292

Int. J. Organisational Design and Engineering, Vol. 1, No. 4, 2011

Collaboration trumps competition in high-tech project teams Michel Avital*

University of Amsterdam, 1018 WB Amsterdam, The Netherlands E-mail: michel@avital.net *Corresponding author

Baldev Singh
Aviva plc, 611 5th Ave, Des Moines, IA 50309, USA E-mail: baldev.singh@avivausa.com
Abstract: The tension between competition and collaboration in teams raises a question how to balance between these two different strategies to enhance the performance of teams. It has been established that both competition and collaboration have a positive impact on team performance. However, given that measures that enhance competition tend to jeopardise collaboration and vice versa, it remains unclear how to balance the two in order to attain an optimal impact on performance. Building on quantitative data from 176 IT projects in Motorola and its affiliates, we analysed the influence of intra-team collaboration and competition on project performance. The results suggest a strong and broad positive influence of intra-team collaboration on project performance; but only a limited influence of intra-team competition. Subsequently, we discuss further implications to forming and managing effective teamwork in IT projects and potentially in other functional areas. Keywords: collaboration; competition; teamwork; IS projects; success factors; innovation teams; team performance. Reference to this paper should be made as follows: Avital, M. and Singh, B. (2011) Collaboration trumps competition in high-tech project teams, Int. J. Organisational Design and Engineering, Vol. 1, No. 4, pp.292314. Biographical notes: Michel Avital is an Associate Professor of Information Management at University of Amsterdam and Visiting Professor at Viktoria Institute. The study of how information technology promotes innovation and extraordinary outcomes is a central theme of his work. Building on positive modalities of inquiry, his research focuses on information and organisation with an emphasis on the social aspects of information technologies. He has published articles on topics such as generative systems design, creativity, innovation, collaboration and competition, green IT and sustainable value. He is an editorial board member of seven leading IS journals and serves in various organising capacities in ICIS, AOM, ECIS and other topical conferences. Further and current information is available at http://avital.net. Baldev Singh is the Senior Vice President and Chief Architect at Aviva plc, where he brings an international background in business development to enterprise architecture, process and technology innovation, as well as Copyright 2011 Inderscience Enterprises Ltd.

Collaboration trumps competition in high-tech project teams

incubation of new concepts to maturity. In his previous role as Chief Architect at Motorola, he was responsible for the enterprise architecture and knowledge management. Previously, he managed strategy and operations for enterprise business development and led cross-sector teams to explore growth opportunities in emerging markets. He was a lead researcher in various companies where he was involved in the development of numerous technologies and systems, which led to several products and spin-off ventures. He received his Doctor in Management from Case Western Reserve University and MBA from Kellogg School of Management at Northwestern University.

293

Introduction

The tension between competition and collaboration in teams raises a question, how to balance these two different strategies to enhance the performance of teams. Collaboration and competition are not mutually exclusive. They can co-exist. For example, players in professional sports teams compete for opportunities to boost their personal records but at the same time must collaborate to score as a team. Neither high personal record in a trailing team, nor poor personal record in a winning team reflects positively on any team member. Team members must engage in competition and at the same time collaborate not only in sports teams; it is common in project teams, sales teams, and almost any team formation in the organisational context. It has been established that both competition and collaboration have a positive impact on team performance. However, given that measures that enhance competition tend to jeopardise collaboration and vice versa, it remains unclear how to balance the two in order to attain an optimal impact on performance. This seemingly paradoxical contention was the springboard for the following empirical study that explored the role of collaboration and competition among team mates in the context of IT development. Specifically, we conceptualised intra-team collaboration and competition as two distinct and independent constructs, and subsequently assessed and qualified the unique effect of each on project performance. Collaboration is considered as a source of performance. Many researchers and practitioners have addressed the significance of collaboration and its significant role in the execution of business transformations (e.g., Austin, 2000; Evans and Wolf, 2005; Huxham and Vangen, 2005; Smith, 2004). Collaboration and teamwork are organisational imperatives in response to the complexity of work and the diverse and changing business environment. Most organisational initiatives require collaboration among individuals across various functional areas. Collaboration has become particularly essential in light of globalisation, when individuals from multiple cultures, regions, and functional areas work together on projects, and inevitably bring divergent interests, perspectives, differing orientations towards goals, interpersonal relations, and key external constituents. As a result, there is increasing emphasis on collaboration within and across teams to enhance participation and information sharing for streamlining operational processes and ensuring superior competitive results, while simultaneously increasing organisational responsiveness and agility (Veneeva, 2006). Competition too is generally considered as a source of performance. Competition is believed to reduce slack, provide incentives for efficient organisation of production, and enhance innovation. It is associated with overall productivity growth of a firm (Nickell, 1996).

294

M. Avital and B. Singh

Collaboration and competition are alternative structuring approaches for project teams. While collaboration may be useful in understanding, sharing and joint action of individuals as well as team processes and project effectiveness, competition may explain the motivation, drive, and creativity of the individual team members (Gladstein, 1984). This raises the questions: Can project teams structured for competitive behaviour be productive as well? Can teamwork prevail in a competitive environment? Whereas competitive and collaborative behaviours have been often viewed as two opposite ends (e.g., Lado et al., 1997), qualitative analysis of collaboration among business units in a multi-unit organisation (Singh, 2005) revealed competitive collaboration (Hamel, 1991) in intra-team relationships, i.e., the collaborating people or sub-units were simultaneously engaged in competition. For example, the heads of business units, while advocating collaboration for leveraging resources across business units, were also engaged in competition for a single CEO position. In this paper, we examine a particular form of collaboration and competition that are conceptualised as two distinct and independent constructs. The collaboration examined is the propensity of project team members to share information, resources, and tasks as the proportion of members reporting high levels of sharing increases, so will the collaborative behaviour among project members, and ultimately project performance. The competition examined is the propensity of project team members for individual contributions, lack of reciprocity, conflict, and pursuit of self-interest as the proportion of members reporting high levels of self-interest pursuit increases, so will the competitive behaviour among project team members, and ultimately project performance. Preliminary evidence suggests that such propensities for collaboration and competition among project team members may be helpful in understanding project performance. The study is based on quantitative data from 176 IT projects in Motorola and its affiliates that were analysed for the influence of intra-team collaboration and competition on project performance. IT development in a global organisation is an excellent context for this study because of its inherent collaborative nature in leveraging common elements of the infrastructure for reducing costs, while simultaneously competing to serve individual and situated needs. The results suggest a strong and broad positive influence of intra-team collaborative behaviour on project performance; but only a limited influence of intra-team competition. Though derived from an IT projects context, the study provides generally applicable insights that can potentially serve as guidelines in structuring effective project teams in other functional areas.

Theoretical background and model development

2.1 The nature of competition and collaboration

Collaboration and competition are dual approaches for structuring the relationships among individuals or groups in a collective and orienting their behaviour toward performance-related goals. In terms of the level of analysis, competition and collaboration are viewed as modes of interaction among various social units, such as nations, organisations, groups, teams and individuals. This paper focuses on competition and collaboration among members of project teams who engage in the pursuit of the project objectives. Another key parameter is the nature of competition and collaboration among the project team members. At any particular level of analysis, both competition

Collaboration trumps competition in high-tech project teams

295

and collaboration can occur in various ways. For example, individual team members collaborate for leverage through information, resources, and task sharing, and compete for self-interest, scarce resources (e.g., financial rewards, political, and physical resources), as well as attention (Thibaut and Kelley, 2004). The underlying assumption is that financial resources for rewards are limited, and thus should be allocated to the most worthwhile opportunities and individuals who are more likely to ensure positive project performance. The degree to which projects should employ competition or collaboration among team members is an ongoing dilemma. Competition emphasises individual performance by rewarding high performers while sanctioning low performers, in an attempt to motivate and stimulate individuals to outperform one another in achieving project objectives. However, competition may be counterproductive because the competing individuals place self-interest ahead of the project interests, and thus, gains achieved by one member are not only achieved at the expense of another, but might negate the efforts of others. In contrast, collaboration emphasises group accomplishments and minimises any individual performance-based distinctions that may impede teamwork, sharing, and cooperation. It is often argued that individual-based competitive rewards are suited best for individual contributor members working alone, while team-based rewards are suited for situations when members are interdependent. Clearly, collaborative rewards promote team cohesion and mutual supportive behaviours that support project performance. However, the boundaries between individual and team work are fuzzy and information shared during collaboration may be used for other purposes than the pursuit of common interests (Khanna et al., 1998). The tension between collaboration and competition with respect to rewards is especially salient if organisations transform gradually to teamwork-orientation over the years, but have left their reward structure unchanged, resulting in possible dysfunctional teams and failure of team-based initiatives (Hackman, 1998). Building on the metaphor of internal market (Williamson, 1975), the underlying concern is whether to apply a form of competition in which two or more team members undertake duplicate activities to compete for the limited rewards, or a form of collaboration in which they share and leverage synergies. In its extreme case, competition among team members would involve parallel work streams housed within the same project, with their outputs directly competing with each other; while collaboration among team members would involve synergistically coordinated networks of interdependent work streams jointly leading to a non-redundant and project outcome that meets or exceeds expectations. Competition is beneficial especially when trying out unconventional ideas or when speed is critical, while the cost of duplication is insignificant (Birkinshaw and Lingblad, 2005). The beneficial aspects of team collaboration are well documented (Galbraith, 1973; Hoegl and Gemuenden, 2001). However, it should be noted that intra-team collaboration and competition may have both positive and negative consequences: collaboration could lead to groupthink (Esser, 1998), thereby limiting output to known solutions, while competition could lead to extreme waste through redundant outputs (Tsai, 2002). Although the literature is abundant with theories of competition and collaboration, an empirical study of intra-team competition and collaboration faces two a priori methodological obstacles. First, each of the concepts has multiple interpretations that are

296

M. Avital and B. Singh

partly inconsistent. For example, competition may be viewed as having a negative effect (Birkinshaw and Lingblad, 2005) or positive effect on performance (Vickers, 1995). Second, the literature is also inconsistent with respect to the relationship between competition and collaboration. Whereas in some instances it treats the two as ends of the same bipolar scale (Axelrod, 2006; Cox et al., 1997), in other instances it treats them as two independent phenomena that could coexist (Eisenhardt and Gahmic, 2000; Brandenburger and Nalebuff, 1996).

2.2 Conceptual model

The study was designed to assess the effects of intra-team collaboration and competition on project performance. Overall, the conceptual model is based on the established framework that holds a direct relationship between critical success factors and project performance in the context of IT development. Collaboration and competition were theorised as distinct and independent constructs that play a significant role in that context, as illustrated in Figure 1 and detailed in the remainder of this section.
Figure 1 The role of collaboration and competition

Critical Success Factors Goal & Strategy Partner Involvement Team Competencies Management Support Competition

Project Performance Project Expectations Innovation Team Performance Business Value

Collaboration

2.3 Project performance

Project performance is conceptualised in terms of its components: project expectations, innovation, team performance, and business value. Project expectations refers to the operational aspects of project performance such as project completion within schedule, budget, and meeting customer requirements (Kerzner, 2003), as well as meeting the expectation of the management and the project team. Innovation, however, is the level of creativity and innovation of the project deliverables (Anderson and West, 1998). Team performance is the teams responsiveness to customer needs, adaptability to changing requirements, overall efficiency, and team reputation (Ancona and Caldwell, 1992). Business value is the resulting boost to company performance from project execution and deliverables (Mukhopadhyay et al., 1995). The previously mentioned literature suggests a relationship between intra-team collaboration and team performance as well as between competition and team performance. This leads to the following hypotheses:

Collaboration trumps competition in high-tech project teams H1 Intra-team collaboration has a positive effect on project performance.1 H1a Intra-team collaboration has a positive effect on project expectations. H1b Intra-team collaboration has a positive effect on innovation. H1c Intra-team collaboration has a positive effect on team performance. H1d Intra-team collaboration has a positive effect on business value. Intra-team competition has a positive effect on project performance. H2a Intra-team competition has a positive effect on project expectation. H2b Intra-team competition has a positive effect on innovation. H2c Intra-team competition has a positive effect on team performance. H2d Intra-team competition has a positive effect on business value.

297

H2

2.4 Critical success factors

Information systems project failures are still prevalent in the IT industry. The recent 2009 edition of the Chaos Report from Standish Group suggests that only 32% of the IT projects surveyed can be categorised as successful, that is, they were completed on time, within budget, and met the requirements. Further, the Standish Group study identified that 24% of the projects were outright failures (Galorath, 2009). Subsequently, many studies were dedicated to identifying critical success factors of IS projects (e.g., Holland and Light, 1999; Rockart, 1979). For example, research identified critical success factors for different objectives in construction projects (Chua et al., 1999), provided a cultural framework for project success (Kendra and Taplin, 2004), and benchmarked the critical success factors in product innovation (Johne and Snelson, 1988). Insights gained from these and other studies have identified collectively a small set of critical success factors, such as user involvement and participation, clarity of goals and strategy, team competencies, partner involvement, and management support and commitment. User involvement and participation in the system development process reflects a set of behaviours or activities that potential users or their representatives perform, including providing advice as well as participating in project execution. Indeed, there is strong support in the literature for the impact of user involvement and participation on project performance (Barki and Hartwick, 1994; Ives and Olson, 1984). However, in the projects selected for this study, all users are internal members of the organisation, who do not reflect the user involvement and participation that is discussed in the literature, and therefore this success factor was excluded from the study. Goal interdependence theory predicts that team members beliefs about how their goals are related determine how they interact, which in turn affects their individual and project performance (Deutsch, 1949). Respectively, the level of understanding of the project goals, objectives, and adopted strategy, is likely to influence the behaviours of project team members, and in turn to impact project performance (Ancona and Caldwell, 1992; Gladstein, 1984). This leads to the following hypotheses: H3 Clarity of goals and strategy has a positive effect on project performance. H3a Clarity of goals and strategy has a positive effect on project expectations. H3b Clarity of goals and strategy has a positive effect on innovation. H3c Clarity of goals and strategy has a positive effect on team performance.

298

M. Avital and B. Singh H3d Clarity of goals and strategy has a positive effect on business value.

With the increasing use of commercial off-the-shelf (COTS) applications in IS projects, vendor involvement has become critical. The importance of this emerging success factor is illustrated, for example, in Goodmans (2003) account of the impact of vendor involvement in the development of a shuttle navigation software system. Vendor and other related partner involvement during project execution is essential for the projects success. This leads to the following hypotheses: H4 Partner involvement has a positive effect on project performance. H4a Partner involvement has a positive effect on project expectations. H4b Partner involvement has a positive effect on innovation. H4c Partner involvement has a positive effect on team performance. H4d Partner involvement has a positive effect on business value.

Team competencies reflect a set of the combined fundamental competencies of all project team members. The appropriate mix of technical and managerial skills, as well as necessary resource capacity, is essential to project execution (Cockburn and Highsmith, 2001; Evans and Dion, 1991; Guinan et al., 1998). This leads to the following hypotheses: H5 Team competencies have a positive effect on project performance H5a Team competencies have a positive effect on project expectations. H5b Team competencies have a positive effect on innovation. H5c Team competencies have a positive effect on team performance. H5d Team competencies have a positive effect on business value.

Management support and commitment has been long recognised as essential to project success. Significant evidence exists in the literature showing that the support and commitment of management has a direct impact on project success (Cooper and Kleinschmidt, 1995a; Wooldridge and Floyd, 1990). This leads to the following hypotheses: H6 Management support has a positive effect on project performance. H6a Management support has a positive effect on project expectations. H6b Management support has a positive effect on innovation. H6c Management support has a positive effect on team performance. H6d Management support has a positive effect on business value.

Finally, the four underlying success factors can be also linked to intra-team collaboration and competition behaviours. The level of understanding of the project goals, objectives, and strategy adopted affects the ability of team mates to coordinate work and engage in meaningful competition, let alone collaboration, toward the achievement of the desired outcome of a project (Axelrod, 2006). Team competencies reflect a set of the combined fundamental competencies of all project team members, including their ability to engage in competitive collaboration (Singh, 1995). Vendor involvement was found to influence the relationship among the team members and may encourage either competitive or collaborative behaviours, as documented in the account of a large R&D consortium

Collaboration trumps competition in high-tech project teams

299

(Gibson and Rogers, 1994). Last but not the least, management is instrumental in structuring and sustaining a mix of competitive and collaborative behaviours (Galbraith, 1973). This leads to the following hypotheses: H7 H8 H9 Clarity of goals and strategy has a positive effect on intra-team collaboration. Clarity of goals and strategy has a positive effect on intra-team competition. Partner involvement has a positive effect on intra-team collaboration.

H10 Partner involvement has a positive effect on intra-team competition. H11 Team competencies have a positive effect on intra-team collaboration. H12 Team competencies have a positive effect on intra-team competition. H13 Management support has a positive effect on intra-team collaboration. H14 Management support has a positive effect on intra-team competition.
Figure 2 Hypothesised model, (a) hypotheses critical success factors to project performance (b) hypotheses connecting critical success factors to collaboration/competition
Critical Success Factors Goal & Strategy Partner Involvement Team Competencies Management Support Collaboration
H3 H1

Project Performance Project Expectations Innovation Team Performance

H2

H4

H5

H6

Competition

Business Value

(a)
Critical Success Factors Goal & Strategy H8 Partner Involvement H10 Team H11 Competencies Management Support
H13 H12 H14 H7 H9

Project Performance Collaboration Project Expectations Innovation Team Performance Competition Business Value

(b)

300

M. Avital and B. Singh

Figure 2 extends the conceptual model in Figure 1, and summarises the hypothesised research model as discussed above.

Measurements and data collection

3.1 Measurements
All measurement scales were developed with items borrowed from relevant literature, synthesised, and adapted for the current research context. Multiple researchers had utilised different scales in diverse research settings. They were all normalised to the seven-point Likert scale (from 1 = strongly disagree to 7 = strongly agree). Given that objective measures of performance across diverse projects have yet to be developed, we used subjective measures comparable with those used in similar studies. Respondents were asked to indicate the degree to which they agreed or disagreed with the subject project having met project performance in terms of project expectations, innovation, team performance, and business value. The project expectations scale measured operational aspects of the project, such as meeting schedule, budget, reliability, and customer satisfaction with items adopted from project performance scales in Katz (1982), and Katz and Allen (1985), and the expectations of customers, management, and the team, with items adopted from product success scale in Lynn et al. (1999). The innovation scale measured the degree of creativity and innovation of the project deliverable with items adopted from the performance scale in Tushman and Katz (1980). The team performance scale measured the responsiveness, efficiency, and reputation for excellence, with items adopted from team performance and stakeholder-rated performance scales in Ancona and Caldwell (1992), and Guinan et al. (1998). The business value scale measured internal impact on the organisation in terms of advanced strategy, improved decision-making, streamlined business processes, and operating flexibility and external impact in terms of enhanced product/service value, customer relations, time to market and competitive advantage adopted from process planning and support, production and operations, product and service enhancement, and customer relations scales in Tallon et al. (2000). The critical project success factors formed the independent variables: team competencies, clarity of goal and strategy, partner involvement, and management support and commitment. The team competencies scale was modified from the staff scale in Cooper and Kleinschmidt (1995b) and measured project team members perception of team expertise, right mix of skills, relevant knowledge and experience, and resource sufficiency. The clarity of goal and strategy scale measured the level of an individual project team members understanding of project goals and strategy for it to be actionable. It was adopted from the strategic focus scales used in Cooper and Kleinschmidt (1995b), and Lynn et al. (1999). Partner involvement was modified from the user involvement scale used in Barki and Hartwick (1994) and the supplier relations scale in Tallon et al. (2000). It measured the level of vendor, supplier, and other partner involvement in project execution as perceived by the project team members. Finally, the managerial support scale measured the extent of managerial support and commitment to the project as perceived by the project team members. It was adopted from the senior managerial commitment and accountability scales used in Cooper and Kleinschmidt (1995b), and Guinan et al. (1998).

Collaboration trumps competition in high-tech project teams

301

The measures of collaboration and competition in particular required the development of new items through blending original work with the contributions of several researchers. Collaboration and competition has been studied mostly at the organisational unit level, and only in a limited form at the project level. Hence, the collaboration scale in this study measured the perception of intra-team collaborative behaviour in terms of communication, information, resources, expectations, and task sharing. It was synthesised from scale items used in Frey et al. (2004) for collaboration, Dooms and Oijen (2005) for resource sharing, Luukkonen et al. (1992) and Baggs (1994) for information sharing, Pinto et al. (1993) for cooperation, and Nobeoka and Cusumano (1994) for coordination. Moving on, the competition scale measured the level of perceived intra-team competitive behaviour in terms of conflict and contention for resources and rewards. It was developed with modified items that were adopted from inter-unit competition for resources in Tsai (2002), and individual self-interest, and rewards in Beersma et al. (2003).

3.2 Sample
The research model was tested with survey data that were collected from projects within Motorola and some of its partner organisations. The target sample included projects or clearly bounded project phases that were completed up to two years prior to data collection. The population included different project size, type of project, and functional areas involved to account for and test the impact from such differences. The survey was web-hosted and participation was voluntary. As the unit of analysis was a project, the survey participants were individual project team members representing a particular project. Lists of projects completed in the last two years were collected from the central project registries, and the participation of associated project leaders was solicited using e-mail. Project leaders in turn solicited the participation of their respective project team members. Out of the 675 projects solicited, 409 were eliminated due to overlapping member participation. Of the remaining 266 projects, a total of 566 surveys were fully completed. The complete responses were later sorted by project name and synthesised into the dataset for the study. All individual responses for a project (ranging from 2 to 11 people per project) were averaged to yield a single project data point. Thus, 566 responses when synthesised resulting in 176 projects data points, yielding a response rate of 66%. The demographics of the respondents reflect a highly educated, diverse, and balanced sample of respondents. The projects surveyed had been distributed globally, and the geographical distribution reflects the distribution of the survey population. Survey respondents were primarily based in North America (69%), followed by Asia Pacific (21%). This distribution is reflective of the workforce distribution of Motorola and its partners. Overall, 77% of the respondents were male and 23% female; and 99% of the respondents had some college degrees. Due to the modification or adaptation of the original items used from the relevant literature, considerable development-through-validation rigor was applied to the development of the measures. Consequently, the resulting measurement reflects strong composite reliabilities (.795 to .933, average .885) and provides evidence of convergent and discriminant validity for all constructs. Following thorough testing and a pilot run, the full survey consisting of 84 questions was deployed through online hosting for eight weeks to obtain responses from an ultimate sample of 176 projects.

302

M. Avital and B. Singh

3.3 Measurement model invariance analysis

Half-split measurement invariance analysis was conducted to verify instrument equivalence, testing that there would be no significant difference between responses from the two halved datasets at the survey item level. This work was meant to lend methodological insight into the ramifications of measurement error and unequal reliability, overall error rate and construct equivalence. To ascertain construct equivalence, we expected no differences in the scales of measurement for all constructs utilised in this study, as well as factorial pattern similarity (i.e., scale items load on the same factor or construct) and factorial equivalence (i.e., each scale item has the same loading within statistical bounds and on the same factor) (Singh, 1995). Next, variance extraction analysis was performed with a disaggregated CFA model. Forty models were initially run (unconstrained, fully constrained, each item independently constrained), identifying only four items with p values < .05 (CM2, CM4, GS3, and TP3). Four successive models were then run, constraining all but one of the four items of concern. In each instance, p > .05 indicated that our model displayed measurement invariance. This cleared the way to create the composites and to proceed with structural model specification and hypothesis testing. Table 1 summarises the uni-variate statistics of the composites and the bi-variate inter-correlations among the constructs for the 176 projects.
Table 1 Descriptive statistics and inter-correlations among variables
CL Collaboration (CL) Competition (CM) Goals and strategy (GS) Partner involvement (PI) Team competencies (CC) Project expectation (PP) Innovation (IN) Team performance (TP) Business value (BV) Mean Standard deviation Reliability 1 0.532 CM 1 1 1 1 1 1 1 1 1 5.35 1.05 5.79 1.13 GS PI CC MS PP IN TP BV

0.463 0.377

0.489 0.212 0.397

0.622 0.434 0.586 0.465

Management support (MS) 0.573 0.366 0.476 0.481 0.599

0.518 0.384 0.421 0.241 0.485 0.411

0.392 0.248 0.390 0.372 0.258 0.280 0.227

0.664 0.501 0.517 0.458 0.607 0.533 0.663 0.319 5.57 1.07 3.13 1.09 5.55 1.10 4.92 1.18 5.40 1.11 5.11 1.17 5.61 1.26 5.23 1.42

0.588 0.262 0.576 0.441 0.520 0.536 0.373 0.318 0.480

0.933 0.838 0.891 0.826 0.795 0.916 0.928 0.938 0.892 0.888

Analysis and findings

4.1 Fit and reliability

Analytical methods were employed to examine possible data quality or model specification issues. Specifically, we checked for confounding effects of measurement error, pursuing evidence of acceptable construct reliability, as well as convergent and discriminant validity. With an overall sample size of 176, a key concern was the presence of random error that could bias the estimation of structural paths in an unpredictable fashion. Subsequently, we tested for possible effect of control variables, such as project

Collaboration trumps competition in high-tech project teams

303

type, size, and duration model. However, no significant effects were found. We also examined potential for misspecification bias that could occur if some of the effects not hypothesised were significant, yet not included in the empirical analysis. The theoretical model was used as a baseline to ensure model fit. Then, any new paths not hypothesised were tested through incremental increases while examining changes in individual coefficients and model fit indices. Systematic implementation of this procedure to test potential effects mitigated concerns for misspecification bias in addition to testing for the significance of partial mediation effects. Using the full dataset, the hypothesised model (Figure 2) was tested and encountered no estimation problems. A fully disaggregated CFA measurement model with all observed indicators was estimated to ensure the measures corresponded only to their hypothesised constructs (Fornell and Larcker, 1981; Ramaswami and Singh, 2003). The measurement model included 41 of the 69 original items that remained after the initial EFA and CFA process and all individual measures loaded only on a single factor as expected. The initial fit indicators (CFI = 0.9, NFI = 0.797, RMSEA = 0.065, PCFI = 0.809) suggested that the hypothesised measurement model is a reasonable representation of data. Model fine-tuning and re-specification2 yielded improved relative and absolute fit indicators as follows: Chi-square = 25.253, d.f. = 4, CFI = 0.974, NFI = 0.971, PCFI = 0.082, RMSEA = 0.174 and SRMR = 0.0323 as well as reasonable R2. Moreover, the variance extracted for the ten constructs are all above the desired 0.50 threshold and higher than the highest shared variance (Hair et al., 1998). The composite reliabilities were between 0.795 and 0.938. The standardised factor loadings, without exception, were highly significant statistically (t-values > 6.74 and p < 0.001) and substantially large (with the majority between 0.6 and 0.933, except for CM2 = 0.498 and CM4 = 0.526). The Appendix provides additional details and statistics.

4.2 Hypotheses testing and structural model

The path analysis of the structural model is summarised in details in Table 2 and the significant paths are highlighted in Figure 3. Further details about each hypothesis are provided in the remainder of this section.

4.2.1 The effect of collaboration and competition on project performance (H1 to H2)
While collaboration (CL) has a positive influence on project expectations (PP), innovation (IN), team performance (TP), and business value (BV) with coefficients of 0.305, 0.248, 0.320, and 0.362, respectively, competition (CM) has a negative influence on team performance (TP) with a coefficient of 0.158 and a positive influence on business value (BV) of 0.147. Among the H1 and H2 hypotheses groups, only hypothesis H2a and H2b are not significant. In summary, while collaboration (CL) has a global positive effect on project performance, competition (CM) has a mixed effect.

4.2.2 The effect of success factors on project performance (H3 to H6)

Goal and strategy (GS) has significant impact on project expectation (PP), innovation (IN), team performance (TP), and business value (BV) with coefficients of 0.156, 0.279, 0.133, and 0.342 respectively. Thus, H3 hypotheses group is valid. The influence of partner involvement (PI) is significant only for innovation (IN) with a coefficient of

304

M. Avital and B. Singh

0.216. Therefore, hypothesis H4b is valid while hypotheses H4a, H4c and H4d are not valid. Team competencies (CC) have significant impact only on team performance (TP) with a coefficient of 0.168 but had no significant impact on the other performance indicators. Hence, only hypothesis H5c is valid. Similarly, management support (MS) has significant influence only on business value (BV) with a coefficient of 0.174 and as a result only hypothesis H6d is valid. In summary, whereas goal and strategy (GS) has positive impact on all aspects of performance, team competencies (CC) has an impact only on team performance (TP), partner involvement (PI) has an impact only on innovation (IN), and management support (MS) has an impact only on business value (BV).

4.2.3 The effect of critical success factors on collaboration and competition (H7 to H14)
Goal and strategy (GS) has no significant effect on collaboration (CL), thus hypothesis H7 is invalid. However, goal and strategy (GS) has a significant negative effect with a coefficient of 0.168 on competition (CM). Though not positive as expected, hypothesis H8 is valid for negative effect. Partner involvement (PI) has a significant effect with a coefficient of 0.181 on collaboration (CL) but has no significant effect on competition (CM), thus hypothesis H9 is valid but H10 is invalid. Team competencies (CC) has significant affect on both collaboration (CL) and competition (CM) with coefficients of 0.353 and 0.270, respectively. Hence, hypothesis H11 is valid but hypothesis H12 is negative and not positive as expected. Finally, management support (MS) has significant effect with a coefficient of 0.241 on collaboration (CL), and no significant effect on competition (CM). Hence, hypothesis H13 is valid while hypothesis H14 is invalid. In summary, the critical success factors have a positive or neutral effect on collaboration (CL), and in contrast, the critical success factors have a negative or neutral effect on competition (CM). Table 2 and Figure 3 summarise the results of the hypothesis testing as discussed above.
Figure 3 Final estimated model3
Critical Success Factors Goal & Strategy Partner Involvement Team Competencies Management Support
0.181**

Project Performance Collaboration

0.156* 0.305***

0.353*

0.133*

0.279*** 0.216**

0.248*

Project Expectations Innovation

0.241*** 0.342*** 0.168* -0.168* -0.270** 0.174* 0.320***

0.362*** -0.158*

Team Performance Business Value

Competition
Positive Effect Negative Effect

0.147*

Collaboration trumps competition in high-tech project teams

Table 2
Hs

305

Hypothesis tests and structural coefficients4

IVs DVs Coefficient 0.305 t-value/p-value 3.406*** Comments Valid

H1a H2a H1b H2b H1c H2c H1d H2d H3a H3b H3c H3d H4a H4b H4c H4d H5a H5b H5c H5d H6a H6b H6c H6d H7 H8 H9 H10 H11 H12 H13 H14

CL CM CL CM CL CM CL CM GS GS GS GS PI PI PI PI CC CC CC CC MS MS MS MS GS GS PI PI CC CC MS MS

PP PP IN IN TP TP BV BV PP IN TP BV PP IN TP BV PP IN TP BV PP IN TP BV CL CM CL CM CL CM CL CM

0.087
0.248

1.185
2.555*

Invalid
Valid

0.045
0.320 0.158 0.362 0.147 0.156 0.279 0.133 0.342

0.561
4.259*** 2.560* 4.599*** 2.283* 2.004* 3.325*** 2.051* 5.016***

Invalid
Valid

Valid
Valid Valid Valid Valid Valid Valid

0.106
0.216

1.424
2.695**

Invalid
Valid

0.098 0.065 0.175 0.173

0.168

1.583 1.000 1.935 1.772

2.216*

Invalid Invalid Invalid Invalid

Valid

0.023 0.085 0.011 0.080

0.174

0.294 1.033 0.121 1.163

2.409*

Invalid Invalid Invalid Invalid

Valid

0.069
0.168 0.181

0.999
1.991* 2.796**

Invalid Valid
Valid

0.052
0.353 0.270 0.241

0.656
4.595*** 2.888** 3.347***

Invalid
Valid

Valid
Valid

0.149

1.701

Invalid

Note: ***p 0.001, **p 0.01, * p 0.05

4.3 Mediation
Testing mediation can provide further insights regarding the role of collaboration and competition in the context of IT projects and the mechanism through which the success factors affect project performance. Mediation analysis (Baron and Kenny, 1986)

306

M. Avital and B. Singh

demonstrates that collaboration (CL) mediates, at least in part, the effects of partner involvement (PI), team competencies (CC) and management support (MS) on all project performance indicators. However, competition (CM) mediates with marginal effects only in the influence of goal and strategy (GS) and team competencies (CC) on team performance (TP) and business value (BV). Table 3 provides the mediation and indirect effects estimates.
Table 3 Mediation and indirect effects estimates
Mediated via Project expectations Innovation Team performance Business value

Goal and strategy Partner involvement Team competencies Management support

Collaboration Collaboration Collaboration Collaboration

Mediated via

0.055 0.108 0.074

Project expectations

0.045

0.058
0.113

0.066 0.128
0.087 Business value 0.025

0.088 0.060
Innovation

0.077
Team performance 0.027

Goal and strategy Partner involvement Team competencies Management support

Competition Competition Competition Competition

0.043

0.040 -

Notes: Bold = full mediation; italics = partial mediation

Discussion

5.1 Critical success factors

Overall, the findings provide further insights regarding the impact of critical success factors on project performance. While the evidence for relationships between critical success factors and project performance is prevalent in the literature, the analysis provides some interesting new insights. Clarity of goals and strategy had significant effects on all aspects of project performance as expected. But the influence of team competencies was limited to team performance. Team competencies in terms of right mix of technical, managerial, and people skills should enable the team to self-manage for meeting project performance. Yet, the findings did not support such causality. Similarly, the analysis found the influence of partner involvement limited to innovation and not significantly affecting any other aspect of project performance. It is likely that the external involvement brings a diverse perspective adding to the richness of project team knowledge, and facilitating out-of-the-box thinking that paves the way to innovation. From this perspective it is understandable that the diversity injected into the project through partner involvement would not be helpful in meeting project expectations, requiring a narrower and more focused effort. Furthermore, partner involvement may seem to be unrelated directly to team performance. However, the lack of significant relations between partner involvement and business value is a surprise.

Collaboration trumps competition in high-tech project teams

307

Perhaps the partner involvement is restrictive in the sample population, and thus does not rise to a level appropriate to the generation of business value. The analysis also suggests that the influence of management support is limited to business value. Management support in terms of participating in critical decision-making, coordination with other groups, protection from external interference, and managing external communication, help ensure focus and proper alignment to generate desired business value, which is in line with the findings. However, such management support also protects the project team from external distractions, thus allowing it to remain focused on the task at hand in better meeting project expectation, innovation, and yields better team performance. Clearly, these findings are not in line with the literature, and may be a peculiar attribute of the study population.

5.2 Collaboration and competition

Next, examining the influence of collaboration and competition on project performance, we find that collaboration has a reasonably strong influence on project performance. However, influence of competition is limited just to team performance and business value. The negative influence of competition on team performance is understandable since competition among individuals is bound to impact negatively on the collective team performance. From these analyses it is clear that collaboration is preferred over competition. There could be two biases for this finding: the projects surveyed were more collaborative by nature than competitive; i.e., the tasks involved required collaborative behaviour. The prevalence of collaborative norms may be another explanation for why people were less likely to report on competitive behaviour, as it was considered counter-productive and not normative, especially in light of the continuous communications in Motorola that urge people to collaborate and behave as one team. Nonetheless, when collaboration and competition are combined with the critical success factor, the analyses revealed interesting observations with regard to the impact of critical success factors on collaboration and competition and the mediation effects. First, it is interesting to note the lack of influence from goals and strategy on collaboration. One would expect that clarity of project goals and strategy would facilitate identification of collaborative and competitive tasks; for example, what to share when and where for the common good, and when and where to compete for private rewards and self-interest. Second, the positive influence of team competencies and partner involvement on collaboration is as expected, but negative influence of goals and strategy and team competencies on competition dampens individual competition. The lack of influence on competition from management support refers to the collaborative norms that were nurtured by management in the tested population. Similarly, the lack of influence on competition from partner involvement seems reasonable, because partner involvement is inherently collaborative. Examination of the effects of mediation reveals interesting insights with respect to the role of collaboration and competition in the context of IT projects and the mechanism through which the success factors affect project performance. Collaboration appears to have a significant role in mediating the effect of the success factors on all project performance indicators. In contrast, competition appears to have a marginal role in that context. Clearly, many previous studies that examined success factors fell short of accounting for the confounding effect of collaboration on project performance.

308

M. Avital and B. Singh

The revealed differences in the nature of the influence of collaboration and competition on project performance, and in particular the noteworthy mediation effect of collaboration, reinforce the underlying presupposition that competition and collaboration are two independent and different constructs, each on its own unique continuum. Moreover, in an attempt to test the effect of coopetition (Tsai, 2002) on project performance, we operationalised coopetition as an interaction variable of collaboration and competition (i.e., CL CM) but its effect on project performance was insignificant. Subsequently, the collective findings suggest that high propensity for collaboration among project team members would lead to high project performance in terms of project expectations, innovation, team performance, and business value. On the other hand, high propensity for competition among project team members would lead to high project performance only in terms of business value but would hamper team performance. Finally, high propensity for both collaboration and competition concurrently would lead to high project performance in terms of high business value, project expectation, and innovation, but at the expense of team performance. In conclusion, structuring teams with high propensity for collaboration is crucial when team performance is a high priority, for example, in mission critical tasks with low tolerance for error. However, structuring teams with a propensity for both collaboration and competition, that is for competitive collaboration is likely to yield higher business value at the organisation level. Although the study population was limited to Motorola and some of its partner firms, there is nothing inherently restrictive about these findings, and hence they could serve as guidelines generally applicable to team-based projects at large.

5.3 Challenges and suggestions for future research

Several limitations of this study are noteworthy and should be taken into account in interpreting the results. First, a sample size of greater than 300 would be more appropriate for the hypothesised model and the small sample in this study, n = 176, is a limitation. However, a larger sample represents a significant data collection challenge, given that the model is based on a project level and that each data point is an averaged perception of a number of unique project team members greater than two. A second limitation of the study is the span of data collection, limited to Motorola and its partner organisations. Although the survey included data from several geographical regions, it was still limited to Motorola. A more robust data collection should include companies from diverse industries representing projects spanning more diverse functional areas. Differentiating further the types of projects (e.g., dealing with front-end or back-end office) may provide even greater insight. A third limitation is the self-reported mode of data collection, which may be skewed based on self-perceptions. Additionally, the association among constructs may be inflated as a result of common method variance resulting from the use of a single questionnaire. For a robust data collection, reliance on different data sources is needed, with performance data being collected from project managers, line managers and users, while the data for remaining constructs, especially collaboration and competition, is collected from the project team members. This would also ensure a more balanced perspective among project leaders and workers, thus addressing possible prestige bias.

Collaboration trumps competition in high-tech project teams

309

Nonetheless, the studys findings point to several areas for future study. First, the relationship between collaboration and competition deserves further exploration. The non-recursive modelling in this study suggested a moderate negative correlation between collaboration and competition. However, further exploration of this phenomenon requires recursive modelling that treats the antecedent of competition and collaboration independently from one another. Second, although recognising that collaboration and competition may have negative as well as positive impact on project performance, the scope of this study was limited to exploring just the positive aspects. An area for further investigation would be the exploration of negative aspects, as well as the combination of negative and positive aspects. Third, an extended area of study is the introduction of project life cycle to explore when collaboration, collaboration, or their combination is appropriate with respect to the project life cycle. For example, collaboration alone may be more effective during initial phases of the project life cycle, with competition following in the later phases. A fourth area for further investigation is the exploration along the collaboration life cycle developed in the earlier qualitative research (Singh, 2005). Different forms of collaboration and collaboration are appropriate for different types of problems and projects. Hence, the findings from one form of collaboration may not necessarily be applicable to other forms. These aspects were left for future research.

Conclusions

Although collaboration and competition have often been treated as opposite ends of the same scale, both have also been considered beneficial for team performance. This seemingly paradoxical contention motivated this study which explored the role of collaboration and competition among teammates in the context of IT development. Specifically, the objective of the study was to explore the influence of collaborative and competitive behaviours among team members in Motorola on project performance. Based on the presupposition that collaboration and competition are not mutually exclusive, they were conceptualised and operationalised as distinct and independent constructs. The results suggested a strong and broad positive influence of intra-team collaboration behaviour on project performance; but only a limited influence of intra-team competition. We also encountered some surprises, especially in terms of the influence of critical success factors on project performance. The influence of goals and strategy on all aspects of project performance was confirmed by the study, as expected from the literature. However, team competencies influenced significantly only team performance; partner involvement influenced only innovation; and management support influenced only business value, which is not in line with the expectations from the literature. Nonetheless, collaboration appears to have a significant role in mediating the effect of the latter three success factors on all project performance indicators. Given that in contrast, competition appears to have a marginal role in that context, we conclude that intra-team collaboration trumps competition at least as far as project performance is concerned. Overall, with the increasing global competition as well as the variety of forms of collaboration and sharing, understanding the dynamics of collaboration and competition is fundamental to organisations, teams and individuals that have to sort out among a growing number of competing opportunities and demands in order to attain their respective goals.

310

M. Avital and B. Singh

References
Ancona, D.G. and Caldwell, D.F. (1992) Demography and design: predictors of new product team performance, Organization Science, Vol. 3, No. 3, pp.321341. Anderson, N.R. and West, M.A. (1998) Measuring climate for workgroup innovation: development and validation of the team climate inventory, Journal of Organizational Behavior, Vol. 19, No. 3, pp.235258. Austin, J.E. (2000) The Collaboration Challenge, Jossey-Bass Publishers, San Francisco. Axelrod, R. (2006) The Complexity of Cooperation: Agent-Based Models of Competition and Collaboration, Princeton University Press Princeton, New Jersey. Baggs, J.G. (1994) Development of an instrument to measure collaboration and satisfaction about care decisions, Journal of Advanced Nursing, Vol. 20, No. 1, pp.176182. Barki, H. and Hartwick, J. (1994) Measuring user participation, user involvement and user attitude, MIS Quarterly, Vol. 18, No. 1, pp.5982. Baron, R.M. and Kenny, D.A. (1986) The moderator-mediator variable distinction in social psychological research: conceptual, strategic and statistical considerations, Journal of Personality and Social Psychology, Vol. 51, No. 6, pp.11731182. Beersma, B., Hollenbeck, J.R., Humphrey, S.E., Moon, H., Conlon, D.E. and Ilgen, D.R. (2003) Cooperation, competition, and team performance: toward a contingency approach, Academy of Management Journal, Vol. 46, No. 5, pp.572590. Birkinshaw, J. and Lingblad, M. (2005) An evolutionary theory of intra-organizational competition, Organization Science, Vol. 16, No. 6, pp.674686. Brandenburger, A.M. and Nalebuff, B.J. (1996) Co-optition: A Revolution Mindset that Combines Competition and Cooperation: The Game Strategy thats Changing the Game of Business, Doubleday, New York, NY. Chua, D.K.H., Kog, Y.C. and Loh, P.K. (1999) Critical success factors for different project objectives, Journal of Construction Engineering and Management, Vol. 125, No. 3, pp.142150. Cockburn, A. and Highsmith, J. (2001) Agile software development, the people factor, Software Management, Vol. 34, No. 11, pp.131133. Cooper, R.G. and Kleinschmidt, E.J. (1995a) Benchmarking the firms critical success factors in new product development, Journal of Product Innovation Management, Vol. 12, No. 5, pp.374391. Cooper, R.G. and Kleinschmidt, E.J. (1995b) New products: what separates winners from losers?, The Journal of Product Innovation Management, Vol. 4, No. 3, pp.169184. Cox, W., Julie, R., Mann, L. and Samson, D. (1997) Benchmarking as a mixed metaphor: disentangling assumptions of competition and collaboration, Journal of Management Studies, Vol. 34, No. 2, pp.285314. Deutsch, M. (1949) A theory of cooperation and competition, Human Relations, Vol. 2, No. 2, pp.129152. Dooms, E. and Oijen, A.V. (2005) Control differentiation, resource sharing, and performance of business units, Schmalenbach Business Review, Vol. 57, No. 4, pp.320331. Eisenhardt, K.M. and Gahmic, D.C. (2000) Coevolving: at last, a way to make synergies work, Harvard Business Review, Vol. 78, No. 1, pp.91101. Esser, J.K. (1998) Alive and well after 25 years: a review of groupthink research, Organizational Behavior and Human Decision Processes, Vol. 73, No. 23, pp.116141. Evans, C.R. and Dion, K.L. (1991) Group cohesion and performance: a meta-analysis, Small Group Research, Vol. 22, No. 2, pp.175186. Evans, P. and Wolf, B. (2005) Collaboration rules, Harvard Business Review, Vol. 83, No. 7, pp.96104. Fornell, C. and Larcker, D.F. (1981) Evaluating structural equation models with unobservable variables and measurement error, Journal of Vocational Behavior, Vol. 18, No. 1, pp.3950.

Collaboration trumps competition in high-tech project teams

311

Frey, B.B., Lohmeier, J.H., Lee, S.W., Tollefson, N. and Johanning, M.L. (2004) Measuring change in collaboration among school safety partners, Persistently Safe Schools: Proceedings of the National Conference of the Hamilton Fish Institute on School and Community Violence, George Washington University, Washington, DC. Galbraith, J. (1973) Designing Complex Organizations, Addison-Wesley, Reading, MA. Galorath, D. (2009) 2009 Standish chaos report...Software going downhill, Dan on Estimating, available at http://www.galorath.com/wp/2009-standish-chaos-report-software-goingdownhill.php (accessed on 24 July 2011). Gibson, D.V. and Rogers, E.M. (1994) R&D Collaboration on Trial, Harvard Business School Press, Boston, MA. Gladstein, D.L. (1984) Groups in context: a model of task group effectiveness, Administrative Science Quarterly, Vol. 29, No. 4, pp.499517. Goodman, J.J. (2003) The space shuttle and GPS: a safety-critical navigation system, Proceedings of the Second International Conference on COTS-Based Software Systems, Spinger-Verlag, London, UK. Guinan, P.J., Cooprider, J.G. and Faraj, S. (1998) Enabling software development performance during requirements definition: a behavioral versus technical approach, Information Systems Research, Vol. 9, No. 2, pp.101125. Hackman, J.R. (1998) Why teams dont work, in Tindale, S.R. and Heath, L. (Eds.): Theory and Research on Small Groups, pp.245265, Plenum, New York. Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1998) Multivariate Data Analysis, 5th ed., Prentice Hall, Upper Saddle River, NJ. Hamel, G. (1991) Competition for competence and inter-partner learning within international strategic alliances, Strategic Management Journal, Vol. 12, No. S1, pp.83103. Hoegl, M. and Gemuenden, H.G. (2001) Teamwork quality and the success of innovative projects: a theoretical concept and empirical evidence, Organization Science, Vol. 12, No. 4, pp.435449. Holland, C.P. and Light, B. (1999) A critical success factors model for ERP implementation, IEEE Software, Vol. 16, No. 3, pp.3036. Huxham, C. and Vangen, S. (2005) Managing to Collaborate, Routledge-Taylor & Francis Group, London and New York. Ives, B. and Olson, M.H. (1984) User involvement and MIS success: a review of research, Management Sciences, Vol. 30, No. 5, pp.586603. Johne, F.A. and Snelson, P.A. (1988) Success factors in product innovation: a selective review of the literature, Journal of Product Innovation Management, Vol. 5, No. 2, pp.114128. Katz, R. (1982) The effects of group longevity and project communication and performance, Administrative Science Quarterly, Vol. 27, No. 1, pp.81104. Katz, R. and Allen, T.J. (1985) Project performance and the locus of influence in the R&D matrix, Academy of Management Journal, Vol. 28, No. 1, pp.6787. Kendra, K. and Taplin, L. (2004) Project success: a cultural framework, Project Management Journal, Vol. 35, No. 1, pp.3045. Kerzner, H. (2003) Project Management: A Systems Approach to Planning, Scheduling, and Controlling, John Wiley & Sons, Hoboken, NJ. Khanna, T., Gulati, R. and Nohria, N. (1998) The dynamics of learning alliances, competition, cooperation, and relative scope, Strategic Management Journal, Vol. 19, No. 3, pp.193210. Lado, A.A., Boyd, N.G. and Hanlon, H.C. (1997) Competition, cooperation, and the search for economic rents: a syncretic model, Academy of Management Review, Vol. 22, No. 1, pp.110141. Luukkonen, T., Persson, O. and Sivertsen, G. (1992) Understanding patterns of international scientific collaboration, Science, Technology & Human Values, Vol. 17, No. 1, pp.101126. Lynn, G.S., Skov, R.B. and Abel, K.D. (1999) Practices that support team learning and their impact on speed to market and new product success, Journal of Product Innovation Management, Vol. 16, No. 5, pp.439454.

312

M. Avital and B. Singh

Mukhopadhyay, T., Kekre, S. and Kalathur, S. (1995) Business value of information technology: a study of electronic data exchange, MIS Quarterly, Vol. 19, No. 2, pp.137156. Nickell, S.J. (1996) Competition and corporate performance, Journal of Political Economy, Vol. 104, No. 4, pp.724746. Nobeoka, K. and Cusumano, M.A. (1994) Multi-Project Management: Inter-Project Interdependency and Organizational Coordination in New Product Development, The MIT Japan Program Center for International Studies, Massachusetts Institute of Technology. Pinto, M.B., Pinto, J.K. and Prescott, J.E. (1993) Antecedents and consequences of project team cross-functional cooperation, Management Science, Vol. 39, No. 10, pp.12811297. Ramaswami, S.N. and Singh, J. (2003) Antecedents and consequences of merit pay fairness for industrial salespeople, Journal of Marketing, Vol. 67, No. 4, pp.4666. Rockart, J.F. (1979) Chief executives define their own data need, Harvard Business Review, Vol. 57, No. 2, pp.8993. Singh, B. (2005) Collaboration and competition tensions in multi-unit organizations facing volatile business environments, Working paper, Case Western Reserve University, Cleveland, OH. Singh, J. (1995) Measure issues in cross-national research, Journal of International Business Studies, Vol. 26, No. 4, pp.597619. Smith, K.A. (2004) Teamwork and Project Management, McGraw Hill, New York. Tallon, P.P., Kraemer, K.L. and Gurbaxani, V. (2000) Executives perceptions of the business value of information technology: a process-oriented approach, Journal of Management Information Systems, Vol. 6, No. 4, pp.145173. Thibaut, J.W. and Kelley, H.H. (2004) The Social Psychology of Groups, Transaction Publishers, New Brunswick, New Jersey. Tsai, W. (2002) Social structure of coopetition within a multiunit organization: coordination, competition, and intraorganizational knowledge sharing, Organization Science, Vol. 13, No. 2, pp.179190. Tushman, M.L. and Katz, R. (1980) External communication and project performance: an investigation into the role of gatekeepers, Management Science, Vol. 26, No. 11, pp.10711085. Veneeva, V. (2006) Ensuring effective teamwork in organization, Ezine@articles, available at http://ezinearticles.com/?Ensuring-Effective-Teamwork-in-Organization&id=232585 (accessed on 24 June 2011). Vickers, J. (1995) Concepts of competition, Oxford Economic Papers, Vol. 47, No. 1, pp.123. Williamson, O.E. (1975) Markets and Hierarchies: Analysis and Antitrust Implications, Free Press, New York. Wooldridge, B. and Floyd, S.W. (1990) The strategy process, middle management involvement, and organizational performance, Strategic Management Journal, Vol. 11, No. 3, pp.231241.

Notes
1 2 3 The top-level hypotheses, H1 to H6, are operationalised by their respective components marked H1a, H1b, etc. Based on the modification indices, we also specified correlation between the error terms of CL and CM as well as PP, IN and TP. Number on the shown path is the best-fit estimated standardised coefficient accounting for misspecification bias. Coefficients are significant at p 0.001 indicated by ***, p 0.01 indicated by **, and p 0.05 indicated by *. Number on the shown path is the best fit estimated standardised coefficient accounting for misspecification bias. Coefficients are significant at p 0.001 indicated by ***, p 0.01 indicated by **, and p 0.05 indicated by *. The remaining coefficients are not significant.

Appendix

Analysis Description Loading Highest 0.506 0.252 Average 0.740 0.684 0.722 0.800 0.960 0.933 0.806 0.894 1.024 0.782 0.895 0.877 0.768 0.656 0.811 0.740 0.795 0.751 9.364*** 12.573*** 11.030*** 12.195*** 11.266*** 11.627*** 14.270*** 0.572 0.426 0.302 0.888 14.730*** 11.995*** 0.737 18.004*** 0.524 0.392 0.892 14.450*** 0.883 0.534 12.823*** 0.285 0.938 16.291*** 17.138*** 12.683*** 10.925*** 11.307*** 0.928 t-value/ p-value Reliability Variance extracted Variance shared R**2 Schedule Budget Customer satisfaction Teams expectations Managements expectations Met/exceeded similar projects Innovative project Creative outcome Responsive Efficient Reputation for excellence Improved decision making Streamlined business processes Enhanced operating flexibility Enhanced product/service value Customer attraction/retention Reduced time to market

Factors

Items

Project expectation (PP)

PP1

PP2

PP4

PP6

PP7

PP8

Innovation (IN)

IN1

IN2

Team performance (TP)

TP1

TP2

Project performance

TP3

Business value (BV)

BV2

BV3

Collaboration trumps competition in high-tech project teams

Factor loadings and measurement properties of constructs used

BV4

BV5

BV7

BV8

313

314

Analysis Items Loading Highest 0.523 0.353 0.933 Average 0.775 0.881 0.806 0.844 0.891 0.821 0.498 0.524 0.331 0.526 0.842 0.935 0.707 0.901 0.951 0.656 0.805 0.722 0.817 0.740 0.817 0.689 0.811 0.817 0.862 0.817 0.830 10.209*** 12.051*** 10.594*** 12.166*** 9.585*** 12.706*** 12.870*** 13.993*** 12.890*** 13.177*** 0.687 0.481 0.291 0.916 0.565 0.524 0.373 0.795 11.772*** 9.533*** 0.614 0.327 0.214 0.826 16.318*** 14.959*** 0.739 10.386*** 0.461 0.267 0.891 15.733*** 13.390*** 7.217*** 6.741*** 13.098*** 0.199 0.838 14.953*** 13.685*** 12.720*** 14.652*** 12.013*** 0.698 t-value/ p-value Reliability Variance extracted Variance shared R**2 CL2 CL6 CL8 CL9 CL10 CL11 CM2 CM4 CM5 CM6 CM7 GS1 GS2 GS3 PI1 PI2 PI3 CC1 CC2 CC3 MS3 MS4 MS5 MS6 MS8 Inform others Scan threats Protection from outside interference Coordination w. external groups Project decision involvement Sufficient resources Right mix of technical and managerial skills Team expertise Partner team membership Partner sign-off authority Partner advice Strategic focus areas Role clarity Goal clarity Focus on self-interest Persistent conflict Criticism Competition for resources Preference for working alone Switching responsibilities Shared tasks Shared suggestions Shared expectations Shared resources Shared information Description

Factors

M. Avital and B. Singh

Intra-team collaboration

Collaboration/competition

Intra-team competition (CM)

Goals and strategy (GS)

Partner involvement (PI)

Critical success factors

Team competencies (CC)

Factor loadings and measurement properties of constructs used (continued)

Management support (MS)