The Emerald Research Register for this journal is available at www.emeraldinsight.

com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0309-0566.htm

EJM 38,3/4

COMMENTARY

Action research in marketing

Chad Perry
Southern Cross University, Gold Coast, Australia, and

310

Evert Gummesson
Stockholm University, Stockholm, Sweden
Keywords Action learning, Marketing, Learning methods, Marketing strategy Abstract Develops a denition of action research that is particularly suitable for marketing and based on the articles in this issue of European Journal of Marketing, emphasising the breadth of action research in marketing and its distinctive interest in analytic generalisation, that is, in building a theory that extends beyond the particular situation that is being action researched to other situations.. The three sections of this commentary include: denition of traditional action research, action learning and case research. Second, drawing of four implications from the articles within this special issue about how action research can be done in marketing. Finally, presents a broad denition of action research in marketing.

European Journal of Marketing Vol. 38 No. 3/4, 2004 pp. 310-320 q Emerald Group Publishing Limited 0309-0566 DOI 10.1108/03090560410518567

The term action research was invented by the eminent social scientist Kurt Lewin over half a century ago (Lewin, 1946). Since then, it has become acclaimed and criticised. Reasons for controversy are that the label of action research is rather broad, is often left undened, or it used in different ways (Coghlan and Brannick, 2001). For example, Gummesson (2000) distinguishes four types of action research for management: societal action science (the traditional type where researchers help underprivileged groups to solve problems), management action science (where the purpose is to understand organizations, markets and customers better, usually to make an operation more efcient), real-time action science (working in a research project planned for action research), and retrospective action science (letting past experience and action through later scholarly reection become data in a research project). Thus the aim of this commentary is to develop a denition of action research that is suitable for marketing in particular. That categorisation is based on the articles in this issue and emphasises the breadth of action research in marketing and its distinctive interest in analytic generalisation, that is, in building a theory that extends beyond the particular situation that is being action researched to other situations. This editorial has three sections. First, it denes traditional action research, action learning and case research. Then four implications are drawn from the articles in this special issue about how action research can be done in marketing. Finally, a broad denition of action research in marketing is presented.

Traditional action research Consider the traditional form of action research rst. There are four elements of a traditional action research project (this discussion is based on Carson et al., 2001 and the references listed there). That is, traditional action research involves: (1) a group of people who use spiralling cycles of activities that involve planning, acting, observing and reecting upon what had happened, shown diagrammatically in Figure 1; (2) to try to improve workgroup processes of action; (3) that help to solve complex, practical problems about which little is known; and (4) produces at least one report to the workgroups organisation about what was found. During these spiralling cycles, what is the relationship between an individual researcher and the other members of the action research group? How do an action researcher and others in the group interact? A key point is that an action researcher and his or her clients differ in knowledge. Clients are the problem owners and they have experience-based knowledge from their actual context.

Action research in marketing

311

Figure 1. The spiralling cycles of activities of traditional action research

EJM 38,3/4

312

In contrast, the researcher has her or his theorybased knowledge, but such knowledge can be crucial to more precisely identify actual problems, clarify implicit assumptions, and through interaction and training change a clients perspective on the need to undertake actions for improvements (Argyris, 1983). In more detail, there are three levels of researcher participation in an action research project: technical, practical and emancipatory (Carr and Kemmis, 1986). In the rst, technical level of participation the action researcher is merely a technical expert, a consultant who tells other people what to do. This is the normal form of a consultants project; for example, a technical agribusiness consultant is working in a grain development project in a developed country and simply transfers the technology across. This is the simplest form of action research and may not even meet the requirements of traditional action research noted above. The second, practical level of participation by a researcher is like the starting point of a process consultant (Schein, 1990), where the researcher has a Socratic role, encouraging participation and reection about processes so that others can learn about learning about doing, and not just learn about doing. The researcher helps the client understand of how he or she ts into a system. The third, emancipatory level of researcher participation is the ideal according to some action researchers (Carr and Kemmis, 1986). Here the researcher becomes a co-researcher with the other people, for responsibility for the project is shared equally among everyone. In emancipatory action research, the researchers aim to change the whole context of the problem and thus liberate themselves from its causes, including their mental context. That is, this type of participation:
. . . aims not only at technical and practical improvement [technical] and the participants better understanding [practical] . . . but also at changing the system itself and/or those conditions which impede desired improvement in the system or organisation. It also aims at the participants empowerment and self condence (Zuber-Skerritt, 1996, p. 5).

This third level of participation is indeed an ideal one and is probably only really achieved by revolutionaries who change the structure within their whole organisation or community. In other words, this type of action researcher becomes a transformative intellectual who transforms the view that people had of their world and so emancipates them from their mental prison bars. When emancipatory action research is being done in Latin America, for example, it might aim to lead to revolutions to liberate the poor (Freire, 1972). When emancipatory action research is done in education, it can lead to more democratic classrooms. When it is done in business, it may not be so dramatic but it can lead to new ways of thinking that restructure processes and save costs. For example, action researchers from several departments can come together to look at functional interrelationships affecting what was at rst thought to be just a marketing problem. This traditional form of action

research is rare in marketing. One reason may be that marketing is highly controlled by the market and its external forces, above all customers and competitors, as noted in this special issues commentary In marketing, the companys external environment is always more important than the internal . . . The external environment is neither particularly knowledgeable nor interested in the company and its development. (Gummesson, 2000, p. 105). In contrast, traditional action research is more focused on employee resolution of internal affairs. A second reason is that marketing research as taught and practised in business school today, emphasises the positivist survey methodology as the highway to rigorous research and generalizable results. This positivistic, theory testing research retards theory development in marketing. Moreover, it is too often oriented to techniques rather than to useful results that can improve understanding of essential phenomena and have an impact on enterprises and society, as shown by the extreme disinterest in business school research by managers. In one survey of marketing managers, clear evidence was obtained that academic marketing journals are neither read nor recognised by the great bulk of the sample (Mckenzie et al., 2002, p. 1196). Similarly, a thorough investigation of senior managers, new MBA recruits into rms, their superiors and consultants, found that business school research was denitely not useful to practising managers:
... as far as we could tell, many key managers and executives pay little or no attention to such research and ndings [of business academics] . . . the direct impact appears nil . . . not a single [manager] . . . who was interviewed cited the research of business schools as either their most important strength or their major weakness.. The business world is . . . ignoring the research coming out of business schools (Porter and McKibbin, 1988, p. 180; emphases added).

Action research in marketing

313

A third reason is that marketing researchers are ignorant of action research and even if they have read about it there are few academic environments where they are encouraged to use it. The action research project becomes high risk Where can we nd examiners and reviewers that understand it and will accept it? and is consequently hazardous to an academic career. A fourth reason is that action research could be considered to be too demanding. Traditional action research requires involvement, a secure personality and creativity as well as an initial preunderstanding of business practices beyond those provided in marketing textbooks (Gummesson, 2000, p. 15). It also requires an ability to make decisions, take action, and balance the split personality of a simultaneously involved actor and detached scholar. These demands of traditional action research are explored in articles in this special issue. In brief, a key difference between traditional action research and other methods like surveys is that the action researcher is both an actor and a researcher and attempts to contribute both to practice and academe. It requires the involvement of the researcher as a consultant and expert, a facilitator or an

EJM 38,3/4

interventionist in a change process, and a desire to dig into complex mechanisms by living them. Action learning is a sub-set of action research The traditional action research above can be distinguished from action learning (Zuber-Skerritt, 2001).Traditional action research necessarily focuses on a workgroup within an organisation or community, all of whom are involved in joint cycles of planning/acting/observing/reecting. In contrast, action learning emphasises individual learning from experience, and taking action as a result of this learning. Admittedly, this action learning takes place within a group of people that is called a set of learning associates or comrades in adversity (Revans, 1982), but each individual within that group learns from separate experiences that do not necessarily involve other associates, and the separate experiences may not even involve workgroups. Action research involves action learning, but not vice versa, because action research is more deliberate, systematic and rigorous. Case research is also included in this special issue Finally, consider case research about a persons or a groups actions that occurred in the past and that can affect our present understanding and knowledge, and can in turn affect future actions. Essentially, case research (based on Perry, 1998; 2001 and the references listed there) is: . an investigation of a contemporary, dynamic phenomenon and the emerging body of knowledge about that phenomenon; . within the phenomenons real-life context where the boundaries between the phenomenon and its context are unclear; . when explanation of causal links are too complex for survey or experimental methods so that single, clear outcomes are not possible; and . the use of interviews, observation and other multiple sources of data that could include the diaries and minutes of meetings that are the sources of data in project action research. Recent examples of case research in marketing that have been published in journals are Alam (2002), Alam and Perry (2002), Batonda and Perry (2003), Madden and Perry (2003), Riege et al. (2001) and Riege and Perry (2000). Essentially, these articles have a literature review that develops some research issues or objectives. Then the methodologys data collection and analysis processes (based on the considerable literature about them) are described, along with a justication for its use. Then the data are analysed, based on the research objectives, and a nal theory is built. Case research is not wide spread or well understood in scholarly marketing, although it has been used in PhD programs in Sweden, Finland and Australia.

314

That is, case research offers empirical data to be used for creating concepts, categories, models and general theories, even testing theories; its not just for class-room exercises, examples, illustrations, and trendy success stories in textbooks. It demands a lot from the researcher in both understanding the techniques of case study research and accepting its hardships. Case study research takes a systemic, holistic stance recognizing reality as it is, not just settling for descriptions but adding value through conceptualization. It does not assume away complexity, chaos, ambiguity, fuzziness, uncertainty and dynamic forces for the convenience of the researcher and his or her analysis. It is primarily qualitative and interpretive, although quantitative research can be part of it. An observation, which is both amusing and scary, is that quantitative research starts and ends with qualitative assumptions and subjective interpretation, and even in its most regulated and systematic collection and processing of numbers, is dependent on judgment calls and inter-subjective agreements. This issues concern about theory building in action research is distinctive Case research is not usually placed within the umbrella of action research, but it is for this special issue, because it raises important points about theory building from particular situations for the emerging body of knowledge about a phenomenon. The discussion of action research and action learning above emphasised the particular situation within which it occurs, but what about other situations in the rm or in similar types of rms, and in other rms in the same industry or other industries? That is, a traditional action research project may enhance learning within one organisation, but it rarely addresses the question: how can the action research project also make a contribution to a body of knowledge that interests academics and consultants interested in principles that can apply in other organisations? That is, a researcher faces two goals or imperatives (McKay and Marshall, 2001, p. 46). One goal is to solve a practical problem within an organisation, and the second is to generate new knowledge and understanding about other organisations. How to address both these goals has been addressed rarely in the action research literature and there is little direct guidance on how to do it (McKay and Marshall, 2001, p. 49) (although guidance for thesis writers is in Perry and Zuber-Skerritt, 1994; Zuber-Skerritt and Perry, 2002). Because there is this little guidance in action research literature in other disciplines, it is somewhat surprising that both these goals of addressing the particular situation and generalising from it to other situations, is discussed in nearly all the articles in this special issue! Indeed, it is this feature that provides the cutting-edge feature of this collection of action research articles. Perhaps this feature is caused by the fact that many marketing phenomena are clearly determined by the external reality of a marketplace, as noted above. In contrast,

Action research in marketing

315

EJM 38,3/4

316

action researchers in other disciplines like management can emphasise just the one organisational situation because their internal, particular world is so powerful. These two goals of the particular and the general reect the different paradigms of organisational action researchers and the readers of academic journals and theses. Those two goals can be best understood by distinguishing three worlds of phenomena that scientists can investigate. For each world, scientists share a worldview or paradigm that is internally consistent, rational and logical. In more detail, we consider three worlds identied by Popper and their corresponding scientic paradigms (based on Magee, 1985 and Guba and Lincoln, 1994): (1) World 1 ts the positivism paradigm consisting of objective, material things. Here, in essence, reality is a straightforward concept that is easy to measure. We need not consider this physical sciences world further here because marketing is a social science. (2) World 2 ts the critical theory and constructivism paradigms and consists of the subjective world of minds, that is, of meanings. The two paradigms of this world are sometimes combined into one paradigm called the hermeneutic or interpretive paradigm (Gummesson, 2000; Zuber-Skerritt, 2001). In this world, perception is reality. Traditional action research ts into this world 2 and the critical theory paradigm; action learning ts into this world 2 and the constructivism paradigm. (3) World 3 ts the realism paradigm and consists of abstract things that are born of peoples minds but exist outside and independently of any one person. These abstract things are the phenomena investigated by many social scientists. Here, a persons perceptions are important because they are a window on to that blurry, external reality, rather than being important for their own sake. Case research often ts into this world and the realism paradigm. This is the world and paradigm of readers of academic journals and theses, and also the world of theories within a consultants mind when he or she approaches a new action project. Case research has addressed the important issue of bridging the two different worlds of traditional action research and academic/consultant knowledge. This bridge can be called analytic generalisation (Yin, 1994). This analytic generalisation is theory building, that is, the development of a holistic framework out of past, particular situations that can be used to plan action in future, other situations (Gummesson, 2000); in contrast, statistical generalisation that is the cornerstone of positivism and tests if a random sample ts a population. In-depth case interviews, action learning or traditional action research can all be used as the data to be used in the process of theory building, for all three provide the in-depth understanding of structure, process

and driving forces that are required to construct the frameworks that will be the basis of future action. Indeed, the major difference between the three types of research considered in this special issue is that the researcher is personally involved in action research and action learning, and the researcher learns about this involvement in action through other people in case research. In other words, traditional action research is allied to case research although traditional action research is the most demanding and far-reaching method of doing case study research (Gummesson, 2000, p. 116).

Action research in marketing

317

Four implications of the special issues articles for marketing action research This collection of articles about traditional action research, action learning and case research has four implications for readers of EJM and other marketing researchers. First, action research does have a role in many areas of the marketing discipline. Wherever a workgroup of marketing people has to grapple with its work processes, traditional action research can indeed be used to improve those processes. That action research can consist of a researcher working within a group for as long as six months (as in the article by OLeary, Rao and Perry) and even for longer times (as in Ballantyne), or in shorter workshops (as in Daniel and Wilson). Second, what seems to mark action research in marketing is a greater emphasis on the external, outside world of the market place; in contrast, action research in other settings usually emphasises the internal, subjective world of the participants or the social world within an organisation. That is, action research in marketing is more concerned with generalising from a particular situation to many others than it is in some other disciplines. An understanding of scientic paradigms can help this generalisation to be done and can liberate some researchers from the one paradigm of their personal preferences and training; in brief, this understanding allows action research to be a methodology rather than an ideology. Thus marketers should be careful that their action research projects are not judged by action researchers from other disciplines like management who sometimes appear to ignore the importance of phenomena in Poppers world 3 and look at only the subjective elements of a constructed, world 2 reality. Examiners of marketing research theses and reviewers of marketing journal articles should be chosen with care, to ensure they appreciate that phenomena from both of Poppers worlds 2 and 3 can interest social scientists. Next, note how all the articles deal with marketing managers rather than consumers. Marketing management phenomena are legitimate topics for marketing research, but the methodologies to investigate them have to focus on complex activities in an in-depth way that quantitative methods cannot do. As well, competitive secrecy makes response rates to surveys too small. Thus

EJM 38,3/4

318

these three forms of action research are effective and efcient methods to investigate marketing management phenomena. The nal implication is that action research is an umbrella term in marketing that can cover the three methods of project action research, action learning and case research. These three methods cover a persons or a groups actions that occurred in the past and that can affect our present understanding and knowledge, and can in turn affect future actions. Thus they represent three closely related ways of exploring the social science world of marketing in an appropriate way. In brief, these four implications suggest action research in marketing should be called marketing action research or interactive marketing research or marketing action science (Argyris et al., 1985) to reect its greater breadth than traditional action research, for action research in marketing can cover action learning and case research. This breadth is required because of the importance of the eternal reality of customers and competitors to marketing management, and because of the related importance that marketing researchers place upon analytic generalisation. This marketing research has the merit of privileged access to reality (Clark, 1972), and does not just rely upon the claims of respondents and their perceptions in a survey. Anthropological approaches, using direct or participant observation, are useful but they do not take you as far as marketing action research can do. There is always the risk in this broad approach to action research that the methodology of traditional action research will be deprived of its core and that its practice will be diluted; but that risk has to be balanced against the risk in a fundamentalist strategy for applying the methodology that it will be useless except in some very select cases. Some kind of compromise and balance between different and conicting demands is needed. For further discussion of methodological issues for research in marketing and in management where action research plays a role, see Gummesson (2000; 2001; 2002). Conclusion In conclusion, action research in marketing must address the importance of the eternal reality of customers and competitors to marketing management, and the related importance that marketing researchers place upon analytic generalisation. What should be recognised under the label of action research in marketing need not be conclusively established here. But this special issue of EJM is a rst step towards an understanding of what that label could cover. That rst step is that marketing action research is about a persons or a groups involvement in actions related to a market place that occurred in the past and that can affect our present understanding and knowledge, which can in turn affect future actions. This research can use data sources ranging from the emancipatory participation of a researcher in a traditional action research project through to case research. The cutting-edge examples of this kind of

action research in this special issue make us hope that its use will increase. If that happens, practising marketing managers will begin to read reports of business school research, and marketing students and researchers will master management action competencies as well as academic ones.
References Alam, I. (2002), An exploratory investigation of user involvement in new service development, Journal of the Academy of Marketing Science, Vol. 30 No. 3, pp. 250-61. Alam, I. and Perry, C. (2002), A customer-oriented new-service development process, Journal of Services Marketing, Vol. 16 No. 6, pp. 515-34. Argyris, C. (1983), Action science and intervention, The Journal of Applied Behavioural Science, Vol. 19 No. 2, pp. 115-40. Argyris, C., Putnam, R. and Smith, D.M. (1985), Action Science, Jossey-Bass, San Francisco, CA. Batonda, G. and Perry, C. (2003), Approaches to relationship development processes in inter-rm networks, European Journal of Marketing, Vol. 37 No. 10, pp. 1457-84. Carr, W. and Kemmis, S. (1986), Becoming Critical: Education, Knowledge and Action Research, Deakin University Press, Geelong. Carson, D., Gilmore, A., Gronhaug, K. and Perry, C. (2001), Qualitative Research in Marketing, Sage, London. Clark, P.A. (1972), Action Research and Organizational Change, Harper & Row, London. Coghlan, D. and Brannick, T. (2001), Doing Action Research in Your Own Organisation, Sage, London. Freire, P. (1972), Pedagogy of the Oppressed, Penguin, Harmondsworth. Guba, E.G. and Lincoln, Y.S. (1994), Competing paradigms in qualitative research, in Denzin, N.K. and Lincoln, Y.S. (Eds), Handbook of Qualitative Research, Sage Publications, Thousand Oaks, CA, pp. 105-17. Gummesson, E. (2000), Qualitative Methods in Management, 2nd rev. ed., Sage, Thousand Oaks, CA. Gummesson, E. (2001), Are current research approaches in marketing leading us astray?, Marketing Theory, Vol. 1 No. 1, pp. 27-48. Gummesson, E. (2002), Practical value of adequate marketing management theory, European Journal of Marketing, Vol. 36 No. 3, pp. 325-49. Lewin, K. (1946), Action research and minority problems, Journal of Social Issues, Vol. 2 No. 4, pp. 34-46. McKay, J. and Marshall, P. (2001), The dual imperatives of action research, Information Technology and People, Vol. 14 No. 1, pp. 45-59. Mckenzie, C.J., Wright, S., Bal, D.F. and Baron, P.J. (2002), Commentary: the publications of marketing faculty who are we really talking to?, European Journal of Marketing, Vol. 36 No. 11/12, pp. 1196-208. Magee, B. (1985), Popper, 3rd ed., Fontana, London. Madden, K. and Perry, C. (forthcoming), How do customers of a nancial services institution judge its communications?, Journal of Marketing Communication. Perry, C. (1998), Processes of a case study methodology for postgraduate research in marketing, European Journal of Marketing, Vol. 32 No. 9/10, pp. 785-802.

Action research in marketing

319

EJM 38,3/4

320

Perry, C. (2001), Case research in marketing, The Marketing Review, No. 1, pp. 303-23, available at www.themarketingreview.com Perry, C. and Zuber-Skerritt, O. (1994), Doctorates by action research for senior practising managers, Management Learning, Vol. 25 No. 2, pp. 341-64. Porter, W.M. and McKibbin, L.E. (1988), Management Education and Development: Drift or Thrust into the 21st Century?, McGraw-Hill, New York, NY. Revans, R.W. (1982), The Origins and Growth of Action Learning, Studentlitteratur, Lund. Riege, A. and Perry, C. (2000), National marketing strategies in international travel and tourism, European Journal of Marketing, Vol. 34 No. 11/12, pp. 1290-304. Riege, A.M., Perry, C. and Go, F.M. (2001), Partnerships in international travel and tourism marketing: a systems-oriented approach between Australia, New Zealand, Germany and the United Kingdom, Journal of Travel and Tourism, Vol. 11 No. 1, pp. 59-78. Schein, E.A. (1990), A general philosophy of helping: process consulting, Sloan Management Review, Vol. 31 No. 3, pp. 57-64. Yin, R.K. (1994), Case Study Research Design and Methods, Sage, Thousand Oaks, CA. Zuber-Skerritt, O. (1996), New Directions in Action Research, The Falmer Press, London. Zuber-Skerritt, O. (2001), Action learning and action research: paradigm, praxis and programs, in Sankaran, S., Dick, B., Passeld, R. and Swepson, P. (Eds), Effective Change Management Using Action Research and Action Learning: Concepts, Frameworks, Processes and Applications, Southern Cross University Press, Lismore, pp. 1-20. Zuber-Skerritt, O. and Perry, C. (2002), Action research within organisations and university thesis writing, Organisational Learning, Vol. 9 No. 4, pp. 171-9. Further reading Perry, C., Riege, A. and Brown, L. (1999), Realisms role among scientic paradigms in marketing research, Irish Marketing Review, Vol. 12 No. 2, pp. 16-23.