context, to suit the various needs and
practices of communicating individuals.
How Technology Affects Participation Dynamics in Electronic
Knowledge Networks
In EKNs, although technology is adapted to meet the needs of the users, the primary communica-
tion channel of asynchronous computer-mediated communication does infuence how knowledge
is shared and exchanged. First, knowledge must be codifed in the form of a text-based message
and posted to an electronic discussion forum. Participation generally includes posting questions
with various purposes, such as seeking a specifc answer or general advice or opening a debate.
Participation can also include posting of announcements, messages of gratitude, follow-up ques-
tions, or messages intended to infame others. However, the principal source of value in EKNs
comes from the response messages posted to address the concerns of forum participants. Posting
responses provides a dual value. First, it directly helps the knowledge seeker. Additionally, pas-
sive knowledge seeking occurs through lurkingreading the postings of others without actively
engaging through public posting. Messages and their contents are typically made available to all
participants in the network regardless of an individuals participation in the original exchange.
The messages are often archived for future reference, thereby creating a searchable electronic
help desk. It is also worth noting that this visibility provides network members with complete
information about the online conduct of other network members and sharply contrasts with the
ephemeral, typically private conversations between a limited number of individuals that occur in
face-to-face communication.
Another unique aspect of EKNs is that they are generally self-organizing, in that they consist
of individuals who voluntarily choose to participate. The technology underlying EKNs supports
communication between thousands, even tens of thousands of people (Sproull and Faraj, 1995).
EKNs also provide a high level of anonymity to participants. Individuals are able to choose what
personal information they reveal, and there is little way to know if individuals portray accurate
online depictions. Individuals also choose when they participate and how often and the level of
effort they are willing to invest. These characteristics of voluntary and relatively anonymous
participation create situations in which the barriers to entry and exit are low, enabling individu-
als to actively participate for a time and then withdraw, or to ask a question without intending to
become an invested participant.
The technology underlying EKNs creates obvious benefts to knowledge seekers. The personal
costs of posting a question to an EKN are generally small, as there is both relative anonymity
and little social stigma incurred by seeking advice from others in these forums. EKNs provide
seekers with access to expertise distributed throughout the world, regardless of physical location
or personal familiarity. The effort required to post a message electronically is low, and experts
typically have no expectations that seekers help others, or even return to the EKN. In sharp con-
trast, there are potentially high costs associated with seeking knowledge from personal ties, such
as expectations of reciprocity or a sense of embarrassment about admitting ignorance on a topic
(Borgatti and Cross, 2003). Research indicates that knowledge seekers are typically motivated to
participate in EKNs for personal gain, by receiving answers to a specifc question, having ques-
tions answered quickly, and fnding valuable information that is more up to date than other sources
(Wasko and Faraj, 2000). Not only do individuals exchange advice in EKNs, but they also share
tools and computer code that are not available through other means, often for free (Wasko and
Faraj, 1999, 2000).
However, the primary problem with seeking knowledge on EKNs is that it requires depend-
ing upon the kindness of strangers (Constant et al., 1996). Because participation in EKNs is
open and voluntary, participants are typically strangers. Knowledge seekers have no control over
who responds to their questions or the quality of the responses. There are also signifcant costs
associated with contributing knowledge to others. Contributing knowledge eventually causes the
possessor to lose their unique value relative to what others know (Thibaut and Kelley, 1959) and
benefts all network members except the contributor (Thorn and Connolly, 1987). Knowledge
contributors have no assurances that those they are helping will ever return the favor, while seekers
and lurkers may draw upon the knowledge of others without contributing anything in return. As
management in many organizations has discovered, the availability of electronic communication
technologies is no guarantee that knowledge sharing will actually take place (Alavi and Leidner,
1999; Orlikowski, 1996). While interest in the organizing processes of EKNs continues to grow,
we know surprisingly little about how or why these networks are sustained (Desanctis and Monge,
1999; Lin, 2001). As noted above, it is easier to understand why individuals turn to EKNs when
they need advice. However, the primary source of value in EKNs comes from the propensity of
individuals to provide their valuable knowledge and insights to strangers (Kollock and Smith, 1996;
Rheingold, 1993; Wasko and Faraj, 2000), but why individuals participate in this activity when
there is no obvious beneft to them is not well understood. The purpose of the following section
is to investigate why people invest their valuable time and efforts to help strangers in EKNs.
WHY PEOPLE CONTRIBUTE KNOWLEDGE AND HELP OTHERS IN
ELECTRONIC KNOWLEDGE NETWORKS
Theories of collective action and public goods provide an overarching framework for the ex-
amination of why people contribute knowledge in EKNs. One unique characteristic of EKNs is
that the interaction, posting and responding to messages, produces and maintains a continuous
stream of relevant practice knowledge that all participants may access. These messages become
a public good, available to everyone with an interest in the network. Public goods are typically
associated with two characteristics: nonrivalry and nonexcludability. The most basic defnition of
a public good is that it is a good that is nonrival, meaning a good that is not used up or depleted
in its consumption (Shmanske, 1991). Typical examples of public goods include tangible goods
such as public parks and lighthouses and intangible or information goods such as public televi-
sion and radio. The second characteristic often associated with public goods is nonexcludability
(Head, 1962). Nonexcludability occurs when noncontributors cannot be excluded from consum-
ing the good. The nonrival nature of a public good allows the good and its beneft to be offered
to everyone in the collective, and nonexcludability infuences individual decision making about
participation in the production, maintenance, or consumption of the public good. In particular, the
issue of nonexcludability results in the tendency to free-ride, that is, to consume the public good
without contributing to its production or maintenance. In fact, the optimal individual decision is
to free-ride and consume the public good without contributing anything in return.
However, if everyone decided not to contribute, the public good would not be created and ev-
eryone in the collective would be worse off. The paradox that rational individual decisions lead to
collectively irrational outcomes suggests that it is impossible for rational individuals to cooperate
and engage voluntarily in the creation of public goods. As with the production of any public good
by a collective, the ability of individuals to free-ride on the efforts of others without contribut-
ing in return poses a major threat to the development and sustainability of EKNs. Thus, the key
factor underlying the success and sustainability of these electronic networks is the propensity of
individuals to voluntarily participate and provide their valuable knowledge and insights to others,
even though there are no assurances that anyone will reciprocate the favor (Kollock and Smith,
1996; Rheingold, 1993; Wasko and Faraj, 2000).
EKNs and other public goods are often created and maintained through collective action, even
in the face of self-interest and the ability to free-ride. Collective action theories attempt to explain
why individuals forego the tendency to free-ride and instead contribute to collective benefts. In
EKNs, the collective action is exhibited through the activity of posting and responding of messages
to the network. Collective action is described as action based solely on the voluntary cooperation
of individuals (Marwell and Oliver, 1993) and typically involves the production of a public or
semi-public good (Heckathorn, 1996). In contrast to theories that focus on the voluntary actions
of individuals in isolation (TAM), collective action theories take into account that individuals
and their actions are interdependent with others. These interactions stemming from individuals
engaged in a shared purpose become a signifcant source of social capital that can be leveraged
to sustain collective action.
Social capital is typically defned as resources embedded in a social structure that are ac-
cessed and/or mobilized in purposive action (Lin, 2001, p. 29). In recent years, social capital
concepts have been offered as explanations for a variety of prosocial behaviors, including col-
lective action, community involvement, and differential social achievements, that the concept
of individual-based capital (such as human or fnancial capital) is unable to explain (Coleman,
1990). There are two widely recognized constructs underlying social capital, structural social
capital, and relational social capital (Nahapiet and Ghoshal). Structural social capital refects the
overall structure of ties in the network, and relational social capital takes into account that these
structural ties vary in their relational strength or quality (Granovetter, 1983). Relational social
capital is important to network dynamics because it infuences the development of common
understandings, norms, and cooperation among network members (Coleman, 1990). The key
difference between social capital and other forms of capital is that social capital is embedded in
the social realm. While other forms of capital are based on assets or individuals, social capital
resides in the fabric of relationships between individuals and in individuals connections with
their communities (Putnam, 1995).
In EKNs, the posting and responding to messages represent interactions between individuals,
creating this sense of mutual interest and engagement in a shared practice. Therefore, in addition
to the sum of the attributes of the individuals in the network, collective action theories focus on the
additional importance of network infuences in encouraging voluntary contributions. This suggests
that in order to understand the dynamics of knowledge exchange in EKNs, we must take a multilevel
approach and examine both individuals and their attributes and the unique relational and social
structures that emerge at the network level. Collective action theories focused on understanding
the production of a public good break down into three interrelated areas: (1) the attributes of the
individuals in the collective; (2) the development of the relational dimension of social capitalfor
example, commitment and cohesion; and (3) the structural component of social capital specifying
the pattern of ties between individuals in the collective creating an overall network structure. Each
of these areas and how they pertain to EKNs are reviewed in the following sections.
Individual Motivations and Resources
The frst stream of collective action research proposes that the populations heterogeneity of re-
sources and interests affects collective action and the production of public goods (Hardin, 1982;
Olson, 1965; Oliver, Marwell, and Teixeira, 1985). In most volunteer networks, individuals have
differing levels of interest in helping others as well as differing motives for participating, which in
turn affects their levels of contribution (Marwell and Oliver, 1988). Research in this area focuses
on the attributes of the individuals within the collective, such as an individuals level of motiva-
tions in seeing the good realized and the amount of resources that he or she can leverage toward
the effort.
One of the primary motivations for participating in the creation of a public good is when
there is no other alternative available. In other words, individuals are more likely to actively
volunteer and engage in collective action when no other private solutions are available (Hardin,
1982). In EKNs, this suggests that individuals who do not have access to knowledge and advice
through their personal, private networks will be more likely to participate and share knowledge
with strangers. There is some evidence from prior research that supports this, where the ability
to access and interact with colleagues has been found to be an especially important motivator for
individuals who are geographically isolated from others, creating a unique opportunity to remain
in contact through EKNs (Wasko and Faraj, 2000). Another study indicates that individuals who
did not have access to co-located colleagues were more likely to contribute knowledge in EKNs
(Wasko and Teigland, 2002). These studies suggest that individuals are more likely to invest time
and effort to sustain the network by contributing knowledge, because they have no other way to
access practice-related knowledge.
There is evidence that individuals receive tangible returns for their participation in EKNs.
Grounded in Olsons (1965) premise that collective action may be enhanced by providing selective
incentives, recent work suggests that there are a variety of potential private returns for participating
in EKNs (von Hippel and von Krogh, 2003). For instance, prior research has consistently found
that enhancing ones reputation within the network is an important source of motivation (Lakhani
and von Hippel, 2003; Wasko, 2002). Active participation may also be motivated by a desire to
gain recognition (Gu and Jarvenpaa, 2003) or to infuence others in the network (Firth, 2004).
Another source of motivation comes from a desire for self-actualization and to gain intrinsic
benefts associated with participation in EKNs. People who are motivated by self-actualization seek
opportunities where they are able to use their skills and knowledge to reach their creative potential
(Ghiselli, 1971). According to Deci (1975, 1980), people are born with a need for competence
and self-determination, suggesting that these needs motivate people to seek novelties, challenges,
and incongruities to solve. Banduras social-cognitive theory (1986) suggests that self-evaluation,
based on competence and social norms, is an important source of intrinsic motivation and engage-
ment in activities for the activity itself, rather than extrinsic rewards. There is ample evidence that
intrinsic motivations are an important driver of participation in EKNs. Researchers have found
that individuals contributed knowledge simply because they enjoyed helping others (Constant et
al., 1996; Wasko and Faraj, 2005). Other sources of intrinsic motivation stem from the percep-
tions that participation is entertaining and fun, providing answers to others helps refne ones own
thinking and provides an opportunity to receive feedback on ideas, and because participation in
EKNs is a great way to learn new things (Constant et al., 1996; Lakhani and von Hippel, 2003;
Wasko and Faraj, 2000).
Finally, research suggests that not only do people participate in EKNs out of self-interest or self-
actualization but also people are concerned with their EKNs as a community and participate out of a
sense of moral duty (Schwartz, 1970). Etzioni (1988) argues that individual behavior is signifcantly
affected by moral consideration, not simply personal gain, and that people forego the tendency to
free-ride out of a sense of public duty and fairness. Prosocial behavior occurs when people wish for
good outcomes not only for themselves but also for others, whereby people weigh the social good of
sharing more heavily than their personal cost (Constant et al., 1994). There is evidence that individu-
als are motivated by a sense of moral duty or fairness, because they are interested in helping advance
the professional community as a whole, and because the world is a better place when people help
otherseverybody wins through the open sharing of ideas (Wasko and Faraj, 2000).
In addition to motivations, the distribution of individual resources is also an important factor
supporting collective action (Hardin, 1982). For networks focused on creating and maintaining
a public good, individuals within the network must have access to the required resources. These
resources generally include money, energy, and infuence (Oliver et al., 1985). In EKNs, prior
research has found that people who have higher levels of professional expertise and organizational
tenure are more likely to contribute knowledge (Constant et al., 1996). There is also evidence
that tenure in the shared practice is a strong predictor of contributing knowledge to the network
(Wasko and Faraj, 2005) and individuals who thought their expertise was inadequate were less
likely to answer others questions (Wasko and Faraj, 2000). Other resources might include time,
access to technology, or position in the shared practice.
Relational Social Capital
People who participate and contribute knowledge to EKNs are not engaged in direct, reciprocal
relationships. Rather, they are making their participation visible and open to the network as a whole.
In addition, these participants are members of overlapping communities, the EKN, the organiza-
tional community, and the wider occupational community. Participation in an EKN connotes not
only the process of taking part but also the development of relations through mutual engagement
with others in the network (Wenger, 1998). This mutual engagement results in relationships, and
these relationships can be a signifcant resource for social action due to the emergence of relational
capital, which enables prosocial behavior and collective action. While other forms of capital are
based on assets or individuals, social capital resides in the fabric of relationships between individu-
als and in individuals connections with their communities (Putnam, 1995).
When networks are characterized by a high degrees of trust, obligation, and identifcation
between individuals and the network as a whole, individuals within the network are more likely
to view their interests as aligned with those of the network, leading to more positive interactions
(Jones et al., 1997). Additionally, the identity of individual network members is more likely to
matter, thus compelling members to engage with each other with at least some minimal level of
courtesy and consideration (Williamson, 1991). Prior research in EKNs indicates that trust in oth-
ers ability, benevolence, and integrity is related to a desire to both give and receive information
(Ridings, Gefen, and Arinze, 2002). Individuals with a strong sense of identifcation and attach-
ment to electronic networks are more likely to participate and assist others (Wellman and Gulia,
1997). There is also evidence that individuals are more likely to help others when they feel a
sense of identifcation with the network and the networks goals (Lakhani and von Hippel, 2003).
However, in one EKN, prior research indicates that individuals who contributed the most helpful
answers reported lower levels of commitment to the network (Wasko and Faraj, 2005). This may
simply imply that participants who receive the most help are more likely to feel committed to the
network, indicating that the infuence of social capital should be examined with regard to different
forms of participation in the network.
Structural Social Capital
In EKNS, individual motivations and resources are distributed throughout the network. Individuals
in EKNs voluntarily and individually determine how they participate and what they contribute.
Individuals with collective interests may beneft from creating and sustaining a public good; how-
ever, they may not necessarily act collectively to pursue that good. What seems to set collective
action in motion and sustain it is the formation of a critical mass of motivated and resourceful
individuals (Marwell and Oliver, 1993). Borrowing from nuclear physics, the use of the term criti-
cal mass refers to the idea that a certain threshold of participation or action has to form before a
social movement may come to exist and produce the public good (Oliver and Marwell, 1988). This
core of activists is necessary because most other participants may not have the effcacy to engage
in collective action and may free-ride, hoping that others will create the public good. Marwell and
Olivers critical mass theory (1993) begins with the assumption that free-riding does not necessar-
ily prohibit collective action in large groups. Because the public good is nonrival, free-riders are
not a burden to the collective. This suggests that some minimum number of individuals is needed
to attract the participation and contributions of others in order to sustain collective action, and it
is this subset that constitutes the critical mass.
Critical mass is particularly relevant for collectives where the costs of providing the good are
fxed regardless of the number of individuals contributing towards its provision. In other words,
collective action may be relatively easier to achieve in cases where the good can be created and
sustained through the efforts of only a few, rather than equal participation by all. This is the case
in EKNs. The costs for posting a message are the same, regardless of the number of individuals
who beneft, and knowledge contribution may be sustained through the efforts of a few, active
individuals. Therefore, it is not necessary that all or even most members of the network actively
participate. The essential issue is whether there are enough individuals who are interested and
resourceful enough to provide the public good for others (Marwell and Oliver, 1993). This results
in what Olson refers to as the exploitation of the great by the small (Olson, 1965). This situation
may occur when at least some subset of interested members has an incentive to see the public good
provided, even if they have to bear the full burden of providing it themselves (Olson, 1965).
Individuals constituting the critical mass are likely to have access to more resources and more per-
sonal ties and are more likely to be in a position to beneft from those ties, taking on central positions
within the network (Burt, 1997; Hansen, Nohria, and Tierney, 1999). Central individuals have been
shown to be important determinants of helping behaviors and knowledge sharing in organizations
(Borgatti and Cross, 2003). Studies in informal organizational structures indicate that central individu-
als were perceived to be more innovative (Ibarra, 1993), had better work reputations (Krackhardt,
1990), and had high degrees of interpersonal infuence (Friedkin, 1993). In addition, individuals that
hold central positions in the network are more likely to have access to timely information and to
know where to go for specifc advice (Burt, 1992). The infuence of a critical mass may be even more
pronounced in EKNs because formal structures are weak or nonexistent, and the higher resources of
the critical mass attracts additional individuals to engage in collective action. For instance, Ahuja et
al. (2003) found that an individuals network centrality in an electronic R&D group was associated
with helping behavior and performance, and Gu and Jarvenpaa (2003) found that contributions to
an electronic network of practice increased the more other users contributed. Additional research
indicates that the structure of one EKN resembled a star, with a critical mass in the center providing
the majority of responses, surrounded by a group of peripheral members engaged equally in posting
and responding to questions, but the vast majority of individuals simply posted once or twice, fitting
in and out of the network to seek specifc advice (Wasko and Teigland, 2002).
In summary, we have offered insight from theories of social capital and critical mass to suggest
alternative explanations to individual motivation as an explanation of why individuals participate,
share knowledge, and sustain the EKN over time. Specifcally, the voluntary nature of the ex-
change, the unequal distribution of motivations and resources and the ability of the network to be
sustained by a small subset or critical mass of participants, has interesting implications for how
EKN networks are structured and how these structures generate value and sustain themselves.
Network structure and implications for EKN success are examined in the next section.
NETWORK ASPECTS OF ELECTRONIC KNOWLEDGE NETWORKS
Our review thus far has offered a social view of knowledge exchange, focusing on the motivations
and social infuences of individual participants in EKNs. In general, these psychological/social
views emphasize an individualistic perspective that gives primacy to what participants feel or at
least say about their experience rather than what they do. In contrast, the structural approach
gives primacy to the relationships between participants and the resulting structures formed by
those relationships. This complementary perspective seeks to identify and explain how persistent
patterns of interactions between participants consistently arise within EKNs.
The sheer size and scope of the phenomenon complicate this approach. The theories and meth-
ods most consistent with a structural perspective, those drawn from social network analysis, are
historically based on small-group interactions (e.g., Hawthornes bank wiring room studies). Until
the late 1990s, the majority of social science research has been limited to networks numbering
no more than a few hundred members due to the diffculty in gathering larger data sets and to the
computational limitations of then-current technology.
In recent years, the proliferation of advanced information technology has not only facilitated the
collection of large-scale network data but also increased the availability of large-scale network data
analysis tools and techniques. This has led to an explosion of interest in large network research in
felds spanning biology, physics, mathematics, and social sciences. Popularized in books by Bara-
basi (2002), Strogatz (2003), and Watts (2003), this new science of networks explores structural
characteristics common across all types of large-scale networksnot only social networks, but also
informational networks, technological networks, and biological networks (cf. Newman, 2002).
Due to their prevalence across a wide range of domains, there are two network properties in par-
ticular that have drawn a great deal of attention in the new science of networks: small-world networks
and scale-free networks. Researchers have empirically established the presence of both properties in
real-world networks of metabolic reactions, food webs, transportation networks, board-of-director
interlocks, scientifc collaborations, and World Wide Web link structures (Watts, 2004).
The small-world phenomenon occurs when all pairs of nodes in a network are connected by
shorter-length paths than would be expected in a randomly formed network of the same size, and
the network also has a high degree of local clustering (Watts and Strogatz, 1998). This is con-
ceptually similar to the famous studies by the sociologist Milgram (1967) that asked people to
get a postcard to a stranger in another part of the United States. Milgram found that the average
number of intermediaries in successfully completed paths was sixa surprisingly small number
considering the total population of the United States. Two practical implications of the small-world
phenomenon are that smaller-world networks are more vulnerable to transmission of infections
and, more positively, can more readily diffuse knowledge than a larger-world network.
Small-world networks are particularly relevant in settings where network visibility is restricted.
For example, when participant interactions are private and knowledge of network participants is
transmitted through existing relationships, then small-world properties are quite salient. In those
settings, the shortest connection between two network nodes (measured by path length) and the
likelihood of redundant paths (measured by clustering coeffcients) are indicative of probabilities
for successful communication. Alternatively, in the open settings more characteristic of EKNs,
where current and past participation is typically visible to all participants and active participants
are readily identifable, these small-world properties are less indicative of the effciency or ef-
fectiveness of future communication patterns.
The second major characteristic of large scale-free networks is a power-law distribution
1
in terms
of node degree. The degree of a node refers to the number of different connections existing with
other nodes. Like many other networks that are characterized by power-law distributions, it turns
out that in EKNs a small number of participants have an extremely large number of links to other
participants. Graphically, a power law distribution is readily identifable when a log-log trans-
formation of a rank-ordered list yields a straight line. Figure 14.1 shows both the untransformed
and log-transformed distribution of the number of messages posted by 821 participants of a typical
EKN. These power-law distributional properties of participation on EKNs have so far been found
in every group studied so far by the authors. Examples of power-law distributions related to EKNs
include the frequency distribution of links to Web sites (Adamic and Huberman, 2000), recipients
of e-mail (Wu et al., 2004), and participants on message boards (Adamic et al., 2003; Ravid and
Rafaeli, 2004). Taken together, these studies confrm the power-law nature of the size distribution
of EKNs and of participation within EKNs. The implications of such distributions for EKN research
and management are explored in later sections.
Electronic Knowledge Network Empirical Research
Empirical studies adopting the social view of knowledge exchange have typically researched EKNs
at the individual-level of analysis. Most studies to date have been qualitative studies or surveys of
participants from a single EKN. As a whole, the empirical results and theoretical developments
described in the frst half of this chapter provide a helpful starting point for research based on a
structural perspective. That is, studies from a social perspective bolster theory-based hypothesis
on the structural characteristics of EKNs.
Figure 14.1 Power Law DistributionLog-Log Transformation to Linear Representation
Some early studies have looked at the structure of EKNs as a dependent variable rather than
as context. Butler (2001) investigated the participation dynamics of 284 EKNs (all e-mailbased
listservs). He found that as listservs gained members, they were more attractive to new mem-
bers, yet also had higher rates of membership loss. The net effect of these opposing forces, a net
membership gain, is consistent with Barabasi and Alberts model of preferential attachment and
the formation of a power law distribution. Although the distribution of group membership size
is unreported by Butler (2001), the descriptive data and the log-transformation in measurement
models are both consistent with a power law distribution. As represented in Figure 14.2, the rank-
ordered distribution of the 100 largest Web-based bulletin boards using commercially available
hosting technology (http://www.big-boards.com) follows a power law distribution (R
2
= 0.98).
This is further evidence that the relative popularity of EKNs (as measured by membership size)
follows a power law distribution.
Jones, Ravid, and Rafaeli (2004) analyzed the structural characteristics of 600 EKNs (Usenet
newsgroups). Focusing on the relationship between the volume of group member interaction and
message complexity, they theorize that groups have a maximum stress point where a high volume
of group interaction creates cognitive information overload for its group members. They fnd that
group members prefer to send and receive shorter messages at a point of information overload.
Also, they fnd that group members are more likely to cease participating at the maximum stress
point, a fnding consistent with Butlers (2001). In summary, although larger EKNs may be more
attractive to new participants than smaller ones, participating in a large EKN is a qualitatively
different experience than participating in a smaller one.
Nature of Network Interactions
An important open question in the structural perspective of EKN is to understand the nature of
the interactions that explain the development of stable structural properties. Recent research on
EKN formation has emphasized the competing role of three network formation mechanisms:
Figure 14.2 100 Largest Web-Based Message Boards Form Power-Law Distribution
mutuality, generalized exchange, and preferential attachment (Faraj and Johnson, 2005; Johnson
and Faraj, 2005).
Mutuality is the inclination to respond to a participant with whom an individual has previously
communicated. This motivation to reciprocate has a strong theoretical grounding in social exchange
theory (e.g., Blau, 1964). Similarly, from a structural perspective, we can speak of the degree to
which the participation patterns of a network as a whole exhibit a tendency toward mutuality.
That is, when considering all of the interactions between participants in an EKN, do reciprocated
interactions occur with a higher, lower, or equal likelihood, as expected for a random network of
the same size? There is strong theoretical evidencebased not only on human behaviors in similar
social situations but also on the study of individual EKN participation motivationthat lead us to
predict that mutuality is indeed a structuring element in EKN formation.
Another motivation for electronic knowledge participation is generalized exchange. The nature
of online exchange, in particular, can make it diffcult for participants to directly reciprocatepar-
ticipants have unequal abilities and opportunities to contribute. Therefore participants may be
motivated to contribute to the common good as much as to directly reciprocate back to individual
member of an EKN. Again, from a structural perspective, we can speak of the degree to which an
EKN as a whole exhibits this tendency.
Finally, another motivation of particular interest from a structural perspective is preferential
attachment. It is an important concept in large-scale network formation, being proposed by Bara-
basi and Albert (1999) as the common mechanism for the formation of scale-free networks. The
core elements of their model have withstood numerous critiques and enhancements (cf., Newman,
2002; Watts, 2004). The key conditions for scale-free network formation are (1) an open grow-
ing system, where (2) new entrants decide on preferences based on the expressed preferences of
previous entrants.
An example of preferential attachment in a consumer setting is when you go to a music store
and decide what CD to buy not from listening to music or researching artists but, instead, by au-
tomatically picking the top-selling album from the previous week. In this way, the preference that
you express through your purchase is based on the preferences expressed by other buyers through
their past purchases. In the most extreme case, where there is a very large number of new buyers
each week that follow this strategy (compared to the number of independent-minded buyers), a
winner-takes-all market emerges where a single album would forever dominate future sales.
The fundamental insight of Barabasi and Alberts model is that just these two simple conditions
together, an open growing system (e.g., new buyers each week) and preferential attachment (e.g.,
picking the previous top-seller), can combine to form power law distributions.
Of particular relevance to EKNs is the relationship between preferential attachment and critical
mass. To sustain an EKN, a critical mass of core members is required to maintain continuity and
ensure that the group retains currency and relevance. How is a group of core members established?
The preferential attachment process is one such mechanism. The majority of EKNs are voluntary
organizations that are typically established in an ad hoc process. There is no preexisting set of core
off-line elites who are already known to new members, nor is there a large number of participants
paid to be active contributors. The far more likely scenario is that as new members join a group
they observe who the most experienced and tenured members are. Indeed, many EKNs provide
status cues and explicit indications of member reputation, join dates, or number of contributions.
Thus, as new members join, they are most likely to be drawn into interaction with the more active
existing participants, thereby reinforcing those existing participation patterns. This preferential
attachment process can lead not only to a power law distribution but can also build and reinforce
a core set of participants to form the sustaining critical mass.
Emerging Research Directions
As social communication networks, EKNs share common characteristicsboth structural and
otherwisewith other social networks (Monge and Contractor, 2003). Earlier sections of this
chapter explored the social aspects of EKNs as social networks. Likewise, as large-scale networks,
EKNs share similarities such as power-law distributions with other technical, informational and
biological networks (Newman, 2003). The complexity of the rigorous study of EKNs calls for a
multi-theoretical multi-level approach (Monge and Contractor, 2003). In this section we describe
an ongoing stream of research by two of the chapters co-authors (Johnson and Faraj) that answers
this call. Ibarra et al. (2005) noted the emerging trends of connecting individuals and collectives in
organizational network research. By adopting a structuralist perspective that is heavily informed
by the fndings of studies of individual participation motivations, we are moving from viewing
the EKN as merely a context for research to a multi-level object of study.
Determinants of Network Size
Much of the research described earlier in this chapter describes the motivations for individuals
to participate in EKNs. The implicit assumption behind this research question is that partici-
pants have already identifed a network in which to participate. A less-researched question is
what attracts members to choose among competing EKNs. What are the predictors of electronic
network size?
The most thorough treatment of this question to date (Butler, 2001) adopts a resource-based
model. Butler (2001) found that potential members are, indeed, attracted to sites that provide
members with more resources. Unfortunately, members are also turned off by larger sites, result-
ing in higher levels of membership losses as sites grow (Butler, 2001). Participation structures,
mechanisms for reducing cognitive complexity, are suggested as one method for reducing
membership losses (Butler, 2001). For example, message threads, digests, and topic forums are
technology features for structuring participation in mailing lists and Web-based bulletin boards.
Community-specifc practices, such as message title naming practices, are another method of
organizing participation.
Our own study of 576 EKNs both supports and expands upon Butlers (2001) fndings. Whereas
earlier studies have focused on e-mailbased EKNs, we studied ones formed through Web-based
discussion boards; in this case, a set of comparable EKNs sharing a common technology engine
(vBulletin by Jelsoft). We analyzed the characteristics associated with differences in inception-
to-date membership statistics. We found that larger EKNs tend to have been in existence longer,
have higher levels of administration intensity, have higher levels of participation intensity, and
utilize more participation structures. Surprisingly, we found no association between EKN size
and moderation intensity. Specifcally, we found no statistically signifcant relationship between
the inception-to-date membership of an EKN and the number of site moderators or the likelihood
that a site used an optional feature that allowed individual users to report objectionable content
to site moderators.
Whereas Butler (2001) clearly demonstrates that large groups have both positive and negative
effects, our study provides hope that EKN managers can act in ways to limit those negative ef-
fects. For example, we found that more message threads and more message forums (collections
of related threads) were both associated with larger groups. And, if an EKN sponsor is faced with
a decision between investing in site administration or content moderation resources, our fndings
are that active site administration has a much stronger association with EKN size.
Structuring Mechanisms: Early Empirical Evidence
A challenge to adopting a multitheoretical, multilevel approach to network research is identifying
appropriate empirical methods. Monge and Contractor (2003) advocate the p* method for multilevel
multitheoretical hypothesis testing of communications networks. The MultiNet software applica-
tion provides researchers with a low-cost Windows package that can effectively test hypothesis on
networks with thousands of nodes (Richards and Seary, 2004; Seary and Richards, 2000).
The p* method, a form of Logit regression modeling for network data, is based on a comparison of
observed network characteristics against a Markov random-graph with the same number of nodes and
relationships. Statistical tests in a p* analysis provide both an overall measure of ft (used to compare
competing models) as well as signifcance levels for individual coeffcients. Faraj and Johnson (2005)
detail the use of the p* method for testing the tendency towards mutuality, generalized exchange, and
preferential attachment in EKNs. Based on the methods detailed in Johnson and Faraj (2005), the
results of p* analysis of three weeks of message history for three EKNs are described next. The four
structural tendencies that we analyze are choice (the nonrandom formation of links), reciprocity (the
likelihood that there is a link from B to A when there is a link from A to B), generalized exchange (the
likelihood that there is a link from B to C when there is a link from A to B), and preferential attachment
(the likelihood that there is a link from C to B when there is a link from A to B).
Our analysis shows a consistent pattern across these EKNs. First, sequential tests of nested
models, whereby each network tendency is added one at a time (frst, choice, then reciprocity,
then generalized exchange, and, fnally, preferential attachment) support the inclusion of all four
tendencies in a full model. That is, a model with all four structural mechanisms has greater ex-
planatory power than models with one or more of the mechanisms excluded.
Next, as summarized in Table 14.1, the signifcant coeffcient (p < 0.01) for the choice tendency
for all three networks provides evidence that these EKNs form nonrandomly. The signifcant coef-
fcient (p < 0.01) and positive sign for all three EKNs for the reciprocity coeffcient (b = 1.83, 3.92,
1.12) supports the presence of reciprocation as a participation motivation in each EKN. During the
3-week observation period, participants in all three groups were far more likely to send messages
to someone who had sent them a message than to participants who had not. This tendency was
particularly pronounced in the CNEW group. The signifcant coeffcient (p < 0.01) and positive
sign for the generalized exchange coeffcient (b = 0.46, 0.17, 0.22) support the presence of gen-
eralized exchange as a participation motivation. When a participant received a response to their
own message, he or she was more likely to also respond to a third party.
Finally, a surprising result is the lack of empirical support for the presence of preferential at-
tachment. Among the four mechanisms, the inclusion of preferential attachment provided only
a marginal improvement to the overall model ft. It is signifcant (p < 0.01) in only two of the
three EKNs (PRES, DEEG) and in those groups the coeffcient is negative (b = 0.47, 0.16).
We conclude from these results that when controlling for generalized exchange there is no evi-
dence of the presence of a tendency for preferential attachment and, indeed, that there is tentative
evidence suggesting a preference away from preferential attachment in two of the three groups.
These fndings demonstrate the power of using a sophisticated multilevel, multitheoretic method
to analyze networks.
Structuring Mechanisms: Agent-Based Simulation
The publication of articles in Nature (Watts and Strogatz, 1998) and Science (Barabasi and Albert,
1999) were followed by rapid advances in the study of large-scale networks (cf. Newman, 2002).
Early on, by focusing on characteristics considered universal across wide domains, researchers were
able to make simplifying assumptions that led to mathematically solvable sets of equations. Then
formation models grew more complex, including, for example, multiple generation mechanisms.
Researchers also sought to test their models against multiple outcomes, for example, to match
both the range of small-world and power-law distribution values observed in the real world. When
dealing with networks, even a small increase in complexity can quickly render a model intractable.
Many studies, therefore, also rely on simulation methods to test network formation models.
When applied to EKNs the strength of the classic large-scale network formation models, their
attempt to be universal, is also their weakness. The most common technology for EKNsthe
Web-based discussion boardinstitutes constraints on participants that limit network formation.
For example, whereas a generic network model typically assumes that any node can connect to
any other node at any time, in a threaded Web-based discussed board a connection can only be
made by responding to a message that has not already been responded to. Furthermore, for the
sake of computational simplicity, large-scale network formation models have focused on simple
outcome measures that can be directly calculated or reasonably estimated through tractable solu-
tions. To overcome these limitations, Johnson and Faraj (2005) use an agent-based simulation to
test EKN formation models.
Thus, they can investigate to what extent participant behaviors in EKNs are motivated by
structural characteristics of the network. Specifcally, their study tests the role of preferential
Table 14.1
p-star Analysis of Three Electronic Knowledge Networks
Electronic Knowledge Network
Tendency Measure PRES CNEW DEEG
No. of nodes 36 58 55
No. of links 57 155 143
Choice B 4.67 5.26 4.40
Exp(B) 0.01 0.01 0.01
SE 0.31 0.23 0.16
PL Wald 232 545 749
p <0.01 <0.01 <0.01
Reciprocity B 1.83 3.92 1.12
Exp(B) 6 50 3
SE 0.39 0.25 0.27
PL Wald 22 246 17
p <0.01 <0.01 <0.01
Generalized exchange B 0.46 0.17 0.22
Exp(B) 1.59 1.18 1.24
Std. Err 0.06 0.02 0.02
PL Wald 67 81 143
p <0.01 <0.01 <0.01
Preferential attachment B 0.47 0.04 0.16
Exp(B) 0.63 0.96 0.85
SE 0.09 0.02 0.03
PL Wald 29 3 24
p <0.01 <0.10 <0.01
PRES = programmersresource.com; CNEW = codenewbie.com; DEEG = deegruenig.com/forum.
attachment and mutuality in EKN formation. They develop a simulation model of a thread-
based asynchronous EKN. By applying variance reduction techniques (Law and Kelton, 2000),
they isolate the relative impact of mutuality, preferential attachment, and random links on three
outcome variables (power law distribution, clustering coeffcient, and mutuality index). Figure
14.3 provides an example representation of a communication network derived from a simulated
threaded-discussion board.
Johnson and Faraj (2005) develop three related representations of a threaded EKN. First, they
develop an agent-based model that describes the behaviors, rules, and tendencies that govern a
thread discussion group. Next, they propose an analytical model, including specifc functional forms
and distributions for a threaded EKN. Because of the limited amount of existing research in this
area, they use empirical observations from reference groups to derive some of the analytical model
and simulation parameters. Their third representation of a threaded EKN is the implementation of
the analytical model in an experimental simulation design. They run this experimental model 330
times to simulate over 3.6 million messages generated by almost 39,000 authors.
By using a simulation method and focusing on a specifc network platform, Johnson and Faraj
(2005) extend previous mathematical models of network formation to a richer, more realistic set-
ting. Their fndings are visually represented in Figures 14.4 and 14.5. They fnd both mutuality
and preferential attachment to have strong effects on the formation of network structure. Unsur-
prisingly, changes in mutuality (the agent-level preference) have the most signifcant impact on
mutuality index (a network-level measure).
More surprising, however, is the fnding that mutuality is also important for the emergence of
power law distributions observed in EKNs. One explanation for this fnding is that in the constrained
Figure 14.3 Network Representation of Agent-Based Simulation
environment of a thread discussion boardwhere each message can directly receive only a single
responsemutuality may play an important role in generating more opportunities to interact
with the most popular participations. If the group norm is strong toward direct and immediate
reciprocation, that provides a powerful fuel for the rich get richer cycle of reinforced popularity.
This fnding demonstrates the power of context-specifc modeling as well as the importance of
considering both strength and symmetry for mutuality.
CONCLUSION
This chapter has presented current research thinking and early fndings on EKNs. EKNs are
emerging as alternative sources for knowledge exchange for organizational actors. Because the
primary communication channel for EKNs is asynchronous computer-mediated communication,
the technology has a profound infuence on how knowledge is shared and exchanged. From an
organizational standpoint, EKNs offer unique challenges to management in that they extend across
organizational boundaries and are characterized by the free exchange of knowledge. Attempts
to restrict participation by organizational members have a dual edge: knowledge not only fows
out from organizations but also fows in. To date, we know little about whether organizations
beneft or suffer from their members participation in such settings. For organizational actors that
increasingly view their loyalty and identity to be part of a specialization or a small community of
practice, the continuous development of EKNs seems to indicate that participation in these free
fowing knowledge forums is valuable.
Some specifc fndings reported in this paper require further elaboration. Most important, a num-
ber of theories related to social capital, collective action, and critical mass appear to have emerged
as complements to traditional individual views of why people participate. Some individuals may
be motivated by notions of gain and reputation building but evidence is accumulating about the
Figure 14.4 Relative Effects on Mutuality Index (MI)
Figure 14.5 Relative Effects on Power-Law Distribution Coeffcient
importance of social ties and interpersonal issues such as norms and obligations in support of the
continued operation of EKNs. These results have implications regarding how EKN managers may
want to manage online interactions in order to allow the positive aspects of online community to
emerge. It may also mean that the early phase of the development of a community is absolutely
critical and that there is a need to nurture a critical mass of givers who will sustain the group
during its diffcult periods.
The fndings related to EKN structure are signifcant in terms of understanding interaction
patterns over time. Empirical evidence suggests that EKNs are scale-free networks and thus are
characterized by highly skewed distributions in terms of sizes as well as highly skewed participa-
tion within by individuals. An implication is that EKN managers need to differentiate between
participants. Some posters are more equal than others. This is confrmed by the emergent practice
by EKN managers to often give titles and recognizable signatures to frequent posters. Clearly, if
20% of the posters are responsible for 80% of the posts, then it becomes crucial to ensure that the
constructive lead posters are recognized and rewarded while the negative lead posters are quickly
brought under control.
Our fndings have implications for the study and management of emergent Web technologies
for knowledge publishing, sharing, and linking. The development of publicly edited collaboratories
such as Wikis and networked self-publishing such as blogs are changing the nature of the Web
experience but are little understood so far. For example, hundreds of thousands of people have
contributed to Wikipedia and there are more than 20 million active blogs. The theories discussed
here offer a starting point to understand these highly social technologies. The structural analysis
and methods used to study EKNs are also highly relevant to study Wiki participation and blog
connectivity because in the fnal analysis, no matter the changing underlying technology, the
interactions are social interactions.
This chapter has offered early evidence of the interaction mechanisms that take place in EKNs.
Explanations built on reciprocity, generalized exchange, and preferential attachment have been
tested empirically as well as through simulation studies. The fndings regarding the importance
of reciprocity and generalized exchange are consistent with the theories of social capital and col-
lective action that developed from the direct study of participants. Together, these various strands
of EKN explorations have opened an important window on how and why individuals participate
in EKNs. Yet, as EKNs grow in importance and encompass ever more diverse areas, exciting new
research possibilities abound.
NOTE
1. The terms scale-free network and power-law distribution both frequently appear in large-scale
network literature to refer to the same phenomenon. We use these terms synonymously.
REFERENCES
Adamic, L.A.; Buyukkokten, O.; and Adar, E. 2003. A social network caught in the Web. First Monday, 8,
6, 6.
Adamic, L.A., and Huberman, B.A. 2000. Power-law distribution of the World Wide Web. Science, 287,
2115a.
Ahuja, M.; Galletta, D.; and Carley, K. 2003. Individual centrality and performance in virtual R&D groups:
An empirical study. Management Science, 49, 1, 2138.
Alavi, M., and Leidner, D. 1999. Knowledge management systems: Issues, challenges and benefts. Com-
munication of the Association for Information Systems, 1, 128.
Bandura, A. 1986. Social Foundations of Thought and Action. Englewood-Cliffs, NJ: Prentice-Hall.
Barabasi, A.-L. 2002. Linked: The New Science of Networks. Cambridge, MA: Perseus Publishing.
Barabasi, A.-L., and Albert, R. 1999. Emergence of scaling in random networks. Science, 286, 509512.
Blau, P.M. 1964. Exchange and Power in Social Life. New York: Wiley.
Borgatti, S.P., and Cross, R. 2003. A relational view of information seeking and learning in social networks.
Brown, J.S., and Duguid, P. 1991. Organizational learning and communities-of-practice: Toward a unifed
view of working, learning, and innovation. Organization Science, 2, 1, 4057.
Burt, R.S. 1992. Structural Holes: The Social Structure of Competition. Cambridge, MA: Harvard University
Press.
. 1997. The contingent value of social capital. Administrative Science Quarterly, 42, 339365.
Coleman, J.S. 1990. Foundations of Social Theory. Cambridge, MA: Belknap Press of Harvard University
Press.
Constant, D.; Kiesler, S.; and Sproull, L. 1994. Whats mine is ours, or is it? A study of attitudes about
Culnan, M.J. 2005. The Pro-Social Behavior of Springsteen Fans: A Case Study of the BTX Online Commu-
nity. Presented at the Glory Days: A Bruce Springsteen Symposium, Monmouth University, September
911, 2005.
Daft, R.L., and Lengel, R.H. 1986. Organizational information requirements, media richness and structural
design. Management Science, 32, 5, 355366.
Deci, E.L. 1975. Intrinsic Motivation. New York: Plenum Press.
. 1980. The Psychology of Self-determination. Lexington, MA: Lexington Books.
DeSanctis, G., and Monge, P. 1999. Communication processes for virtual organizations. Organization Sci-
ence, 10, 6, 693703.
Etzioni, A. 1988. The Moral Dimension. London: The Free Press.
Faraj, S., and Johnson, S.L. 2005. Reciprocity or generalized exchange? Structuring of electronic knowledge
networks. Presented at Academy of Management Conference, Honolulu, Hawaii.
Firth, D. 2004. Is free information really free? Information supply into an IT-based organizational memory
system. Presented at Twenty-Fifth International Conference on Information Systems, Association for
Information Systems, Washington, DC.
Ghiselli, E.E. 1971. Explorations in Managerial Talent. Pacifc Palisades, CA: Goodyear Publishing Company.
Gu, B., and Jarvenpaa, S. 2003. Online discussion boards for technical support: The effect of token recog-
nition on customer contributions. Presented at Twenty-Fourth International Conference on Information
Systems, Seattle, WA.
Hardin, R. 1982. Collective Action. Baltimore, MD: Johns Hopkins University Press.
Head, J.G. 1962. Public goods and public policy. Public Finance, 17, 3, 197219.
Heckathorn, D.D. 1996. The dynamics and dilemmas of collective action. American Sociological Review,
41(April), 250277.
Ibarra, H. 1993. Personal networks of women and minorities in management: A conceptual framework.
Academy of Management Journal, 18, 5687.
Ibarra, H.; Kilduff, M.; and Tsai, W. 2005. Zooming in and out: Connecting individuals and collectives at
the frontiers of organizational network research. Organization Science, 16, 4, 359371.
Jarvenpaa, S., and Majchrak, A. 2005. Developing individuals transactive memories of their ego-centric
networks to mitigate risks of knowledge-sharing: The case of professionals protecting cybersecurity.
Presented at The Twenty-sixth International Conference on Information Systems, Las Vegas, NV, De-
cember 1114.
Johnson, S.L., and Faraj, S. 2005 Preferential attachment and mutuality in electronic knowledge networks.
Presented at The Twenty-sixth International Conference on Information Systems, Las Vegas, NV, De-
cember 1114.
Jones, C.; Hesterly, W.S.; and Borgatti, S.P. 1997. A general theory of network governance: Exchange condi-
tions and social mechanisms. Academy of Management Review, 22, 4, 911945.
Jones, Q.; Ravid, G.; and Rafaeli, S. 2004. Information overload and the message dynamics of online
interaction spaces: A theoretical model and empirical exploration. Information Systems Research, 15,
2, 194210.
Kollock, P., and Smith, M.A. 1996. Managing the virtual commons: Cooperation and confict in computer
communities. In S. Herring (ed.), Computer-Mediated Communication: Linguistic, Social and Cross
Cultural Perspectives. Amsterdam: John Benjamins, pp. 109128.
Lakhani, K., and von Hippel, E. 2003. How open source software works: Free user-to-user assistance.
Research Policy, 32, 6, 923943.
Lin, N. 2001. Social Capital. Cambridge, UK: Cambridge University Press.
Marwell, G., and Oliver, P. 1988. Social networks and collective action: A theory of the critical mass, III.
American Journal of Sociology, 94, 3, 502534.
. 1993. The Critical Mass in Collective Action: A Micro-social Theory. New York: Cambridge Uni-
versity Press.
Milgram, S. 1967. The small world problem. Psychology Today, 2, 6067.
Monge, P.R., and Contractor, N.S. 2003. Theories of Communication Networks. New York: Oxford Univer-
sity Press.
Newman, M.E.J. 2003. The structure and function of complex networks. SIAM Review, 2, 167256.
Oliver, P.E., and Marwell, G. 1988. The paradox of group size in collective action: A theory of the critical
mass, II. American Sociological Review, 53, 18.
Oliver, P.E., Marwell, G., and Teixeira, R. 1985. A theory of critical mass, I: Group heterogeneity, interde-
pendence and the production of collective goods. American Journal of Sociology, 91, 522556.
Olson, M. 1965. The Logic of Collective Action. Cambridge, MA: Harvard University Press.
Orlikowski, W.J. 1996. Learning from Notes: Organizational issues in groupware implementation. In R. Kling
(ed.), Computerization and Controversy. New York: Academic Press, pp. 173189.
Putnam, R. 1995. Tuning in, tuning out: The strange disappearance of social capital in America. Political
Science and Politics, December, 664683.
Ravid, G. and Rafaeli, S. 2004. Asynchronous discussion groups as small world and scale free networks.
First Monday, 9, 9.
Rheingold, H. 1993. The Virtual Community: Homesteading on the Electronic Frontier. Reading, MA: Ad-
dison Wesley.
Richards, W.D., and Seary, A.J. 2004. MultiNet for Windows 4.55. Burnaby, British Columbia: Simon Fraser
University.
Ridings, C.M.; Gefen, D.; and Arinze, B. 2002. Some antecedents and effects of trust in virtual communities.
Journal of Strategic Information Systems, 11, 271295.
Schwartz, S.H. 1970. Moral decision making and behavior. In J. Macauley and L. Berkowitz (eds.), Altruism
and Helping Behavior. New York: Academic Press, pp. 127141.
Seary, A.J., and Richards, W.D. 2000. Fitting to p* models in MultiNet. Connections, 23, 1, 84101.
Shmanske, S. 1991. Public Goods, Mixed Goods, and Monopolistic Competition. College Station, TX: Texas
A&M University Press.
Stewart, T.A. 2000. Software preserves knowledge, people pass it on. Fortune, 142, 5, 390392.
Strogatz, S. 2003. Sync: The Emerging Science of Spontaneous Order. New York: Theia.
Sutton, R. 2000. Knowledge management is not an oxymoron. Computerworld, 34, 1, 28.
Thibaut, J.W., and Kelley, H.H. 1959. The Social Psychology of Groups. New York: John Wiley.
Thorn, B.K., and Connolly, T. 1987. Discretionary data bases. A theory and some experimental fndings.
Communication Research, 14, 5, 512528.
von Hippel, E., and von Krogh, G. 2003. Open source software and the private-collective innovation model:
Issues for organization science. Organization Science, 14, 2, 209223.
Wasko, M., and Faraj, S. 1999. Webs of knowledge: An examination of knowledge types and knowledge fows
in an electronic community of practice. Academy of Management Conference, Chicago, Ill.
. 2000. It is what one does: Why people participate and help others in electronic communities of
practice. Journal of Strategic Information Systems, 9, 155173.
. 2005. Why should I share? Examining knowledge contribution in electronic networks of practice.
MIS Quarterly, 29, 2 & 3, 123.
Wasko, M., and Teigland, R. 2002. The provision of online public goods: Examining social structure in a
network of practice. Presented at The Twenty-third Annual International Conference on Information
Systems, Barcelona, Spain.
Watts, D.J. 2003. Six Degrees: The Science of The Connected Age. New York: W.W. Norton & Company.
. 2004. The new science of networks. Annual Review of Sociology, 30, 1, 243270.
Watts, D.J., and Strogatz, S.H. 1998. Collective dynamics of small-world networks. Nature 393, 440
442.
Wellman, B., and Gulia, M. 1997. Net surfers dont ride alone. In P. Smith (ed.), Communities in Cyberspace.
Berkley, CA: University of California Press.
Wenger, E. 1998. Communities of Practice. Cambridge, UK: Cambridge University Press.
Williamson, O.E. 1991. Comparative economic organization: The analysis of discrete structural alternatives.
Wu, F.; Huberman, B.A.; Adamic, L.A.; and Tyler, J.R. 2004. Information fow in social groups. Physica
A, 337, 327335.
292
ChaptEr 15
AD HOC INTERORGANIZATIONAL
COLLABORATION
Safeguards for Balancing Sharing and Protection of Knowledge
yonGsUK Kim, sirKKa l. jarVEnpaa, and ann majChrZaK
Abstract: This chapter sets a stage for future research on the tension between knowledge sharing and
protection in ad hoc interorganizational collaboration. Managing the tension between knowledge
sharing and protection is critical for successful interorganizational collaboration, joint action,
and long-term learning but is particularly challenged in volatile and virtual settings. Ad hoc col-
laborations bring together geographically distributed individuals with specialized knowledge to
solve nonroutine and dynamically changing problems. The volatility of the ad hoc collaboration
makes it impossible to predict which collaboration process, when and why it is performed, who
is involved, and which support tools are used. The virtuality of the ad hoc collaboration reduces
shared face-to-face experiences and makes it challenging to monitor and evaluate others actions.
We review selected literature on interorganizational systems and interorganizational relations to
identify safeguards for managing the tension between sharing and protection. We highlight that
the safeguards are built on the assumption of stable and hierarchical interorganizational rela-
tions, and, thus, are questioned or challenged by the ad hoc collaborations that are volatile and
virtual. Future research needs to combine multiple theoretical perspectives to account for the
unique complexity of knowledge sharing and protection with volatile and virtual contexts. We also
suggest possible roles for information technology.
Keywords: Knowledge Management, Ad hoc Collaboration, Interorganizational Collaboration,
Knowledge Sharing, Knowledge Protection, Virtuality, Volatility
TENSION OF KNOWLEDGE SHARING AND KNOWLEDGE
PROTECTION IN AD HOC COLLABORATION
In the modern era of alliances, outsourcing, and community innovation, there has been increasing
emphasis on frms need to share knowledge with their various partners, customers, suppliers, and
even the public at large. Collaboration tools such as e-mail and virtual workspaces allow rapid
assembly of members from multiple organizations to a virtual knowledge team that spans across
organizations (Majchrzak et al., 2004). We refer to these teams as ad hoc interorganizational
collaborations because they are rapidly assembled specifcally in response to a problem, they
cross organizations, and they disband as soon as the problem is solved. The collaborations may
last hours, days, or months but their problem-focus distinguishes them from traditional teams
AD HOC INTERORGANIZATIONAL COLLABORATION 293
and their cross-organizational nature distinguishes them from task forces. Also referred to as
cross-boundary emergent groups, ad hoc interorganizational collaborations have been observed
in emergency response to natural disasters (Tierney, Lindell, and Perry, 2001), new product de-
velopment (Engwall and Svensson, 2001), process manufacturing (Hagel and Brown, 2005), and
security threats (Majchrzak and Jarvenpaa, 2005). For example, professionals working on cyber
security attacks are members that strongly rely on ad hoc collaborations with professionals from
other organizations and various governmental entities (Goodman and Wilson, 2000). Responding
to these attacks requires coordinating and integrating diverse knowledge in order for joint and
local actions to be taken.
In the process of integrating knowledge, members of collaborations are exposed to vulner-
abilities including knowledge leaks and anticompetitive accusations (Norman, 2002). We consider
vulnerable collaborations to be those where participating individuals must continually balance
the tradeoff between sharing and protection. For example, cross-organizational R&D networks
raise the possibility of both value creation and harming through information sharing (Heiman and
Nickerson, 2002). The possibility of mixed motives creates a situation when frms must protect
themselves from sharing proprietary or confdential information. Such sensitive information may
be related to trade secrets (Hannah, 2005), security of critical infrastructure (Trim, 2005), personal
information on employees (Kaupins and Minch, 2005), and customer information (Culnan and
Armstrong, 1999). Sharing of such confdential information may lead to short-term disastrous
fnancial effects on frms (Campbell, Gordon, Loeb, and Zhou, 2003), longer-term diffculties in
appropriating strategic resources (Norman 2002), long-lasting threats to organizational reputation
(Scott and Walsham, 2005), and physical/material harm to employees and assets.
Although ad hoc collaborations are likely to include members that are perceived as generally
trustworthy or at least associated with trustworthy organizations, members must be on guard that
certain knowledge might pose a vulnerability to the member or the members organization if later
shared with other third parties. Such sharing can be a result of purposeful betrayal or accidental
disclosure. An example of a purposeful betrayal was told to us by the chief security offcer of a
large multinational frm:
Two frms (Firm A and Firm B) were engaged in a joint R&D collaboration in which the
representative for Firm A spent time at the facilities of Firm B. The Firm A representative,
during lunch hours, spent time with a variety of Firm B employees, not just those specif-
cally associated with the joint R&D activity. Six months later, the CEO of Firm A called
the CEO of Firm B and announced that Firm B had developed a new product (unrelated
to the joint R&D venture) that worked remarkably similarly to one under development by
Firm A and that if Firm A was not willing to collaborate with Firm B on the remaining de-
velopment, Firm A would get to the market frst. In retrospect, it seems that the lunch hours
spent by the Firm A representative were probably for competitive intelligence gathering,
not for socializing.
Such potential vulnerabilities posed by sharing sensitive information often result in frms in-
troducing policies regarding sharing and not sharing. Such internal policies include what can and
cannot be shared and required process workfows for clearing information for release to other
organizations. For example, since the aftermath of antitrust accusations among Finnish forestry
enterprises, employees now have to report any contact they have had with employees of competi-
tors including those related to nonbusiness or recreational activities (Majchrzak and Jarvenpaa,
2005). However, such policies can cause enormous delays if followed and endanger the success of
294 KIM, JARVENPAA, AND MAJCHRZAK
the entire collaboration (Hoecht and Trott, 1999). In fact, internal policies exacerbate knowledge
coordination problems rather than remedy them (Sitkin and Bies, 1993). They have an adverse
effect on the motivation of individual employees to share information outside of their organizations
(Majchrzak and Jarvenpaa, 2005). Such decreased motivation to share can be counterproductive
particularly when knowledge collaboration involves emergent knowledge (i.e., knowledge that is
created in the interaction among different organizational actors) (Markus, Majchrzak, and Gasser,
2002). Such policies are rendered particularly unworkable in situations involving highly distributed
collaboration in volatile and virtual settings (Majchrzak and Jarvenpaa, 2005).
Other potential safeguards that we discuss in this paper are also challenged in volatile and
virtual situations. Risks of knowledge leaks as well as the diffculty in deciding what to share
without creating too much vulnerability are both intensifed in virtual and volatile situations be-
cause both exacerbate uncertainty. Both volatility and virtuality are dimensions of postbureaucratic
organizational forms (Kellogg, Orlikowski, and Yates, 2006). Volatility refers to the dynamic
and blurred division of tasks and roles and promotes emergent knowledge in action. In ad hoc
collaborations, specifying ex ante group membership, roles, tasks, knowledge, and support tools
is not feasible. Authority is decentralized and team boundaries are fuzzy. Members belong to
multiple interorganizational teams, and memberships often change (Tierney, Lindell, and Perry,
2001). This volatility exacerbates uncertainty and creates additional challenges for managing the
tension between knowledge sharing and protection. Common ground is likely to be lacking leading
to misinterpretation and miscommunication (Cramton, 2001). The lack of any commitment to a
foreseeable future working relationship is likely to lead to less reciprocity. The lack of a common
organization is likely to yield members working under different norms concerning what to share,
how to share, proper ways to use knowledge that has been shared by others, and even defni-
tions about what is considered sensitive or confdential knowledge (Carlile, 2004; GAO Report,
2006). The context of what is sensitive information also changes as the roles and organizational
memberships of individuals change. A member may move from a role as a collaborator to a role
as a competitor. How these dynamics change the balance of sharing and protection is often hard
to predict in advance.
The other challenge, virtuality, refers to geographic distance or time spent working apart and
the use of information technology in team communication (Crisp and Jarvenpaa, 2006; Griffth,
Mannix, and Neale, 2003). Technology has made it possible for distributed agents to work together
regardless of the temporal gaps and geographical locations of the members, providing scalability
and fexibility in spanning individuals, teams, and organizations to form ad hoc collaborations.
The increased use of collaborative tools such as e-mail and virtual workspaces in such networks
speeds the information fow among distributed workers, providing the advantages of a greater
base of knowledge and a larger network of potential participants. However, interorganizational
collaborations that rally around conversations and dialog discourse are particularly challenged
across distance (Hardy, Lawrence, and Grant, 2005; Majchrzak, Malhotra, and John, 2005). The
virtual network makes it especially diffcult to monitor and evaluate others actions and therefore
predict reliably the potential of a knowledge breach (i.e., the inappropriate sharing of sensitive
information). And collaborations with virtual workspace, e-mail, and instant message archives
leave a digital trail that can be captured, altered, combined with other data, transferred to parties,
and used for harmful purposes with high rates of speed.
In this chapter, we limit our consideration of vulnerable collaborations in virtual and volatile
contexts that assume horizontal relationships with little power differentials. Studies reveal that
powerful frms may force powerless partners to adopt a set of practices to facilitate information
exchange (Hart and Saunders, 1997; Subramani, 2004). Powerful frms may be less concerned
about the partners privacy and exploit it to achieve competitive advantage by justifying its use
for gaining better partner insight (Greenway and Chan, 2005). Therefore, asymmetric power
relationships violate the give-and-take relationships assumed in post bureaucratic structures we
are interested in.
Although the need to manage the tradeoff between sharing and protection is well acknowledged
in practice, there has been relatively little discussion in the knowledge management and knowledge
management systems literature about the tradeoff between protection and sharing, especially in ad
hoc collaboration contexts where both volatility and virtuality are high. Research on knowledge
management has primarily focused on transfer, storage, retrieval, and application of knowledge
within organizations where the organizational boundaries and the division of tasks and roles are
relatively stable with established organizational norms and routines (see Alavi and Leidner, 2001,
for details). Where protection issues are discussed in the knowledge management literature, they
are often made with attributions to knowledge workers who are unwilling to share because of
self-interest rather than frm interest or due to unsupportive organizational policies and structures
that fail to create a conducive social context for knowledge sharing (Brown and Duguid, 1991;
Markus, 2001; Nonaka and Konno, 1998). In ad hoc interorganizational collaborations, managing
the tension between knowledge sharing and protection is a cognitive activity that must be continu-
ously managed regardless of the self-interest or social context.
While the knowledge management literature has generally ignored the tradeoff between protec-
tion and sharing, research on interorganizational information systems (IOS) and interorganizational
relations (IOR) has, implicitly or explicitly, revealed this tradeoff to some extent. Below, we briefy
review this research by frst describing the mechanisms that have been suggested to manage the
tension between sharing and protecting knowledge, distinctively or jointly, in IOS and IOR lit-
erature. This review reveals that the mechanisms suggested by the literature are suitable in stable
and hierarchical IOR only, which are the traditional empirical settings in past studies. We reveal
vulnerabilities in the mechanisms in volatile and virtual settings. We conclude this chapter by
calling for new integrative perspectives on managing the tension between sharing and protection
in ad hoc interorganizational collaborations.
SHARING AND PROTECTION IN
INTERORGANIZATIONAL RELATIONS
In our review of the IOR and IOS literature, we have included only those articles that we could fnd
that make references to both sharing and protection of knowledge in interorganizational settings.
These articles can be categorized by their perspective on knowledge exchange between frms.
Five perspectives have been identifed: (1) transactive cost economics, (2) knowledge-based view
(KBV) of the frm, (3) relational-based perspective, (4) modularity, and (5) transactive memory
systems (TMSs). In the sections that follow, we explore each perspective, identify mechanisms
the perspective suggests for managing the knowledge sharing/protection tension, and explore the
potential effectiveness of the perspective for managing the tension when collaborations are both
volatile and virtual.
Transaction Cost Economics Perspective
The alliance literature has extensively studied the choice of contractual agreements or alliance
governance structures against partners opportunistic behaviors using the theoretical base of
transaction cost economics (TCE) (Pisano, 1989; Williamson, 1991; Oxley, 1997, 1999). It has
been argued that, due to their concerns about partners opportunism, frms incur higher transaction
costs in specifying contracts, monitoring partners behaviors, and enforcing the agreements. This
is particularly so in interorganizational collaboration that involves volatility and hence requires
emergent knowledge. It is infeasible to specify in contracts the details of task processes and related
knowledge sharing in advance so that it is legally clear to everyone what to share and not to share
during collaboration. Moreover, extraction of more knowledge by the partners than anticipated ex
ante is very likely (Heiman and Nickerson, 2002). According to TCE, frms choose appropriate
contractual mechanisms (i.e., governance structure) to minimize the transaction costs. One such
mechanism, equity-based governance, allocates the rights to profts equally to partners through
joint ownership.
The mechanism of equity-based governance has been suggested as one that alleviates
the hazards of opportunism for two reasons. First, shared equity aligns the incentives of the
partners as the return of their mutual fnancial investments hinge on the outcomes of their
IOR. Knowing that opportunistic behaviors would decrease the value of the outcomes pro-
vides partners with the motives to behave in a more mutually benefcial fashion (Dyer and
Singh, 1998; Pisano, 1989). Second, equity-based governance structures promote effective
monitoring of the partners activities. Close working interactions enable joint activities to be
more transparent and contingencies to be managed more easily (Kale, Singh, and Perlmutter,
2000). Equity-based governance is also suggested to minimize the coordination costs that
arise from (1) the task interdependence across organizational boundaries and the complexity
of coordinating individual or joint activities and (2) the related extent of communication and
decisions (Gulati and Singh, 1998).
While the TCE equity-based governance structure may alleviate opportunism in stable IOR,
such a structure is diffcult to use in volatile situations where volatility may arise from dynami-
cally changing tasks and changing membership. Moreover, in virtual settings, it is diffcult to
assess whether what is observed refects the actual behavior, a critical factor needed to make such
governance structures work. Although it is argued that information technology (IT) has the ability
to lower coordination costs by supporting communication and decision making (e.g., Clemons,
Reddi, and Row, 1993), the jury is still out on what effect the capabilities of technology-mediated
monitoring have on opportunism costs in volatile and virtual settings. It is particularly unclear
how IT-supported equity-based governance structure could appropriately account for the constant
infux and effux of membership in ad hoc collaborations.
Knowledge-Based Perspective
Firms develop and sustain their competitive advantage by keeping their resources and capabilities
hard to obtain, imitate or replicate (Barney, 1991). The KBV extends the resource-based view of
the frm, by arguing that knowledge is the fundamental resource of the frm (Kogut and Zander,
1992). In interorganizational collaboration, partners may acquire both easily codifed and tacitly
understood knowledge. Codifed knowledge such as best practices and operating procedures are
more easily transferred because the causal ambiguity is reduced (Szulanski, 2003). Firms interested
in sharing their knowledge with partners often expend much effort in codifying knowledge. How-
ever, codifed knowledge is more susceptible to unwanted transfer as well, including inadvertent
leakage or unwanted imitation by other frms in ad hoc relationships (Winter, 1997; Zander and
Kogut, 1995). To protect vulnerability of unwanted transfer of codifed knowledge, contractual
agreements on intellectual property are recommended (Teece, 2000).
Knowledge that is not as easily codifed is referred to as sticky or tacit knowledge (Polanyi,
1962; von Hippel, 1988). For transfer, tacit knowledge requires observation and learning through
relational ties (Inkpen and Dinur, 1998; Jordan and Lowe, 2004; Nonaka and Takeuchi, 1995).
Close and repeated interaction with others using discursive conversations is recommended as ways
to surface tacit assumptions (Boland, Tenkasi, and Teeni, 1994). Thus, if the intention is not to
share tacit knowledge, one should avoid discursive conversations with others in the collaboration,
such as by accepting information conveyed by others rather than challenging it. This suggests
that a mechanism for managing the tradeoff between protection and sharing of tacit knowledge
may lie in how a discourse is handled. Limiting discourses in which tacit knowledge is shared to
trustworthy parties may be one way to manage the tension between sharing and protection. How
to accomplish this in volatile and virtual situations is, however, far from clear.
While contractual agreements and limited discourse may alleviate opportunism in stable rela-
tionships, volatility creates challenges. In volatile situations, members lack a common ground for
sharing a common interpretation of the knowledge; as such, contracts governing the dissemination
and use of the knowledge are easily misinterpreted. Volatility also raises questions about the value
of avoiding discourse. When membership changes, new thought worlds enter into the collaboration
(Dougherty, 1992), requiring the need for discourse to overcome the differences. In the discursive
process, then, tacit knowledge may be inadvertently released.
Virtual contexts also create challenges to mechanisms in the KBV. In virtual settings, it can be
diffcult to monitor and assess whether others are adhering to contractual obligations for the use
and dissemination of the protected knowledge. What one observes may or may not be what the
other party is doing or not doing. The virtual context is very susceptible to incorrect representa-
tion of the circumstances of a social exchange. The other party may make up information about
something that simply does not exist or re-label or categorize information to mislead (Grazioli and
Jarvenpaa, 2003). Hence, the virtual context complicates the exercise of contractual agreements
and increases the need to limit discourses to trustworthy parties.
Relational-Based Perspective
Fundamental to the relational-based view of IOR is the role of trust (Ring and Van de Ven, 1992).
Trust refers to ones belief and expectation about others positive motive in situations entailing
risk (Deutsch, 1973) Trust allows effective knowledge sharing to occur through close, intense,
and iterative interactions between individual members in interorganizational collaboration (Kale,
Singh, and Perlmutter, 2000; Kumar and Becerra-Fernandez, 2007). Firms tend to be less depen-
dent on formal structures to manage relationship once they have developed high levels of trust
(Gulati, 1995). Moreover, frms are reluctant to share valuable know-how with each other unless
they have high levels of mutual trust (Norman, 2002).
Numerous studies have suggested a relationship between relational governance (i.e., trust, repu-
tation) and formal controls. Poppo and Zenger (2002) extensively review literature on relational
governance and formal contracts as substitutes and as complements. According to their review,
if both partners trust each other, there is no need for formal contracts. On the other hand, if they
use formal contracts, it indicates a lack of trust and may even result in adverse consequences such
as encouraging opportunistic behaviors in areas not specifed by the contracts. In discussions
of substitution between relational governance and formal contracts, Poppo and Zanger (2002)
conclude that
the net effects of formal contracts and relational governance on exchange performance are
ambiguous. While relational governance and formal contracts may have positive direct
effects on exchange performance, because they function as replacements for one another
(or in the case of formal contracts causally damage the other), the net effect on exchange
performance is, at a minimum, reduced and potentially negative (p. 711).
The authors empirically corroborate the argument that the combination of formal and informal
safeguards would work more effectively than either alone when facing severe hazards. Formal
contracts promote relational governance by specifying a long-term commitment to knowledge
sharing and by articulating punishments that limit opportunism. Relational governance in return
enables the refnement of contracts and promotes stable knowledge exchanges between partners.
Das and Teng (2001) also discuss the complementary roles of trust and control and argue that
the relational risk, including opportunistic behaviors, can be minimized through goodwill trust,
behavior control, and social control all together. The complementarity of formal controls and
relational governance has been widely accepted (Crisp and Jarvenpaa, 2006).
Research on trust in ad hoc collaborations notes that an effective management of the tension
between knowledge sharing and protection may require managed trust (Meyerson, Weick, and
Kramer, 1996). Trust that is too low or too high can present risks. When trust is too low due to a
limited history of collaboration and little prospect for repeat collaboration in the future, shared
norms of obligation and responsibility are underdeveloped, and individuals will be reluctant to
share their core knowledge with each other. Too high of a trust increases dependence on others
and might blind the person for others exploitation of sensitive information (Bouty, 2000).
In ad hoc collaborations in which new members are constantly arriving and there is a lack of
prior relationships and reciprocity, trust is likely to be low. Moreover, the lack of social control
to encourage normative actions may keep trust low in volatile groups. One mechanism that le-
verages the volatility to manage the knowledge sharing/protection tension is hedging. Hedging
is aimed to reduce mutual dependence and its perceived costs by creating an apparent trusting
attitude, yet having a backup strategy just in case (Meyerson, Weick, and Kramer, 1996). That
is, individuals behave in a trusting manner but also hedge to reduce the vulnerability. A good
example of hedging is a credit card limit (Fink, 2000). While reducing the risk of fraud, a credit
card holder can still make a series of small payments. Similarly, hedging may be an agreed-upon
practice in an ad hoc collaboration whereby frms maintain trusted interfrm level relations while
intentionally assigning new individuals to collaborate with others in different problem situations.
By collaborating with different people, they may be able to manage the tension of sharing and
protection more effectively because they do not feel overly indebted to any particular set of in-
dividuals and yet they are reasonably active in mutual sharing because of the presence of trust at
the interorganizational level.
Virtuality may also impact the degree to which trust can be used as a mechanism to manage
the knowledge sharing/protection tension. Most frms today rely on face-to-face communication
to establish trust, periodically reinforce trust, and rebuild trust when perceived violations occur
(Majchrzak et al., 2004). This raises the fundamental question about whether trust is an appropriate
mechanism for managing the tension between knowledge sharing and protection in virtual set-
tings. Lack of face-to-face interactions intensifes miscommunication, confict, and opportunistic
behaviors, decreasing trust and harming knowledge sharing (Hinds and Bailey, 2003). Yet, Kumar
and Becerra-Fernandez (2007) describe how an information technology-based process supporting
conversation for action helped to build relationships and trust between business partners. Hence,
the proper use of IT to support virtual communication may help foster trust virtually and allow it
to continue as an effective mechanism for managing tension of sharing and protection in virtual
settings.
Modularity-Based Perspective
The IOS literature suggests that one way to manage the tension of sharing and protection in vir-
tual teams is by modularizing the work. Modularity, originating from the property of computer
programming, refers to the extent to which a system can be composed of smaller components,
or modules. Modularity creates a high degree of independence or a loose coupling between
components, which mutually interact through standardized component interface specifcation
(Sanchez and Mahoney, 1996). Modularity extends beyond technological product design to the
design of interorganizational collaboration structures (Langlois, 2002, Malhotra, Malhotra, and
John, 2005). The modularity of an interorganizational process is defned as the breaking up of
complex processes into subprocesses (activities) that are performed by different enterprises inde-
pendently (such that subprocesses occur simultaneously) with specifed outputs managed through
embedded coordination enabled by standardized process interfaces and information exchange
(Malhotra, Gosain, and El Sawy, 2005, p. 155). Modularity creates task partitioning, and task par-
titioning reduces the need for frequent communications, resulting in limited cross-frm knowledge
sharing by design (Sobrero and Roberts, 2001). Simultaneously, individual frms are able to
freely share intra-frm knowledge without concern about knowledge misappropriation by other
partners in the chain. As such, a focal frm, or the network leader (Subramani, 2004) can protect
its core knowledge regarding the architectural knowledge of interlinking partitioned tasks because
each supplier is restricted to a component level and limited interface learning. Jordan and Lowe
(2004) conducted feld studies on knowledge protection practices in joint development projects and
pointed out that task partitioning emerged as the projects evolved: we see modularity emerging
unselfconsciously as a way of dealing with complexity and the risks of knowledge appropriation
and as a process that is leading frms towards greater specialization rather as consequent upon it
(p. 257). Thus, modularity can be a useful mechanism for protecting confdential knowledge in
interorganizational collaborations.
Modularity provides fexibility against change in membership because tasks and their inter-
actions are all specifed in details independent of who are assigned. However, the assumption
underlying modularity is that the ad hoc collaboration is not harmed by partitioning the tasks.
The assumption of a well-defned and stable division of tasks generally does not hold in ad hoc
collaborations where emergent knowledge work is assumed. Well-defned modularization becomes
challenging in a dynamic setting where emergent knowledge collaboration such as joint R&D
makes it hard to partition a project in advance as the knowledge to acquire or share is continu-
ously changing. In such an interorganizational collaboration, remodularization inevitably follows
as loosely coupled learning proceeds (Sanchez and Mahoney, 1996). Therefore, decisions on
communication, coordination structures, and staffng of the interfaces have to be continuously
aligned with the decisions on task partitioning (Jordan and Lowe, 2004), inducing high coordina-
tion costs. Another issue is that task partitioning may be harmful since it removes the possibility of
serendipitous learning of divergent perspectives, a factor critical to formulating creative responses
(Majchrzak, Cooper, and Neece, 2004).
Virtuality may also impact the utility of modularity as a safeguard mechanism, depending on
the divisibility of the task. With divisible tasks, virtuality may actually help modularity by the
use of such IT safeguards as project-specifc electronic workspaces, access controls, document
version control, and document check in/check out tools. With tasks less divisible, virtuality may
make it more diffcult to identify improperly defned interfaces between modules and the reasons
for missed coordination cues (Majchrzak et al., 2004b). This may lead to the need to exchange
more tacit knowledge which may, in turn, increases risks of inadvertent knowledge leaks.
Transactive Memory Theory and Systems Perspective
Transactive memories play critical roles in helping group members pool their differentiated
knowledge to effciently solve shared problems (Lewis, 2003). Transactive memory refers to the
information that resides within a group about who knows what and how to access those knowledge
stores (Anand, Manz, and Glick, 1998; Wegner, 1995). The TMS of a group contains three types
of meta-knowledge about the group: memory differentiation (who knows what), task coordina-
tion (how to coordinate the people), and knowledge credibility (whose expertise can be trusted)
(Wegner, 1995). Shared group memories are maintained and updated as members share training
and work experiences, exchange feedback on work practices, and frequently interact either in
person or via information technology in the organization. A well-developed TMS at the team level
can improve knowledge sharing, knowledge integration and overall performance of formal teams
(Faraj and Sproull, 2000; Lewis, 2003). TMS can help anticipate the knowledge needs and shape
linking specialized knowledge to ft to the problem at hand.
In an interorganizational context, TMS may also serve as a safeguard, although empirical
research on this is in its infancy (Jarvenpaa and Majchrzak, 2005). In high-risk situations, know-
ing who knows what may affect what information should not be shared to avoid anyone knowing
too much. For example, an individual from one organization may learn of the best practices of
another organization, but as long as that individual is not aware of which best practices were not
found as useful in different contexts, the competitive advantage of the organization with the best
practices is maintained.
TMS has been mostly applied to stable relationships (e.g., Lewis, 2004; Lewis, Lange, and
Gallis, 2005). Recently, though, Moreland and Argote (2003) suggest how TMS could be applied
to more volatile relationships. A TMS could be developed for the organization from which tempo-
rary teams are drawn, containing an expertise directory at the organizational level (Anand, Manz,
and Glick, 1998). This expertise directory might then serve as an initial TMS when a temporary
team is created.
When collaborations cross organizations, the notion of a single organization-wide TMS would
need to be supplanted. One idea advanced by Jarvenpaa and Majchrzak (2005) is that the single
TMS is supplanted by a TMS that the individual applies to her own personal network of contacts
from which collaborative partners are drawn. During collaborations, then, an individual could
decide what information to share and withhold based on what the individual believes others in the
network are likely to know. This concept of a personal network-based TMS fails to address the
question of how such a TMS is developed within a network. TMS literature has concluded that
shared experiences among all team members is the most effective way to develop a TMS for the
team (Lewis, Lange, and Gallis, 2005). Yet, ad hoc collaborations drawn from personal networks
will not include all members of the network. Consequently, TMS may not be shared between
members of a collaboration, which will limit the TMSs effectiveness in coordinating knowledge
integration. Therefore, while a personal network-based TMS may help to explain why one person
may not share with others, having multiple unshared TMS may harm knowledge sharing.
Moreover in highly volatile situations, a TMS will need frequent updates to be maintained
(Brandon and Hollingshead, 2004; Griffth and Neal, 2001). Akgun, Byrne, Keskin, and Lynn
(2006) found that in volatile situations, members in a new product development team were not
necessarily able to update their TMS; yet they continued to rely on obsolete TMS. Because of the
prior habit and routines to rely on others knowledge, members continued to use obsolete TMS
without critically evaluating the reliability. This adversely affected team learning.
In addition to volatility, virtuality may also impact how TMS develops and serves as a safeguard.
While Yoo and Kanawattanachai (2001) demonstrated the development of TMS virtually, virtuality
may make the validation of TMS more challenging because of deception and misrepresentation in
virtual settings. In virtual settings, it is more diffcult to ensure that knowledge or expertise that is
assumed to exist is in fact the knowledge that will be acted on. As such, TMS in virtual settings
may be less credible and hence may have limited utility as a safeguard. Virtual collaborations may
require extra effort not only in maintaining the TMS but in developing strategies such as redundancy
(i.e., using multiple sources of knowledge) to compare and validate the accuracy of TMS.
In volatile and virtual settings, information technology helps to coordinate access to resources
and expertise among dispersed members. IT can play a critical role by supplying and supporting
in real-time the building blocks of transactive memory (who knows what, knowledge coordination,
and credible knowledge) (Griffth, Mannix, and Neale, 2003), with repositories, version histories,
and transparent access to members responses to requests for knowledge. Nevo and Wand (2005)
developed information architecture for directories of meta-memories that support TMS at the
broader organization and community level. Given that the tension of sharing and protection is an
individual-level issue, the IT architectures must also support the individuals view of the collective
TMS and resolve conficts among the various levels of TMS in real-time.
We summarize our review in Table 15.1.
CONCLUSION
Our literature review suggests that the existing literature on IOS and IOR does not adequately address the
conditions of high-risk interorganizational collaborations that occur in volatile and virtual settings.
We have suggested that formal contractual arrangements from TCE are likely to be of limited
value as parties and tasks to the contracts change and as the virtual setting impedes transparent
monitoring for contract violations. Formal agreements protecting intellectual property, inferred
from the KBV of the frm, will befall a similar fate. An alternative safeguard from the KBV, of
limiting discourse, may protect tacit knowledge leakage but will also limit suffcient knowledge
exchange for an effective collaboration. Relational-based view establishes trust, reciprocity, and
control as potential safeguards. However, each of these require enforcement through the promise
of future interactions among members, a mechanism not often present in volatile and virtual set-
tings. The modularity perspective of task partitioning may protect leakage between modules or
tasks, but may simultaneously limit needed knowledge exchange in virtual and volatile situations.
Finally, the TMS perspective offers a safeguard of knowing what not to share with others based on
knowing what others know. Yet, volatility means that the cognitive structure of the TMS may not
be shared among the collaborating members or the TMS may not be suffciently updated, which
may limit the ability of TMS to facilitate knowledge coordination. Moreover, virtuality of an ad
hoc collaboration may impede the validation of TMS resulting in an unreliable safeguard.
Thus, safeguards for managing the knowledge sharing/protection tension provided by each
perspective appear to be of limited value in volatile and virtual settings. We believe this limitation
of the current literature is due to a lack of a theory describing how ad hoc collaborators decide
what knowledge to share and how to share it. Ad hoc collaborative knowledge sharing is an indi-
vidual-level phenomenon infuenced by, but not acted by, organizations. As an individual action,
social-cognitive infuences become prevalent. Communication norms, feedback, and processes
shape the social-cognitive processes. Finally, as an individual operating within a groupno matter
how temporarygroup dynamics affect the individual. A theory is needed that combines social
cognition, group dynamics, and communication theories to explain management of the knowledge
sharing/protection tension in volatile and virtual contexts.
302
T
a
b
l
e

1
5
.
1
S
u
m
m
a
r
y

o
f

P
e
r
s
p
e
c
t
i
v
e
s
P
e
r
s
p
e
c
t
i
v
e
C
h
a
r
a
c
t
e
r
i
s
t
i
c
s
S
u
g
g
e
s
t
e
d

S
a
f
e
g
u
a
r
d
I
m
p
a
c
t

o
f

V
o
l
a
t
i
l
i
t
y

o
n

S
a
f
e
g
u
a
r
d
I
m
p
a
c
t

o
f

V
i
r
t
u
a
l
i
t
y

o
n

S
a
f
e
g
u
a
r
d
T
r
a
n
s
a
c
t
i
o
n

c
o
s
t

e
c
o
n
o
m
i
c
s

(
T
C
E
)
T
h
e

c
h
o
i
c
e

o
f

a
p
p
r
o
p
r
i
a
t
e

c
o
n
t
r
a
c
t
u
a
l

s
a
f
e
g
u
a
r
d
s

o
r

g
o
v
e
r
n
a
n
c
e

s
t
r
u
c
t
u
r
e

m
a
y

r
e
d
u
c
e

t
h
e

r
i
s
k

o
f

o
p
p
o
r
t
u
n
i
s
t
i
c

b
e
h
a
v
i
o
r
s
.
E
q
u
i
t
y
-
b
a
s
e
d

g
o
v
e
r
n
a
n
c
e

m
a
y

b
e

a

p
r
o
t
e
c
t
i
v
e

g
o
v
e
r
n
a
n
c
e

s
t
r
u
c
t
u
r
e

t
o

s
a
f
e
g
u
a
r
d

a
g
a
i
n
s
t

k
n
o
w
l
e
d
g
e

m
i
s
a
p
p
r
o
p
r
i
a
t
i
o
n

w
h
i
l
e

e
n
s
u
r
i
n
g

k
n
o
w
l
e
d
g
e

s
h
a
r
i
n
g
.
T
h
e

v
o
l
a
t
i
l
i
t
y

o
f

m
e
m
b
e
r
s
h
i
p

f
a
i
l
s

t
o

m
e
e
t

t
h
e

s
t
a
b
i
l
i
t
y

a
s
s
u
m
e
d

i
n

t
h
e

c
o
n
t
r
a
c
t
u
a
l

g
o
v
e
r
n
a
n
c
e

s
t
r
u
c
t
u
r
e
;

m
o
n
i
t
o
r
i
n
g

b
e
h
a
v
i
o
r

i
s

d
i
f
f
c
u
l
t

w
h
e
n

m
e
m
b
e
r
s

a
n
d

t
a
s
k
s

c
o
n
s
t
a
n
t
l
y

c
h
a
n
g
e
.
T
h
e

e
f
f
e
c
t
i
v
e

a
n
d

t
r
a
n
s
p
a
r
e
n
t

m
o
n
i
t
o
r
i
n
g

a
s
s
u
m
e
d

i
n

t
h
e

e
q
u
i
t
y
-
b
a
s
e
d

g
o
v
e
r
n
a
n
c
e

f
a
c
e
s

c
h
a
l
l
e
n
g
e
s

i
n

v
i
r
t
u
a
l

c
o
n
t
e
x
t
,

a
n
d

i
t

i
s

e
v
e
n

m
o
r
e

d
i
f
f
c
u
l
t

t
o

d
o

w
h
e
n

h
a
v
i
n
g

c
o
n
s
t
a
n
t

c
h
a
n
g
e

i
n

m
e
m
b
e
r
s
h
i
p
.
K
n
o
w
l
e
d
g
e
-
b
a
s
e
d

v
i
e
w

(
K
B
V
)
C
o
d
i
f
e
d

k
n
o
w
l
e
d
g
e

i
s

e
a
s
y

t
o

t
r
a
n
s
f
e
r
,

f
a
c
i
l
i
t
a
t
i
n
g

k
n
o
w
l
e
d
g
e

e
x
c
h
a
n
g
e
,

b
u
t

a
l
s
o

i
n
c
r
e
a
s
i
n
g

v
u
l
n
e
r
a
b
i
l
i
t
y

o
f

u
n
i
n
t
e
n
d
e
d

k
n
o
w
l
e
d
g
e

l
e
a
k
a
g
e
;

t
a
c
i
t

k
n
o
w
l
e
d
g
e

t
r
a
n
s
f
e
r
r
e
d

t
h
r
o
u
g
h

o
b
s
e
r
v
a
t
i
o
n

a
n
d

l
e
a
r
n
i
n
g

t
h
r
o
u
g
h

e
x
t
e
n
d
e
d

r
e
l
a
t
i
o
n
a
l

t
i
e
s

a
n
d

c
l
o
s
e

a
n
d

r
e
p
e
a
t
e
d

i
n
t
e
r
a
c
t
i
o
n
s
.
C
o
n
t
r
a
c
t
u
a
l

a
g
r
e
e
m
e
n
t
s

s
u
c
h

a
s

i
n
t
e
l
l
e
c
t
u
a
l

p
r
o
p
e
r
t
y

r
i
g
h
t
;

l
i
m
i
t
i
n
g

d
i
s
c
o
u
r
s
e
s

i
n

w
h
i
c
h

t
a
c
i
t

k
n
o
w
l
e
d
g
e

i
s

s
h
a
r
e
d

o
n
l
y

t
o

t
r
u
s
t
w
o
r
t
h
y

p
a
r
t
i
e
s
.
T
h
e

i
n
s
t
a
b
i
l
i
t
y

o
f

m
e
m
b
e
r
s
h
i
p

h
a
r
m
s

t
h
e

s
h
a
r
e
d

c
o
n
t
e
x
t

r
e
q
u
i
r
e
d

t
o

h
a
v
e

a

s
h
a
r
e
d

u
n
d
e
r
s
t
a
n
d
i
n
g

a
b
o
u
t

t
h
e

k
n
o
w
l
e
d
g
e
,

c
a
l
l
i
n
g

i
n
t
o

q
u
e
s
t
i
o
n

t
h
e

c
o
n
t
r
a
c
t
u
a
l

s
a
f
e
g
u
a
r
d
s
.

V
o
l
a
t
i
l
i
t
y

r
e
q
u
i
r
e
s

s
o
m
e

d
i
s
c
o
u
r
s
e

t
o

a
c
h
i
e
v
e

a

c
o
m
m
o
n

g
r
o
u
n
d

d
u
r
i
n
g

w
h
i
c
h

t
a
c
i
t

k
n
o
w
l
e
d
g
e

m
a
y

b
e

i
n
a
d
v
e
r
t
e
n
t
l
y

r
e
l
e
a
s
e
d
.
V
i
r
t
u
a
l
i
t
y

m
a
k
e
s

i
t

d
i
f
f
c
u
l
t

t
o

m
o
n
i
t
o
r

a
d
h
e
r
e
n
c
e

t
o

a
g
r
e
e
m
e
n
t
s

a
n
d

i
n
c
r
e
a
s
e
s

t
h
e

p
r
o
b
a
b
i
l
i
t
y

o
f

m
i
s
i
n
t
e
r
p
r
e
t
a
t
i
o
n

w
i
t
h
o
u
t

d
i
s
c
o
u
r
s
e
.
R
e
l
a
t
i
o
n
a
l
-

b
a
s
e
d

v
i
e
w
T
r
u
s
t

p
r
o
m
o
t
e
s

k
n
o
w
l
e
d
g
e

e
x
c
h
a
n
g
e

a
n
d

r
e
d
u
c
e
s

o
p
p
o
r
t
u
n
i
s
t
i
c

b
e
h
a
v
i
o
r
s
.

T
o
o

h
i
g
h

t
r
u
s
t

m
a
y

p
r
o
d
u
c
e

o
p
p
o
r
t
u
n
i
s
t
i
c

b
e
h
a
v
i
o
r
.

T
r
u
s
t

a
n
d

c
o
n
t
r
o
l

w
o
r
k

e
f
f
e
c
t
i
v
e
l
y

a
s

c
o
m
p
l
e
m
e
n
t
s
.
U
s
e

c
o
m
b
i
n
a
t
i
o
n

o
f

f
o
r
m
a
l

(
c
o
n
t
r
o
l
)

a
n
d

i
n
f
o
r
m
a
l

(
i
.
e
.
,

t
r
u
s
t

a
n
d

r
e
c
i
p
r
o
c
i
t
y
)

a
s

s
a
f
e
g
u
a
r
d
s
.
T
r
u
s
t

r
e
q
u
i
r
e
s

p
o
s
s
i
b
i
l
i
t
y

o
f

f
u
t
u
r
e

i
n
t
e
r
a
c
t
i
o
n
s

w
h
i
c
h

i
s

m
i
s
s
i
n
g

i
n

v
o
l
a
t
i
l
e

i
n
t
e
r
a
c
t
i
o
n
s
.

H
e
d
g
i
n
g

m
a
y

p
r
o
v
i
d
e

a

m
e
a
n
s

o
f

u
s
i
n
g

v
o
l
a
t
i
l
i
t
y

t
o

m
a
n
a
g
e

t
e
n
s
i
o
n
.
F
e
w
e
r

f
a
c
e
-
t
o
-
f
a
c
e

i
n
t
e
r
a
c
t
i
o
n
s

i
n
t
e
n
s
i
f
y

m
i
s
c
o
m
m
u
n
i
c
a
t
i
o
n
,

o
p
p
o
r
t
u
n
i
s
t
i
c

b
e
h
a
v
i
o
r

a
n
d

c
o
n
f
i
c
t

l
e
a
d
i
n
g

t
o

l
e
s
s

t
r
u
s
t
.
303
M
o
d
u
l
a
r
i
t
y
M
o
d
u
l
a
r
i
t
y

c
r
e
a
t
e
s

t
a
s
k

p
a
r
t
i
t
i
o
n
i
n
g

a
n
d

r
e
d
u
c
e
s

t
a
s
k

i
n
t
e
r
d
e
p
e
n
d
e
n
c
y
,

e
n
s
u
r
i
n
g

m
a
x
i
m
u
m

i
n
t
r
a
f
r
m

k
n
o
w
l
e
d
g
e

s
h
a
r
i
n
g

w
h
i
l
e

k
e
e
p
i
n
g

n
e
c
e
s
s
a
r
y

i
n
t
e
r
f
r
m

s
h
a
r
i
n
g

a
t

m
i
n
i
m
u
m
;

O
f
t
e
n

e
m
e
r
g
e
s

a
s

a

w
a
y

t
o

d
e
a
l

w
i
t
h

c
o
m
p
l
e
x
i
t
y

a
n
d

t
h
e

r
i
s
k
s

o
f

k
n
o
w
l
e
d
g
e

a
p
p
r
o
p
r
i
a
t
i
o
n
.
M
o
d
u
l
a
r
i
z
i
n
g

t
a
s
k
s

m
a
y

h
e
l
p

m
a
n
a
g
e

t
h
e

t
e
n
s
i
o
n
.
M
o
d
u
l
a
r
i
t
y

p
r
o
v
i
d
e
s

f
e
x
i
b
i
l
i
t
y

a
g
a
i
n
s
t

i
n
s
t
a
b
l
e

m
e
m
b
e
r
s
h
i
p

b
e
c
a
u
s
e

t
a
s
k
s

a
n
d

t
h
e
i
r

i
n
t
e
r
a
c
t
i
o
n
s

a
r
e

a
l
l

s
p
e
c
i
f
e
d

i
n
d
e
p
e
n
d
e
n
t
l
y

o
f

w
h
o

i
s

a
c
c
o
u
n
t
a
b
l
e

f
o
r

t
h
e
m
;

h
o
w
e
v
e
r
,

t
h
e

a
s
s
u
m
p
t
i
o
n

o
f

w
e
l
l
-
d
e
f
n
e
d

a
n
d

s
t
a
b
l
e

d
i
v
i
s
i
o
n

o
f

t
a
s
k
s

g
e
n
e
r
a
l
l
y

d
o
e
s

n
o
t

h
o
l
d

i
n

a
d

h
o
c

c
o
l
l
a
b
o
r
a
t
i
o
n
s

w
h
e
r
e

e
m
e
r
g
e
n
t

k
n
o
w
l
e
d
g
e

w
o
r
k

i
s

a
s
s
u
m
e
d
.
V
i
r
t
u
a
l
i
t
y

w
i
t
h

m
o
d
u
l
a
r
i
t
y

m
a
k
e
s

i
t

d
i
f
f
c
u
l
t

t
o

s
u
r
f
a
c
e

m
i
s
i
d
e
n
t
i
f
e
d

i
n
t
e
r
f
a
c
e
s

b
e
t
w
e
e
n

m
o
d
u
l
e
s
.
T
r
a
n
s
a
c
t
i
v
e

m
e
m
o
r
y

s
y
s
t
e
m
s

(
T
M
S
)
T
M
S

c
o
o
r
d
i
n
a
t
e
s

w
h
a
t

i
s

k
n
o
w
n

a
b
o
u
t

w
h
a
t

o
t
h
e
r
s

k
n
o
w

a
n
d

w
h
e
n

o
t
h
e
r
s

k
n
o
w
l
e
d
g
e

i
s

n
e
e
d
e
d
,

a
n
d

h
e
l
p
s

f
n
d

t
h
a
t

k
n
o
w
l
e
d
g
e

a
n
d

a
s
s
e
s
s

t
h
e

c
r
e
d
i
b
i
l
i
t
y

o
f

t
h
e

k
n
o
w
l
e
d
g
e
;

b
y

k
n
o
w
i
n
g

w
h
a
t

o
t
h
e
r
s

k
n
o
w

o
r

m
a
y

k
n
o
w
,

v
u
l
n
e
r
a
b
i
l
i
t
i
e
s

m
a
y

b
e

p
r
o
t
e
c
t
e
d

b
y

n
o
t

s
h
a
r
i
n
g

i
n
f
o
r
m
a
t
i
o
n

t
h
a
t

m
a
y

l
e
a
d

t
o

c
o
r
e

k
n
o
w
l
e
d
g
e

l
e
a
k
a
g
e
.
I
f

l
e
v
e
r
a
g
e
d

a
p
p
r
o
p
r
i
a
t
e
l
y
,

T
M
S

c
a
n

e
n
h
a
n
c
e

p
r
o
t
e
c
t
i
o
n

a
n
d

m
a
n
a
g
e

t
h
e

t
r
a
d
e
o
f
f

b
e
t
w
e
e
n

s
h
a
r
i
n
g

a
n
d

p
r
o
t
e
c
t
i
o
n

i
n

a
d

h
o
c

c
o
l
l
a
b
o
r
a
t
i
o
n
.
T
M
S

a
s
s
u
m
e
s

s
h
a
r
e
d

e
x
p
e
r
i
e
n
c
e
s

a
m
o
n
g

s
t
a
b
l
e

m
e
m
b
e
r
s
h
i
p
;

w
i
t
h
o
u
t

s
h
a
r
e
d

e
x
p
e
r
i
e
n
c
e
s
,

T
M
S

m
a
y

n
o
t

b
e

s
h
a
r
e
d
,

t
h
u
s

l
i
m
i
t
i
n
g

r
a
t
h
e
r

t
h
a
n

h
e
l
p
i
n
g

e
f
f
e
c
t
i
v
e

k
n
o
w
l
e
d
g
e

i
n
t
e
g
r
a
t
i
o
n
.
V
i
r
t
u
a
l
i
t
y

a
m
o
n
g

d
i
v
e
r
s
e

d
i
s
t
r
i
b
u
t
e
d

k
n
o
w
l
e
d
g
e

s
o
u
r
c
e
s

w
i
t
h

r
e
s
p
o
n
s
i
b
i
l
i
t
y

f
o
r

h
a
v
i
n
g

t
h
e

l
a
t
e
s
t

i
n
f
o
r
m
a
t
i
o
n

a
b
o
u
t

t
h
e

c
o
n
s
t
i
t
u
e
n
c
y

t
h
e
y

r
e
p
r
e
s
e
n
t

c
r
e
a
t
e
s

n
e
e
d

f
o
r

c
o
n
s
t
a
n
t

T
M
S

u
p
d
a
t
i
n
g

a
n
d

p
e
r
s
u
a
d
i
n
g

m
e
m
b
e
r
s

t
h
a
t

T
M
S

r
e
p
r
e
s
e
n
t
s

c
u
r
r
e
n
t

a
n
d

r
e
l
e
v
a
n
t

e
x
p
e
r
t
i
s
e

o
f

t
h
e

g
r
o
u
p
.
Future theory development should include the role of information systems. Is it possible to
develop a vigilant information system (Walls, Widmeyer, and El Sawy, 1992) that collabora-
tively monitors the sharing of documents and knowledge to provide proactive feedback to users
who are potentially violating confdentiality? Are there analogues in industry and government of
people self-policing their collaborative interactions to avoid breaches that can inform us of how to
structure similar self-policing during ad hoc collaborations? One possible analogue may be found
in studying internet games. System administrators emerge in these seemingly lawless games
such as CounterStrike, encouraging game players to act according to evolving norms through
withdrawal of game privileges. Examining the characteristics of these emergent administrators
and the communities that foster their emergence may provide additional suggestions on designing
information systems to facilitate managing the balance between knowledge sharing and protection.
Can intelligent agents be developed that can scan documents for their tightness of coupling and
complex interactivity with other documents, and then use this information to suggest a rating of
the risk to the organization if the document was shared?
In conclusion, the objective of this chapter has been to examine different mechanisms proposed
in the literature to manage the tension between knowledge sharing and knowledge protection in
ad hoc interorganizational collaboration. We reviewed different streams of literature to identify
potential safeguards for managing the tension between sharing and protection. We highlighted
how the virtual and volatile context of ad hoc collaboration challenges the potential safeguards.
We suggested potential roles for information technology in accounting for the unique complexity
of knowledge sharing and protection with volatile and virtual contexts. With this chapter, we hope
to stimulate future research on this important problem.
REFERENCES
Akgun, A.E.; Byrne, J.C.; Keskin, H.; and Lynn, G.S. 2006. Transactive memory system in new product
development teams. IEEE Transactions on Engineering Management, 53, 1, 95111.
Anand, V.; Manz, C.; and Glick, W. 1998. An organizational memory approach to information management.
Barney, J.B. 1991. Firm resources and sustained competitive advantage. Journal of Management, 17, 1,
99120.
Boland, R.; Tenkasi, R.; and Teeni, D. 1994. Designing information technology to support distributed cogni-
tion. Organization Science, 5, 3, 456475.
Bouty, I. 2000. Interpersonal and interaction infuences on informal resource exchanges between R&D re-
searchers across organizational boundaries. Academy of Management Journal, 43, 1, 5066.
Brandon, D.P., and Hollingshead, A.B. 2004. Transactive memory systems in organizations: Matching tasks,
expertise, and people. Organization Science, 15, 6, 633644.
Brown, J.S., and Duguid, P. 1991. Organizational learning and communities of practice: Toward a unify-
ing view of working, learning, and innovation. In M.D. Cohen and L.S. Sproull (eds.), Organizational
Learning. London: Sage, pp. 5982.
Carlile, P.R. 2004. Transferring, translating and transforming: An integrative framework for managing
knowledge across boundaries. Organization Science, 15, 5, 555568.
Campbell, K.; Gordon, L.A.; Loeb, M.P.; and Zhou, L. 2003. The economic cost of publicly announced
information security breaches: Empirical evidence from the stockmarket. Journal of Computer Security,
11, 3, 431448.
Clemons, E.K.; Reddi, S.P.; and Row, M.C. 1993. The impact of information technology on the organiza-
tion of economic activity: The move to the middle hypothesis. Journal of Management Information
Systems, 10, 2, 935.
Cramton, D. 2001. The mutual knowledge problem and its consequences in geographically dispersed teams.
Crisp, C.B., and Jarvenpaa, S.L. 2006. Team regulation in virtual teams: Reciprocity of control and trusting
beliefs. Working paper, Indiana University, Bloomington.
Culnan, M.J., and Armstrong, P.K. 1999. Information privacy concerns, procedural fairness and impersonal
trust: An empirical investigation. Organization Science, 10, 1, 104115.
Das, T.K., and Teng, B. 2001. Trust, control, and risk in strategic alliances: An integrated framework. Or-
ganization Studies, 22, 251283.
Deutsch, M. 1973. The Resolution of Confict: Constructive and Destructive Processes, New Haven, CT:
Yale University Press.
Dougherty, D. 1992. Interpretive barriers to successful product innovation in large frms. Organization Sci-
ence, 3, 2, 170202.
Dyer, J.H., and Singh, H. 1998. The relational view: Cooperative strategy and sources of interorganizational
competitive advantage. Academy of Management Review, 23, 660679.
Engwall, M., and Svensson, C. 2001. Cheetah teams. Harvard Business Review, 79, 2, 2021.
Faraj, S., and Sproull, L. 2000. Coordinating expertise in software development teams. Management Sci-
ence, 46, 12, 15541568.
Fink, D. 2000. Developing trust for e-commerce. In L. Janczewski (ed.), Internet and Intranet Security
Management: Risks and Solutions. London: Idea Group Publishing.
GAO Report. 2006. Homeland Security and Information Sharing, GAO-06-385. Washington, DC: U.S.
Government Accountability Offce.
Greenway, K.E., and Chan, Y.E. 2005. Theoretical explanation for frms information privacy behaviors.
Journal of Association for Information Systems, 6, 6, 171198.
Goodman, P.S., and Wilson, J.M. 2000. Substitutes for socialization and exocentric teams. In M.A. Neale, E.A.
Mannix, and T.L. Griffth (eds.), Research on Managing Groups and Teams. Stanford, CT: JAI Press.
Grazioli, S., and Jarvenpaa, S. 2003. Consumer and business deception over the Internet: Content analysis
of documentary evidence. International Journal of Electronic Commerce, 7, 4, 93118.
Griffth, T.L., and Neal, M.A. 2001. Information processing in traditional, hybrid, and virtual teams: From
nascent knowledge to transactive memory. Research in Organizational Behavior, 23, 379422.
Griffth, T.L.; Mannix, E.A.; and Neale, M.A. 2003. Confict in virtual teams. In S.G. Cohen and C.B. Gibson
(eds.), Virtual Teams That Work. San Francisco, CA: Jossey-Bass, pp. 335352.
Gulati, R. 1995. Familiarity breeds trust? The implications of repeated ties on contractual choice in alliances.
Academy of Management Journal, 38, 85112.
Gulati, R., and Singh, H. 1998. The architecture of cooperation: Managing coordination costs and appropria-
tion concerns in strategic alliances. Administrative Science Quarterly, 43, 781814.
Hagel, J., and Brown, J.S. 2005. The Only Sustainable Edge: Why Business Strategy Depends on Productive
Friction and Dynamic Specialization. Boston: Harvard Business School Publishing.
Hannah, D.R. 2005. Should I keep a secret? The effects of trade secret protection procedures on employees
obligations to protect trade secrets. Organization Science, 16, 1, 7184.
Hardy, C.; Lawrence, T.B.; and Grant, D. 2005. Discourse and collaboration: The role of conversations and
collective identity. Academy of Management Review, 30, 1, 5877.
Hart, P., and Saunders, C. 1997. Power and trust: Critical factors in the adoption and use of electronic data
interchange. Organizational Science, 8, 1, 232254.
Heiman, B., and Nickerson, J.A. 2002. Towards reconciling transaction cost economies and the knowledge-
based view of the frm: The context of interfrm collaborations. International Journal of the Economies
of Business, 9, 1, 97116.
Hinds, P.J., and Bailey, D.E. 2003. Out of sight, out of sync: Understanding confict in distributed teams.
Hoecht, A., and Trott, P. 1999. Trust, risk, and control in the management of collaborative technology devel-
opment. International Journal of Innovation Management, 3, 3, 257270.
Hollingshead, A.B. 1998a. Communication, learning, and retrieval in transactive memory systems. Journal
of Experimental Social Psychology, 34, 423442.
. 1998b. Distributed knowledge and transactive processes in groups. In M.A. Neale, E.A. Mannix,
and D.H. Gruenfeld (eds.), Research on Managing Groups and Teams. Greenwich, CT: JAI Press, pp.
103123.
Inkpen, A.C., and Dinur, A. 1998. Knowledge management processes and international joint ventures. Or-
ganization Science, 9, 4, 454468.
Jarvenpaa, S.L., and Majchrzak, A. 2005. Developing individuals transactive memories of their ego-centric
networks to mitigate risks of knowledge sharing: The case of professionals protecting cyberSecurity.
In David E. Avison, and Dennis F. Galletta (eds.), Proceedings of the 26th International Conference of
Information Systems (ICIS), December 1114, Las Vegas, NV, pp. 273285.
Jordan, J., and Lowe, J. 2004. Protecting strategic knowledge: Insights from collaborative agreements in the
aerospace sector. Technology Analysis & Strategic Management, 16, 2, 241259.
Kale, P.; Singh, H.; and Perlmutter, H. 2000. Learning and protection of proprietary assets in strategic alli-
ances: Building relational capital. Strategic Management Journal, 21, 3, 217237.
Kaupins, G., and Minch, R. 2005. Legal and ethical implications of employee location monitoring. In Pro-
ceedings of 38th Hawaii International Conference on Systems Sciences, January 36.
Kellogg, K.C.; Orlikowski, W.J.; and Yates, J. 2006. Life in the trading zone: Structuring coordination across
boundaries in postbureaucratic organizations. Organization Science, 17, 1, 2244.
Kogut, B., and Zander, U. 1992. Knowledge of the frm, combinative capabilities, and the replication of the
technology. Organization Science, 3, 3, 383397.
Kumar, K., and Becerra-Fernandez, I. 2007. Interaction technology: Speech-act based information technology
support for building collaborative relationships and trust. Decision Support Systems, 43, 2, 584606.
Langlois, R.N. 2002. Modularity in technology and organization. Journal of Economic Behavior and Or-
ganization, 49, 1937.
Lewis, K. 2003. Measuring transactive memory systems in the feld: Scale development and validation.
Journal of Applied Psychology, 88, 4, 587604.
Lewis, K. 2004. Knowledge and performance in knowledge-worker teams: A longitudinal study of transactive
memory systems. Management Science, 50, 11, 15191533.
Lewis, K.D.; Lange, D.; and Gallis, L. 2005. Transactive memory systems, learning, and learning transfer.
Majchrzak, A.; Cooper, L.; and Neece, O. 2004. Knowledge reuse for innovation. Management Science,
50, 2, 174188.
Majchrzak, A.; Malhotra, A.; Stamps, J.; and Lipnack, J. 2004. Can absence make a team grow stronger?
Harvard Business Review, 82, 5, 131137.
Majchrzak, A., and Jarvenpaa, S.L. 2005. Information security in cross-enterprise collaborative knowledge
work. Emergence: Complexity & Organization (E:CO), 6, 4, 48.
through information technology-enabled contextualization: Evidence from distributed teams. Information
Malhotra, A.; Gosain, S.; and El Sawy, O.A. 2005. Absorptive capacity confgurations in supply chains:
Gearing for partner-enabled market knowledge creation. MIS Quarterly, 29, 1, 145187.
Markus, M.L.; Majchrzak, A.; and Gasser, L. 2002. A design theory for systems that support emergent
knowledge processes. MIS Quarterly, 26, 3, 179212.
Meyerson, D.; Weick, K.; and Kramer, R.M. 1996. Swift trust and temporary groups. In R.M. Kramer and
T.R. Tyler (eds.), Trust in Organizations: Frontiers of Theory and Research. Thousand Oaks, CA: Sage,
pp. 166195.
Moreland, R.L. 1999. Transactive memory: Learning who knows what in work groups and organizations. In
L.L. Thompson, J.M. Levine, and D.M. Messick (eds.), Shared Cognition in Organizations. Hillsdale,
NJ: Lawrence Erlbaum, pp. 331.
Moreland, R.L., and Argote, L. 2003. Transactive memory in dynamic organizations. In R.S. Peterson and
E.A. Mannix (eds.), Leading and Managing People in the Dynamic Organization. Mahwah, NJ: Lawrence
Erlbaum, pp. 135162.
Nevo, D., and Wand, Y. 2005. Organizational memory information systems: A transactive memory approach.
Decision Support Systems, 39, 4, 549562.
Nonaka, I., and Konno, N. 1998. The concept of ba: Building a foundation for knowledge creation. Cali-
the Dynamics of Innovation. New York: Oxford.
Norman, P.M. 2002. Protecting knowledge in strategic alliances: Resource and relational characteristics. The
Journal of High Technology Management Research, 13, 2, 177202.
Oxley, J.E. 1997. Appropriability hazards and governance in strategic alliances: A transaction cost approach.
Journal of Law, Economics and Organization, 13, 2, 387409.
. 1999. Institutional environment and the mechanisms of governance: The impact of intellectual
property protection on the structure of interfrm alliances. Journal of Economic Behavior and Organiza-
tion, 38, 3, 283309.
Polanyi, M. 1962. Personal Knowledge: Towards a Post-Critical Philosophy. New York: Harper & Row.
Poppo, L., and Zenger, T. 2002. Do formal contracts and relational governance function as substitutes or
complements? Strategic Management Journal, 23, 8, 707725.
Pisano, G. 1989. Using equity participation to support exchange: Evidence from the biotechnology industry.
Journal of Law, Economics and Organization, 5, 1, 109126.
Ring, P.S., and Van de Ven, A.H. 1992. Structuring cooperative relationships between organizations. Strategic
Management Journal, 13, 483498.
Sanchez, R., and Mahoney, J.T. 1996. Modularity, fexibility and knowledge management in product and
organization design. Strategic Management Journal, 17, 10, 6376.
Scott, S.V., and Walsham, G. 2005. Reconceptualizing and managing reputation risk in the knowledge
economy: Toward reputable action. Organization Science, 16, 3, 308322.
Sitkin, S.B., and Bies, R.J. 1993. The legalistic organization: Defnitions, dimensions, and dilemmas. Orga-
Sobrero, M., and Roberts, E.B. 2001. The trade-off between effciency and learning in interorganizational
relationships for product development. Management Science, 47, 4, 493511.
Subramani, M. 2004. How do suppliers beneft from information technology use in supply chain relation-
ships. MIS Quarterly, 28, 1, 4573.
Szulanski, G. 2003. Sticky Knowledge: Barriers to Knowing in the Firm. London: Sage.
Teece, D.J. 2000. Managing Intellectual Capital: Organizational, Strategic, and Policy Dimensions. New
York: Oxford University Press.
Tierney, K.J.; Lindell, M.K.; and Perry, R.W. 2001. Facing the Unexpected: Disaster Preparedness and
Response in the United States. Washington, DC: Joseph Henry Press.
Trim, P.R. 2005. Managing computer security issues: Preventing and limiting future threats and disasters.
Disaster Prevention and Management, 14, 4, 493505.
Von Hippel, E. 1988. The Sources of Innovation. New York: Oxford University Press.
Walls, J.G.; Widmeyer, G.R.; and El Sawy, O.A. 1992. Building an information system design theory for
vigilant EIS. Information Systems Research, 3, 1, 3659.
Wegner, D.M. 1995. A computer network model of human transactive memory. Social Cognition, 13, 3,
319339.
Winter, S.G. 1987. Knowledge and competence as strategic assets. In D. Teece (eds.), The Competi-
tive ChallengeStrategies for Industrial Innovation and Renewal. Cambridge, MA: Ballinger, pp.
159184.
Zander, U., and Kogut, B. 1995. Knowledge and the speed of the transfer and imitation of organizational
capabilities: An empirical test. Organization Science, 6, 1, 7692.
Yoo, Y., and Kanawattanachai, P. 2001. Developments of transactive memory systems and collective mind
in virtual teams. International Journal of Organizational Analysis, 9, 2, 187208.
309
ABOUT THE EDITORS AND CONTRIBUTORS
Maryam Alavi is the vice dean at the Goizueta Business School of Emory University and John
M. and Lucy Cooks Chair in Information Strategy. She has served as the interim dean and the
dean of faculty and research at Goizueta from 2002 to 2005. Dr. Alavi has authored numerous
scholarly papers in the areas of information technology implementation, decision support systems,
and, more recently, technology-mediated learning and knowledge management. Her published
articles appeared in Management Science, IEEE Software, MIS Quarterly, Information Systems
Research, Journal of Management Information Systems, Academy of Management Journal, and
Organization Science among others. She has served on the editorial boards of several scholarly
information systems journals, including Management Information Systems Quarterly, Information
Systems Research, and Journal of Management Information Systems.
Irma Becerra-Fernandez is the Knight Ridder Research Professor of Management Information
Systems at Florida International University College of Business Administration. Her research focuses
on KM, KM systems, enterprise systems, disaster management, and IT entrepreneurship. She has
studied and advised organizations, in particular, NASA, about KM practices. She founded the FIU
KM Lab and has obtained funding as principal investigator for over $1.8 million from the National
Science Foundation, NASA, and the Air Force Research Lab. She has published extensively in lead-
ing journals, including the JMIS, Decision Sciences, IEEE Transactions on Engineering Manage-
ment, and ACM Transactions on Internet Technology, among others. She is an author of the book
Knowledge Management: Challenges, Solutions, and Technologies. She has delivered many invited
presentations at federal sites, universities around the world, and international conferences. She was
the recipient of the 2004 FIU Outstanding Faculty Torch Award. She serves as the faculty director for
the masters in management information systems (MIS) PhD program, and is on the editorial board
of International Journal of Knowledge and Learning and International Journal of Mobile Learning
and Organization and the International Journal of Doctoral Studies.
Tom Beckman is principal of Beckman Associates, a frm that provides advanced educational,
consulting, research, and assessment services. Until his retirement from the Internal Revenue
Service, he held positions as portfolio manager for IT infrastructure, division chief information
offcer, enterprise architect, systems engineer, chief methodologist for business reengineering, and
artifcial intelligence specialist. In addition, Mr. Beckman has 35 years of experience in success-
fully developing and managing projects that rely on innovative IT and advanced systems design.
Recently, he infused advanced IT and KM principles into a portfolio of projects, as well as conceived
of, organized, and managed the Large Case Analysis data mining project that resulted in an ROI
of 50 to 1 in under 1 year. With Dr. Liebowitz, Beckman co-authored the book on applying KM
to business opportunities, Knowledge Organizations: What Every Manager Should Know, as well
as authoring six chapters, including Creating Business Value from Knowledge Management: The
310 ABOUT THE EDITORS AND CONTRIBUTORS
Fusion of Knowledge and Technology in Knowledge Management: The Catalyst for Electronic
Government. Recent workshops include Engineering Intelligent Organizations: Designing and
Architecting Business Systems at ProjectWorld2005 and Increase the Value of Your Knowledge:
Eight Ways for Organizations to Improve their KM at the CIO Councils KM Working Group. Re-
cent presentations include Knowledge Representation for Semantic Technology at the Semantic
Interoperability Conference and Enterprise Information Management: Designing an Enterprise
Architecture at Braintrust2006 KM Summit. Finally, as adjunct associate professorial lecturer, Mr.
Beckman develops and teaches graduate courses in artifcial intelligence, MIS, KM, management
of innovation, and business reengineering at The George Washington University.
Matteo Bonifacio is a researcher in organization and management sciences on the Faculty of
Economics of the University of Trento in Italy, where he teaches KM and organizational learning.
Currently, he is involved in a team of experts as policy developer at the Directorate General for Edu-
cation and Culture of the European Commission, where he is working for the European Institute of
Technology. From 1996 to 2001, he has been senior consultant at Andersen, where he was involved
in projects dealing with KM and process management in both organizational and technological
terms. From 2001 to 2003, he worked as program manager in the research sector at the Institute
for Scientifc and Technological Research. In particular, he dealt with the organization of funding,
research, and innovation activities in the area of automated reasoning and semantic technologies.
Paul F. Clay is an assistant professor of IS at Washington State University. He received a doc-
torate in MIS from the Kelley School of Business at Indiana University. His research interests
include KM, structuration, and issues of ft among business processes, technology, and individual
workers. His research has appeared in ISR, JAIS, and the proceedings of the Hawaii International
Conference on Systems Sciences.
Alan R. Dennis is professor of IS and holds the John T. Chambers Chair of Internet Systems in
the Kelley School of Business at Indiana University. He received a bachelor of science degree in
computer science from Acadia University, a master of business administration degree from Queens
University, and a doctorate in MIS from the University of Arizona. His research focuses on the
use of computer technologies to support teams, on KM, and on the use of the Internet to improve
business and education. He is the author of more than 100 research papers and has won numerous
awards for his theoretical and applied research. He is the author of four books and serves as the
publisher of MIS Quarterly Executive and as a senior editor for MISQ.
Samer Faraj is associate professor in the Decision and Information Technologies Department at
the Smith School of Business, University of Maryland. He received his doctorate in MIS from the
Boston University School of Management and a master of science degree in technology and policy
from MIT. Prior to getting his doctorate, he spent a decade working in a variety of consulting and IS
positions. His research interests include the coordination of expertise in software development and
health care, the development of online knowledge communities, and the organizational impact of
IT. His work has appeared in journals such as ISR, Management Science, MISQ, Journal of Applied
Psychology, IEEE Transactions on Engineering Management, Journal of Strategic Information
Systems, and Information Technology & People. He is senior editor at Organization Science, an
associate editor of ISR, and on the editorial board of Information and Organization. In addition,
he is currently co-editing the special issue of Organization Science on IT and organizational form
and function. He has received three National Science Foundation grants and a Fulbright award.
ABOUT THE EDITORS AND CONTRIBUTORS 311
Patrick S.W. Fong is an associate professor in the Department of Building and Real Estate at
The Hong Kong Polytechnic University. He was an adjunct professor at the Faculty of Design,
Architecture and Building, University of Technology, Sydney, Australia. His doctorate in indus-
trial and business studies was obtained from the Warwick Business School in England. He has a
multidisciplinary background and his research interests include KM and project management. He
has authored or co-authored many refereed research papers and received citations for excellence
in research, practice, and originality. He co-edited a book entitled Management of Knowledge
in Project Environments (2005) and has contributed chapters in several leading KM books in
construction as well as encyclopedias in KM. Dr. Fong maintains close links with industry and
was vice president of the Hong Kong Knowledge Management Society. He is also a member of
the international body CIB W102-Information and Knowledge Management in Building and a
visiting research scholar of the Royal Institution of Chartered Surveyors.
Thomas Franz is a doctoral student in the Department of Computer Science at the University
of Koblenz-Landau, Germany. His research interests include personal information management,
in particular, knowledge integration and management with respect to computer-mediated com-
munication. He received his master of science degree in computer science from the University
of Freiburg, Germany, and studied at Chalmers University of Technology, Sweden, and Clemson
University, United States. In 2002, he worked as a scientifc assistant in the Knowledge Manage-
ment Research Group of the University of Applied Sciences Cologne, Germany.
Sirkka L. Jarvenpaa is professor and the James Bayless/Rauscher Pierce Refsnes Chair in
business administration at McCombs School of Business at the University of Texas at Austin.
She is the director of the Center for Business, Technology and Law at McCombs School. Dr.
Jarvenpaa has served as the editor-in-chief of the Journal of Association for Information Sys-
tems, co-editor in chief of the Journal of Strategic Information Systems, and the senior editor
for MISQ and ISR.
Richard John is associate professor of psychology and Chair of the Institutional Review Board
at the University of Southern California. His research is on human judgment and decision making
and decision and risk analysis. Dr. John has published in leading journals, including Manage-
ment Science, Organizational Behavior and Human Decision Processes, ISR, and Interfaces. He
received his doctorate in quantitative psychology from the University of Southern California in
1984, master of science degree in applied mathematics from the University of Southern Califor-
nia in 1983, and bachelor of science degree in applied mathematics from the Georgia Institute of
Technology in 1976.
Steven L. Johnson is an IS doctoral candidate in the Decision and Information Technologies
Department at the Robert H. Smith School of Business, University of Maryland, College Park. He
earned a bachelor of science degree in computer science and a master of business administration
degree from the College of William & Mary in Williamsburg, Virginia. Prior to pursuing doctoral
studies, he spent 15 years working for a variety of software and IT frms, including Platinum Tech-
nology and Siebel Systems. His research interests include KM in online communities, large-scale
social network dynamics, and the impact of IT on organizations. His research has been presented
at the International Conference on Information Systems, the Academy of Management Annual
Meeting, the Americas Conference on Information Systems, and the INFORMS Computing
Society Conference.
Gerald C. (Jerry) Kane is a doctoral candidate at the Goizueta Business School of Emory Uni-
versity. His research interests include KM, social networks, and IT in the health care industry. He
has presented research at the International Conference of Information Systems (ICIS) and at the
Academy of Management Annual Meeting. He has published his research in DATABASE.
Timothy R. Kayworth is an associate professor of IS at Baylor University. His research interests
center on the management of IT in organizations. Recent research projects have included such
topics as leadership in global virtual teams, IT security governance, the impact of organizational
culture on KM practice, and the role of culture in IS research. His work has been published in
such journals as MISQ, JMIS, Data Base for Advances in Information Systems, and European
Management Journal. Dr. Kayworth currently teaches in Baylor University full-time as well as
executive MBA programs.
Yongsuk Kim is a doctoral student in McCombs School of Business at University of Texas at Austin.
He received his master of science degree in human computer interaction from the University of
Michigan at Ann Arbor and his bachelor of science degree from Yonsei University, Seoul, Korea.
His research interests include KM, virtual collaboration, and social computing.
Dong-Gil Ko is an assistant professor in the Kelley School of Business at Indiana University. He
holds a doctorate in MIS from the Katz Graduate School of Business. His current research focuses
on the transfer of knowledge, the exercise of control, and the use of systems for improving the
management, performance, and impact of IS-related projects. His published research articles ap-
pear in scholarly journals including Management Science and MISQ. Dr. Ko is an active member
of the Academy of Management and the International Conference on Information Systems, and
has presented the results of his research in several conferences. He has over 8 years of industry
experience as an IS consultant to large organizations.
Dorothy E. Leidner is the Randall W. and Sandra Ferguson Professor of Information Systems at
Baylor University. She has broad international experience, having previously served as associate
professor at INSEAD and as visiting professor at ITESM in Monterrey, Mexico, at the University
of Caen, France, and at the University of Mannheim, Germany. She has published in a variety of
journals, including MISQ, ISR, and Organization Science. Dr. Leidner is currently serving as co-
editor-in-chief of the journal Data Base for Advances in Information Systems. She also is serving
as a senior editor for MISQ and Journal of Strategic Information Systems and as associate editor
for Decision Sciences and Decision Support Systems.
Jay Liebowitz is a full professor in the Carey Business School at Johns Hopkins University. He
is the program director for the graduate certifcate in competitive intelligence at Johns Hopkins
University, and the MS-ITS Capstone Coordinator. Prior to joining Hopkins, he was the frst
knowledge management offcer at NASA Goddard Space Flight Center. Before NASA, Dr. Li-
ebowitz was the Robert W. Deutsch Distinguished Professor of Information Systems at the Uni-
versity of Maryland- Baltimore County, professor of management science at George Washington
University, and chair of artifcial intelligence at the U.S. Army War College. He is the founder and
editor-in-chief of Expert Systems with Applications: An International Journal. He has published
over 35 books and 200 articles dealing with knowledge management, expert/intelligent systems,
and IT management. He also completed a study on Bridging the Knowledge and Skills Gaps:
Tapping Federal Retirees for the IBM Center for the Business of Government. Dr. Liebowitz
is a Fulbright Scholar, IEEE-USA Federal Communications Commission Executive Fellow, and
Computer Educator of the Year (International Association for Computer Information Systems).
He has consulted and lectured worldwide for numerous organizations.
Ann Majchrzak received her doctorate in social psychology from UCLA in 1980 and is professor
of IS at the Marshall School of Business, University of Southern California. Her research is on
the organizational impacts and design of computer-automated work environments with specifc
interests in knowledge sharing, innovation, and distributed cognition. Her recent publications
can be found in ISR, Management Science, MISQ, Harvard Business Review, Sloan Management
Review, and Journal of Knowledge Management.
Arvind Malhotra is an associate professor of entrepreneurship at the University of North
Carolinas Kenan-Flagler Business School. His research interests include innovation, KM, virtual
teams, and interorganizational information sharing. Dr. Malhotras papers have been published
in leading academic journals such as ISR, JMIS, MISQ, Communications of the ACM, Harvard
Business Review, and Sloan Management Review. Dr. Malhotra received his doctorate in business
administration and master of science degree in manufacturing engineering from the University
of Southern California.
Naren Peddibhotla is an assistant professor of Management Information Systems at the School
of Business, State University of New York Institute of Technology at Utica. He received his PhD
in Business Administration from the University of Minnesota in 2006. He also holds a Bachelor of
Technology (B. Tech) degree in Civil Engineering from the Indian Institute of Technology, Bombay,
and a Post Graduate Diploma in Management (PGDM) from the Indian Institute of Management,
Calcutta. He has worked for Asian Paints and Bajaj Electricals in India for a decade in the areas
of sales and information systems. His research interests are in the areas of organizational use of
information technologies for competitive advantage, knowledge management, and the role of
information technologies in individual work. His work has been published in Organization Stud-
ies and presented at the Academy of Management annual meeting and the Behavioral Decision
Research in Management conference.
Rajiv Sabherwal is University of Missouri System Curators Professor and the Emery C. Turner
Professor of Information Systems at University of Missouri, St. Louis. His research focuses on KM,
IS strategy, and social aspects of systems development. It has been published, or is forthcoming,
in journals such as IEEE Transactions on Engineering Management, Management Science, ISR,
Journal of Management Information Systems, MISQ, Organization Science, California Manage-
ment Review, Communications of the ACM, and Decision Sciences. Dr. Sabherwal is a departmental
editor for IEEE Transactions on Engineering Management and serves on the editorial boards for
ISR, Journal of MIS, Management Science, and Journal of AIS. He recently completed a term as a
senior editor at MISQ. He earned his doctorate from the University of Pittsburgh, and a PGDM
from Indian Institute of Management, Calcutta.
Ulrike Schultze is associate professor in IT and operations management at Southern Methodist
University. Her research focuses on the impact of IT on work practices, and she is particularly
interested in knowledge work. Dr. Schultze has written on hard and soft information genres, in-
formation overload, and KM. Her more recent research projects are in the area of Internet-based
technologies and their implications for customer co- and peer-production. Dr. Schultze frequently
relies on multimethod research designs, which include ethnographic observations, interviews,
and surveys. Her research has been published in, among others, ISR, MISQ, and Information &
Organizations. She has served as an associate editor at MISQ and ISR.
Steffen Staab is a professor of databases and IS in the University of Koblenz-Landaus Institute
for Computer Science heading the research group on IS and the Semantic Web (ISWeb). His re-
search interests include technologies for KM, such as the Semantic Web, text mining, ontologies,
peer-to-peer computing, and service management via semantic descriptions. He coordinated the
FP5 EU IST research project SWAPSemantic Web and Peer-to-Peer, and he is co-director of
the FP6 EU IST integrated project X-MediaKnowledge Sharing and Re-use Across Media
and is on the steering board of several other large EU projects. Dr. Staab co-edited The Handbook
on Ontologies (Springer, 2004). He has been co-chair of the German conference on KM, he is a
co-chair of EKAW-2006 (European Conference on Knowledge Engineering and Management)
and SAMT-2006 (Semantic and Digital Media Technology, Athens, Greece), and he is general
chair of ACM IUI-2008 (International Conference on Intelligent User Interfaces).
David Schwartz is a senior lecturer and head of the Information Systems Division of the Graduate
School of Business Administration at Bar-Ilan University, Israel. His research focuses on computer-
mediated communications, KM, and Internet-based systems. Since 1998 he has served as editor of
the internationally acclaimed journal Internet Research. Dr. Schwartzs research has appeared in
publications such as IEEE Intelligent Systems, International Journal of Human-Computer Studies,
IEEE Transactions on Professional Communications, Kybernetes, and Journal of Organizational
Behavior. His books include Encyclopedia of Knowledge Management (2006) and Internet-Based
Organizational Memory and Knowledge Management (2000), for which he served as editor, and
Cooperating Heterogeneous Systems (1995). Dr. Schwartz received his doctorate from Case West-
ern Reserve University, master of business administration degree from McMaster University, and
bachelor of science degree from the University of Toronto, Canada.
Mani Subramani is an associate professor in the Information and Decision Sciences Department
at the Carlson School of Management, University of Minnesota. He received his undergraduate
degree in Electronics Engineering from BITS, Pilani, his Masters degree from the Indian Institute
of Management, Bangalore and his doctoral degree from Boston University. His research focuses
on the strategic role of information technology within the organization and in inter-organizational
relationships. His current areas of research are knowledge management and the leveraging of
organizational capabilities using information technologies. His work has been published in the
Academy of Management Journal, Communications of the ACM, Information Systems Research,
Journal of Management Information Systems, MIS Quarterly, and Sloan Management Review.
He serves as Associate Editor on the editorial boards of MIS Quarterly and Information Systems
Research.
Dov Teeni is professor of IS at Tel-Aviv University. He is also the chairman of Meitalthe national
center for E-learning in Higher Education. Dr. Teeni studies several related areas of IS: human-
computer interaction, support for communication, KM, systems design, and nonproft organiza-
tions. His research usually combines model building, laboratory experiments, and development
of prototypes like Spider and kMail. He serves as senior editor for MISQ and AE for Journal of
AIS, Information and Organizations, and Internet Research. He has published in journals such as
Management Science, MISQ, Organization Science, EJOR, and Communications of ACM and in
more specifc journals of human-computer interaction such as IJHCS, Behaviour and Information
Technology, Computers in Human Behavior, and IEEE Transactions. He is conference co-chair
of ICIS2008 in Paris. He is also co-author of a new book on human-computer interaction for
organizations.
Molly Wasko is an assistant professor in the Department of Management Information Systems at
Florida State University, where she teaches primarily strategic information technologies, corporate
information security, and research methods. She received her doctorate in MIS from the University
of Maryland, College Park, and she holds a master of business administration degree from Averett
University. Prior to getting her doctorate, she spent 8 years working in production and operations
management. Her research interests include the intersection of digital and social networks, social
network analysis, the development of online communities, and open source software projects. Her
work has appeared in the Journal of Strategic IS, Decision Sciences, MISQ, JAIS, and JITTA and
has been presented at ICIS, AOM, and AMCIS. She is a member of the Academy of Management,
AIS, and INFORMS.
Youngjin Yoo is associate professor in MIS and Irwin L. Gross Research Fellow at the Fox School
of Business and Management School of Management at Temple University. Prior to joining
Temple University, he was Lewis-Progressive Chair of Management at the Weatherhead School of
Management at Case Western Reserve University. He holds a doctorate in IS from the University
of Maryland. He received his master of business administration and bachelor of science degrees
in business administration from Seoul National University in Seoul, Korea. His work covers a
wide range of topics broadly related to IT-enabled innovation and organization transformation.
In particular, he is interested in integrating design approaches in managing innovations and IT,
KM, and ubiquitous computing. He was also a research associate at NASA Glenn Research Cen-
ter in Cleveland, OH. His work was published at leading academic journals such as ISR, MISQ,
Organization Science, Communications of the ACM, Academy of Management Journal, Journal
of Strategic Information Systems, Journal of Management Education, and Information Systems
Management. He also edited two books on ubiquitous computing and wrote several book chapters.
He is an associate editor of ISR and on the editorial board of Organization Science, Journal of
AIS, and Information and Organization.
317
SERIES EDITOR
Vladimir Zwass is the University Distinguished Professor of Computer Science and Management
Information Systems at Fairleigh Dickinson University. He holds a PhD in Computer Science from
Columbia University. Professor Zwass is the Founding Editor-in-Chief of the Journal of Manage-
ment Information Systems, one of the three top-ranked journals in the feld of Information Systems;
the Journal has celebrated 20 years of its publication. He is also the Founding Editor-in-Chief
of the International Journal of Electronic Commerce, ranked as the top journal in its feld. More
recently, Dr. Zwass has been the Founding Editor-in-Chief of the monograph series Advances
in Management Information Systems, whose objective is to codify the knowledge and research
methods in the feld. Dr. Zwass is the author of six books and several book chapters, including
entries in the Encyclopaedia Britannica, as well as of a number of papers in various journals and
conference proceedings. Vladimir Zwass has received several grants, consulted for a number of
major corporations, and is a frequent speaker to national and international audiences. He is a former
member of the Professional Staff of the International Atomic Energy Agency in Vienna, Austria.
319
INDEX
A
abstraction, 154
ad hoc interorganizational collaboration, 9,
292307
cost economics perspective, 29596
interorganizational information systems (IOS),
295
interorganizational relations (IOR), 295
knowledge-based perspective, 29697
modularity-based perspective, 299
relational-based perspective, 29798
transactive memory theory, 300301
and knowledge protection, 29295
interorganizational relations, 295301
and knowledge sharing, 29295
sticky or tacit knowledge, 29697
summary of perspectives, 3023
trust, 298
virtuality, 294, 298, 299
volatile situations, 294
vulnerable collaborations, 29394
adaptation, 154
agent-based simulation, electronic knowledge
networks, 28487
Answer Garden, 236
ARPANET, 91
automation (or guiding) of reasoning, 91
Autonomy Agentware, 88, 97, 99, 101
B
Barcelona Olympics (1992), 138
behaviorist theory, 242
benchmarking, 170
Bibster system, 115
BTX Web site, 272
business strategy, K3M. See Management
maturity model (K3M)
C
capability building, projects. See Projects,
management of knowledge and learning
capability management (CM), 16768
centrality, social networks, 7577
access, 75
communication support capabilities (CSC), 77
knowledge-brokers, 76
referrals, 75
timing, 75
centralized knowledge, 107
KM systems, centralized, 114
knowledge sharing, 105
four-layer model for information technology,
1089
Churchill, Winston, 158
CL. See Common LISP language
cliques, 74
CMC. See Computer-mediated communication,
organizational form and
CMC impact cycle. See Computer-mediated
communication, organizational form and
CMS. See Content management systems (CMS)
collaborative systems, 233
collective action, 275, 278
collectivist organizations, 5455
combination processes, 2829
common knowledge, 91
Common LISP language, 88, 9194, 99100
schematic, 93
communication support capabilities (CSC),
6768
320 INDEX
communication support capabilities (CSC)
(continued)
centrality, 77
density in, 7475
future research, 7879
shared background knowledge, 78
tie strength in, 7172, 79
trust, social networks, 78
competence management systems (CMS), 9799
competence-in-the-making, 97
competence-in-use, 97
implicit competence, 99
strategic competence development, 99
competency modeling, 16768
complex, yet repeatable decisions, 91
computer-mediated communication
electronic knowledge networks, 272
organizational form and. See Computer-
mediated communication, organizational
form and
computer-mediated communication,
organizational form and, 8, 14562
CMC impact cycle, 146, 149, 15758
CMC, KT, OF triumvirate, 146
interplay, 150
core concepts, 14749
functional interdependencies, 149, 151, 158
impact of CMC, 14951
CMC impact cycle, 146, 149, 15758
interplay between OF, CMC and KT, 150
interpersonal relations, 149, 151, 156, 157
knowledge, 14849, 151
knowledge communication, concept of, 152
knowledge transfer, relationship
abstraction, 154
adaptation, 154
context, 15354
generally, 147, 15657
levels of abstraction, 154
organizational memory, 15455
relationship, 15152
role of communication in knowledge
transfer, 15253
role of knowledge transfer in
communication, 15355
memetics, 145
organizational form, role of
generally, 15556
interpersonal relationships, 156, 157
shared understanding, 156, 157
related work, 14749
computer-mediated communication,
organizational form and (continued)
space, 148, 151
time, 148, 151
congenital learning, 216
content management systems (CMS), 114
context, knowledge and communication,
15354
contextual information, 153
contextualization, 15354
core competences, 97
critical mass theory, 278
CSC. See Communication support capabilities
(CSC)
CYC project, 101
Cycorp, 101
D
data, described, 4
decentralized KM systems, 11415
decentralized knowledge sharing, 10910
decision making, organizational, 4344
decision support systems (DSS), 41, 4344
Deep Dive process, IDEO, 13334
density, social networks, 7275
cliques, 74
7475
environmental context, 74
knowledge sharing goals, 74
network contexts, 74
supported social networks, 7475
Department of Defense
ARPANET, 91
CL. See Common LISP language
design gestalt, 142
deutero-learning, 216
Dewey, John, 129
differentiation perspective, organizational
culture, 49
DMS. See Document management systems
(DMS)
document management systems (DMS), 114
doubles-loop learning, 216
DSS. See Decision support systems (DSS)
DTMS. See Dynamic team memory systems
dyadic analysis, 66
dynamic team memory systems, 9, 24869
INDEX 321
dynamic team memory systems (continued)
conceptual development, 24953
integrity of knowledge, maintaining over
time, 25051
knowledge sharing effectiveness, 249, 25253
pattern matching, 251
virtual workspaces, organizational memory
system framework, 25052
integrity of knowledge, maintaining over time,
25051
model summary, 253
pattern matching, 251
research methods, 25358
control variables, 257
discriminant validity, assessing, 257
expertise location within team, 257
exploratory factor analysis of key latent
constructs, 258
knowledge sharing effectiveness, 254
means, standard deviations, and item
loadings of indicators, 259
measures, 25457
sample, 25354
task analyzability, 257
use of DTMS, 25557
virtual workspaces, 25557
results, 25863
assessing measurement model, 25859
average variance extended, 260
construct correlations, 260
control variables, 260, 26263
direct effects, 262
discriminant validity, 25859
expertise location, 26263
fnal PLS, 263
internal consistency, 260
nested PLS models, path coeffcients and
explained variances for, 261
structural model, 26063
task analyzability, 26263
virtual workspaces, 264
task analyzability, 257, 26263, 26566
virtual workspaces
organizational memory system framework,
25052
research methods and, 25557
E
ego network analysis, 66
EIS. See Executive information systems (EIS)
EKNs. See Electronic knowledge networks
electronic knowledge networks, 9, 27091
agent-based simulation, 28487
collective action, 275, 278
computer-mediated communication, 272
critical mass theory, 278
cues fltered out approach, 272
electric networks of practice, 271
electronic knowledge network empirical
research, 28081
helping others, reasons for, 27479
individual motivations and resources,
27677
relational social capital, 277
structured social capital, 27879
knowledge, reasons for contributing, 27479
large-scale networks, 280
larger-world networks, 279
mutuality index (MI), relative effects, 287
nature of network interactions, 28182
network aspects, 27987
electronic knowledge network empirical
research, 28081
large-scale networks, 280
larger-world networks, 279
mutuality index (MI), relative effects, 287
nature of network interactions, 28182
network size, determinants, 283
100 largest web-based message boards form
power-law distribution, 281
p-star analysis of three electronic knowledge
networks, 285
power-law distributionlog-log
transformation to linear representation,
280
power-law distribution coeffcient, relative
effects, 287
research directions, 293
scale-free networks, 280
small-world networks, 279
structuring mechanisms, 28487
network size, determinants, 283
100 largest web-based message boards form
power-law distribution, 281
networks, 285
participation dynamics, technology and,
27374
power-law distribution
coeffcient, relative effects, 287
322 INDEX
electronic knowledge networks (continued)
power-law distribution
log-log transformation to linear
representation, 280
100 largest web-based message boards, 281
research directions, 293
scale-free networks, 280
small-world networks, 279
structuring mechanisms, 28487
early empirical evidence, 284
technology, role of, 27274
participation dynamics, 27374
virtual groups, 271
embedding of knowledge management, 2122
Enterprise 2.0, 21
evolutionary model of shared knowledge,
11012
exchange, group learning, 2728
executive information systems (EIS), 41, 4344
experience economy, challenges, 12831
experiential learning, 216
expert knowledge, 91
externalization
group level, 26, 27, 28
individual level, 23, 25
virtual communities within organizations,
23435
F
Farmaco. See Pharmaceutical industry,
knowledge management systems
frst-generation knowledge management, 31
Fish Sculpture, 1992 Barcelona Olympics, 138
FORTRAN, 97
four-layer model for information technology, 7,
10425
Bibster system, 115
centralized KM systems, 114
centralized knowledge, 107
combination of knowledge, 120
content management systems (CMS), 114
decentralized samples, 11415
dimensions for knowledge sharing, 11112
document management systems (DMS), 114
dynamic access control, 118
dynamic individual knowledge structures,
supporting, 12021
evolution of organization, 1067
four-layer model for information technology
(continued)
evolutionary knowledge management, 11718
dynamic access control, 118
individual, support for, 11718
interconnecting information, 118
multiple knowledge sources, integration of,
117
Gnowsis systems, 113
individual ontologies, automating creation of,
11819
individual, support for, 11718
interconnecting information, 118
knowledge management today, 11221
knowledge sharing with individual ontologies,
11920
local knowledge, 107
local knowledge management, 11214
multiple knowledge sources, integration of,
117
networking, degrees of, 10712
omitting knowledge sharing, 116
ontology matching example, 120
organization as network, 1045
Peer-to-Peer (P2P) systems, 11415, 121
personal management (PIM) applications,
11213
personalization lacking, 117
requirements, 10712
research and technologies, 11821
combination of knowledge, 120
dynamic individual knowledge structures,
supporting, 12021
individual ontologies, automating creation
of, 11819
knowledge sharing with individual
ontologies, 11920
methods and tools, 119
ontology matching example, 120
trust to control access, 121
security issues, posing, 117
sharing knowledge, layered model of, 10711
centralized sharing of knowledge, 1089
decentralized sharing, 10910
dimensions of knowledge sharing systems,
112
11012
individual knowledge management, 108
shortcomings of current systems, 11517
omitting knowledge sharing, 116
INDEX 323
shortcomings of current systems (continued)
personalization lacking, 117
security issues, posing, 117
single sourced systems, 11617
single sourced systems, 11617
trust to control access, 121
use cases, 10712
Xarop, 115
functional interdependencies
computer-mediated communication, 149, 151,
158
G
gatekeeping, 94
Gehry, Frank O., 128, 13233, 13539
Gehry Partners, 13233, 13539
Fish Sculpture, 1992 Barcelona Olympics,
138
implications of example, 13941
Lewis building, early sketch, 137
models and scales, 136
3D models, examples, 137
globalization, 8, 89
Glymph, Jim, 138
Gnowsis systems, 113
goodness-of-ft
pharmaceutical industry, knowledge
management systems, 196
grafting, 216
grapeVINE, 88, 9496, 99101
schematic, 95
group learning. See Individual, group, and
organizational learning
H
Hewlett-Packard Consulting, 17172
I
ICT. See Information and communication
technologies (ICT)
IDEO, 132, 13335
Deep Dive process, 13334
implications of example, 13941
Patient Journey Map, 133
prototypes, 135
IMC. See Information management capabilities
(IMC)
individual, group, and organizational learning,
6, 1339
embedding of knowledge management, 2122
frst-generation knowledge management, 31
knowledge management systems and, 2230
exchange, 27
externalization, 23, 25, 26, 27
group level, 2628, 3134
internalization, 23, 25
organizational level, 2830
routines, 27
perspective making, 3033
perspective taking, 2021, 3033
second-generation knowledge management,
3132
third-generation knowledge management,
3233
individual knowledge structures, supporting,
12021
Information and communication technologies
(ICT), 1415, 2230
information culture, knowledge management
and, 51
information, described, 45
information management capabilities (IMC),
7880
information systems
advances in, 4147
executive information systems (EIS), 41,
4344
knowledge management systems (KMS),
4446
management information systems (MIS),
4243, 46
organizational structure, 4243
summary of, 46
information technology
four-layer model, 7, 10425
knowledge management technology, 7, 86103
organizational culture and, 42
role in knowledge management, 67
social networks and, 7, 6385
integration perspective, organizational culture, 49
intelligent agent and expert system forums, 21
interaction design, 13941
virtual communities within organizations, 235
324 INDEX
interorganizational collaboration. See Ad hoc
interorganizational collaboration
interorganizational information systems (IOS),
295
interorganizational relations (IOR)
ad hoc interorganizational collaboration, 295
interpersonal relationships, computer-mediated
communication, 149, 151, 156, 157
interpretation, 215
interpretive reading forums, 21
interproject focus, 212
interproject learning, 219220
interdependency of intraproject and
intraproject focus, 209, 212
intraproject learning, 21718
IS. See Information systems
IT. See Information technology
J
justifed true belief, xiii
K
Kazaa, 114
KDEs Kontact, 113
KMMM. See Management maturity model
(K3M)
KMS. See Knowledge management systems
(KMS)
knowledge. See also specifc topic
defned, viiviii
social networks, 6465
described, 5
embedded view of, 65
history of, 6667
support of, use of IT, 6769
knowledge application systems, 7, 26
knowledge-based perspective
ad hoc interorganizational collaboration,
29697
knowledge-brokers, 76
knowledge capture systems, 6, 26
knowledge communication, concept of, 152
Knowledge-Creating Company (Nonaka), 131
knowledge creation, organizational culture and,
5051
knowledge discovery systems, 6
knowledge exploitation, 21011
knowledge management. See also specifc topic
evolvement of discipline, 3
generally, 67
history, 8890, 13133
overview, 6, viixi
knowledge management systems (KMS)
build-and-they-will-come approach, 7
computer-mediated communication. See
Computer-mediated communication
individual, group, and organizational learning,
2230
exchange, 27
externalization, 23, 25, 26, 27
routines, 27
and information and communication
technologies (ICT), 1415
K3M. See Management maturity model (K3M)
learning, 2230
SECI process, 22
learning and, 2230
mechanisms and technologies, 24
SECI process, 22
mobilizing knowledge. See Mobilizing
knowledge
and organizational learning, 13, 15
pharmaceutical industry. See Pharmaceutical
industry, knowledge management
systems
projects. See Projects, management of
knowledge and learning
and virtual communities within organizations,
229
Yu-Gi-Oh! cards, as example of mobilizing
knowledge, 8, 127, 142
INDEX 325
knowledge management technology, 7, 86103
automation (or guiding) of reasoning, 91
Autonomy Agentware, 88, 97, 99, 101
common knowledge, 91
Common LISP language, 88, 9194, 99100
schematic, 93
competence management systems (CMS),
9799
competence-in-the-making, 97
competence-in-use, 97
implicit competence, 99
strategic competence development, 99
complex, yet repeatable decisions, 91
core competences, 97
CYC project, 101
expert knowledge, 91
gatekeeping, 94
grapeVINE, 88, 9496, 99101
schematic, 95
Listservs, 88, 9194
Lotus Notes groupware, 88, 9496
MYCIN, 88, 9091, 99101
schematic of, 91
Volvos VIP prototype, 9799, 100
schematic, 98
knowledge protection
29295
knowledge representation forums, 21
knowledge sharing. See also Knowledge sharing
effectiveness
29295
four-layer model for information technology,
10711
decentralized sharing, 10910
dimensions of knowledge sharing systems,
112
11012
goals, 74
with individual ontologies, 11920
management maturity model (K3M), 174, 176
organizational culture and, 50
social networks and, 80
shared background knowledge, 77, 78
knowledge sharing (continued)
systems, 6
23740
knowledge sharing effectiveness
research methods, 254
knowledge transfer
computer-mediated communication. See
Computer-mediated communication,
organizational form and
group-level aspects, 239
23740
KSE. See Knowledge sharing effectiveness
K3M. See Management maturity model (K3M)
L
learning
knowledge management systems and, 2230
SECI process, 22
legitimate peripheral participation, 235
Lenant, Doug, 101
lessons learned systems (LLS), 21213
Lewis building, early sketch, 137
Linux, 235
LISP CoP. See Common LISP language
Listservs, 88, 9194
LLS. See Lessons learned systems (LLS)
local knowledge, 107
Lotus Notes, 235
groupware, 88, 9496
loyal use, pharmaceutical industry. See
Pharmaceutical industry, knowledge
management systems
M
Machlup, Fritz, vii
management information systems (MIS), 4243,
46
management maturity model (K3M), 8, 16379
ad hoc level, 172
analysis model, 173
benchmarking, 170
business strategy, 16468
alternative approaches to formulation, 166
326 INDEX
management maturity model (K3M)
business strategy (continued)
capability management (CM), 16768
competency modeling, 16768
formulation, 16568
frameworks, 165, 167
key stakeholders, identifcation of, 167
mapping learning, knowledge, and business
strategies, 176
taxonomies, 165, 167
unifed business strategy formulation
methodology, 169
unifed formulation methodology, 168
value propensities, developing, 167
defned level, 172
developing K3M model, 17477
effective implementation, 173
indoctrination learning style, 174
instructor level, 174
interlinking of assets or resources, 170
key indicators for levels, 175
knowledge sharing profciencies, 174, 176
managed level, 17172
mapping learning, knowledge, and business
strategies, 176
maturity levels, 17172, 174, 175
maturity model defned in terms of knowledge
forms, 176
optimized level, 171
penetration stage, 174
repeatable level, 172
review of maturity models, 17174
strategy background, 16871
unifed business strategy formulation
methodology, 169
utility stage, 174
memetics, 145
Meta-knowledge, 141
Microsoft Outlook, 113
MIS. See Management information systems (MIS)
mobilizing knowledge, 8, 12744
experience economy, challenges, 12831
and future research, 14142
Gehry Partners example, 13233, 13539
implications, 13941
IDEO example, 132, 13335
implications, 13941
and interaction design, 13941
modularity-based perspective
Mozilla Suite, 113
multilevel cultural analysis, 5456
multiple knowledge sources, integration of, 117
mutuality index (MI), electronic knowledge
networks, 287
MYCIN, 88, 9091, 99101
schematic, 91
N
Napster, 114
Networked/Social Semantic Desktop, 121
networks and networking. See also Electronic
knowledge networks; Social networks
classifcation by degrees of networking,
10712
dimensions for knowledge sharing, 11112
organization as network, 1045
sharing knowledge, layered model of,
10711
use cases, 10712
Nintendo Gameboy, 127
nodes, social network research, 66, 79
Nonaka, Ikujiro, 131
O
OCI. See Organizational Culture Inventory
(OCI)
organizational culture, 6, 4060
collectivist organizations, 5455
cooperatively disposed individuals and, 55
differentiation perspective, 49
diversity, culture, 4950
effectiveness of knowledge management,
culture and, 5152
empirical research, 53
executive information systems (EIS), 41
framework, 4750
individual personality, relationship between, 55
information culture, knowledge management
and, 51
information systems, advances in, 4147
executive information systems (EIS), 41,
4344
knowledge management systems (KMS),
4446
management information systems (MIS),
4243, 46
INDEX 327
organizational culture
information systems, advances in (continued)
organizational structure, 4243
summary of, 46
integration perspective, 49
knowledge creation, 5051
knowledge management evolution and, 52
longitudinal perspective, 56
multilevel cultural analysis, 5456
Organizational Culture Inventory (OCI), 51
research, future directions, 5456
three-level model, 4748
values approach, 48
work practices, 49
Organizational Culture Inventory (OCI), 51
organizational form and computer-mediated
communication. See Computer-mediated
communication, organizational form and
organizational knowledge management, ix
organizational learning, 21516. See also
Individual, group, and organizational
learning
citations in literature, 1719
combination processes, 2829
evolution of knowledge management, 6, 1422
evolution to knowledge management, 1520
organizational memory, ix
information systems framework, 266
and knowledge and communication, 15455
virtual workspaces, dynamic team memory
systems, 25052
organizational resources, 23334
organizational structure and information
systems, 4243
organizations. See also Organizational culture
knowledge management systems, role of
computer-mediated communication, 8,
14562
generally, 78
management maturity model (K3M), 8,
16379
mobilizing knowledge, 8, 12744
pharmaceutical industry, 8, 180203
projects, management of knowledge and
learning, 8, 20426
teams, role within and across, 89
virtual communities within. See Virtual
communities within organizations
P
networks, 285
Patient Journey Map, IDEO, 133
pattern matching, dynamic team memory
systems, 251
PDCA learning loop, 216
Peer-to-Peer (P2P) systems, 11415, 121
personal management (PIM) applications,
11213
perspective making, 3033
perspective taking, 2021, 3033
management systems, 8, 180203
description of Farmaco, 181
developing system, 18185
Farmaco, described, 181
frst feld sales KMS, 18182
second feld sales KMS, 18284
T-structure, 184
third feld sales KMS, 18485
ease of use, 191, 19899
extrinsic motivation, 19192
Farmaco. See lines throughout this topic
goodness-of-ft, 196
impact of knowledge on performance,
18586
information quality, 192
instrument pretesting and refnement, 19394
item development, 193
knowledge content, 199
knowledge quality, 19293, 19798
lessons learned from Farmacos experiences,
197200
loyal use, 190
lessons learned from Farmacos
experiences, 19798
measurement properties, assessment of,
19597
method, factors affecting use, 19397
goodness-of-ft, 196
instrument pretesting and refnement,
19394
item development, 193
measurement properties, assessment of,
19597
model testing, results of, 195
procedure, 195
results, 197
survey items, 194
328 INDEX
management systems (continued)
method, usage management, 18688
analysis technique, 18788
data, 18687
results, 188, 189
model testing, results of, 195
perceived ease of use, 191
perceived usefulness, 19091, 199
survey items, 194
system quality, 192
lessons learned, 19798
T-structure, 184
use and performance of KMS, 18581
impact of knowledge on performance,
18586
method, 18688
use, factors infuencing, 18897
extrinsic motivation, 19192
information quality, 192
knowledge quality, 19293
loyal use, 190
method, 19397
perceived ease of use, 191, 198
perceived usefulness, 19091, 199
system quality, 192
voluntariness, 191, 200
voluntariness, 191, 200
PIM applications. See Personal management
(PIM) applications
podcast, 129
portal technologies, 31
profts
improvement, pharmaceutical industry. See
management systems
project capability building. See Projects,
management of knowledge and learning
Project Management Body of Knowledge
(PMBOK), 209
Project Management Institute (PMI), 2067
learning, 8, 20426
defnition of project, 2067
features of project-based organizations, 2078
future directions, 222
interproject focus, 212
intraproject focus, 209, 212
knowledge exploration, 211
knowledge management, generally, 21113
learning (continued)
knowledge perspective, 21011
lessons learned systems (LLS), 21213
organization of projects, 20611
perspectives of projects, differing, 20811
intraproject focus, 209
knowledge exploration, 211
knowledge perspective, 21011
resource perspective, 2089
project-based learning, 21621
intraproject learning, 21718
project capability building, 21421
congenital learning, 216
deutero-learning, 216
doubles-loop learning, 216
experiential learning, 216
grafting, 216
interpretation, 215
organizational learning, 21516
PDCA learning loop, 216
project-based learning, 21621
searching and noticing, 216
single-loop learning, 216
vicarious learning, 216
resource perspective, 2089
stakeholders involved in project, 206
types of knowledge practices, 213
unique, defned, 206
P2P systems. See Peer-to-Peer (P2P) systems
R
reengineering movement, 89
referrals, 75
relational-based perspective
29798
relational social capital, 277
routines, group learning, 2728
S
Scheins three-level model of organizational
culture, 4748
searching and noticing, 216
SECI process, 22
INDEX 329
second-generation knowledge management,
3132
shared background knowledge, social networks,
77, 78
ShareNet, 242
Simon, Herbert, 3
single-loop learning, 216
social capital
relational, 277
structured, 278
social infrastructure of frm, 23334
social learning theory, 242
social networks, information technology and, 7,
6385
active approach, research, 66
centrality, 7577
6768, 7879
components of social networks, 6566
density, 7275
dyadic analysis, 66
features of knowledge, 6978
future research, 7880
generative rules, research, 66
history of knowledge, 6667
information management capabilities (IMC),
7879
knowledge, defned, 6465
knowledge sharing, 77, 78, 8081
nodes, research, 66, 79
passive perspective, research, 66
research foundations, 6469
active approach, 66
6768
components of social networks, 6566
dyadic analysis, 66
generative rules, 66
history of knowledge, 6667
knowledge, defned, 6465
nodes, 66, 79
passive perspective, 66
support of knowledge, use of IT, 6769
technology diffusion, 69
whole network analysis, 66
shared background knowledge, 77, 78
support of knowledge, use of IT, 6769
technology diffusion, 69
social networks, information technology and
(continued)
tie strength, 6972, 79
trust, 7778
socialization
learning, group level, 26, 28
virtual communities within organizations, 235
Springsteen, Bruce, 272
sticky or tacit knowledge, 29697
storage of knowledge
23637
structured social capital, 27879
T
T-structure, 184
task narrative forums, 2021
team memory systems. See Dynamic team
memory systems
teams within and across organizations, 89
292307
electronic knowledge networks, 9, 27091
virtual communities within organizations, 9,
22947
technology. See also Information technology
diffusion, social networks, 69
electronic knowledge networks, role in,
27274
participation dynamics, 27374
theory building forums, 21
third-generation knowledge management,
3233
three-level model of organizational culture,
4748
tie strength
in communication support capabilities, 7172
defned, 6970
effects of, 70
weak/strong ties, characteristics between, 72
transactive memory, 141
interorganizational relations (IOR), 300301
trust
access control and, 121
social networks, 7778
78
330 INDEX
U
U.S. Department of Defense. See Department of
Defense
user loyalty, pharmaceutical industry. See
management systems
V
VAX computers, 89
vicarious learning, 216
VIP. See Volvos VIP prototype
virtual communities within organizations, 9,
22947
application of knowledge gained, 24041
background, 23134
and behaviorist theory, 242
boundary objects, 234
collaborative systems, 233
combination, 235
creation of knowledge, 234
externalization, 23435
forums for communication and resource
leverage, 232
future research directions, 24244
internal markets for knowledge, 234
internalization, 235
knowledge application, 24344
knowledge processes, 23442
application of knowledge gained, 24041
combination, 235
externalization, 23435
internalization, 235
knowledge creation, as forums for,
23436
knowledge transfer, 23740
outcomes linked to virtual communities,
24142
socialization, 235
storage of knowledge, 23637
knowledge transfer, 23740
level of analysis, 23234
virtual communities within organizations
(continued)
methodological approaches
examination of virtual communities, 232
prior research, 232
organizational resources, 23334
outcomes linked to virtual communities,
24142
social infrastructure of frm, 23334
and social learning theory, 242
socialization, 235
sources, list of, 231
storage of knowledge, 23637
suppliers, 234
technologies supporting, 230
users of knowledge, 234
virtual groups
electronic knowledge networks, 271
virtual workspaces
dynamic team memory systems
organizational memory system framework,
25052
research methods, 25557, 264
virtuality
299
voluntariness
management systems, 191, 200
Volvos VIP prototype, 9799, 100
schematic, 98
W
Web 2.0 technologies, 141
WiKi, 141
X
Xarop, 115
XCON, 89
Y
experience economy, 12930

Knowledge Management 9780765616371

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Knowledge Management 9780765616371

Uploaded by

Copyright:

Available Formats

KNOWLEDGE

You might also like