In response to high levels of intense competition and economic uncertainty, numerous firms have adopted team-based structures to survive and gain competitive advantage (Gilson & Shalley, 2004; Sundstrom, 1999). It has been argued that, in many circumstances, teams are more effective than individuals because team members can share workloads, monitor their teammates’ behaviors, and coordinate their different areas of expertise (Mathieu, Heffner, Goodwin, Salas, & Cannon-Bowers, 2000). Given the popularity and potential benefits of teamwork, it is important to examine how teams operate effectively and the influences on team effectiveness.

Coaches help people perform tasks well. Previous research has examined coaching in the context of training (Hackman & Wageman, 2005), and focused almost exclusively on individual skill acquisition (Fournies,   1978), athletic coaching, and executive coaching (Berglas, 2002; Elder & Skinner, 2002; Feldman & Lankau, 2005; Johnson, 2004). There are a number of differences between individual task accomplishment and team accomplishment, but there has been relatively little research on team coaching (TC), with the exception of athletic coaching. The concept of team coaching has been a significant recent development in understanding the conditions necessary for team effectiveness
(Hackman & Wageman).

TC is a set of leadership behaviors designed to enhance team ability and performance. According to Hackman and Wageman (2005), team leaders interact with team members frequently, so they need to have the ability to help team members focus on interacting as a team so they successfully complete the task at hand. This, in turn, enhances team members’ satisfaction with their job and their work relationships, thus increasing team effectiveness. Team Coaching Theory can be applied to Research and Development (R&D) teams for the following reasons: (1) although R&D team members are typically given a great deal of autonomy, the key decisions are frequently directed to the team leader; (2) according to Manz and Sims (1987), leaders of self-managed teams are ideally situated to perform certain activities, including encouraging the team, managing team boundaries (Ancona, 1990), and dealing with unexpected problems or events; and (3) R&D team leaders are involved in day-to-day task performance, including progress monitoring and discussions with team members on how to effectively proceed with the task. Despite these points, little is known about how team leaders manage teams and coach them to foster effectiveness.

Hackman and Wageman (2005) developed a theory of team coaching, and made several propositions regarding the influence of various types of TC on team effectiveness, including how timing (project lifecycle) and circumstances (organizational context) influence TC effects. This study was an investigation of team coaching theory in the context of Taiwanese R&D teams, and was aimed at improving understanding of team coaching theory and the practical applications of TC.

How Team Coaching Functions Influence Team Performance Processes

Coaching is defined as a “process of equipping people with the tools, knowledge, and opportunities they need to develop themselves and become more effective” (Peterson, 1996, p. 85). Unlike the terms executive coaching and workplace coaching, which typically describe one-on-one relationships, TC is defined as “direct interaction with a team intended to help members in the coordinated and task-appropriate use of their collective resources in accomplishing the team’s work” (Hackman & Wageman, 2005, p. 269). This may occur collectively, such as during team meetings, or through one-to-one interactions, usually between the team leader and a team member. Also, leadership becomes even more important in a changing environment (Zaccaro, Rittman, & Marks, 2001) or when a team
experiences disruption (Pirola-Merlo, Hartel, Mann, & Hirst, 2002).

Team leaders engage in many different behaviors intended to increase team effectiveness including structuring the team, establishing team goals, arranging resources required to accomplish team tasks, removing organizational impediments, helping individual members strengthen their personal contributions to the team, and working with the team to help members effectively use their collective resources to pursue team goals (Hackman & Wageman, 2005). However, only the last two of these behaviors can be considered TC behaviors. Hackman and Wageman suggest TC functions (motivational, consultative, and educational) have a positive impact on the following team performance processes: (1) the level of team effort involved in accomplishing a task; (2) the appropriateness of the strategies used by the group to perform their task; and (3) the skills and knowledge that members bring to bear on a task. Also, team effectiveness is the function of team performance processes.

As mentioned earlier, the three TC functions are described as motivational, consultative, and educational. Motivational TC aims to minimize free riding or “social loafing” and maximize the shared commitment of team members to the team or their task. Coaching by the team leader can motivate members to devote themselves to teamwork and share workload (Parker, 1994). The aim of consultative TC is to minimize mindless adoption or execution of task performance routines in uncertain or changing task environments and to stimulate the invention of ways of work that are particularly well aligned with task requirements. Denison, Hart, and Kahn (1996, p. 1012) found that successful leaders “facilitated flexible problem-solving and team development”. Educational TC aims to optimize the weighting of member contributions and increase the team members’ knowledge and skills. Edmondson (1999) found that team leader coaching increased team psychological safety, which in turn increased learning behaviors. TC enhances the learning behaviors of team members, thereby improving team member skill and knowledge. These three functions address and enhance effort, strategy, and skills and knowledge, respectively. On the basis of the above, the following hypothesis was formed:
Hypothesis 1 TC functions positively influence the team performance process of effort, strategy, and skills and knowledge.
Hypothesis 1a TC functions positively influence the team performance process
of effort.
Hypothesis 1b TC functions positively influence the team performance process of strategy.
Hypothesis 1c TC functions positively influence the team performance process of skills and knowledge.

Effects of TC on Team Effectiveness (TE)

Hackman and Wageman (2005) defined team effectiveness using a three-dimensional concept, comprising: (1) the productive outcomes of the team meeting or exceeding the team client’s standards of quantity, quality, and timeliness; (2) the social processes used by the team to enhance member ability to cooperate in the future; and (3) the contribution of the group experience to the learning and personal well-being of individual team members. Any input associated with each of the three performance processes is an opportunity for a “process loss” (Steiner, 1972) or a “process gain” (Hackman & Wageman). The interactions of group members can enhance the collective effort, generate appropriate strategies, and utilize the talent of group members, all of which are referred to as process gains, as opposed to process losses. Hackman and Wageman address the nature of the interventions that inhibit process losses and increase process gains for each of the three performance processes. On this basis the following hypothesis was proposed:
Hypothesis 2 The team performance processes of effort, strategy, and skills and knowledge positively influence TE.

Team coaching theory suggests that TC functions impact on team performance processes, and that TE is the outcome of team performance processes. The following hypothesis was therefore proposed:
Hypothesis 3 Team performance processes mediate the relationship between TC functions and TE.

The Black Box of Team Performance Processes

Innovation arises from two sources:       (1) the knowledge available for an innovative activity (e.g., Ahuja, 2000; Powell, Koput, & Smith-Doerr, 1996); and (2) the ability of individuals and teams to apply the available knowledge (Brown & Duguid, 1991; Tripsas, 1997; Von Hippel, 1988). Previous research suggests that by encouraging team members to participate in solving team problems, team leaders will enable members to engage in group information processing (Zaccaro et al.,    2001). Through generating ideas and sharing knowledge, the team members will be able to find a better strategy for the team. Leader motivation coaching therefore enhances information sharing and processing within the team by minimizing any free riding by the members. If team members believe their team is capable of achieving its goals, they become more likely to engage in the tasks (Zaccaro, 1996; Zaccaro, Blair, Peterson, & Zazanis, 1995). They are then more willing to devote time and effort to finding strategies to accomplish the team tasks. The following hypothesis was therefore identified:
Hypothesis 4 The level of team members’ effort is positively related to task appropriate performance strategies.

Since, by definition, R&D teams generally face novel events: (a) they need to develop a shared understanding of team problem parameters and processing; and (b) they must utilize individual and shared knowledge structures to find alternative solutions (Forsyth, 1990; Moreland & Levine, 1992). It follows that the more skills and knowledge that members utilize in the team task, the better the ability of the team to define the problems and identify appropriate strategies for application to team tasks. As mentioned earlier, leader education coaching is positively associated with members’ skill and knowledge contributions to team
tasks. The following hypothesis was therefore identified:
Hypothesis 5 The level and contribution of team members’ skill and knowledge to team task is positively related to task-appropriate performance strategies.

Method

Sample

Between 1991 and 1996 the government of Taiwan proposed, approved, and began to implement a development plan to make Taiwan into a key R&D center in Asia by establishing several science parks for high-tech companies. Taiwan high-tech industries face a fiercely competitive environment and shrinking product life cycles and consequently, they need to develop new products or enhance production processes to rapidly reduce production costs. R&D teams have become part of the basic organizational structure in the Taiwanese high-tech industry. Normally, the identities of team members depend on the tasks involved, with professionals capable of completing the required tasks being recruited to the team. Also, teams often do not remain constant. Team leaders are responsible for choosing team members, determining the team structure, obtaining required resources, helping team members maximize their contributions to the team, and ensuring the team successfully completes its tasks. The coaching behaviors of team leaders can therefore be observed and measured to test the effects of TC. The sample consists of firms involved in the high-tech industry and value-adding manufacturing firms.

Participants and Procedures

Of the 142 high-tech companies in Taiwan Science Parks, 87 (61%) agreed to participate in the study. Responses were obtained from 137 teams, comprising 763 individuals. After removing the incomplete questionnaires, the final sample consisted of 133 teams. The criteria for selecting the R&D teams were as follows: (1) the team had to have at least three members, in addition to a team leader; (2) the team task had to involve original innovation or improvement innovation; and (3) the team structure needed to be compatible with the definition used by Hackman and Wageman (2005), that is, it had to be an actual group, have one or more group tasks to perform, and operate in a social system context. The majority of the teams were obtained from the following industries: semiconductors (29%), electricity (16%), gas and oil (14%), computers (9%), raw material industries (8%), and chemical (5%). The remaining teams were from the biotechnology, medical, transport, finance, and high-tech machinery industries. To avoid common method variance (CMV), three different types of questionnaires were used: one for team members (rating TC functions, effort of team performance processes); one for team leaders (rating effort, strategy, and skills and knowledge of team performance processes); and one for department managers (rating TE), respectively. To ensure an effective sample and for statistical purposes, at least three-quarters of the members of a team (if the team comprised more than 5 members including the team leader), or three members of each team, were asked for responses. The questionnaires were mailed together with reply-paid envelopes to team leaders or department managers, who were asked to distribute the questionnaires to the team members in their teams or departments. The completed questionnaires were then directly and anonymously returned to the researchers by mail. Each team was assigned a unique code for purposes of identification.

Measures

All scales used in the questionnaire were in English in their original versions. They were then translated into Chinese by some of the study’s authors (Chin-Yun Liu, Chih Huang, and Chin-Ann Yang). To validate (including face and content validity) the Chinese versions, they were back translated into English by an HR manager of a high-tech company in Taiwan, who also held an MBA and is an American expert in adult learning, fluent in both English and Chinese. The back translation ensured an accurate prose translation rather than a literal English language translation (Brislin, 1980; Werner & Campbell, 1970). The Chinese questionnaire was pilot tested three times on different types of teams, including R&D teams in the South Taiwan National Science Centre and back-up service R&D teams in high-tech companies. All of the respondents were asked about the clarity of the wording of the items and the questionnaire was altered accordingly

All items were measured on a 5-point scale (ranging from 1 = strongly disagree to 5 = strongly agree). The items were based at team-level and individual responses were aggregated into team averages where necessary (George & James, 1993). Both the intraclass correlation coefficients (ICCs; Shrout & Fleiss, 1979), and the within-group interrater agreement measure (r_wg; James, Demaree, & Wolf, 1993) were used to estimate the appropriateness of aggregation. Confirmatory Factor Analysis (CFA) was used to validate the fitness between observed measured items and their corresponding latent variables. The hypotheses were then tested with structural equation modeling (SEM) using AMOS5.

Team Coaching To measure TC, the authors adapted the Team Diagnostic Survey (Wageman, Hackman, & Lehman, 2005). Two items were added to each of the motivational, consultative, and educational TC scales to meet the criteria of the measurement model. Cronbach’s α for the overall TC scale at team level was 0.92 and the correlation coefficients for the three subscales ranged from 0.88 to 0.94 (based on SEM estimate). This study first tested the measurement model by conducting CFA to validate the fitness between the observed measured items and their corresponding latent variables. Based on the high correlations among the three subscales of TC functions, the three indicators (motivational TC, consultative TC, and educational TC) were used in the CFA to test the construct validity. The 3-subscale measurement model exhibited adequate fit (χ²(40) = 56.77, CFI = 0.99, GFI = 0.93, NFI = 0.95, RMSEA = 0.06, SRMR =  0.04). Furthermore, all the factor loadings were significant. The results of model fit and factor loading analysis provided evidence of construct, discriminant, and convergent validity. ICC (1) and ICC (2) were 0.17 and 0.49 for motivational TC, 0.11 and 0.36 for consultative TC, and 0.17 and 0.48 for educational TC.  According to James (1982), the value of ICC (1) is usually between 0 and 0.5. The average r_wg was 0.98 for motivational TC, 0.92 for consultative TC, and 0.94 for educational TC.

Team Performance Processes This measurement consists of three scales: effort, strategy, and skills and knowledge. The scales were adapted for the study from the original scales by Wageman et al. (2005), and each included three items (e.g., “Members demonstrate their commitment to our team by putting in extra time and effort to help it succeed.”). In the adapted version three items were added to the original strategy scale (including two items from Hirst & Mann, 2004: e.g., “The team has chosen appropriate courses of action to meet project requirements.”) and one item was added to the skills and knowledge scale. CFA was then conducted on the data and consequently one item was removed from both the effort and skills and knowledge scales and three items were removed from the strategy scale. The indices of model fit were as follow: χ² (17) = 26.79, CFI = 0.97, GFI = 0.95, NFI = 0.93, RMSEA = 0.07, TLI = 0.95, SRMR = 0.04. The factor loadings were all significant. The results revealed good construct, discriminant, and convergent validity. Furthermore, the ICC (1) and ICC (2) for effort were 0.17 and 0.53 respectively. In addition, the average r_wg for effort was 0.87. To avoid common method variance, the ratings from both team leaders and team members were used for effort, while only the ratings of the leader were used for strategy and skills and knowledge. The researchers decided that team leaders could objectively appraise how the team applied the appropriate strategy to team tasks and how team members contributed their skills and knowledge to achieving team tasks. So since strategy and skills and knowledge were measured by the responses of the team leader only, there was no need to aggregate the data and so the ICC (1), ICC(2), and r_wg for strategy and skills and knowledge were not calculated. Cronbach’s α for effort, strategy, and skills and knowledge at team level, were 0.73, 0.82, and 0.64 respectively and for overall team performance processes was 0.83.

Team Effectiveness TE was rated by the manager. Since there was just one source there was no need to aggregate data to the team level. The scales for goals, customers, and timeliness were developed by Gibson, Zellmer-Bruhn, and Schwab (2003). The Cronbach’s α for the overall TE scale was 0.82, while for goals it was 0.91, customers, 0.83, and timeliness, 0.82. CFA was performed, and the indices of model fit were identified as follows: χ²(92) = 163.08, CFI = 0.93, GFI = 0.87, NFI = 0.86, RMSEA = 0.08, SRMR = 0.07. All of the factor loadings were significant. The results demonstrated reasonable construct, discriminant, and convergent validity.

Control Variables This study includes team size and team member teamwork tenure as control variables for TE. Team size influences team performance, and according to Wageman (1995) the optimal team size is between four and seven. An excessively large team will harm communication among members. Team member knowledge of how to perform well within a team will increase with their length of tenure. As member knowledge improves, conflicts will reduce and the team will become better able to solve team tasks (Wageman).

Results

Full Measurement Model

CFA was used to test the full measurement model and examine the psychometric properties of the latent constructs. A single composite was used as the manifest indicator for each team performance process, with the λ value set as √α and the error variance as (1-α) × δ² where δ² denotes the variance (Cortina, Chen, & Dunlap, 2001; Williams & Hazer, 1986). The intercorrelations between the three TC functions range from 0.88 to 0.94, and are the same construct of TC (Wageman et al., 2005); this study combined the items of each function basedon the mean, and thus used three parallel indicators for TC functions, namely motivation, consultative and education coaching.

The indices of model fit of full measurement model (χ²(28) = 50.84, CFI = 0.96, GFI = 0.94, NFI = 0.92, RMSEA = 0.08, SRMR = 0.05, TLI = 0.93) indicated that the model was adequate. Furthermore, the intercorrelations of every variable in this investigation were significant for all of the study variables, except the covariance of team size, and average tenure in the teamwork (Table 1).

Structural Model

Since the measurement model exhibited good model fit, the structural model was further examined using structural equation modeling (SEM) for hypothesis testing. The examination involved a series of comparisons of nested models, as listed in Table 2.

Considering the ratio of sample size to variables (Loehlin, 1992), the estimated parameters were reduced and thus the examined model was simplified when performing the structural modeling. The measured items for each variable were combined into between one and three parallel indicators (Fitzgerald, Drasgow, Hulin, Gelfand, & Magley, 1997; Kirkman, Rosen, Tesluk, & Gibson, 2004).¹ Teams’ TC and TE thus each included three parallel indicators, while three team performance processes each included one composite indicator. The results shown in Figure 1 illustrate the results. The parameters of model fit are as follows: χ²(38) = 68.62, GFI = 0.91, CFI = 0.95, RMSEA = 0.08, NFI = 0.90, TLI = 0.93, SRMR = 0.07. The results demonstrated that the hypothesized model has adequate fit.

Figure 1 shows the regression weights for each of the paths derived from the final model, and assesses the individual hypothesized relationships. The results statistically supported all of the hypotheses, except H1b.

(¹ Combinations of measured items may produce “balanced indicators” and maximize the variances shared among indicators of a construct (Fitzgerald et al., 1997, p. 583), a method adopted by a number of seminal works to test the measurement model (e.g., Rahim & Magner, 1995).

The Main Effects in the Hypothesized Model

Figure 1 shows that the standardized direct effects of TC functions on effort, strategy, and skills and knowledge are 0.72, -0.10, and 0.35 respectively, with the effect on strategy failing to reach significance. Thus hypotheses 1a and 1c are supported, but hypothesis 1b is not supported. TC functions enhance the level of team member effort and the contributions from member skills and knowledge. However, TC functions could not influence team application of the appropriate strategy to task accomplishment directly. The direct effect of TC functions on strategy thus still requires further study. As shown in Figure 1, effort is significantly related to TE (standardized direct effect is 0.23, p < 0.1). Similarly the relationship between strategy and TE is significant (standardized direct effect is 0.37, p < 0.05).

However, the relationship between skills and knowledge and TE is nonsignificant. Therefore the results only partially support hypothesis 2. The level of team member effort and whether the team applied the appropriate strategy thus foster TE, but team member skills and knowledge contributions do not directly influence TE.

The Effects of Effort and Skills and Knowledge on Strategy

To test whether the other two variables of team performance processes (effort and skills and knowledge) directly affect strategy, this study established three alternative models and compared them with the hypothesized model (model 1). The second model removed both of the relationships, the third model removed the effect of effort on strategy, and the 4th model removed the relationship between skills and knowledge and strategy. Δχ² s, and Δdfs are listed in Table 3. Table 4 lists the indices of model fit.

We first tested the differences between the hypothesized model and the null model Δχ² = 581, Δdf = 17, p < 0.005. It is worthwhile to examine nested comparison as well. Table 3 shows that all of the differences between the hypothesized model and alternative models were significant. Since model 2 differs significantly from the hypothesized model, it is impossible to free the effects of both effort and skills and knowledge on strategy. To further test which relationship the significance results from, we compared model 3 (in which the relationship between effort and strategy was removed), model 4 (in which the relationship between skills and knowledge and strategy was removed), and the hypothesized model, respectively. The results demonstrated that model 3 and the hypothesized model differed significantly, as did model 4 and the hypothesized model. Thus neither the relationship of strategy with effort, nor the relationship of strategy with skills and knowledge can be freed. The hypothesized model is the best of the various indices of alternative and hypothesized model fits. Since the alternative models all differ significantly from the hypothesized model, hypotheses 4 and 5 are both supported. The direct effect of effort on strategy is 0.32 (p < 0.1), and that of skills and knowledge on strategy is 0.65 ( p < 0.0001) which means a higher level of team member effort is associated with the team easily identifying the appropriate strategy for working tasks, while greater contributions of team member skills and knowledge will foster the effectiveness
of the application of team strategy to task performance. Based on the parameters, skills and knowledge exert a greater effect on strategy than effort does.

The Mediating Effect of Team Performance Processes

To test whether TC functions directly impact TE, the hypothesized model was compared with the model in which TC function directly influences TE, with all other variables remaining the same as in the hypothesized model. Δχ² = 0.10, Δdf = 1, which indicates a nonsignificant result (p > 0.05), and thus TC function does not directly affect TE, supporting hypothesis 3. The team performance processes mediate the relationship between TC function and TE.

Discussion

Implications for Team Coaching Theory

This study empirically tested team coaching theory (Hackman & Wageman, 2005). The results of this study partially support the propositions of the team coaching theory. Hackman and Wageman proposed that team coaching (TC) functions increase TE through team performance processes. However, the results of this study have indicated that not all the team performance processes of effort, strategy, and skills and knowledge impact the relationship between TC and TE. These findings advance the understanding of TC and TE at the team level by demonstrating how their relationship differs across team performance processes. The results showed that TC directly affected team member effort and skills and knowledge, but did not lead teams to apply the appropriate strategy for task completion. The results showed that team performance processes cannot function independently, with all three processes being moderately to highly positively correlated (see Table 1). As mentioned earlier, team innovation comes from the collective information processing that occurs when teams confront problems in task completion. Some of this processing includes (a) sharing understanding of team problems and processing objectives, (b) utilizing member knowledge and skills to define solution alternatives, (c) evaluating and reaching consensus on acceptable solutions, (d) planning and implementing, and (e) monitoring implementation and outcomes (Forsyth, 1990; Moreland & Levine, 1992). This suggests that for teams to identify the appropriate strategy to complete a task, idea sharing and information exchange are necessary. Team member effort and skills and knowledge contribute to information gathering and sharing. The most important mediator in the TE framework was strategy; the reason may be that the effort and skills and knowledge of members foster TE through strategy while the direct relationship is quite weak. Our research answers calls to incorporate coaching behavior into studies of task teams and offers a potential explanation for why the positive relationship between TC and strategy is not stronger.

The results of this study showed the TC functions (motivational, consultative, and educational) were all closely correlated (the correlation coefficients based on SEM estimates were from 0.88 to 0.94, significant at p < 0.0001). A possible explanation for this may be that teams in Taiwan often remain together for a relatively short period, with three months being a typical timeframe. In such a short time, it is difficult to divide the team life cycle into beginning, middle, and end points. Therefore, team leaders engage in coaching behaviors including three functions concurrently to help team members.

Practical Implications

The skills and knowledge of the individual team members is a key factor in a team’s success, especially for innovative teams (e.g., Ahuja, 2000; Powell et al., 1996). This study has shown effort and skills and knowledge influence how teams select and apply strategy in teamwork. The contributions of skills and knowledge of team members did not foster TE directly, but rather helped members identify good strategies for team tasks. Moreover, it could be said that to enhance TE, it is necessary for team members to use their skills and knowledge to produce their team strategy. The above considerations provide a more detailed insight into the relationship between TC and TE which previous research has explored by defining team performance processes as a single entity.

Team leaders are responsible for performing TC functions to help teams accomplish tasks, but leader coaching alone cannot guarantee good team performance. Leaders need to focus on enhancing the interactions among team members and their involvement in task-accomplishment activities, such as information sharing and meetings. Team member skills and knowledge are also important for team strategy.

Team leaders must be careful in recruiting team members. It should be noted that a diverse team can be helpful in terms of information sharing, but can create communication problems. Also, those individuals who have a tendency to “free ride” in a group situation can create perceptions of unfairness and produce job dissatisfaction among team members.

Limitations and Future Research

This study had some limitations which should be addressed in future research.

(1) Timing and supporting organizational context. This study did not explore the most effective timing of the various TC functions nor the organizational support needed for each function. In TC theory, TC functions work under specific conditions and at specific times. For example, regardless of how well team members interact in terms of effort, skills and performance, and strategy, tasks may not be completed if insufficient material resources are available. The efficacy of team coaching also depends on the time in the group’s life cycle when team coaching functions are performed. Motivational and consultative coaching are most helpful at the beginning and midpoint of a performance period respectively. Educational coaching is most helpful when provided after performance activities have been finished.

(2) Generalizing to other types of teams. Different types of teams encounter different problems and interact via varying processes. So future research is needed to investigate how TC functions affect different types of teams.

(3) Cultural impact. This study only examined high-tech companies in Taiwan. However, Asian and Western countries are very different culturally and this impacts their team leadership styles, and their attitude to individualist versus collectivist values. Therefore it would be worthwhile to conduct cross-cultural comparisons.

(4) Common method variance. Although this investigation made considerable efforts to address the issue of common method variance by collecting data from different sources (i.e., team members and team leaders), it is likely that it persists in the model since certain relationship coefficients remained quite large (e.g., the relationship between TC functions and effort). Future studies could collect data on various variables from different sources or simultaneously utilize perceptual data and objective data in a single model (Podsakoff, Mackenzie, Lee, & Podsakoff, 2003).

(5) I-P-O team performance framework. The classic works of McGrath (1984), Hackman (1987), and Steiner (1972) addressed the nature of team performance using the classic systems model in which inputs result in processes that in turn lead to outcomes (I-P-O model). The I-P-O framework tends to suggest a linear progression of main effect influences proceeding from one category to the next. However, recent research has begun to move beyond this framework. Interactions have been demonstrated between various inputs and processes, between various processes, and between inputs or processes and emergent states (Colquitt, Noe, & Jackson, 2002; De Dreu & Weingart, 2003; Dirks, 1999; Janz, Colquitt, & Noe, 1997; LePine, Hollenbeck, Ilgen, & Hedlund, 1997; Simons, Pelled, & Smith, 1999; Simons & Peterson, 2000; Stewart & Barrick, 2000; Taggar, 2002). Describing effectiveness via an I-P-O lens not only stresses the importance of interactions between inputs, processes and resulting outputs, but also recognizes that team outputs can be fed back into input variables, and that teamwork does not occur in a vacuum (Tannenbaum, Beard, & Salas, 1992). Previous outcomes of teamwork affect the inputs of the next team life cycle. Longitudinal research is needed to study this process.

Ahuja, G. (2000). Collaboration networks, structural holes, and innovation: A longitudinal study.  Administrative Science Quarterly, 45, 425-455.

Ancona, D. G. (1990). Outward bound: Strategies for team survival in an organization. Academy of Management Journal, 33, 334-365.

Berglas, S. (2002). The very real dangers of executive coaching. Harvard Business Review, 80, 86-93.

Brislin, R. W. (1980). Translation and content analysis of oral and written material. In H. C. Triandis & J. W. Berry (Eds.), Handbook of cross-cultural psychology: Methodology (pp. 389-444).  Boston: Allyn & Bacon.

Brown, J. S., & Duguid, P. (1991). Organizational learning and communities of practice: Toward a unified view of working, learning and innovation. Organization Science, 2, 40-57.

Colquitt, J. A., Noe, R. A., & Jackson, C. L. (2002). Justice in teams: Antecedents and consequences of procedural justice climate. Personnel Psychology, 55, 83-109.

Cortina, J. M., Chen, G., & Dunlap, W. P. (2001). Testing interaction effects in LISREL: Examination and illustration of available procedures. Organizational Research Methods, 4, 324-360.

De Dreu, C. K. W., & Weingart, L. R. (2003). Task versus relationship conflict, team performance, and team member satisfaction: A meta-analysis. Journal of Applied Psychology, 88, 741-749.

Denison, D. R., Hart, S. L., & Kahn, J. A. (1996). From chimneys to cross-functional teams: Developing and validating a diagnostic model. Academy of Management Journal, 39, 1005-1023.

Dirks, K. T. (1999). The effects of interpersonal trust on work group performance. Journal of Applied Psychology, 84, 445-455.

Edmondson, A. (1999). Psychological safety and learning behavior in work teams.Administrative Science Quarterly, 44, 350-383.

Elder, E., & Skinner, M. (2002). Managing executive coaching consultants effectively. Employment Relations Today, 29, 1-8.

Feldman, D. C., & Lankau, M. J. (2005). Executive coaching: A review and agenda for future research. Journal of Management, 31, 829-848.

Fitzgerald, L. F., Drasgow, F., Hulin, C. L., Gelfand, M. J., & Magley, V. J. (1997). Antecedents and consequences of sexual harassment in organizations: A test of an integrated model. Journal of Applied Psychology, 82, 578-589.

Forsyth, D. R. (1990). Group dynamics (2nd ed.). Pacific Grove, CA: Brook Cole.

Fournies, F. F. (1978). Coaching for improved work performance. Bridgewater, NJ: Van Nostrand Reinhold.

George, J. M., & James, L. R. (1993). Personality, affect, and behavior in groups revisited: Comment on aggregation, levels of analysis, and a recent application of within and between analysis. Journal of Applied Psychology, 78, 798-804.

Gibson, C. B., Zellmer-Bruhn, M. E., & Schwab, D. P. (2003). Team effectiveness in multinational organizations: Evaluation across contexts. Group & Organization Management, 28, 444-474.

Gilson, L. L., & Shalley, C. E. (2004). A little creativity goes a long way: An examination of teams’ engagement in creative processes. Journal of Management, 30, 453-470.

Hackman, J. R. (1987). The design of work teams. In J. W. Lorsch (Ed.), Handbook of organizational behavior (pp. 315-342). Englewood Cliffs, NJ: Prentice-Hall.

Hackman, J. R., & Wageman, R. (2005). A theory of team coaching. Academy of Management Review, 30, 269-287.

Hirst, G., & Mann, L. (2004). A model of R&D leadership and team communication: The relationship with project performance. R&D Management, 34, 147-160.

James, L. R. (1982). Aggregation bias in estimates perceptual agreement. Journal of Applied Psychology, 67, 219-229.

James, L. R., Demaree, R. G., & Wolf, G. (1993). r_wg: An assessment of within-group interrater agreement. Journal of Applied Psychology, 78, 306-309.

Janz, B. D., Colquitt, J. A., & Noe, R. A. (1997). Knowledge worker team effectiveness: The role of autonomy, interdependence, team development, and contextual support variables. Personnel Psychology, 50, 877-904.

Johnson, H. (2004). The ins and outs of executive coaching. Training, 41, 36-41.

Kirkman, B. L., Rosen, B., Tesluk, P. L., & Gibson, C. B. (2004). The impact of team empowerment on virtual team performance: The moderating role of face-to-face interaction. Academy of Management Journal, 47, 175-192.

LePine, J. A., Hollenbeck, J. R., Ilgen, D. R., & Hedlund, J. (1997). Effects of individual differences on the performance of hierarchical decision-making teams: Much more than g. Journal of Applied Psychology, 82, 803-811.

Loehlin, J. C. (1992). Latent variable models. Hillsdale, NJ: Lawrence Erlbaum Publishers.

Manz, C. C., & Sims, H. P. (1987). Leading workers to lead themselves: The external leadership of self-managing work teams. Administrative Science Quarterly, 32, 106-128.

Mathieu, J. E., Heffner, T. S., Goodwin, G. F., Salas, E., & Cannon-Bowers, J. A. (2000). The influence of shared mental models on team process and performance. Journal of Applied Psychology, 85, 273-283.

McGrath, J. E. (1984). Groups: Interaction and performance. Englewood Cliffs, NJ: Prentice-Hall.

Moreland, R. L., & Levine, J. M. (1992). Problem identification by groups. In S. Worchel, W. Wood, & J. A. Simpson (Eds.), Group processes and productivity (pp. 17-47). Newbury Park, CA: Sage.

Parker, G. M. (1994). Cross-functional teams: Working with allies, enemies and other strangers. San Francisco: Jossey-Bass.

Peterson, D. B. (1996). Executive coach at work: The art of one-on-one change. Consulting Psychology Journal: Practice & Research, 48, 78-86.

Pirola-Merlo, A., Hartel, C., Mann, L., & Hirst, G. (2002). How leaders influence the impact of affective events on team climate and performance in R&D Teams. Leadership Quarterly, 13(5), 561-581.

Podsakoff, P. M., Mackenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method biases in organizational research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879-903.

Powell, W. W., Koput, K. W., & Smith-Doerr, L. (1996). Interorganizational collaborations and the locus of innovation: Networks of learning in biotechnology. Administrative Science Quarterly, 41, 116-145.

Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin, 86(2), 420-428.

Simons, T. L., Pelled, L. H., & Smith, K. A. (1999). Making use of difference: Diversity, debate, and decision comprehensiveness in top management teams. Academy of Management Journal, 42, 662-673.

Simons, T. L., & Peterson, R. S. (2000). Task conflict and relationship conflict in top management teams: The pivotal role of intragroup trust. Journal of Applied Psychology, 85, 102-111.

Steiner, I. D. (1972). Group process and productivity. New York: Academic Press.

Stewart, G. L., & Barrick, M. R. (2000). Team structure and performance: Assessing the mediating role of intrateam process and moderating role of task type. Academy of Management Journal, 43, 135-148.

Sundstrom, E. (1999). Supporting work team effectiveness. San Francisco: Jossey-Bass.

Taggar, S. (2002). Individual creativity and group ability to utilize individual creative resources: A multilevel model. Academy of Management Journal, 45, 315-330.

Tannenbaum, S. T., Beard, R. L., & Salas, E. (1992). Team building and its influence on team effectiveness: An examination of conceptual and empirical developments. In K. Kelley (Ed.), Issue, theory, and research in industrial/organizational psychology (pp. 117-153). Amsterdam: Elsevier.

Tripsas, M. (1997). Surviving radical technological change through dynamic capability: Evidence from the typesetter industry. Industrial and Corporate Change, 6, 341-377.

Von Hippel, E. (1988). The sources of innovation. New York: Oxford University Press.

Wageman, R. (1995). Interdependence and group effectiveness. Administrative Science Quarterly, 40, 145-180.

Wageman, R. (2001). How leaders foster self-managing team effectiveness: Design choices versus hands-on coaching. Organization Science, 12, 559-577.

Wageman, R., Hackman, J. R., & Lehman, E. (2005). Team diagnostic survey: Development of an instrument. Journal of Applied Behavior Science, 41, 373-398.

Werner, O., & Campbell, D. (1970). Translating, working through interpreters, and the problem of decentering. In R. Naroll & R. Cohen (Eds.), A handbook of method in cultural anthropology (pp. 398-420). New York: Natural History Press.

Williams, L. J., & Hazer, J. T. (1986). Antecedents and consequences of satisfaction and commitment in turnover models: A reanalysis using latent variable structural equation methods. Journal of Applied Psychology, 71, 219-231.

Zaccaro, S. J. (1996). Social contextual considerations of efficacy beliefs: Defining multiple forms of social efficacy. Keynote address presented at the annual meeting of the Association for the Advancement of Applied Sport Psychology.

Zaccaro, S. J., Blair, V., Peterson, C., & Zazanis, M. (1995). Collective efficacy. In J. Maddux (Ed.), Self-efficacy, adaptation, and adjustment (pp. 305-328). New York: Plenum.

Zaccaro, S. J., Rittman, A. L., & Marks, M. A. (2001). Team leadership. The Leadership Quarterly, 12, 451-483.

Appreciation is due to anonymous reviewers.

Chin-Yun Liu, Department of Finance, Tainan University of Technology, 529 Jhongjheng Rd., Yongkang, Tainan 71002, Taiwan, ROC. Mobile: +886 931 711 279; Fax: +886-6-2421293; Email: [email protected]