Rapid Evidence Assessment (REA)
December 2019
ATTRIBUTES OF EFFECTIVE TEAMS AND INTERVENTIONS INCREASING TEAM EFFECTIVENESS
Culture Review Implementationour journey of positive change
a summary of scientific literature
Acknowledgement of Country
ACT Health Directorate acknowledges the Traditional Custodians of the land, the Ngunnawal people. The Directorate respects their continuing culture and connections to the land and the unique contributions they make to the life of this area. It also acknowledges and welcomes Aboriginal and Torres Strait Islander peoples who are part of the community we serve.
Accessibility
The ACT Government is committed to making its information, services, events and venues as accessible as possible.
If you have difficulty reading a standard printed document and would like to receive this publication in an alternative format such as large print, please phone 13 22 81 or email [email protected]
If English is not your first language and you require a translating and interpreting service, please phone Access Canberra on 13 22 81.
If you are deaf, or have a speech or hearing impairment and need the teletypewriter service, please phone 13 36 77 and ask for 13 22 81.
For speak and listen users, please phone 1300 555 727 and ask for 13 22 81. For more information on these services visit www.relayservice.com.au
© Australian Capital Territory, Canberra, July 2020.
This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced by any process without written permission from the Territory Records Office, ACT Government, GPO Box 158, Canberra City ACT 2601.
Enquiries about this publication should be directed to the ACT Health Directorate, Communications and Government Relations, GPO Box 825, Canberra City ACT 2601.
www.health.act.gov.au | www.act.gov.au
Enquiries: Canberra 13ACT1 or 13 22 81
This REA was produced by the Center for Evidence Based Management (CEBMa). The ACT Government acknowledges and thanks the CEBMa for allowing ACT Health to reproduce and redesign the content of their REA.
Any enquiries in relation to the content of this REA should be directed to CEBMa through their website: www.cebma.org
CEBMa center forEvidence-Based Management
Contents
What is a Rapid Evidence Assessment (REA)?
Main question: What does this REA answer?
Background
5
6
4
Search strategy: How was the research evidence sought? 6
Critical appraisal: How were the quality of the included studies judged? 7
Main findings 8
Conclusion 18
Limitations 19
Selection process: How were the studies selected?
References
7
20
Appendices 24
4
Novartis, one of the largest pharmaceutical companies, with more than 125,000 employees, is involved in several projects to enhance organisational effectiveness and performance. The effectiveness of teams and workgroups is assumed to be one of the most significant factors affecting a company’s performance. For this reason, Novartis approached the Center for Evidence Based Management (CEBMa) to undertake a Rapid Evidence Assessment (REA) to understand what is known in the scientific literature about the attributes of effective teams and workgroups and the effectiveness of interventions aimed at increasing team effectiveness. This review presents an overview of the findings.
Background
5
Evidence reviews come in many forms. One of the best-known types is the conventional literature review, which provides an overview of the relevant scientific literature published on a topic. However, a conventional literature review’s trustworthiness is often low: Clear criteria for inclusion is often lacking and studies are selected based on the researcher’s personal preferences.
As a result, conventional literature reviews are prone to severe bias. This is why ‘rapid evidence assessments’ (REAs) are used. REAs use a specific research methodology to identify the most relevant studies on a given topic as comprehensively as possible, and to select appropriate studies based on explicit criteria. In addition, the methodological quality of the studies included is assessed by two independent reviewers on the basis of explicit criteria. In contrast to a conventional literature review, REAs are transparent, verifiable, and reproducible, and, as a result, the likelihood of bias is considerably smaller.
What is a Rapid Evidence Assessment (REA)?
6
What is known in the scientific literature about the attributes of effective teams?
Other issues raised, which will form the basis of our conclusion regarding the main question above, are:
1. What constitutes a team?
2. What is team effectiveness?
3. How can the effectiveness of teams be measured?
4. What are the attributes (e.g. characteristics, conditions, composition) of effective teams?
5. What interventions influence team effectiveness?
6. What is known about the reliability and validity of team effectiveness models?
Three databases were used to identify studies: ABI/INFORM Global, Business Source Premier and PsycINFO. The following generic search filters were applied during the search:
1. Scholarly journals, peer-reviewed.
2. Published in the period 2000 to 2019.
3. Articles in English.
A search was conducted using combinations of various search terms, including ‘performance’, ‘effectiveness’, ‘team’, and ‘group’. We conducted 10 different search queries and screened the titles and abstracts of 992 studies. An overview of all search terms and queries is provided in Appendix I.
Main question: What does the REA answer?
Search strategy: How was the research evidence sought?
7
Study selection took place in two phases. First, titles and abstracts of the 993 studies identified were screened for relevance. In case of doubt or lack of information, the study was included. Duplicate publications were removed. This first phase yielded 141 studies. Second, studies were selected based on the full text of the article using these inclusion criteria:
1. Type of studies: Focusing on quantitative, empirical studies.
2. Measurement: Only studies in which relationships among team attributes, interventions and outcomes were quantitatively measured.
3. Context: Only studies related to workplace settings
4. Level of trustworthiness: Only studies that were graded level C or above (see below).
In addition, the following exclusion criteria were applied:
• Studies of ad hoc teams formed for immediate task performance, such as emergency teams.
• Studies of dyadic teams.
• Studies measuring the effect of leader attributes on team effectiveness.
This second phase yielded a total number of 70 studies. An overview of the selection process is provided in Appendix II.
Selection process: How were the studies selected?
The overall quality of the included studies were moderate to high. Of the 70 studies included, 31 studies concerned controlled studies and were therefore graded level B or higher. The remaining studies concerned uncontrolled, longitudinal studies, and were therefore classified as level C or lower. An overview of all studies included and their year of publication, research design, sample size, population, main findings, effect sizes and limitations is provided in Appendix III.
Critical appraisal: How were the quality of the included studies judged?
8
In daily life, a team is simply a group of people working together to achieve a goal. In the domain of social sciences, however, teams have specific characteristics that differentiate them from groups generally. For example, a widely used definition states: “A team is a collection of individuals who are interdependent in their tasks, share responsibility for outcomes, see themselves (and who are seen by others) as a social entity embedded in one or more larger social system (for example, business unit or the corporation), and, who manage their relationships across organisational boundaries.” (Cohen, 1997). Most researchers, however, summarise the minimal defining characteristics of a team as: A group of employees who are 1) formally established; 2) assigned (some) autonomy; and 3) interdependent.
Teams are not automatically more effective than individual employees. Working in teams may impede performance because of the potential conflict between individual and group interests. In addition, a team’s performance may decline due to a phenomenon known as social loafing: The tendency of team members to get by with less effort than what they would have put when working alone (also referred to as the free-rider effect). Although the term ‘team effectiveness’ is widely used in the research literature, it is rarely specified. In fact, even some of the meta-analyses and high quality studies included in this review fail to provide a clear definition of team effectiveness. Most studies included in this review consider team effectiveness synonymous to team performance. As such, team effectiveness is broadly defined as task performance, contextual performance, and/or adaptive performance (e.g. learning, creativity, decision making). Some scholars differentiate between performance behaviours and performance outcomes (Beal, 2003). Behaviours are actions that are relevant to achieving goals, whereas outcomes are the consequences or results of performance behaviours (Mathieu, 2008). Examples of performance behaviours include feedback seeking, reflectivity, information sharing, and learning behaviours.
Finally, several authors point out that an effective team is not necessarily an efficient team (Beal, 2003). Whereas team effectiveness concerns merely an evaluation of a team’s results, team efficiency also takes into account the ‘costs’ of achieving those results. For this reason, intra-team processes such as communication, information sharing, conflicts, etc. are often considered an essential element of team effectiveness (Mathieu, 2008).
Main findings
What constitutes a team?
What is team effectiveness?
1.
2.
9
Whether or not a team is effective depends on the applied criteria. In most of the studies included, the criterium is team (task) performance, that is, the degree to which a team accomplishes its goals, as reflected by performance-indicators such as number of units produced, number of items sold, number of clients served, number of innovations, number of errors, number of complaints, etc. In addition, some studies also measure intra-team processes such as team-member exchange, internal communication, level of information sharing, etc., as these are considered relevant indicators for team effectiveness.
As a result, there is no generally accepted instrument that measures team effectiveness - organisations, researchers, and consulting firms often create their own. Instruments developed by consulting firms typically ask members to assess their teams on those dimensions that the consulting firm assumes to be most consequential for team effectiveness (Wageman, 2005). By contrast, scholar-developed instruments that measure team effectiveness tend to focus on variables that are of (research) interest to the scholars who devised them.
This review has yielded a large number of studies examining myriad attributes. To facilitate a better understanding, we have grouped the findings into three main categories: team composition, emergent socio-affective states, and emergent cognitive states. Team composition refers to team member characteristics such as age, gender, level of education and functional background. Team composition variables and their impact on team outcomes have been incorporated into studies of team effectiveness for nearly 60 years (e.g., Mann, 1959).
Emergent states are team attitudes that arise from individual team members’ experiences. As such, they are different from team processes such as membership changes, internal communication or conflicts. Whereas team process describes the nature of team member interaction, emergent states describe conditions that dynamically enable and underlie effective teamwork (DeChurch, 2010). Research on teams has identified a range of emergent states assumed to affect a team’s performance such as confidence, efficacy, cohesion, trust, and shared mental models. In the scientific literature, two main categories of emergent states are distinguished: socio-affective states and cognitive states. While conceptually distinct, socio-affective and cognitive states are correlated and are assumed to in tandem.
It should be noted, however, that team performance is, to a large extent, a compositional construct - it is a direct result of its members’ individual performance. As such, drivers of individual performance - such as goal clarity, supervisory support, employee recognition - should first be taken into account before considering attributes and interventions at the team level (see CEBMa’s REA - Factors Associated with Knowledge Worker Performance, a summary of research literature, July 2019).
How can team effectiveness be measured?
What are the attributes of effective teams?
3.
4.
10
Finding 2: Of the Big Five personality traits, only agreeableness and conscientiousness are (somewhat) positively related to team performance (Level B)
Several meta-analyses with a combined sample size of more than 100 studies found that the higher the level of agreeableness and conscientiousness within teams, the better their performance (Bell, 2007; Hopp, 2012; Peeters, 2006; Prewett, 2009). The effect sizes found, however, were small. Other personality traits such as emotional stability, extraversion, and openness to experience were not related with team performance.
Socio-affective states describe team members’ collective reactions to interpersonal aspects of team functioning. Examples of emergent socio-affective states that have received considerable attention during the past decades include team confidence, social cohesion, collective efficacy, shared feelings, psychological safety, and intra-team trust (Mathieu, 2008). Below an overview is provided of the socio-affective states that were found to have the highest impact on team effectiveness.
It is often assumed that differences between individual members of a team on dimensions such as age, functional background, organisational tenure, gender, race, ethnicity, and experience enhance team effectiveness. As such, team diversity is one of the most researched attributes of effective teams. This review identified eight meta-analyses representing a combined sample size of more than 2,000 teams that measured the correlations between these attributes and team effectiveness or team efficiency (Bell, 2011; Bui, 2019; Guillaume, 2012; Haas, 2010; Horwitz, 2007; Wang, 2019; Webber, 2001; Zhou, 2015). Surprisingly, all meta-analyses demonstrated only small (< .1), zero, or even negative associations, regardless of team size, team type, or task type. As such, it is important to also consider – and compensate for – potential negative consequences of team diversity on communication, cohesiveness, and consequently performance (see also Finding 9).
Finding 1: The link between team effectiveness and team diversity dimensions such as age, gender, ethnicity, religion, functional background, educational background, organisational tenure, and experience is small and sometimes negative (Level AA)
Team composition
11
Finding 3: Intrateam trust is positively related to performance (Level A)
Finding 4: Team trust is most critical when team virtually, task interdependency, authority differentiation, and/or team temporality is high (Level A).
Several meta-analyses and high quality studies have demonstrated that a high level of intra-team trust is an important attribute of effective teams (Breuer, 2016; De Jong, 2016; Webber, 2008). Scholars often distinguish two types of trust: cognition-based trust (a member’s cognitive evaluation of the reliability, integrity, and competence of other members) and affect-based trust (a member’s emotional feelings/evaluation of the reliability, integrity, and competence of other members), which are regarded as functionally distinct, in that they affect a team performance through different causal mechanisms (De Jong, 2016). It was also found that team trust is even more important under conditions that create challenges for teamwork. These include: a high level of task interdependence (i.e. the degree to which team members must rely on each other’s input and resources to perform their tasks effectively); a high level of virtuality (i.e. the degree to which team members do not work in either the same place and/or at the same time, and therefore cannot collaborate face-to-face all of the time); low temporal stability (i.e. the degree to which team members have a history of working together in the past and an expectation of working together in the future); high authority differentiation (decision-making responsibility is distributed across the team); and a high level of skill differentiation (the degree to which teams consist of members with specialised knowledge or skills that make them uniquely qualified and
therefore difficult to substitute).
In addition, it was found that in virtual teams, team familiarity has a positive effect on the development of team trust
(Webber, 2008), whereas negative performance feedback has a substantial negative impact on
team trust (Jaakson, 2019). Finally, a meta-analysis of controlled studies indicate that
team-building has a moderate to large positive effect on a team’s affect-based
trust (Klein, 2009).
Socio-affective states
12
Finding 5: Group-level psychological safety has a moderate to large positive impact on team performance (Level B)
Finding 6: Team cohesion has a moderate to large impact on team performance (Level B)
Finding 7: The cohesion – performance relationship is moderated by team size, type of team, and task interdependence (Level B)
Psychological safety at the group-level refers to the shared belief held by members of a group that the group is safe for ‘interpersonal risk taking’ - a sense of confidence that others will not embarrass, reject or punish someone for speaking up (Edmondson, 1999). Psychological safety is related to ‘intra-team trust’, but the primary difference is that psychological safety concerns a belief about a group norm, whereas trust concerns a belief that one person has about another (Edmondson, 2003). A large, recent meta-analyses including 136 studies with a combined sample size of 5,000 teams indicates that psychological safety has a moderate to large impact on team performance (Frazier, 2017).
Several meta-analyses have demonstrated that cohesion, in particular social cohesion, has a moderate to large impact on a team’s (behavioural) performance (Chiocchio, 2009; Evans, 2012; Mathieu, 2015). Social cohesion refers to a shared liking or attraction to the group, emotional bonds of friendship, caring and closeness among group members, and enjoyment of each other’s company (Chiocchio, 2009). Other constructs that are related to social cohesion, such as relationship building, team familiarity, friendship, social network density, have shown a similar impact on team performance (Chung, 2018; De Jong, 2017). For example, a meta-analysis involving more than 3000 teams found that for newly acquainted team members, informal (social) ties are critical to performance (Balkundi, 2006).
In addition, it was found that the positive effect of social cohesion is stronger within large teams, virtual teams, project teams, and teams with high task interdependency (Lin, 2008; Gully, 2012; Chiocchio, 2009). Finally, social cohesion is not a stable trait; it can (and most likely does) change over time. More specifically, several studies suggest that it takes time for team cohesion to develop and solidify before it positively affects performance. As such, it may be beneficial to try and accelerate the process by engaging in team building and other activities aimed at enhance team familiarity, morale and cohesion (Mathieu, 2015).
13
Finding 9: Team cohesion is strongly associated with team inclusion (Level B)
Finding 10: Team identification has a positive effect on social cohesion and consequently team performance (Level B)
Finding 11: Turnover has a negative effect on social cohesion and consequently team performance (Level C)
Finding 8: The emergence of intra-team trust and social cohesion are critical for virtual teams (Level A)
According to the social inclusion model developed by Shore et al (2011), inclusive teams are expected to be more effective than non-inclusive ones, because inclusion stimulates social cohesion, improves intra-team trust, and reduces chances of conflict in the team. Findings from a recent longitudinal study confirmed the central tenet of the social inclusion model and demonstrated that, indeed, at the team level, perceptions of inclusiveness strongly correlate with team cohesion. In addition, the study’s findings suggest that when a team contains members who all feel included (i.e. accepted and feel that their unique characteristics are valued), the team becomes vastly more cohesive, which in turn has a positive effect on a team’s effectiveness (De Cooman, 2016).
Team identification refers to the extent to which people acknowledge and value being part of a team, share norms and behaviours, and experience a sense of social cohesion and interdependency (Solansky, 2011). Randomised experimental interventions have demonstrated, that team identification leads to increased emotional convergence (the process in which people are affectively influenced by others and become more similar with regard to their socio-affective states), social cohesion, and consequently team performance (Tanghe, 2010). Conversely, turnover of members in a team negatively affects social integration and cohesion and thus negatively impacts team performance (Van der Vegt, 2010).
Team cognition is an emergent state that refers to the way in which knowledge important to team functioning is cognitively organised, represented, and distributed within the team (Kozlowski, 2006). Team cognition is a bottom-up emergent construct, originating from the cognition of individual team members (DeChurch, 2010).
As mentioned above, the positive effect of intra-team trust and social cohesion is stronger within virtual teams. In fact, a meta-analysis of high quality studies shows that these two attributes are critical for the performance of teams with a high level of virtuality (Lin, 2008). This finding was confirmed by several randomised controlled studies that demonstrate that virtual teams with a high level of social cohesion and intra-team trust outperform teams in which trust and social cohesion is low (Capiola, 2019; Fang, 2014; Kennedy, 2010). As pointed out by Kennedy (2010), this finding suggests that managers, when setting up a computer-supported team, may want to require an initial face-to-face session (i.e., more than just a “meet and greet”) to prepare members to work together in the future.
14
Finding 12: Team cognition – in particular information sharing, transactive memory systems, and cognitive consensus – has a large, positive impact on team performance (Level AA).
In the past decade, a large number of high quality studies have consistently demonstrated that team cognition is one of the most important drivers of team effectiveness (DeChurch, 2010; Mesmer-Magnus, 2009; Turner, 2014). The research literature distinguishes several constructs of team cognition, such as shared mental models, team mental models, information sharing, transactive memory systems, cognitive consensus and group learning. Of these constructs, information sharing, transactive memory systems, and cognitive consensus have the largest impact on team performance.
Information sharing (IS) refers to the extent to which a team utilises its individual members’ knowledge or expertise for the team’s benefit. Especially if complex problems have to be addressed, IS is indispensable in that it allows team members to share their knowledge and past experiences and exchange and discuss ideas, which is particularly important for the generation of new ideas (Hulsheger et al., 2009). In addition, sharing information with teammates promotes team trust and social cohesion, which in turn increases team performance. Finally, it was found that IS strongly predicts team performance across all kinds of moderators (team size, team type, etc.).
An important concept related to IS is that of the Transactive Memory System (TMS). TMS within a team refers to a form of knowledge that is embedded in a team’s collective memory. This collective memory works like an indexing system that tells members who knows what. Results from meta-analyses consistently show that TMS has a large, positive effect on team performance (Bachrach, 2019; Mesmer Magnus, 2017, Turner, 2014). Surprisingly, a cross-sectional study suggests that trust in team mates is a strong predictor for the emergence of a TMS, whereas trust in management is not (Robertson, 2013).
Cognitive consensus (CC) refers to refers to similarity among group members regarding how key issues are defined and conceptualised (Mohammed, 2001). CC is not so much about consensus regarding final decisions or solutions, but rather about consensus regarding the interpretation of issues. Put differently, CC is about whether team members attend to, interpret, and communicate about issues in a similar way (Mumford, 2008).
Cognitive states
15
Finding 13: Team learning does not automatically lead to team performance improvement (Level AA)
Finding 14: Team reflexivity moderates the effect of team cognition on team performance (Level C)
Finding 15: Team building has a moderate positive effect on team performance (Level A)
Team learning involves behaviours such as asking questions, challenging assumptions, and discussing errors or unexpected outcomes. Surprisingly, team learning seems to have a rather small impact on team performance (Turner, 2014; Santos, 2015). However, this does not seem to be the case for team reflexivity, which is often considered an important element of (team) learning.
In fact, team reflexivity – the extent to which team members overtly reflect upon the team’s goals, collaboration, decision making processes, internal communication, etc. – seems to moderate the effect of team cognition (Schippers, 2013; Konradt, 2015; Wildman, 2018). Put differently, if teams don’t periodically reflect on how the team is doing, the positive effects of information sharing, a shared memory system, and cognitive consensus on team performance will decrease (see also Finding 16).
In the past decades, numerous studies on the effectiveness of team interventions have been published. Below an overview is provided of interventions that have shown to have moderate to large effects.
Originally designed as a group process intervention (e.g., Schein, 1969, 1999) for improving interpersonal relations and social interactions, team-building interventions are common and popular (Klein, 2009). Although team building interventions encompass a wide range of activities, the term refers to a class of formal and informal team-level interventions that focus on improving social relations and clarifying roles. As such, team building typically does not target skill-based competencies and is often done in settings outside the actual workplace.
A meta-analyses of longitudinal studies shows that, in general, team-building interventions have a moderate, positive indirect effect on team performance, and a moderate to large positive direct effect on trust, social cohesion, and internal communication (Klein, 2009). This meta-analysis confirmed the findings of a previous meta-analyses that included controlled studies and examined the effect of moderating factors (Svyantec, 1999). Results indicate that the effect of team building is larger when:
» the initiator is external (rather than internal to the team).
» the rationale is corrective (rather than preventive).
» when team members are not involved in the planning.
» when the focus is on both the team’s goals and interpersonal relations.
» when the team building is planned together with other interventions.
» when the team building is led by both an internal staff member and external consultant.
» when the focus is on the group (rather than the individuals).
» when the team building is supported by (higher) management.
What interventions influence team effectiveness?5.
16
Finding 16: Teamwork training has a large positive effect on team performance (Level A)
Finding 17: Debriefing sessions and guided team reflexivity have a moderate to large positive effect on team performance (Level A)
In the scientific literature a distinction is made between ‘taskwork’ and ‘teamwork’. In short, taskwork represents what teams are doing, whereas teamwork describes how they are doing it. Teamwork training involves education of team members about the importance of providing social support within the team or promoting ways to manage interpersonal conflict among teammates.
In some types of training team members take part in a group activity in which they discuss the team’s purposes, goals, and performance, or a simulation wherein they experientially enact various teamwork skills, such as interpersonal communication and coordination. A recent meta-analysis that included 51 controlled studies shows that teamwork training, in general, tend to have a large, positive effect on team performance (McEwan, 2017). This study confirmed the findings of previous meta-analyses, that demonstrated that teamwork training not only has a large, positive effect on team performance, but also on a team’s affective, social, and cognitive state (Delise, 2010; Salas, 2008).
Debriefing sessions lead teams through a series of questions that allow its members to reflect on a recent experience, construct their own meaning from their actions, and uncover lessons learned in a non-punitive environment (Tannenbaum, 2013). Debriefing sessions are also referred to as ‘guided team reflexivity’ (see also inding 13). Meta-analyses and randomised controlled studies have found that, when appropriately conducted, debriefing sessions can lead to substantial improvement of a team’s performance (Tannenbaum, 2013; Konradt, 2015). In addition, it was found that debriefs are most effective when the following requirements are met:
» The focus of the debrief should be on learning and improvement, rather than evaluation or judgment. A developmental, non-punitive focus not only yields more honest and accurate feedback, but also enhances experiential learning.
» The debrief should focus on specific activities, episodes or events, rather than performance or results in general.
» The debrief should be informed by a variety of perspectives and evidence sources. For example, the review should include input from multiple participants and at least one additional source of evidence (e.g. organisational data).
Finally, it was found that facilitated and highly structured debriefs have a greater effect on team performance than non-facilitated or loosely structured debriefs.
17
Finding 18: Setting group goals has a moderate to large positive effect on team performance (Level AA)
Finding 19: For most models of team effectiveness the underlying research is inadequate to establish reliability and validity.
Over the past decades, high-quality meta-analyses in a wide range of disciplines (management, medicine, sports, rehabilitation, prevention, etc.) and populations (patients, athletes, managers, senior adults, children, etc.) have demonstrated the positive effects of goal-setting interventions on performance outcomes.
Overall, challenging and specific goals (rather than non-specific ‘do your best’ goals) have a positive effect on performance. Several studies, however, demonstrate that setting goals at the group level may yield even higher performance than individual goals (Kleingeld, 2011). In addition, it is assumed that group goals enhance both social and cognitive group processes such as planning, cooperation, morale-building, communication, and collective efficacy. Finally, a recent cross-sectional study indicates that the effect of group goals is mediated by team reflexivity (Acikgoz, 2018; see also Findings 14 and 17).
In the popular management literature, there are many team effectiveness models available that claim to help teams work together more efficiently. Examples of popular models are Lombardo and Eichinger’s T7 Model, Hackman & Wageman’s 6 Conditions model, the Lencioni Model, the Katzenbach & Smith model, and the Drexler/Sibbet Team Performance Model. Some of these models focus on team composition and structure, while others emphasise intra-team processes such as communication and interaction.
Although some models contain factors that have been shown to be strong predictors of team performance (e.g. social cohesion, goal clarity, trust), the underlying psychometric research is often inadequate to establish the reliability and validity of the model as a whole (Eisele, 2015).
In an attempt to understand how effective teams work, a number of authors have developed models for determining a team’s performance. Typically, these models contain several variables believed to influence team effectiveness. Some of these models have been proposed decades ago, whereas some have been developed in recent years. By 2008, it was found that there were over 130 different models of team effectiveness available (Salas, 2008). As such, a full review of the evidence supporting these models is beyond the scope of this REA. However, based on the studies included in this review, the following general finding emerged:
What is known about the reliability and validity of team effectiveness models?
6.
18
Attributes of effective teams are one of the most widely researched topics in industrial/organisational (I/O) psychology. This review identified a large number of high quality studies that indicate that effective teams are not so much determined by their composition, but rather through the emergence of socio-affective (in particular trust, psychological safety and social cohesion) and cognitive states (in particular cognitive consensus, information sharing and TMS). An overview of the minimal and maximal effect sizes is provided below.
Conclusion
Interventions » team building
» teamwork training
» debriefing sessions
» group goalsetting
» diversity
» personality
» interdependency
» virtuality
» temporality
» authority
» size
» turnover
» reflexivity
» identification
» cognitive consensus
» information sharing
» transactive memory system
» intra-team trust
» psychological safety
» social cohesion
Team processes & activities
Moderators
Composition
Social-affective states
Effectiveness
Cognitive states
19
In addition, findings suggest that level of interdependency, virtuality, team size, team reflexivity, identification, authority, turnover, and temporality are important moderators. Finally, team interventions such as team building, team training, debriefing and goal setting have shown to positively affect both the emergence of socio-affective and cognitive states and consequently team performance. This suggests that team leaders not only have an important role in promoting and stimulating the emergence of socio-affective and cognitive states - they can also (proactively) initiate interventions to enhance team effectiveness.
This REA aims to provide a balanced assessment of what is known in the scientific literature about the attributes of effective teams and interventions increasing team effectiveness by using the systematic review method to search and critically appraise empirical studies. However, in order to be ‘rapid’, concessions were made in relation to the breadth and depth of the search process, such as the exclusion of unpublished studies, the use of a limited number of databases and a focus on empirical research published in the period 2000 to 2019. As a consequence, some relevant studies may have been missed.
A second limitation concerns the critical appraisal of the studies included, which did not incorporate a comprehensive review of the psychometric properties of their tests, scales and questionnaires.
Finally, this REA focused only on high-quality studies, i.e. studies with a control group and/or longitudinal studies. For this reason, cross-sectional studies were excluded. As a consequence, new, promising findings that are relevant for practice may have been missed.
Given these limitations, care must be taken not to present the findings presented in this REA as conclusive.
Limitations
Average minimal and maximal effect sizes
Team diversity Р = -.05 /.10 Soc. cohesion Р = .20/.60 Team building Р = .25/.45
Personality Р = -.20/.25 Cognitive consensus
Р = .40 Teamwork training Р = .35/.55
Team trust Р = .30/.40 Information sharing
Р = .30/.50 Debriefing/reflection d = .30/.70
Psychological safety Р = .40/.50 Transactive Memory system
Р = .30/.50 Group goalsetting d = .55/1.2
20
Açıkgöz, A., & Latham, G. P. (2018). The mediating effect of team reflection on the relationship between a challenging learning goal and new product success. Canadian Journal of Behavioural Science/Revue canadienne des sciences du comportement, 50(3), 136.
Bachrach, D. G., Lewis, K., Kim, Y., Patel, P. C., Campion, M. C., & Thatcher, S. M. B. (2019). Transactive memory systems in context: A meta-analytic examination of contextual factors in transactive memory systems development and team performance. Journal of Applied Psychology, 104(3), 464-493.
Balkundi, P., & Harrison, D. A. (2006). Ties, leaders, and time in teams: Strong inference about network structure’s effects on team viability and performance. Academy of Management Journal, 49(1), 49-68.
Beal, D. J., Cohen, R. R., Burke, M. J., & McLendon, C. L. (2003). Cohesion and Performance in Groups: A Meta-Analytic Clarification of Construct Relations. Journal of Applied Psychology, 88(6), 989-1004.
Bell, S. T. (2007). Deep-level composition variables as predictors of team performance: A meta-analysis. Journal of Applied Psychology, 92(3), 595.
Bell, S. T., Villado, A. J., Lukasik, M. A., Briggs, A. L., & Belau, L. (2011). Getting Specific about Demographic Diversity Variable and Team Performance Relationships: A Meta-Analysis. Journal of Management, 37(3), 709-743.
Breuer, C., Huffmeier, J., & Hertel, G. (2016). Does trust matter more in virtual teams? A meta-analysis of trust and team effectiveness considering virtuality and documentation as moderators. Journal of Applied Psychology, 101(8), 1151.
Bui, H., Chau, V. S., Degl'Innocenti, M., Leone, L., & Vicentini, F. (2019). The Resilient Organisation: A Meta‐Analysis of the Effect of Communication on Team Diversity and Team Performance: International Review of Applied Psychology. Applied Psychology, 68(4), 621-657.
Capiola, A., Alarcon, G. M., Lyons, J. B., Ryan, T. J., & Schneider, T. R. (2019). Collective Efficacy as a Mediator of the Trustworthiness - Performance Relationship in Computer-Mediated Team-based Contexts. The Journal of Psychology, 153(7), 732-757.
Chiocchio, F., & Hélène, E. (2009). Cohesion and Performance: A Meta-Analytic Review of Disparities Between Project Teams, Production Teams, and Service Teams. Small Group Research, 40(4), 382-420.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates.
Cohen, S. G., & Bailey, D. E. 1997. What makes teams work: Group effectiveness research from the shop floor to the executive suite. Journal of Management, 23: 239-290.
DeChurch, L. A., & Mesmer-Magnus, J. R. (2010). The cognitive underpinnings of effective teamwork: A meta-analysis. Journal of Applied Psychology, 95(1), 32.
De Jong, B. A., Dirks, K. T., & Gillespie, N. (2016). Trust and team performance: A meta-analysis of main effects, moderators, and covariates. Journal of Applied Psychology, 101(8), 1134.
De Jong, J. P., & Fodor, O. C. (2017). Attuning to individual work routines and team performance. Team Performance Management, 23(7/8), 385-406.
Delise, L. A., Gorman, C. A. P., Brooks, A. M. P., Rentsch, J. R. P., & Steele-Johnson, D. P. (2010). The effects of team training on team outcomes: A meta-analysis. Performance Improvement Quarterly, 22(4), 53.
Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350_383.
Edmondson, A.C. (2003). Managing the Risk of Learning: Psychological Safety in Work Teams. In West, Michael A.; Tjosvold, Dean; Smith, Ken G. (eds.). International Handbook of Organisational Teamwork and Cooperative Working. New York: Wiley. pp. 255–275.
Eisele, P. (2013). Validation of the Team Diagnostic Survey and a Field Experiment to Examine the Effects of an Intervention to Increase Team Effectiveness. Group Facilitation(12), 53-70.
Eisele, P. (2015). The predictive validity of the team diagnostic survey: Testing a model with performance and satisfaction as output variables. Team Performance Management, 21(5/6), 293-306.
Evans, C. R., & Dion, K. L. (1991). Group cohesion and performance: A meta-analysis. Small Group Research, 22(2), 175.
Evans, C. R., & Dion, K. L. (2012). Group Cohesion and Performance: A Meta-Analysis. Small Group Research, 43(6), 690-701.
Fang, H. M., & Wen-Ching, C. (2014). The Effects of member familiarity, task results visibility and perceived co-worker loafing on technology-supported team performance: social loafing effect perspective. Asia Pacific Management Review, 19(4), 361.
Frazier, M. L., Fainshmidt, S., Klinger, R. L., Pezeshkan, A., & Vracheva, V. (2017). Psychological safety: A meta‐analytic review and extension. Personnel Psychology, 70(1), 113-165.
References
21
Greer, L. L., de Jong, B. A., Schouten, M. E., & Dannals, J. E. (2018). Why and when hierarchy impacts team effectiveness: A meta-analytic integration. Journal of Applied Psychology, 103(6), 591.
Guillaume, Y. R. F., Brodbeck, F. C., & Riketta, M. (2012). Surface- and deep-level dissimilarity effects on social integration and individual effectiveness related outcomes in work groups: A meta-analytic integration. Journal of Occupational and Organisational Psychology, 85(1), 80.
Gully, S. M., Joshi, A., Incalcaterra, K. A., & Beaubien, J. M. (2002). A meta-analysis of team-efficacy, potency, and performance: Interdependence and level of analysis as moderators of observed relationships. Journal of Applied Psychology, 87(5), 819-832.
Gully, S. M., Devine, D. J., & Whitney, D. J. (MA - 2012). A meta-analysis of cohesion and performance: Effects of level of analysis and task interdependence. Small Group Research, 43(6), 702-725.
Haas, H. (2010). How can we explain mixed effects of diversity on team performance? A review with emphasis on context. Equality, Diversity and Inclusion: An International Journal, 29(5), 458-490.
Hackman, J.R. (2002), Leading Teams: Setting the Stage for Great Performance, Harvard Business School Press, Boston.
Hackman, J.R. and Wageman, R. (2005), “A theory of team coaching”, Academy of Management Review, Vol. 30, pp. 269-87.
Hopp, C., & Zenk, L. (2012). Collaborative team networks and implications for strategic HRM. The International Journal of Human Resource Management, 23(14), 2975.
Horwitz, S. K., & Horwitz, I. B. (2007). The effects of team diversity on team outcomes: A meta-analytic review of team demography. Journal of Management, 33(6), 987-1015.
Hulsheger, U. R., Anderson, N., & Salgado, J. F. (2009). Team-level predictors of innovation at work: A comprehensive meta-analysis spanning three decades of research. Journal of Applied Psychology, 94(5), 1128.
Jaakson, K., Reino, A., & McClenaghan, P. B. (2019). The space between – linking trust with individual and team performance in virtual teams. Team Performance Management, 25(1/2), 30-46.
Kennedy, D. M., Vozdolska, R. R., & McComb, S. A. (2010). Team Decision Making in Computer-Supported Cooperative Work: How Initial Computer-Mediated or Face-to-Face Meetings Set the Stage for Later Outcomes. Decision Sciences, 41(4), 933.
Klein, C., DiazGranados, D., Salas, E., Huy, L., Burke, C. S., Lyons, R., & Goodwin, G. F. (2009). Does Team Building Work? Small Group Research, 40(2), 181-222.
Kleingeld, A., van Mierlo, H., & Arends, L. (2011). The effect of goal setting on group performance: A meta-analysis. Journal of Applied Psychology, 96(6), 1289.
Konradt, U., Schippers, M. C., Garbers, Y., & Steenfatt, C. (2015). Effects of guided reflexivity and team feedback on team performance improvement: The role of team regulatory processes and cognitive emergent states. European Journal of Work and Organisational Psychology, 24(5), 777.
Kozlowski, S. W. J., & Ilgen, D. (2006). Enhancing the effectiveness of work groups and teams. Psychological Science in the Public Interest, 7, 77_124.
Lin, C., Standing, C., & Liu, Y.-C. (2008). A model to develop effective virtual teams. Decision Support Systems, 45(4), 1031.
Mann, R. D. (1959). A review of the relationships between personality and performance in small groups. Psychological Bulletin, 56: 242-270.
Mathieu, J., Maynard, M. T., Rapp, T., & Gilson, L. (2008). Team effectiveness 1997-2007: A review of recent advancements and a glimpse into the future. Journal of Management, 34(3), 410-476.
McEwan, D., Ruissen, G. R., Eys, M. A., Zumbo, B. D., & Beauchamp, M. R. (2017). The effectiveness of teamwork training on teamwork behaviors and team performance: A systematic review and meta-analysis of controlled interventions. PLoS ONE, 12(1).
Mesmer-Magnus, J. R., & DeChurch, L. A. (2009). Information sharing and team performance: A meta-analysis. Journal of Applied Psychology, 94(2), 535.
Mesmer-Magnus, J., Niler, A. A., Plummer, G., Larson, L. E., & DeChurch, L. A. (2017). The cognitive underpinnings of effective teamwork: a continuation. Career Development International, 22(5), 507-519.
Mohammed, S., & Dumville, B. C. (2001). Team mental models in a team knowledge framework: Expanding theory and measurement across disciplinary boundaries. Journal of Organisational Behavior, 22, 89–106.
Mumford, M. D., Hunter, S. T., & Bedell-Avers, K. E. (Eds.). (2008). Multi-level issues in creativity and innovation. Emerald Group Publishing Limited.
Peeters, M. A. G., Van Tuijl, H. F. J. M., Rutte, C. G., & Reymen, I. M. M. J. (2006). Personality and Team Performance: A Meta-Analysis. European Journal of Personality, 20(5), 377-396.
22
Petticrew, M., & Roberts, H. (2006). How to appraise the studies: an introduction to assessing study quality. Systematic reviews in the social sciences: A practical guide, 125-163.
Prewett, M., Walvoord, A. G., Stilson, F. B., Rossi, M., & Brannick, M. (2009). The Team Personality-Team Performance Relationship Revisited: The Impact of Criterion Choice, Pattern of Workflow, and Method of Aggregation. Human Performance, 22(4), 273-296.
Robertson, R., Gockel, C., & Brauner, E. (2012). Trust your teammates or bosses? Differential effects of trust on transactive memory, job satisfaction, and performance. Employee Relations, 35(2), 222-242.
Salas, E., DiazGranados, D., Klein, C., Burke, C. S., Stagl, K. C., Goodwin, G. F., & Halpin, S. M. (2008). Does team training improve team performance? A meta-analysis. Human Factors, 50(6), 903-933.
Salas, E., Cooke, N.J. and Rosen, M.A. (2008), “On teams, teamwork, and team performance: discoveries and developments”, Human Factors: The Journal of the Human Factors and Ergonomics Society, Vol. 50 No. 3, pp. 540-547.
Santos, C. M., Uitdewilligen, S., & Passos, A. M. (2015). A temporal common ground for learning: The moderating effect of shared mental models on the relation between team learning behaviours and performance improvement. European Journal of Work and Organisational Psychology, 24(5), 710.
Santos, J. P., Caetano, A., & Tavares, S. M. (2015). Is training leaders in functional leadership a useful tool for improving the performance of leadership functions and team effectiveness? Leadership Quarterly, 26(3), 470.
Schein, E. H. (1969). Process consultation: Its role in organisation development. Reading, MA: Addison-Wesley.
Schein, E. H. (1999). Process consultation revisited: Building the helping relationship. Reading, MA: Addison-Wesley.
Schippers, M. C., Homan, A. C., & van Knippenberg, D. (2013). To reflect or not to reflect: Prior team performance as a boundary condition of the effects of reflexivity on learning and final team performance. Journal of Organisational Behavior, 34(1), 6.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Houghton, Mifflin and Company.
Shaughnessy, J. J., & Zechmeister, E. B. (1985). Research methods in psychology. Alfred A. Knopf.
Shore, L. M., Randel, A. E., Chung, B. G., Dean, M. A., Ehrhart, K. H., & Singh, G. (2011). Inclusion and diversity in work groups: A review and model for future research. Journal of Management, 37, 1262-1289.
Solansky, S. T. (2011). Team identification: a determining factor of performance. Journal of Managerial Psychology, 26(3), 247-258.
Solansky, S. T. P., & Stringer, D. P. (2019). Collective Mind: A Study of Development and Team Performance. Organisation Development Journal, 37(3), 59-69.
Svyantek, D. J., Goodman, S. A., Benz, L. L., & Gard, J. A. (1999). The Relationship Between Organisational Characteristics and Team Building Success. Journal of Business and Psychology, 14(2), 265.
Tanghe, J., Wisse, B., & van der Flier, H. (2010). The Role of Group Member Affect in the Relationship between Trust and Cooperation. British Journal of Management, 21(2), 359.
Tanghe, J., Wisse, B., & van der Flier, H. (2010). The Formation of Group Affect and Team Effectiveness: The Moderating Role of Identification. British Journal of Management, 21(2), 340.
Tannenbaum, S. I., & Cerasoli, C. P. (MA - 2013). Do team and individual debriefs enhance performance? A meta-analysis. Human Factors, 55(1), 231-245.
Turner, J. R., Chen, Q. P., & Danks, S. (2014). Team Shared Cognitive Constructs: A Meta-Analysis Exploring the Effects of Shared Cognitive Constructs on Team Performance. Performance Improvement Quarterly, 27(1), 83.
Van Der Vegt, G. S., Bunderson, S., & Kuipers, B. (2010). Why turnover matters in self-managing work teams: Learning, social integration, and task flexibility. Journal of Management, 36(5), 1168-1191.
Wageman, R. (1995). Interdependence and group effectiveness. Administrative Science Quarterly, 40: 145-180.
Wageman, R., J. R. Hackman, and E. Lehman. (2005). “Team Diagnostic Survey.” Journal of Applied Behavioral Science, 41(4): 373–98.
Wang, J., Grand H‐, L. C., Chen, T., & Leung, K. (2019). Team creativity/innovation in culturally diverse teams: A meta‐analysis. Journal of Organisational Behavior, 40(6).
23
Webber, S. S., & Donahue, L. M. (2001). Impact of highly and less job-related diversity on work group cohesion and performance: a meta-analysis. Journal of Management, 27: 141-162.
Webber, S. S. (2008). Development of cognitive and affective trust in teams: A longitudinal study. Small Group Research, 39(6), 746-769.
Widmann, A., & Mulder, R. H. (2018). Team learning behaviours and innovative work behaviour in work teams. European Journal of Innovation Management, 21(3), 501-520.
Zhou, W., & Rosini, E. (2015). Entrepreneurial Team Diversity and Performance: Toward an Integrated Model. Entrepreneurship Research Journal, 5(1), 31-60.
24
Appendix ISearch terms & hits
ABI/Inform Global, Business Source Elite, PsycINFOpeer reviewed, scholarly journals, November 2019
Search terms ABI BSP PSY
S1: ti(team* OR workgroup* OR group*) 27,086 32,959 69,490
S2: ti(effectiv* OR effic* OR perform* OR innovati* OR learn* OR success* OR collaborati* OR cooperati*
191,527 237,841 252,785
S3: S1 AND S2 4,960 5,031 8,659
S4: filter meta-analyses 62 69 140
S5: ti(antecedents OR attributes OR characteristics OR predictor*)
37,217 30,080 77,497
S6: S3 AND S5 limit > 2010 138 77 71
S7: S3 AND filter high quality studies, limit > 2010
371 384 284
S8: S6 OR S7 431 465 344
S9: ti(team*) AND ti(build* OR interven* OR train* OR develop*)
1,288 1,156 1,147
S10: S9 AND (filter meta-analyses* OR high quality studies**)
11* 15* 101**
25
Appendix IISelection of studies
ABI Inform
n = 504
duplicates
n = 646
excluded
n = 71
excluded
n = 852
BSP
n = 549
Articles obtained from search
n = 1,639
Titles and abstracts screened for relevance
n = 993
critical appraisal & text screened for relevance
n = 141
included studies
n = 70
PsycINFO
n = 585
26
Ap
pen
dix
III
Crit
ical
App
rais
al S
tudi
es T
eam
Effe
ctiv
enes
s
Au
thor
& y
ear
Des
ign
&
sam
ple
siz
eSe
ctor
/ P
opu
lati
onM
ain
fin
din
gs
Eff
ect
size
sLi
mit
atio
ns
Leve
l
Aci
kgoz
, 20
18
cros
s-se
ctio
nal
st
ud
y
n =
194
(7
8 te
ams)
pro
du
ct
dev
elop
men
t te
ams
of 4
3 h
igh
tech
firm
s in
th
e Is
tan
bu
l re
gio
n
The
resu
lts
show
ed t
hat
set
tin
g a
sp
ecifi
c, c
hal
len
gin
g le
arn
ing
goa
l is
asso
ciat
ed w
ith
team
per
form
ance
(new
pro
du
ct s
ucc
ess)
, bu
t th
at t
his
re
lati
onsh
ip is
med
iate
d b
y co
llect
ive
team
refle
ctio
n.
R2
= .6
4n
o se
riou
s lim
itat
ion
sD
Bac
hra
ch,
2019
Met
a-an
alys
is
k =
76
N =
6,8
69
vari
ous
Res
earc
h o
n m
oder
ator
s of
TM
S to
per
form
ance
rela
tion
ship
. It
was
fou
nd
th
at e
nvi
ron
men
tal v
olat
ility
(mar
ket
turb
ule
nce
, tec
hn
olog
y tu
rbu
len
ce,
or e
nvi
ron
men
tal d
ynam
ism
), le
ader
ship
eff
ecti
ven
ess,
an
d te
am h
um
an
cap
ital
(tea
m le
vel k
now
led
ge,
ski
lls, a
bili
ties
) are
pos
itiv
ely
asso
ciat
ed w
ith
TM
S, a
nd
info
rmat
ion
al d
iver
sity
(het
erog
enei
ty o
f wor
k ex
per
ien
ce -
e.g
., or
gan
isat
ion
al te
nu
re, j
ob e
xper
ien
ce- e
du
cati
on le
vel,
edu
cati
on m
ajor
, fu
nct
ion
al b
ackg
rou
nd
) an
d g
end
er d
iver
sity
are
neg
ativ
ely
asso
ciat
ed w
ith
TM
S d
evel
opm
ent.
All-
over
team
p
erfo
rman
ce Р
= .4
5
Task
per
form
ance
Р
= .4
4
Aff
ecti
ve p
erfo
rman
ce
Р =
.58
Cre
ativ
e p
erfo
rman
ce
Р =
.42
Eff
ect
mod
erat
ors:
en
v vo
lati
lity
Р =
.12
lead
er e
ff Р
= .6
0
team
h c
ap Р
= .1
2 in
f div
Р =
-.0
8 g
end
er d
iv Р
= -.
13
Des
ign
of
incl
ud
ed
stu
die
s n
ot
spec
ified
C
27
Bal
kun
di,
20
06
Met
a-an
alys
is
of 3
7 cr
oss-
sect
ion
al
stu
die
s;
N =
3,0
98
team
s
Team
s fr
om
the
follo
win
g
con
text
s:
mili
tary
, top
m
anag
emen
t, p
roje
ct,
pro
du
ctio
n a
nd
se
rvic
e
1. Te
ams
wit
h d
ense
r so
cial
net
wor
ks te
nd
to p
erfo
rm b
ette
r (H
1) a
nd
h
ave
gre
ater
team
via
bili
ty (H
2) in
bot
h, t
eam
’s in
stru
men
tal*
(H1a
, H2a
) an
d e
xpre
ssiv
e* (H
1b, H
2b) s
ocia
l net
wor
ks. M
oreo
ver,
team
’s e
xpre
ssiv
e ti
e d
ensi
ty m
igh
t a
larg
er im
pac
t th
an in
stru
men
tal t
ie d
ensi
ty o
n
team
via
bili
ty (H
3b).
How
ever
, th
e ta
sk p
erfo
rman
ce im
plic
atio
ns
of
inst
rum
enta
l tie
s ar
e n
o d
iffer
ent
from
th
ose
of e
xpre
ssiv
e ti
es (H
3a).
2.
Team
s w
hos
e le
ader
s ar
e ce
ntr
al in
th
e te
am’s
inst
rum
enta
l soc
ial
net
wor
k (H
4),
as w
ell a
s te
ams
that
are
cen
tral
in a
n in
terg
rou
p n
etw
ork
(H5)
, ten
d to
per
form
bet
ter.
3.
A m
ore
inte
gra
tive
net
wor
k st
ruct
ure
(eas
e of
sh
arin
g re
sou
rces
) is
likel
y to
ben
efit
futu
re te
am t
ask
per
form
ance
bu
t is
not
as
likel
y to
refle
ct p
ast
per
form
ance
(H6)
.
4.
For
new
ly a
cqu
ain
ted
or
inex
per
ien
ced
team
mem
ber
s, in
form
al t
ies
seem
to b
e m
ore
crit
ical
to p
erfo
rman
ce. A
s te
am m
emb
ers
gai
ned
ex
per
ien
ce w
ith
on
e an
oth
er a
nd
th
eir
wor
k, e
ffect
s of
th
ose
ties
dec
lined
(H
7).
* Tw
o ty
pes
of t
ies
in s
ocia
l net
wor
ks c
an b
e d
isti
ng
uis
hed
: in
stru
men
tal a
nd
ex
pre
ssiv
e. In
stru
men
tal t
ies
are
pat
hw
ays
of w
ork
rela
ted
ad
vice
. Th
ey m
igh
t em
erg
e fr
om a
form
al re
lati
onsh
ip (e
.g.,
lead
er-s
ub
ord
inat
e), a
nd
th
e p
rimar
y co
nte
nt
exch
ang
ed t
hro
ug
h t
hem
is in
form
atio
n re
sou
rces
or
know
led
ge
that
is re
leva
nt
to c
omp
leti
ng
on
e’s
job
wit
hin
a u
nit
. In
con
tras
t, ex
pre
ssiv
e ti
es re
flect
frie
nd
ship
s. T
hey
are
mor
e af
fect
-lad
en. T
hes
e ti
es a
re im
por
tan
t co
nd
uit
s of
soc
ial s
up
por
t an
d v
alu
es.
H1a
: Р =
.15
H1b
: Р =
.22
H2a
: Р =
.14
H
2b: Р
= .5
5
H3a
: Р =
-0
.08
ns
H3b
: Р =
0.6
3
H4
: Р =
.29
H5:
Р =
.13
H6:
Р =
0.4
1 H
7: Р
= -
0.4
0
No
seri
ous
limit
atio
ns
C
Bay
ley,
20
07
lon
git
ud
inal
st
ud
y
(ass
essm
ent
at 3
poi
nts
in
tim
e:
imm
edia
tely
af
ter,
3 m
onth
s af
ter,
6 m
onth
s af
ter)
47 p
eop
le
from
11 te
ams
Hea
lth
p
rofe
ssio
nal
s in
th
e U
K
Fin
din
g: T
eam
bu
ildin
g d
oes
not
hav
e an
eff
ect
on p
erce
ived
con
dit
ion
s fo
r su
cces
sfu
l tea
mw
ork
(as
mea
sure
d b
y a
team
dev
elop
men
t q
ues
tion
nai
re –
se
e b
elow
).
Team
dev
elop
men
t m
easu
re: i
nst
rum
ent
mea
surin
g e
igh
t fa
ctor
s re
lati
ng
to
succ
essf
ul t
eam
wor
kin
g: a
pp
rop
riat
e w
orkl
oad
, tea
m c
apab
ility
, ass
ocia
tion
w
ith
col
leag
ues
, sh
arin
g o
f act
ivit
ies,
role
sec
uri
ty, p
erso
nal
em
pow
erm
ent,
self-
dire
ctio
n, a
nd
focu
s of
ap
pro
ach
.
not
rep
orte
dn
o se
riou
s lim
itat
ion
sB
28
Bel
l, 20
07
Met
a-an
alys
is
of 3
7 cr
oss-
sect
ion
al
stu
die
s;
k =
24
3 co
rrel
atio
ns
Team
s
(con
text
is
not
cle
ar)
Team
mea
n c
onsc
ien
tiou
snes
s (H
1), t
eam
min
imu
m a
gre
eab
len
ess
(H2)
, ex
trav
ersi
on (H
3), t
eam
ave
rag
e em
otio
nal
sta
bili
ty (H
4), o
pen
nes
s to
ex
per
ien
ce (H
5), c
olle
ctiv
ism
(H6)
, an
d p
refe
ren
ce fo
r te
amw
ork
(H7)
wer
e fo
un
d to
be
rela
ted
to te
am p
erfo
rman
ce in
fiel
d s
tud
ies.
On
ly n
eglig
ible
eff
ects
wer
e ob
serv
ed in
lab
set
tin
gs
for
the
rela
tion
ship
b
etw
een
th
ese
fact
ors
and
team
per
form
ance
. How
ever
, in
lab
set
tin
gs,
te
am m
inim
um
an
d m
axim
um
gen
eral
men
tal a
bili
ty (H
8) a
nd
team
mea
n
emot
ion
al in
telli
gen
ce (H
9) w
ere
rela
ted
to te
am p
erfo
rman
ce. A
lso
in t
he
fiel
d s
etti
ng
GM
A w
as re
late
d to
team
per
form
ance
(H8)
.
The
rela
tion
ship
s b
etw
een
th
e p
erso
nal
ity
fact
ors
and
team
per
form
ance
w
as n
ot re
late
d to
th
e te
am te
nu
re (H
12 n
ot s
up
por
ted
).
H1:
Р =
.30
H
2: Р
= .3
1 H
3: Р
= .1
5 H
4: Р
= .2
1 H
5: Р
= .2
0
H6:
Р =
.40
H
7: Р
= .2
2 H
8: Р
= .3
3; Р
= .2
6 (la
b)
H9:
Р =
.20
(lab
)
No
seri
ous
limit
atio
ns
B
Bel
l, 20
11
met
a-an
alys
is,
k =
92
(274
in
dep
end
ent
corr
elat
ion
s)
1. Th
ere
is a
sm
all p
osit
ive
rela
tion
ship
bet
wee
n fu
nct
ion
al b
ackg
rou
nd
d
iver
sity
in te
rms
of v
arie
ty a
nd
team
per
form
ance
.
2.
Ther
e is
NO
rela
tion
ship
bet
wee
n e
du
cati
onal
bac
kgro
un
d d
iver
sity
in
term
s of
var
iety
an
d te
am p
erfo
rman
ce.
3.
The
pos
itiv
e re
lati
onsh
ip b
etw
een
fun
ctio
nal
bac
kgro
un
d d
iver
sity
in
term
s of
var
iety
an
d te
am p
erfo
rman
ce is
(som
ewh
at) s
tron
ger
wh
en
the
team
per
form
ance
crit
erio
n is
cre
ativ
ity
or in
nov
atio
n ra
ther
th
an
effi
cien
cy.
4.
The
posi
tive
rela
tions
hip
betw
een
func
tiona
l bac
kgro
und
dive
rsity
in te
rms
of v
arie
ty a
nd te
am p
erfo
rman
ce is
(som
ewha
t) s
tron
ger
whe
n th
e te
am is
a
desi
gn
team
or t
op m
anag
emen
t tea
ms
(TM
T) a
s co
mpa
red
to a
noth
er te
am
type
.
5.
The
pos
itiv
e re
lati
onsh
ip b
etw
een
ed
uca
tion
al b
ackg
rou
nd
div
ersi
ty
in te
rms
of v
arie
ty a
nd
team
per
form
ance
is s
tron
ger
wh
en t
he
team
p
erfo
rman
ce c
riter
ion
is c
reat
ivit
y or
inn
ovat
ion
rath
er t
han
effi
cien
cy.
6.
The
pos
itiv
e re
lati
onsh
ip b
etw
een
ed
uca
tion
al b
ackg
rou
nd
div
ersi
ty in
te
rms
of v
arie
ty a
nd
team
per
form
ance
is (s
omew
hat
) str
ong
er w
hen
th
e te
am is
a d
esig
n te
am o
r TM
T as
com
par
ed to
an
oth
er te
am ty
pe.
7.
Ther
e is
NO
rela
tion
ship
bet
wee
n te
am m
ean
ed
uca
tion
al le
vel a
nd
team
p
erfo
rman
ce, r
egar
dle
ss o
f tea
m ty
pe.
8.
Ther
e is
a s
mal
l pos
itiv
e re
lati
onsh
ip b
etw
een
team
mea
n o
rgan
isat
ion
al
ten
ure
an
d te
am p
erfo
rman
ce w
hen
effi
cien
cy is
th
e cr
iterio
n.
9.
Ther
e is
NO
rela
tion
ship
bet
wee
n o
rgan
isat
ion
al te
nu
re d
iver
sity
an
d
team
per
form
ance
wh
en in
nov
atio
n is
th
e cr
iterio
n.
10.
Team
mea
n te
nu
re is
NO
T re
late
d to
team
per
form
ance
wh
en e
ffici
ency
is
th
e cr
iterio
n.
11.
Ther
e is
NO
rela
tion
ship
bet
wee
n s
ex o
r ag
e d
iver
sity
in te
rms
of
sep
arat
ion
an
d te
am p
erfo
rman
ce.
Mos
t E
S ar
e ve
ry s
mal
l
1. Р
= .1
1
2.
Р =
.01
3.
inn
ovat
ion
Р =
.18
effi
cien
cy Р
= .0
3
4.
des
ign
Р =
.16
TMT
Р =
.07
5.
oth
er Р
= -.
01
6.
crea
/inn
o Р
= .2
3 ef
fici
ency
Р =
-.0
2
7.
des
ign
Р =
.07
TMT
Р =
.13
oth
er Р
= -.
05
8.
Р =
.01
9.
Р =
.14
10.
Р =
.04
11.
Р =
.11 n
s an
d -.
04
12.
Р =
.04
ns
Des
ign
of
stu
die
s in
clu
ded
not
sp
ecifi
ed,
bu
t th
e te
xt
sug
ges
ts
that
RC
Ts
wer
e in
clu
ded
.
AA
29
Bre
uer
, 20
16
met
a-an
alys
is
of c
ross
-se
ctio
nal
an
d
lon
git
ud
inal
st
ud
ies
k =
54
N =
12,6
15
(1,8
50 te
ams)
vari
ous
Team
tru
st fa
cilit
ates
coo
rdin
atio
n a
nd
coo
per
atio
n in
team
s, a
nd
is t
her
efor
e p
osit
ivel
y re
late
d w
ith
team
eff
ecti
ven
ess
(att
itu
des
, in
form
atio
n p
roce
ssin
g
and
team
per
form
ance
). Th
e re
lati
onsh
ip b
etw
een
team
tru
st a
nd
team
tas
k p
erfo
rman
ce w
as s
tron
ger
in v
irtu
al te
ams
than
in fa
ce-t
o-f
ace
team
s.
Team
eff
ecti
ven
ess
ov
eral
l Р =
.33
Team
att
itu
des
Р =
.64
Team
inf.
pro
c Р
= .5
4
Team
per
f Р
= .2
7
(tas
k Р
= .2
7,
con
tx Р
= .2
7)
Team
per
form
ance
: vi
rt te
ams
Р =
.33
ftf t
eam
s Р
= .2
2
no
seri
ous
limit
atio
ns
B
Bu
i, 20
19
met
a-an
alys
is,
k =
35va
riou
s
1. Te
am d
iver
sity
mea
sure
d w
ith
SC
att
ribu
tes
hav
e a
pos
itiv
e im
pac
t on
a)
open
nes
s an
d b
) fre
qu
ency
of c
omm
un
icat
ion
.
2.
Team
div
ersi
ty m
easu
red
wit
h K
SA a
ttrib
ute
s h
ave
a n
egat
ive
imp
act
on
a) o
pen
nes
s an
d b
) fre
qu
ency
of c
omm
un
icat
ion
.
3.
Freq
uen
cy (a
) an
d h
igh
op
enn
ess
(b) o
f com
mu
nic
atio
n h
ave
a p
osit
ive
rela
tion
ship
wit
h te
am p
erfo
rman
ce.
Not
e 1:
soci
al-c
ateg
ory
(SC
) dif
fere
nce
s =
race
, eth
nic
ity,
gen
der
, ag
e, re
ligio
n,
sexu
al o
rien
tati
on, a
nd
ph
ysic
al a
bili
ties
; dif
fere
nce
s in
kn
owle
dg
e, s
kills
, an
d
abili
ties
(KSA
) =
edu
cati
on, f
un
ctio
nal
kn
owle
dg
e, in
form
atio
n o
r ex
per
tise
, tr
ain
ing
, exp
erie
nce
, an
d a
bili
ties
).
Not
e 2:
Op
enn
ess
of c
omm
un
icat
ion
is d
efin
ed in
sev
eral
stu
die
s as
'k
now
led
ge
shar
ing
'.
1a:
Р =
.13
1b: Р
= .0
0 n
s
2a:
Р =
.14
ns
2b:
Р =
.11 n
s
3a:
Р =
.20
3b
: Р
= .3
5
des
ign
of
incl
ud
ed
stu
die
s n
ot
spec
ified
C
Cap
iola
, 20
19
RC
T n
= 3
20
(64
team
s)
un
der
gra
du
ate
stu
den
ts a
nd
g
ener
al p
ub
lic
in t
he
US
1. In
div
idu
al-le
vel t
rust
wor
thin
ess
per
cep
tion
s is
pos
itiv
ely
rela
ted
to te
am
per
form
ance
in a
com
pu
ter-
med
iate
d t
ask.
2.
Ind
ivid
ual
-leve
l tru
stw
orth
ines
s p
erce
pti
ons
has
ind
irect
effe
cts
on
team
per
form
ance
in a
com
pu
ter-
med
iate
d t
ask
thro
ug
h g
rou
p-le
vel
colle
ctiv
e ef
fica
cy a
cros
s ti
me.
only
un
stan
dar
diz
ed c
oeffi
cien
ts a
re
rep
orte
d
arti
fici
al
sett
ing
an
d
task
s (a
irpor
t si
mu
lati
on)
A
30
Ch
iocc
hio
, 20
09
met
a-an
alys
is
k =
29
(9,4
16
par
tici
pan
ts
dis
trib
ute
d in
1,5
98 te
ams)
vari
ous
The
coh
esio
n–p
erfo
rman
ce re
lati
onsh
ip is
mod
erat
ed b
y ty
pe
of te
am a
nd
se
ttin
g. P
roje
ct te
ams
in o
rgan
isat
ion
al s
etti
ng
s sh
ow la
rge
effe
ct s
izes
th
an
oth
er t
ypes
of t
eam
s an
d te
ams
in d
iffe
ren
t se
ttin
gs.
In a
dd
itio
n, t
he
coh
esio
n-p
erfo
rman
ce re
lati
onsh
ip is
str
ong
est
for
soci
al
coh
esio
n –
beh
avio
ura
l per
form
ance
.
Not
e: O
utc
ome
per
form
ance
rela
tes
to t
he
end
resu
lts
of t
asks
an
d in
clu
des
m
easu
res
such
as
pro
fits
, sal
es, r
anks
, gra
des
as
wel
l as
sch
edu
le a
nd
cos
t va
rian
ce. B
ehav
iou
ral p
erfo
rman
ce in
clu
des
tw
o ty
pes
of p
erfo
rman
ces:
tas
k an
d c
onte
xtu
al.
pro
j tea
ms:
Р =
.49
pro
d te
ams:
Р =
.14
se
rvic
e te
ams
Р =
.33
soc
coh
–beh
avio
ura
l per
f:
Р =
.65
task
coh
–ou
tcom
e p
erf:
Р
= .3
8 ta
sk c
oh–b
ehav
iou
ral p
erf:
Р
= .3
6 so
cial
coh
–ou
tcom
e p
erf:
Р
= .1
9
des
ign
of
incl
ud
ed
stu
die
s n
ot
spec
ified
(in
clu
des
so
me
lon
git
ud
inal
st
ud
ies)
B
Ch
un
g,
2018
met
a-an
alys
is
k =
26
(1,0
16 g
rou
ps)
vari
ous
Res
ult
s sh
ow t
hat
frie
nd
ship
has
a s
ign
ifica
nt
pos
itiv
e ef
fect
on
gro
up
tas
k p
erfo
rman
ce.
Furt
her
mor
e, t
his
rela
tion
ship
was
mod
erat
ed b
y g
rou
p s
ize
(i.e.
, th
e p
osit
ive
effe
ct o
f fri
end
ship
on
per
form
ance
incr
ease
d w
ith
gro
up
siz
e) a
nd
tas
k fo
cus
(i.e.
, fri
end
ship
gro
up
s p
erfo
rmed
bet
ter
than
acq
uai
nta
nce
gro
up
s on
ta
sks
req
uiri
ng
a h
igh
qu
anti
ty o
f ou
tpu
t, w
her
eas
ther
e w
as n
o p
erfo
rman
ce
ben
efit
on t
asks
req
uiri
ng
a s
ing
le o
r h
igh
-qu
alit
y ou
tpu
t).
Task
inte
rdep
end
ence
did
not
mod
erat
e th
e ef
fect
.
frie
nd
ship
vs
ac
qu
ain
tan
ce g
rou
ps:
d
= .3
1
Des
ign
of
the
incl
ud
ed
stu
die
s n
ot
spec
ified
C
Cor
der
y,
2010
Inte
rru
pte
d
tim
e se
ries
N
= 17
team
s
was
tew
ater
tr
eatm
ent
team
s
1. R
edes
ign
ing
wor
k to
pro
vid
e te
ams
wit
h in
crea
sed
au
ton
omy
resu
lts
in
imp
rove
d te
am p
erfo
rman
ce.
2.
Incr
easi
ng
leve
ls o
f tas
k u
nce
rtai
nty
is a
ssoc
iate
d w
ith
dec
linin
g le
vels
of
team
per
form
ance
.
3.
Task
un
cert
ain
ty a
nd
team
au
ton
omy
inte
ract
, su
ch t
hat
th
e h
igh
er
the
leve
l of t
ask
un
cert
ain
ty, t
he
stro
ng
er t
he
pos
itiv
e im
pac
t of
team
au
ton
omy
on te
am p
erfo
rman
ce.
Not
rep
orte
dN
o m
ajor
w
eakn
esse
sB
De
Coo
man
, 20
16
lon
git
ud
inal
st
ud
y
n =
121
(30
team
s)
colle
ge
stu
den
ts
par
tici
pat
ing
in
a c
ours
e on
str
ateg
ic
man
agem
e n
t in
a la
rge
Du
tch
u
niv
ersi
ty
1. In
div
idu
al-le
vel s
up
ple
men
tary
fit
pos
itiv
ely
corr
elat
ed w
ith
team
co
hes
ion
.
2.
At
the
team
leve
l, th
e ag
gre
gat
e of
su
pp
lem
enta
ry fi
t p
osit
ivel
y co
rrel
ated
wit
h t
he
team
ave
rag
e of
team
coh
esio
n.
3.
The
agg
reg
ate
of c
omp
lem
enta
ry fi
t p
osit
ivel
y co
rrel
ated
wit
h t
he
team
av
erag
e of
team
coh
esio
n (r
= .4
1).
1. r
=.47
2.
r =
.87
3.
r =
.41
no
seri
ous
limit
atio
ns
C
31
Del
ise,
20
10
met
a-an
alys
is
of c
ross
-se
ctio
nal
st
ud
ies
(30
stu
die
s m
easu
red
ou
tcom
es
imm
edia
tely
af
ter
trai
nin
g,
11 h
ad a
tim
e-la
g)
K =
41
(141
3 te
ams)
vari
ous
Inte
rven
tion
: Tea
m t
rain
ing
is d
efin
ed a
s a
pla
nn
ed e
ffor
t d
esig
ned
to
imp
rove
team
per
form
ance
by
assi
stin
g in
div
idu
als
in t
he
acq
uis
itio
n o
f new
in
form
atio
n, s
kills
, an
d a
ttit
ud
es e
ssen
tial
to e
ffec
tive
per
form
ance
in a
team
en
viro
nm
ent.
It is
ad
min
iste
red
to a
n e
nti
re te
am, a
imed
at
enh
anci
ng
th
e p
erfo
rman
ce o
f th
e te
am a
s a
un
it. I
t is
a p
lan
ned
eff
ort
to d
evel
op a
team
’s
task
-sp
ecifi
c co
mp
eten
cies
, th
ereb
y im
pro
vin
g it
s ab
ility
to p
erfo
rm it
s ta
sks
effe
ctiv
ely.
1. Te
am t
rain
ing
is p
osit
ivel
y as
soci
ated
wit
h a
ffect
ive,
cog
nit
ive,
su
bje
ctiv
e ta
sk-b
ased
ski
ll, o
bje
ctiv
e ta
sk-b
ased
ski
ll, a
nd
team
wor
k sk
ill.
2.
Ther
e ar
e n
o d
iffer
ence
s in
effe
cts
of te
am t
rain
ing
in m
ilita
ry v
s. c
ivili
an
sam
ple
s, la
bor
ator
y vs
. fiel
d s
etti
ng
, ad
-hoc
vs.
inta
ct (e
xist
ing
) tea
ms,
te
am-o
rien
ted
vs.
tas
k-or
ien
ted
tra
inin
g, s
hor
t vs
. lon
g t
rain
ing
.
Team
tra
inin
g is
ass
ocia
ted
mor
e st
ron
gly
wit
h im
pro
ved
cog
nit
ive
outc
omes
af
ter
a p
erio
d o
f tim
e p
asse
d fr
om t
he
trai
nin
g t
han
imm
edia
tely
aft
er t
he
trai
nin
g.
1. A
ffect
ive
outc
omes
: d=.
80;
cog
nit
ive:
d=1
.37;
su
bje
ctiv
e ta
sk-b
ased
ski
ll: d
=.88
; ob
ject
ive
task
-bas
ed s
kill:
d=.
76; t
eam
wor
k sk
ill: d
=.64
.
2.
Mili
tary
: d=1
.05,
civ
ilian
: d=.
80;
Lab
: d=.
87, fi
eld
: d=.
76; A
d-h
oc:
d=.
92, i
nta
ct: d
=.62
; no
effe
ct
size
est
imat
es g
iven
for
team
-or
ien
ted
vs,
tas
k-or
ien
ted
tr
ain
ing
an
d s
hor
t vs
. lon
g
trai
nin
g.Im
med
iate
ly a
fter
th
e tr
ain
ing
: d=1
.21;
tim
e la
g a
fter
tr
ain
ing
: d=2
.40
Pu
blic
atio
n
bia
s is
in
corr
ectl
y an
alys
ed
(fu
nn
el p
lot
only
map
s p
osit
ive
effe
ct s
izes
)
C
Eva
ns,
20
12
met
a-an
alys
is
k =
16
(372
gro
up
s)
vari
ous
The
resu
lts
ind
icat
ed a
pos
itiv
e re
lati
onsh
ip w
ith
th
e av
erag
e co
hes
ive
gro
up
p
erfo
rmin
g 18
per
cen
tile
poi
nts
ab
ove
the
aver
age
non
-coh
esiv
e g
rou
p.
r =
.42
not
e: la
rge
95%
CI
limit
ed
sear
ch,
rele
van
t st
ud
ies
may
h
ave
bee
n
mis
sed
d
esig
n o
f in
clu
ded
st
ud
ies
not
sp
ecifi
ed
C
Fan
g,
2014
RC
T n
= 2
85
(95
team
s)
stu
den
ts fr
om
Taiw
an
1. C
omp
ared
to te
ams
in w
hic
h m
emb
ers
are
fam
iliar
wit
h e
ach
oth
er,
team
s in
wh
ich
mem
ber
s ar
e st
ran
ger
s h
ave
low
er p
erfo
rman
ce.
2.
Com
par
ed to
team
s in
wh
ich
tas
k re
sult
s ar
e vi
sib
le, t
eam
s in
wh
ich
th
ese
resu
lt a
re in
visi
ble
do
NO
T h
ave
low
er p
erfo
rman
ce.
3.
Com
par
ed to
team
s in
wh
ich
team
mem
ber
s ar
e n
ot p
erce
ived
as
eng
agin
g in
co-
wor
ker
loafi
ng
, tea
ms
in w
hic
h m
emb
ers
are
per
ceiv
ed
as e
ng
agin
g in
co-
wor
ker
loafi
ng
do
NO
T h
ave
low
er te
am p
erfo
rman
ce.
no
effe
ct s
izes
rep
orte
d
arti
fici
al
sett
ing
an
d
task
s (b
rain
st
orm
ing
)
A
32
Fraz
ier,
20
17
Met
a-an
alys
is
k =
136
N =
> 2
2,0
00
, (5
,00
0 g
rou
ps)
Var
iou
s
Psy
chol
ogic
al s
afet
y im
pac
ts im
por
tan
t or
gan
isat
ion
al o
utc
omes
. It
is p
osit
ivel
y re
late
d to
info
rmat
ion
sh
arin
g, c
itiz
ensh
ip b
ehav
iou
rs
and
tas
k p
erfo
rman
ce. T
her
e ar
e p
erso
nal
ity
trai
ts t
hat
are
pos
itiv
ely
rela
ted
to p
sych
olog
ical
saf
ety.
Th
e re
sult
s in
dic
ate
that
psy
chol
ogic
al
safe
ty is
imp
acte
d b
y p
osit
ive
lead
er re
lati
ons
(e.g
. tra
nsf
orm
atio
nal
le
ader
ship
), w
orkp
lace
su
pp
ort
(e.g
. pee
r su
pp
ort)
, an
d w
ork
des
ign
(e.g
. in
terd
epen
den
ce).
Task
per
form
ance
Р
= .4
3
Info
rmat
ion
sh
arin
g
Р =
.52
OC
B Р
= .3
2
Sear
ch te
rms
not
sp
ecifi
edB
Gar
rett
, 20
19
Qu
asi-
exp
erim
ent
N =
79
team
s (2
4
inte
rven
tion
, 55
con
trol
)
Insu
ran
ce, U
S
The
stu
dy
exam
ines
th
e p
erfo
rman
ce a
nd
beh
avio
ura
l im
pac
t of
team
d
esig
n o
n s
ales
per
form
ance
. Th
e fi
nd
ing
s d
emon
stra
te im
pro
ved
ove
rall
per
form
ance
for
the
team
an
d t
he
ind
ivid
ual
mem
ber
s of
th
e te
am -
the
gai
ns
wer
e p
arti
cula
rly
pro
nou
nce
d w
hen
mem
ber
s h
ave
mod
erat
e le
vels
of
dif
fere
nce
in a
bili
ty, r
ath
er t
han
sm
all o
r la
rge
dif
fere
nce
s in
ab
ility
.
1. A
. In
div
idu
als
in te
am t
asks
per
form
bet
ter
than
th
ose
in in
div
idu
al
task
s. B
. Wea
ker
ind
ivid
ual
s in
team
tas
ks p
erfo
rm b
ette
r th
an t
hos
e in
in
div
idu
al t
asks
. c. S
tron
ger
ind
ivid
ual
s in
team
tas
ks p
erfo
rm b
ette
r th
an
thos
e in
ind
ivid
ual
tas
ks.
2.
A. P
erfo
rman
ce g
ain
s ac
ross
ind
ivid
ual
s in
team
tas
ks, r
elat
ive
to t
hos
e in
ind
ivid
ual
tas
ks, e
xhib
it a
n in
vert
ed U
-sh
aped
rela
tion
ship
wit
h t
he
diff
eren
ce in
ab
ility
of t
he
team
mem
ber
s. B
. Per
form
ance
gai
ns
for
wea
ker
ind
ivid
ual
s in
gro
up
tas
ks, r
elat
ive
to w
eake
r in
div
idu
als
wor
kin
g
alon
e, e
xhib
it a
n in
vert
ed U
-sh
aped
rela
tion
ship
wit
h t
he
diff
eren
ce
in a
bili
ty o
f th
e g
rou
p m
emb
ers.
C. P
erfo
rman
ce g
ain
s fo
r st
ron
ger
in
div
idu
als
in g
rou
p t
asks
, rel
ativ
e to
str
ong
er in
div
idu
als
wor
kin
g a
lon
e,
exh
ibit
an
inve
rted
U-s
hap
ed re
lati
onsh
ip w
ith
th
e d
iffer
ence
in a
bili
ty o
f th
e g
rou
p m
emb
ers.
3.
Inst
rum
enta
lity
to t
he
team
rela
tes
pos
itiv
ely
to t
he
per
form
ance
gai
n
exh
ibite
d b
y th
e sa
les
team
mem
ber
.
4.
Self-
effi
cacy
for
the
task
doe
s n
ot re
late
pos
itiv
ely
to t
he
per
form
ance
g
ain
of t
he
sale
s te
am m
emb
er.
5.
Imp
ress
ion
man
agem
ent
doe
s n
ot re
late
pos
itiv
ely
to t
he
per
form
ance
g
ain
of t
he
sale
s te
am m
emb
er, b
ut
neg
ativ
ely.
6.
The
per
cep
tion
of r
ecei
vin
g c
oach
ing
fro
m a
team
mem
ber
neg
ativ
ely
mod
erat
es t
he
rela
tion
ship
of s
elf-
effi
cacy
to t
he
per
form
ance
gai
n o
f th
e sa
les
team
mem
ber
.
Not
rep
orte
dIn
con
sist
ent
rep
orti
ng
on
h
ypot
hes
es
B
33
Gin
o,
2010
Stu
dy
# 2
: R
CT
n =
36
team
s N
= 2
38
par
tici
pan
ts
Col
leg
e st
ud
ents
Dire
ct t
ask
exp
erie
nce
lead
s to
mor
e h
igh
ly d
evel
oped
tra
nsa
ctiv
e m
emor
y sy
stem
s th
an in
dire
ct e
xper
ien
ce (H
4).
Tran
sact
ive
mem
ory
pos
itiv
ely
influ
ence
s th
e le
vel o
f cre
ativ
ity
of p
rod
uct
s w
ith
in te
ams
(H5)
.
Tran
sact
ive
mem
ory
med
iate
s th
e re
lati
onsh
ip b
etw
een
exp
erie
nce
an
d t
he
leve
l of c
reat
ivit
y of
pro
du
cts
wit
hin
team
s (H
6).
H5:
Ƞ²
= .0
5
H6:
ß =
0.3
3
No
seri
ous
limit
atio
ns
A
Gre
er,
2018
Met
a-an
alys
is
of 5
4 c
ross
-se
ctio
nal
st
ud
ies
N =
13,9
14
team
s
Team
s,
org
anis
atio
nal
se
ttin
g
In g
ener
al, h
iera
rch
y is
like
ly to
hav
e a
neg
ativ
e im
pac
t on
team
eff
ecti
ven
ess
(per
form
ance
an
d v
iab
ility
); th
is e
ffec
t is
med
iate
d b
y in
crea
sed
con
flict
-en
ablin
g s
tate
s (H
2a, H
2b).
The
neg
ativ
e re
lati
onsh
ip b
etw
een
hie
rarc
hy
and
team
per
form
ance
is
exac
erb
ated
by
asp
ects
of t
he
team
str
uct
ure
: mem
ber
ship
inst
abili
ty (H
4a)
an
d s
kill
dif
fere
nti
atio
n (H
4b
), an
d t
he
hie
rarc
hy
itse
lf: m
uta
bili
ty (H
5a).
The
pre
dic
tion
s re
gar
din
g t
he
pos
itiv
e ef
fect
of h
iera
rch
y on
team
via
bili
ty a
s m
edia
ted
by
coor
din
atio
n-e
nab
ling
pro
cess
es (H
1b),
and
of h
iera
rch
y on
team
p
erfo
rman
ce a
s m
edia
ted
by
coor
din
atio
n-e
nab
ling
pro
cess
es (H
1a),
as w
ell
as t
he
mod
erat
ing
role
s of
sev
eral
asp
ects
of t
eam
tas
ks: i
nte
rdep
end
ence
(H
3a) a
nd
com
ple
xity
(H3c
), an
d t
he
hie
rarc
hy:
form
(5b
), w
ere
not
su
pp
orte
d,
wit
h t
he
exce
pti
on t
hat
tas
k am
big
uit
y en
han
ced
th
e p
osit
ive
effe
cts
of
hie
rarc
hy
(H3b
).
Not
e: a
ll E
S lo
w o
r n
s
r =
-.08
(hie
rarc
hy
& p
erfo
rman
ce)
r =
-.11 (
hie
rarc
hy
& v
iab
ility
)
H1a
: not
su
pp
orte
d
H1b
: not
su
pp
orte
d
H2a
: un
clea
r H
2b: u
ncl
ear
H3a
: not
su
pp
orte
d
H3b
: not
su
pp
orte
d
H3c
: not
su
pp
orte
d
H4
a: u
ncl
earH
4b
: un
clea
r
H5a
: un
clea
rH5b
: not
su
pp
orte
d
No
seri
ous
limit
atio
ns
C
34
Gu
illau
me,
20
12
Met
a-an
alys
is
of c
ross
-se
ctio
nal
st
ud
ies
Wor
k g
rou
ps
from
div
ers
ind
ust
ries
1. B
oth
, su
rfac
e-le
vel*
(H1a
) an
d d
eep
-leve
l** (
H1b
) dis
sim
ilarit
y ar
e n
egat
ivel
y re
late
d to
soc
ial i
nte
gra
tion
.
2.
The
neg
ativ
e re
lati
onsh
ip b
etw
een
su
rfac
e-le
vel d
issi
mila
rity
and
soc
ial
inte
gra
tion
is w
eake
r u
nd
er h
igh
team
inte
rdep
end
ence
th
an u
nd
er
low
team
inte
rdep
end
ence
(H2a
). O
n t
he
oth
er h
and
, th
e n
egat
ive
rela
tion
ship
bet
wee
n d
eep
-leve
l dis
sim
ilarit
y an
d s
ocia
l in
teg
rati
on
is s
tron
ger
un
der
hig
h te
am in
terd
epen
den
ce t
han
un
der
low
team
in
terd
epen
den
ce (H
2b).
3.
Ther
e is
a p
osit
ive
rela
tion
ship
bet
wee
n s
ocia
l in
teg
rati
on a
nd
tas
k p
erfo
rman
ce (H
3) a
nd
bet
wee
n s
ocia
l in
teg
rati
on a
nd
con
text
ual
p
erfo
rman
ce (H
4).
Mor
eove
r, so
cial
inte
gra
tion
an
d t
urn
over
are
n
egat
ivel
y re
late
d (H
5).
4.
Un
der
low
team
inte
rdep
end
ence
, soc
ial i
nte
gra
tion
med
iate
s th
e n
egat
ive
rela
tion
ship
bet
wee
n s
urf
ace-
leve
l dis
sim
ilarit
y an
d t
ask
per
form
ance
(H6a
), an
d b
etw
een
su
rfac
e-le
vel d
issi
mila
rity
and
co
nte
xtu
al p
erfo
rman
ce (H
7a).
5.
Un
der
hig
h te
am in
terd
epen
den
ce, s
ocia
l in
teg
rati
on m
edia
tes
the
neg
ativ
e re
lati
onsh
ip b
etw
een
dee
p-le
vel d
issi
mila
rity
and
tas
k p
erfo
rman
ce (H
6b),
and
bet
wee
n d
eep
-leve
l dis
sim
ilarit
y an
d c
onte
xtu
al
per
form
ance
(H7b
).
6.
Un
der
low
team
inte
rdep
end
ence
, soc
ial i
nte
gra
tion
med
iate
s th
e p
ositi
ve
rela
tion
ship
bet
wee
n s
urf
ace-
leve
l dis
sim
ilarit
y an
d tu
rnov
er (H
8a).
Un
der
h
igh
team
inte
rdep
end
ence
, soc
ial i
nte
gra
tion
med
iate
s th
e p
ositi
ve
rela
tion
ship
bet
wee
n d
eep
-leve
l dis
sim
ilarit
y an
d tu
rnov
er (H
8b).
* su
rfac
e-le
vel d
issi
mila
riti
es: i
.e.,
age,
rac
e/et
hn
icit
y, g
end
er, t
enu
re, *
* d
eep
-le
vel d
issi
mila
rity
: i.e
., p
erso
nal
ity,
att
itu
des
, an
d v
alu
es.
Un
clea
r, on
ly g
amm
a’s
ar
e re
por
ted
No
seri
ous
limit
atio
ns
C
Gu
lly,
200
2
Met
a-an
alys
is
of 6
7 cr
oss-
sect
ion
al
stu
die
sC
onte
xt is
not
cl
ear
The
pos
itiv
e re
lati
onsh
ips
bet
wee
n te
am e
ffica
cy a
nd
per
form
ance
, an
d
pot
ency
* an
d p
erfo
rman
ce is
str
ong
er a
t th
e te
am le
vel o
f an
alys
is t
han
at
the
ind
ivid
ual
leve
l (H
3).
At
the
team
leve
l, b
oth
team
-effi
cacy
(H1)
an
d p
oten
cy (H
2) h
ad p
osit
ive
rela
tion
ship
s w
ith
per
form
ance
.
The
rela
tion
ship
bet
wee
n te
am-e
ffica
cy a
nd
per
form
ance
see
ms
to b
e st
ron
ger
wh
en in
terd
epen
den
ce w
as h
igh
th
an w
hen
it w
as lo
w).
Such
m
oder
atin
g e
ffec
t of
inte
rdep
end
ence
was
not
fou
nd
for
the
rela
tion
ship
b
etw
een
pot
ency
an
d p
erfo
rman
ce.
* Pot
ency
refe
rs to
gen
eral
ized
bel
iefs
ab
out t
he
cap
abili
ties
of th
e te
am a
cros
s ta
sks
and
con
text
s (i.
e., o
ur t
eam
will
be
succ
essf
ul n
o m
atte
r wh
at th
e ta
sk).
H1:
Р =
.41
H2:
Р =
.37
H3:
Р =
.39
(tea
m le
vel)
Р =
.20
(in
div
idu
al le
vel)
H4
(tea
m-e
ffica
cy):
Р =
.45
(hig
h in
terd
epen
den
ce)
Р =
.34
(low
in
terd
epen
den
ce)
No
seri
ous
limit
atio
ns
C
35
Gu
lly,
2012
met
a-an
alys
is
k =
46
(51 e
ffec
t si
zes)
vari
ous
Res
ult
s su
gg
est
that
leve
l of a
nal
ysis
an
d t
ask
inte
rdep
end
ence
mod
erat
e th
e co
hes
ion
-per
form
ance
rela
tion
ship
.
low
tas
k in
terd
: r
= .2
1
hig
h t
ask
inte
rd:
r =
.46
limit
ed
sear
ch,
rele
van
t st
ud
ies
may
h
ave
bee
n
mis
sed
des
ign
of
incl
ud
ed
stu
die
s n
ot
spec
ified
C
Haa
s,
2010
met
a-an
alys
is
k =
30
vari
ous
(incl
ud
es n
on-
wor
k se
ttin
gs)
1. Tw
elve
ou
t of
th
e 15
rela
tion
ship
s of
ag
e d
iver
sity
to te
am p
erfo
rman
ce
that
wer
e te
sted
in re
gre
ssio
n m
odel
s d
id n
ot s
how
an
y si
gn
ifica
nce
. Th
e re
mai
nin
g t
hre
e w
ere
neg
ativ
e.
2.
For
gen
der
div
ersi
ty, p
erfo
rman
ce li
nks
(18)
sh
ow w
eak
(pos
itiv
e, n
egat
ive,
an
d n
on-s
ign
ifica
nt)
cor
rela
tion
s.
3.
For
edu
cati
onal
leve
l div
ersi
ty, r
egre
ssio
ns
(4) a
re n
egat
ive
or n
ot
sig
nifi
can
t. T
her
e is
no
pat
tern
of c
onte
xt fa
ctor
s th
at m
igh
t ex
pla
in t
he
diff
eren
ces.
4.
For
eth
nic
div
ersi
ty (n
ote:
com
bin
ed w
ith
nat
ion
al d
iver
sity
!), f
our
stu
die
s sh
owed
a n
egat
ive
and
th
ree
a (w
eak)
pos
itiv
e re
gre
ssio
n re
sult
, lea
vin
g
11 n
on-s
ign
ifica
nt.
Wh
en c
onsi
der
ing
nat
ion
al d
iver
sity
alo
ne,
th
ere
are
no
sig
nifi
can
t re
lati
onsh
ips.
An
ad
dit
ion
al fi
nd
ing
is t
hat
neg
ativ
e ef
fect
s of
eth
nic
div
ersi
ty in
term
s of
cor
rela
tion
coe
ffici
ents
occ
ur
in te
ams
wit
h
mor
e th
an 12
mem
ber
s.
5.
For
fun
ctio
nal
bac
kgro
un
d d
iver
sity
an
d o
rgan
isat
ion
al te
nu
re d
iver
sity
, co
rrel
atio
ns
are
mix
ed (s
ize,
neg
/pos
), b
ut
mos
tly
non
-sig
nifi
can
t
6.
Non
e of
th
e ob
serv
ed re
lati
onsh
ips
bet
wee
n te
am te
nu
re d
iver
sity
an
d
team
per
form
ance
is s
ign
ifica
ntl
y p
osit
ive
or n
egat
ive.
clos
e to
zer
o or
ns
Ver
y lim
ited
se
arch
, m
ain
ly
stu
die
s ci
ted
in
oth
er
revi
ews.
Des
ign
of
stu
die
s in
clu
ded
not
sp
ecifi
ed,
bu
t th
e te
xt
sug
ges
ts
that
exp
lab
st
ud
ies
wer
e in
clu
ded
.
Use
s vo
te
cou
nti
ng
!
B
Hop
p,
2012
Non
-ra
nd
omis
ed
lon
git
ud
inal
st
ud
y 18
team
s (9
7 st
ud
ents
)
Com
pu
ter
scie
nce
st
ud
ents
Team
s th
at e
xhib
it a
hig
her
var
iati
on w
ith
resp
ect
to c
onsc
ien
tiou
snes
s (w
ith
mor
e p
eop
le h
avin
g h
igh
er o
r lo
wer
val
ues
th
an o
ther
team
mem
ber
s)
red
uce
s te
am p
erfo
rman
ce. H
avin
g p
eop
le in
th
e te
am t
hat
dev
iate
on
cr
uci
al p
erso
nal
ch
arac
teri
stic
s is
neg
ativ
e fo
r p
erfo
rman
ce. T
his
eff
ect
is s
up
por
ted
by
the
neg
ativ
e co
effi
cien
t as
soci
ated
wit
h t
he
dev
iati
on o
f as
pire
d p
oin
ts a
mon
g te
am m
emb
ers.
A w
ide
dis
per
sion
is n
egat
ive
for
team
p
erfo
rman
ce. H
avin
g te
am m
emb
ers
wh
o va
ry in
th
eir
goa
ls fr
om t
he
rest
of
the
team
is d
etrim
enta
l for
team
per
form
ance
.
Un
clea
rn
o se
riou
s lim
itat
ion
sC
36
Hor
wit
z,
200
7
met
a-an
alys
is
k =
35
(78
corr
elat
ion
s)
1. A
. Th
ere
is a
pos
itiv
e re
lati
onsh
ip b
etw
een
tas
k-re
late
d d
iver
sity
an
d t
he
qu
alit
y of
team
per
form
ance
. B. :
Th
ere
no
rela
tion
ship
bet
wee
n b
io-
dem
ogra
ph
ic d
iver
sity
an
d t
he
qu
alit
y of
team
per
form
ance
.
2.
The
rela
tion
ship
bet
wee
n te
am d
iver
sity
an
d te
am p
erfo
rman
ce is
st
ron
ger
for
task
-rel
ated
div
ersi
ty t
han
bio
-dem
ogra
ph
ic d
iver
sity
.
3.
A. T
her
e is
a p
osit
ive
rela
tion
ship
bet
wee
n t
ask-
rela
ted
div
ersi
ty a
nd
th
e q
uan
tity
of t
eam
per
form
ance
. B. T
her
e is
a p
osit
ive
rela
tion
ship
bet
wee
n
bio
-dem
ogra
ph
ic d
iver
sity
an
d t
he
qu
anti
ty o
f tea
m p
erfo
rman
ce.
4.
Ther
e is
a (v
ery
smal
l) n
egat
ive
rela
tion
ship
bet
wee
n te
am d
iver
sity
(b
io-d
emog
rap
hic
div
ersi
ty a
nd
tas
k-re
late
d d
iver
sity
) an
d s
ocia
l in
teg
rati
on a
mon
g te
am m
emb
ers.
5.
Task
com
ple
xity
, tea
m ty
pe,
an
d te
ams
size
do
NO
T m
oder
ate
the
rela
tion
ship
bet
wee
n te
am d
iver
sity
an
d te
am p
erfo
rman
ce.
Not
e: te
am p
erfo
rman
ce =
dec
isio
n m
akin
g, c
reat
ivit
y &
inn
ovat
ion
, pro
ble
m
solv
ing
.
All
ES
are
very
sm
all
1a:
Р =
.13
1b:
Р =
.00
3a:
Р =
.07
3b:
Р =
-.0
2
4:
bio
Р =
-.0
4
task
Р =
-.0
2
Des
ign
of
stu
die
s in
clu
ded
not
sp
ecifi
ed,
bu
t th
e te
xt
sug
ges
ts
that
RC
Ts
wer
e in
clu
ded
.
A
37
Hu
, 20
15
1) L
ong
itu
din
al
fiel
d s
tud
y,
N =
67
team
s,
310
team
m
emb
ers
2)
Ran
dom
ised
2x
2 ex
per
imen
t
N =
124
4
-per
son
te
ams
1) W
ork
team
s, d
iver
se
ind
ust
ries
, US
and
Ch
ina
2)
un
der
gra
du
ate
bu
sin
ess
stu
den
ts, U
S
Stu
dy
pro
pos
ed a
th
eore
tica
l mod
el t
hat
lin
ks te
am p
roso
cial
mot
ivat
ion
to
team
eff
ecti
ven
ess
as m
edia
ted
by
team
pro
cess
es.
1. Te
am p
roso
cial
mot
ivat
ion
is in
dire
ctly
an
d p
osit
ivel
y re
late
d to
(a) t
eam
p
erfo
rman
ce a
nd
(b) t
eam
OC
B, b
ut
not
neg
ativ
ely
rela
ted
to (c
) tea
m
volu
nta
ry t
urn
over
via
team
coo
per
atio
n.
2.
Team
pro
soci
al m
otiv
atio
n is
ind
irect
ly a
nd
pos
itiv
ely
rela
ted
to (a
) tea
m
per
form
ance
, an
d n
egat
ivel
y re
late
d to
(c) t
eam
vol
un
tary
tu
rnov
er
via
team
via
bili
ty, b
ut
not
pos
itiv
ely
rela
ted
to (b
) tea
m O
CB
via
team
vi
abili
ty.
3.
Task
inte
rdep
end
ence
mod
erat
es t
he
ind
irect
effe
cts
of te
am p
roso
cial
m
otiv
atio
n o
n (a
) tea
m p
erfo
rman
ce, (
b) t
eam
OC
B, b
ut
not
(c) t
eam
vo
lun
tary
tu
rnov
er v
ia te
am c
oop
erat
ion
, su
ch t
hat
th
ese
rela
tion
ship
s ar
e st
ron
ger
wh
en t
ask
inte
rdep
end
ence
is h
igh
th
an w
hen
tas
k in
terd
epen
den
ce is
low
.
4.
Task
inte
rdep
end
ence
did
not
mod
erat
e th
e in
dire
ct e
ffect
s of
team
p
roso
cial
mot
ivat
ion
on
(a) t
eam
per
form
ance
, an
d (b
) tea
m O
CB
, bu
t d
id
mod
erat
e (c
) tea
m v
olu
nta
ry t
urn
over
via
team
via
bili
ty, s
uch
th
at t
hes
e re
lati
onsh
ips
are
stro
ng
er w
hen
tas
k in
terd
epen
den
ce is
hig
h t
han
wh
en
task
inte
rdep
end
ence
is lo
w.
Stu
dy
1 1a
) ß1 =
.56
ß2
= .4
9 1b
) ß1 =
.56
ß2
= .2
4
1c) ß
1 = .5
6 ß
2 n
s
2a) ß
1 = .5
9 ß
2 =
.24
2b
) ß1 =
.59
ß2
ns
2c) ß
1 = .5
9 ß
2 =
-.59
3a) h
igh
ß =
.26
low
ß
= .0
2 (n
s)
3b) h
igh
ß =
.21 l
ow
ß =
.02
(ns)
3c
) CI i
ncl
0
4a)
CI i
ncl
0
4b
) CI i
ncl
0
4c)
Hig
h ß
= -.
47 lo
w
ß =
-.20
Stu
dy
2 1a
) ß1 =
.26
ß2
= .5
9 1b
) ß1 =
.26
ß2
= .2
5 1c
) ß1 =
.26
ß2
ns
2a) ß
1 = .5
8 ß
2 =
.51
2b) ß
1 = .5
8 ß
2 n
s 2c
) ß1 =
.58
ß2
= -.5
6
3a) ß
dif
f = .0
8 3b
) ßd
iff =
.07
3c) ß
dif
f = n
s
4a)
CI i
ncl
0
4b
) CI i
ncl
0
4c)
ßd
iff =
-.17
no
seri
ous
wea
knes
ses
1) C
2) A
38
Hu
lsh
eger
, 20
09
Met
a-an
alys
is
of 10
4 c
ross
-se
ctio
nal
st
ud
ies.
Team
s, w
ork
con
text
Res
ult
s re
veal
ed t
hat
team
pro
cess
var
iab
les
of s
up
por
t fo
r in
nov
atio
n
(H7)
, vis
ion
(H5)
, tas
k or
ien
tati
on (H
8), a
nd
ext
ern
al c
omm
un
icat
ion
(H10
b)
dis
pla
yed
th
e st
ron
ges
t re
lati
onsh
ips
wit
h c
reat
ivit
y an
d in
nov
atio
n.
Slig
htl
y w
eake
r as
soci
atio
ns
wer
e fo
un
d fo
r co
hes
ion
(H9)
an
d in
tern
al
com
mu
nic
atio
n (H
10a)
. In
con
tras
t, p
arti
cip
ativ
e sa
fety
dis
pla
yed
on
ly a
wea
k,
non
-gen
eral
izab
le p
osit
ive
corr
elat
ion
wit
h in
nov
atio
n (H
6 n
ot s
up
por
ted
). M
oreo
ver,
the
resu
lts
did
not
con
firm
ed a
n a
ssoc
iati
on b
etw
een
tas
k an
d
rela
tion
ship
con
flict
an
d in
nov
atio
n (H
11 n
ot s
up
por
ted
).
Inp
ut
vari
able
s (i.
e., t
eam
com
pos
itio
n a
nd
str
uct
ure
): jo
b-r
elev
ant
div
ersi
ty
(H1a
), ta
sk (H
2a) a
nd
goa
l (H
2b) i
nte
rdep
end
ence
, tea
m s
ize
(H3)
an
d te
am
lon
gev
ity
(H4)
sh
owed
wea
ker
pos
itiv
e as
soci
atio
n w
ith
cre
ativ
ity
and
in
nov
atio
n. B
ackg
rou
nd
div
ersi
ty (H
1b) a
pp
eare
d to
be
neg
ativ
ely
rela
ted
to
inn
ovat
ion
.
Mod
erat
or a
nal
yses
con
firm
ed t
hat
rela
tion
ship
s d
iffe
r su
bst
anti
ally
d
epen
din
g o
n m
easu
rem
ent
met
hod
(sel
f-ra
tin
gs
vs. i
nd
epen
den
t ra
tin
gs
of in
nov
atio
n) a
nd
mea
sure
men
t le
vel (
ind
ivid
ual
vs.
team
inn
ovat
ion
). Te
am
vari
able
s d
isp
laye
d c
onsi
der
ably
str
ong
er re
lati
onsh
ips
wit
h s
elf-
rep
ort
mea
sure
s of
inn
ovat
ion
com
par
ed w
ith
ind
epen
den
t ra
tin
gs
and
ob
ject
ive
crit
eria
. Tea
m p
roce
ss v
aria
ble
s w
ere
mor
e st
ron
gly
rela
ted
to c
reat
ivit
y an
d
inn
ovat
ion
mea
sure
d a
t th
e te
am t
han
th
e in
div
idu
al le
vel.
H1a
: Р =
.16
H1b
: Р =
-.13
H2a
: Р =
.04
H
2b: Р
= .2
8
H3:
Р =
.17
H4
: Р =
.02
(ns)
H5:
Р =
.49
H6
(not
su
pp
orte
d):
Р
= .1
5
H7:
Р =
.47
H8
: Р =
.42
H9:
Р =
.31
H10
a: Р
= .3
6 H
10b
: Р =
.48
H11
a (n
ot s
up
por
ted
):
Р =
.07
H11
b (n
ot s
up
por
ted
):
Р =
-.0
9
No
seri
ous
limit
atio
ns
C
Jaak
son
, 20
19
Lon
git
ud
inal
st
ud
y
n =
71 t
eam
s
inte
rnat
ion
al
virt
ual
stu
den
t te
ams
wor
kin
g
in fo
ur
un
iver
siti
es
in F
inla
nd
, E
ston
ia, L
atvi
a an
d R
uss
ia
Res
ult
s sh
owed
th
at, i
n v
irtu
al te
ams,
rela
tive
ly h
igh
leve
ls o
f in
itia
l tru
st
did
not
ch
ang
e ov
er t
he
per
iod
of t
he
team
s’ p
roje
cts
in g
ener
al, b
ut
in
team
s w
her
e fe
edb
ack
on p
erfo
rman
ce w
as n
egat
ive,
bot
h t
rust
an
d
tru
stw
orth
ines
s d
eclin
ed s
ub
stan
tial
ly.
not
rep
orte
d
(bu
t te
xt s
ug
ges
ts la
rge)
no
seri
ous
limit
atio
ns
C
39
Jin
, 20
17
Met
a-an
alys
is
of 5
2 cr
oss-
sect
ion
al
stu
die
s
N =
55
ind
epen
den
t sa
mp
les
(8,8
92
obse
rvat
ion
s)
Org
anis
atio
nal
se
ttin
g, t
eam
s
in t
he
con
text
of
new
ven
ture
s (h
igh
-an
d lo
w-
tech
ind
ust
ry)
Ag
gre
gat
ed e
ntr
epre
neu
rial
team
com
pos
itio
n c
har
acte
rist
ics
are
pos
itiv
ely
rela
ted
to n
ew v
entu
re p
erfo
rman
ce, s
uch
th
at t
he
gre
ater
th
e ag
gre
gat
ed
char
acte
rist
ics,
th
e g
reat
er t
he
new
ven
ture
per
form
ance
(H1)
. Mor
eove
r, co
ntr
ary
to t
he
exp
ecta
tion
s, t
he
rela
tion
ship
bet
wee
n a
gg
reg
ated
en
trep
ren
euri
al te
am c
har
acte
rist
ics
and
new
ven
ture
per
form
ance
is
stro
ng
er in
low
-tec
h in
du
stri
es t
han
in h
igh
-tec
h in
du
stri
es (H
4a)
.
The
het
erog
enei
ty o
f en
trep
ren
euri
al te
am c
omp
osit
ion
ch
arac
teri
stic
s is
pos
itiv
ely
rela
ted
to n
ew v
entu
re p
erfo
rman
ce, s
uch
th
at t
he
gre
ater
th
e h
eter
ogen
eity
, th
e g
reat
er t
he
new
ven
ture
per
form
ance
(H2)
. Th
is
rela
tion
ship
is s
imila
r in
th
e lo
w-t
ech
an
d h
igh
-tec
h in
du
stri
es (H
4b
).
Entr
epre
neu
rial t
eam
siz
e is
pos
itive
ly re
late
d to
new
ven
ture
per
form
ance
, su
ch
that
the
gre
ater
the
team
siz
e, th
e g
reat
er th
e n
ew v
entu
re p
erfo
rman
ce (H
3).
H1:
r =
.14
H2:
r =
.0
5
H3:
r =
.0
8
H4
a (n
ot s
up
por
ted
): r
= .2
5 (lo
w-t
ech
ind
ust
ries
)
r =
.11 (
hig
h-t
ech
ind
ust
ries
)
No
seri
ous
limit
atio
ns
C
De
Jon
g,
(20
16)
Met
a-an
alys
is
k =
112
N =
7.7
63
vari
ous
Intr
atea
m t
rust
is p
osit
ivel
y re
late
d to
team
per
form
ance
. To
max
imiz
e te
am
per
form
ance
, tru
st b
uild
ing
init
iati
ves
shou
ld fo
cus
on d
evel
opin
g b
oth
co
gn
itiv
e an
d a
ffec
tive
bas
es o
f tru
st w
ith
in t
he
team
, an
d e
nh
ance
team
m
emb
ers’
tru
st b
oth
in e
ach
oth
er a
nd
in t
he
team
lead
er. T
eam
tru
st w
ill b
e m
ost
crit
ical
wh
en te
am m
emb
ers
wor
k in
a h
igh
ly in
terd
epen
den
t m
ann
er,
wit
h o
ther
mem
ber
s w
ho
pos
sess
un
iqu
e sk
ills
and
hav
e d
iffe
ren
t le
vels
of
auth
orit
y in
th
e te
am (s
ee m
oder
ator
s).
Not
e I:
Cog
nit
ion
-bas
ed t
rust
: in
div
idu
als’
cog
nit
ive
eval
uat
ion
s of
th
e re
liab
ility
, in
teg
rity
, an
d c
omp
eten
ce o
f oth
ers.
Aff
ect
bas
ed t
rust
: in
div
idu
als’
fe
elin
gs
of e
mot
ion
al in
volv
emen
t an
d o
ther
s’ g
enu
ine
care
an
d c
once
rn
for
thei
r w
elfa
re. B
esid
es b
ein
g c
once
ptu
ally
dis
tin
ct, c
ogn
itio
n- a
nd
aff
ect-
bas
ed t
rust
are
reg
ard
ed a
s fu
nct
ion
ally
dis
tin
ct, i
n t
hat
th
ey a
ffec
t ou
tcom
es
thro
ug
h d
isti
nct
cau
sal m
ech
anis
ms
and
th
us
un
iqu
ely
con
trib
ute
to
pre
dic
tin
g p
erfo
rman
ce.
Not
e II:
- Ta
sk in
terd
epen
den
ce: t
he
deg
ree
to w
hic
h te
am m
emb
ers
mu
st
rely
on
eac
h o
ther
’s in
pu
t an
d re
sou
rces
to p
erfo
rm t
hei
r ta
sks
effe
ctiv
ely;
- T
eam
vir
tual
ity:
th
e d
egre
e to
wh
ich
team
mem
ber
s d
o n
ot w
ork
in e
ith
er
the
sam
e p
lace
an
d/o
r at
th
e sa
me
tim
e, a
nd
th
eref
ore
can
not
col
lab
orat
e fa
ce-t
o-f
ace
all o
f th
e ti
me;
- Te
mp
oral
sta
bili
ty: t
he
deg
ree
to w
hic
h te
am
mem
ber
s h
ave
a h
isto
ry o
f wor
kin
g to
get
her
in t
he
pas
t an
d a
n e
xpec
tati
on
of w
orki
ng
tog
eth
er in
th
e fu
ture
; - A
uth
orit
y d
iffe
ren
tiat
ion
: how
dec
isio
n-
mak
ing
resp
onsi
bili
ty is
dis
trib
ute
d a
cros
s th
e te
am; -
Ski
ll d
iffe
ren
tiat
ion
: th
e d
egre
e to
wh
ich
team
s co
nsi
st o
f mem
ber
s w
ith
sp
ecia
lised
kn
owle
dg
e or
sk
ills
that
mak
e th
em u
niq
uel
y q
ual
ified
an
d t
her
efor
e d
iffi
cult
to s
ub
stit
ute
.
Team
per
form
ance
ove
rall
Р =
.30
Cog
nit
ive-
bas
ed t
rust
ß
= .2
4
Aff
ect-
bas
ed t
rust
ß
= .1
5
Mod
erat
ors
(lo
w v
s h
igh
):
virt
ual
ity
Р =
.26
vs .3
5
task
inte
rdep
end
ence
Р
= .2
1 vs
.33
tem
por
al s
tab
ility
Р
= .2
3vs
.32
auth
orit
y d
iffe
ren
tiat
ion
Р
= .2
5 vs
.41
skill
dif
fere
nti
atio
n
Р =
.23
vs .3
6
Sear
ch te
rms
not
sp
ecifi
ed
Des
ign
in
clu
ded
st
ud
ies
not
sp
ecifi
ed
(ref
’s s
ug
ges
t so
me
are
lon
git
ud
inal
or
con
trol
led
)
A
40
De
Jon
g,
2017
1) Q
uas
i ex
per
imen
t N
=
35 te
ams
2) F
ield
stu
dy
N =
66
team
s (2
55 te
am
mem
ber
s)
1) M
ain
ly fe
mal
e st
ud
ents
, R
oman
ia
2) W
ork
team
s,
3 co
un
trie
s,
vari
ous
sect
ors
1) A
bov
e an
d b
eyon
d te
am fa
mili
arit
y, t
ran
sact
ive
mem
ory
and
frie
nd
ship
n
etw
ork
den
sity
, cro
ss-a
ttu
nin
g (C
A) h
ave
a p
osit
ive
imp
act
on te
am
per
form
ance
.
2a T
he
gro
up
leve
l ele
vati
on o
f soc
ial s
ensi
tivi
ty (t
eam
soc
ial s
ensi
tivi
ty) i
s p
osit
ivel
y re
late
d to
CA
.
2b C
A m
edia
tes
the
rela
tion
ship
bet
wee
n t
he
gro
up
leve
l ele
vati
on o
f soc
ial
sen
siti
vity
(tea
m s
ocia
l sen
siti
vity
) an
d te
am p
erfo
rman
ce.
2c T
eam
soc
ial s
ensi
tivi
ty is
mor
e p
osit
ivel
y re
late
d to
CA
in s
mal
l tea
ms
wit
h
low
lon
gev
ity
and
in la
rge
team
s w
ith
hig
h lo
ng
evit
y in
com
par
ison
to la
rge
team
s w
ith
low
lon
gev
ity
bu
t n
ot w
ith
sm
all t
eam
s w
ith
hig
h lo
ng
evit
y.
Not
e: C
ross
-att
un
ing
= h
avin
g a
n a
ccu
rate
un
der
stan
din
g o
f an
d a
nti
cip
ate
on o
ne
anot
her
’s w
ork
rou
tin
es.
1) R
2 =
.58
vs .3
5
2a) ß
= .1
4
2b) I
nd
irect
eff
ect
.16
No
maj
or
wea
knes
ses
B
Ken
ned
y,
2010
RC
T n
= 2
94
(98
team
s)
un
der
gra
du
ate
bu
sin
ess
stu
den
ts fr
om
a la
rge
pu
blic
u
niv
ersi
ty in
th
e N
orth
east
ern
U
nit
ed S
tate
s
Res
ult
s in
dic
ate
com
pu
ter-
med
iate
d te
ams
rep
orte
d lo
wer
par
tici
pat
ive
dec
isio
n m
akin
g t
han
face
-to
-fac
e te
ams
afte
r th
e fi
rst
sess
ion
an
d t
he
dis
par
ity
con
tin
ued
at
the
seco
nd
ses
sion
.
Res
ult
s su
gg
est
that
pra
ctit
ion
ers
may
wan
t to
req
uire
an
init
ial
face
-to
-fac
e se
ssio
n (i
.e.,
mor
e th
an ju
st a
mee
t an
d g
reet
) to
pre
par
e m
emb
ers
to w
ork
tog
eth
er in
th
e fu
ture
.
In a
dd
itio
n,
Wh
en s
etti
ng
up
a c
omp
ute
r-su
pp
orte
d te
am, p
ract
itio
ner
s n
eed
to c
onsi
der
how
th
e d
ura
tion
of t
he
team
’s e
xist
ence
may
imp
act
the
team
’s p
roce
ss d
evel
opm
ent
and
ou
tpu
ts. T
eam
s th
at a
re a
ssem
ble
d
to c
omp
lete
a s
pec
ific
task
in a
ver
y sh
ort
per
iod
may
not
hav
e ti
me
to
succ
essf
ully
dev
elop
pro
cess
es a
s w
ould
a te
am w
orki
ng
on
a p
roje
ct o
ver
a m
uch
lon
ger
du
rati
on. I
n s
uch
cas
es, a
ssig
nin
g te
am m
emb
ers
that
are
wel
l ac
qu
ain
ted
wit
h e
ach
oth
er m
ay b
e m
ost
app
rop
riat
e.
no
effe
ct s
izes
rep
orte
dar
tifi
cial
se
ttin
g a
nd
ta
sks
A
41
Kle
in,
200
9
met
a-an
alys
is
of b
efor
e-an
d-
afte
r st
ud
ies
(som
e st
ud
ies
are
also
co
ntr
olle
d)
20 s
tud
ies,
1,5
62 te
ams
Ad
ult
s in
non
-cl
inic
al s
etti
ng
s
Inte
rven
tion
: tea
m b
uild
ing
(a c
lass
of f
orm
al a
nd
info
rmal
team
-lev
el
inte
rven
tion
s th
at fo
cus
on im
pro
vin
g s
ocia
l rel
atio
ns
and
cla
rify
ing
role
s, a
s w
ell a
s so
lvin
g t
ask
and
inte
rper
son
al p
rob
lem
s th
at a
ffec
t te
am fu
nct
ion
ing
). Te
am b
uild
ing
doe
s n
ot t
arg
et s
kill-
bas
ed c
omp
eten
cies
, is
not
sys
tem
atic
in
nat
ure
, an
d is
typ
ical
ly d
one
in s
etti
ng
s th
at d
o n
ot a
pp
roxi
mat
e th
e ac
tual
p
erfo
rman
ce e
nvi
ron
men
t.
1. Te
amb
uild
ing
inte
rven
tion
s h
ave
a m
oder
ate
pos
itiv
e ef
fect
on
team
ou
tcom
es.
2.
Team
bu
ildin
g h
ave
a sm
all p
osit
ive
effe
ct o
n c
ogn
itiv
e te
am o
utc
omes
(e
.g.
dec
lara
tive
kn
owle
dg
e of
team
wor
k co
mp
eten
cies
).
3.
Team
bu
ildin
g h
ave
a m
ediu
m-la
rge
pos
itiv
e ef
fect
on
affe
ctiv
e te
am
outc
omes
(e.g
. tr
ust
, tea
m p
oten
cy).
4.
Team
bu
ildin
g h
ave
a m
ediu
m-la
rge
pos
itiv
e ef
fect
on
pro
cess
team
ou
tcom
es (e
.g.
coor
din
atio
n, c
omm
un
icat
ion
).
5.
Team
bu
ildin
g h
ave
a sm
all-
to-m
ediu
m p
osit
ive
effe
ct o
n te
am
per
form
ance
.
6.
Com
pon
ents
of t
eam
bu
ildin
g (g
oal s
etti
ng
, in
terp
erso
nal
rela
tion
s,
pro
ble
m s
olvi
ng
, an
d ro
le c
larifi
cati
on):
each
of t
hem
ind
ivid
ual
ly h
as a
m
ediu
m e
ffect
on
team
ou
tcom
es.
Not
e: T
eam
siz
e m
oder
ates
th
e ef
fect
s of
team
bu
ildin
g o
n te
am o
utc
omes
: th
e ef
fect
is m
ediu
m fo
r sm
all a
nd
med
ium
-siz
ed te
ams,
an
d la
rge
for
la
rge
team
s.
Р=.
31
Р=.
13
Р=.
44
Р=.
44
Р=.
26
goa
l set
tin
g: Р
=.37
; in
terp
erso
nal
re
lati
ons:
Р
=26;
p
rob
lem
sol
vin
g: Р
=.24
; ro
le c
lari
fica
tion
: Р=.
35
smal
l tea
ms
(<5
mem
ber
s): Р
=.28
; m
ediu
m (5
-10
mem
ber
s): Р
=.27
; lar
ge
(>10
mem
ber
s): Р
=.66
no
seri
ous
limit
atio
ns
A
Kle
ing
eld
, 20
11
3/4
lab
st
ud
ies,
1/4
fi
eld
stu
die
s;
only
stu
die
s w
ith
pre
-tes
t an
d c
ontr
ol
gro
up
k =
49,
N
(gro
up
s) =
73
9;
N (i
nd
ivid
ual
s)
= 29
54
vari
ous
Spec
ific
dif
ficu
lt g
oals
yie
ld c
onsi
der
ably
hig
her
gro
up
per
form
ance
co
mp
ared
wit
h n
onsp
ecifi
c g
oals
. Mod
erat
ely
dif
ficu
lt a
nd
eas
y g
oals
wer
e al
so a
ssoc
iate
d w
ith
per
form
ance
ben
efits
rela
tive
to n
onsp
ecifi
c g
oals
, bu
t th
ese
effe
cts
wer
e sm
alle
r.
Un
exp
ecte
dly
, tas
k in
terd
epen
den
ce, t
ask
com
ple
xity
, an
d p
arti
cip
atio
n d
id
NO
T m
oder
ate
the
effe
ct o
f gro
up
goa
ls.
Ou
r in
ven
tory
of m
ult
ileve
l goa
ls in
inte
rdep
end
ent
gro
up
s in
dic
ated
th
at t
he
effe
ct o
f in
div
idu
al g
oals
in g
rou
ps
on g
rou
p p
erfo
rman
ce w
as
con
tin
gen
t u
pon
th
e fo
cus
of t
he
goa
l: “E
goc
entr
ic” i
nd
ivid
ual
goa
ls, a
imed
at
max
imiz
ing
ind
ivid
ual
per
form
ance
, yie
lded
a p
arti
cula
rly
neg
ativ
e g
rou
p-
per
form
ance
eff
ect,
wh
erea
s “g
rou
p c
entr
ic” g
oals
, aim
ed a
t m
axim
izin
g t
he
ind
ivid
ual
con
trib
uti
on to
th
e g
rou
p’s
per
form
ance
, sh
owed
a p
osit
ive
effe
ct.
Thes
e fi
nd
ing
s d
emon
stra
te t
hat
gro
up
goa
ls h
ave
a ro
bu
st e
ffec
t on
gro
up
p
erfo
rman
ce. I
nd
ivid
ual
goa
ls c
an a
lso
pro
mot
e g
rou
p p
erfo
rman
ce b
ut
shou
ld b
e u
sed
wit
h c
auti
on in
inte
rdep
end
ent
gro
up
s.
over
all
d =
.56
spec
ific
& d
iffi
cult
d
= 0
.80
gro
up
cen
tric
d
= 1.
2
No
seri
ous
limit
atio
ns
AA
42
Kn
igh
t, 20
15
Met
a-an
alys
is
of 3
9 st
ud
ies
N =
2,7
99
gro
up
s
Con
text
is n
ot
clea
r
Grou
p p
osit
ive
affe
ct h
as c
onsi
sten
t p
osit
ive
effe
cts
on s
ocia
l in
teg
rati
on a
nd
ta
sk p
erfo
rman
ce re
gar
dle
ss o
f con
text
ual
ch
arac
teri
stic
s (H
1).
The
effe
cts
of g
rou
p n
egat
ive
affe
ct d
epen
d o
n t
he
con
text
. Sh
ared
neg
ativ
e fe
elin
gs
pro
mot
e so
cial
inte
gra
tion
an
d t
ask
per
form
ance
wh
en s
tem
min
g
from
an
exo
gen
ous
sou
rce
(H2)
or
exp
erie
nce
d in
a 1-
shot
gro
up
(H3)
, bu
t u
nd
erm
ine
soci
al in
teg
rati
on a
nd
tas
k p
erfo
rman
ce w
hen
ste
mm
ing
from
an
en
dog
enou
s so
urc
e (H
2) o
r ex
per
ien
ced
in a
n o
ng
oin
g g
rou
p (H
3).
Un
clea
r, on
ly B
’s in
stea
d o
f ß’s
are
re
por
ted
.
Des
ign
of
the
incl
ud
ed
stu
die
s
not
rep
orte
d
C
Kon
rad
t, 20
15
RC
T n
= 2
94
(98
team
s)
un
iver
sity
st
ud
ents
(Du
tch
an
d
Ger
man
?)
1. R
eflec
tion
is h
igh
er in
team
s (i
rres
pec
tive
of v
irtu
al o
r fa
ce-t
o-fa
ce) t
hat
re
ceiv
e g
uid
ed re
flexi
vity
com
bin
ed w
ith
feed
bac
k, a
s co
mp
ared
to
team
s w
ho
rece
ive
eith
er (a
) gu
ided
refle
xivi
ty w
ith
out
feed
bac
k or
(b)
nei
ther
gu
ided
refle
xivi
ty n
or fe
edb
ack.
2.
Vir
tual
team
s d
o N
OT
show
low
er te
am re
flect
ion
th
an fa
ce-t
o-fa
ce
team
s.
Not
e: G
uid
ed te
am re
flexi
vity
(som
etim
es re
ferr
ed to
as
bri
efin
g/d
ebri
efin
g)
refe
rs to
an
inte
rven
tion
to in
du
ce re
flect
ion
in g
rou
ps.
1. ß
= .3
4 v
s ß
= .2
4
2.
ß =
-.13
ns
arti
fici
al
sett
ing
an
d
task
sA
Lee,
20
14
Lon
git
ud
inal
st
ud
y,
n =
528
(1
32 fo
ur
mem
ber
te
ams)
un
der
gra
du
ate
bu
sin
ess
stu
den
ts a
t a
un
iver
sity
in t
he
US
Freq
uen
t, d
yad
ic in
form
atio
n e
xch
ang
es a
mon
g te
am m
emb
ers
bot
h h
elp
an
d h
ind
er m
emb
ers
lear
n a
bou
t th
e ex
per
tise
of o
ther
mem
ber
s an
d t
hu
s h
elp
an
d h
ind
er t
he
dev
elop
men
t of
a T
MS.
smal
l bet
a’s
sim
ula
tion
,
stu
den
t p
opu
lati
on,
smal
l tea
ms
C
Lin
, 20
08
Met
a-an
alys
is
(k =
50
) an
d R
CT
(n =
20
0)
MA
: var
iou
s
RC
T: A
ust
ralia
n
stu
den
ts
Res
ult
s sh
ow t
hat
bot
h s
ocia
l (e.
g. r
elat
ion
ship
bu
ildin
g a
nd
coh
esio
n) a
nd
ta
sk (e
.g. c
oord
inat
ion
) fac
tors
are
cru
cial
for
imp
rovi
ng
th
e p
erfo
rman
ce a
nd
sa
tisf
acti
on o
f vir
tual
team
s.
SEM
su
gg
est
the
follo
win
g p
ath
s:
1. co
mm
un
icat
ion
> re
lati
onsh
ip b
uild
ing
> c
oord
inat
ion
> p
erfo
rman
ce.
2.
com
mu
nic
atio
n >
coh
esio
n >
coo
rdin
atio
n >
per
form
ance
.
Coo
rd –
Per
f: r
= .5
3 C
omm
– P
erf:
r =
.32
Coh
– P
erf:
r =
.36
Rel
Bu
i – P
ef: r
= .2
1 Tr
ust
– P
erf:
r =
.29
no
seri
ous
limit
atio
ns
AA
43
Mar
low
, 20
18
Met
a-an
alys
is
of 15
0 c
ross
-se
ctio
nal
st
ud
ies;
N =
9,7
02
team
s
Team
s of
st
ud
ents
or
emp
loye
es
(fiel
ds
such
as
man
agem
ent,
sale
s,
rese
arch
an
d
dev
elop
men
t, su
rgic
al te
ams,
se
arch
an
d
resc
ue
team
s,
and
sim
ula
ted
w
ar g
ames
.
Com
mu
nic
atio
n is
pos
itiv
ely
and
sig
nifi
can
tly
rela
ted
to te
am
per
form
ance
(H1)
.
The
rela
tion
ship
bet
wee
n c
omm
un
icat
ion
an
d te
am p
erfo
rman
ce is
str
ong
er
in fa
mili
ar te
ams
than
in u
nfa
mili
ar te
ams
(H2)
, an
d in
face
-to
-fac
e te
ams
com
par
ing
to v
irtu
al te
ams
(H3,
th
e d
iffe
ren
ce b
etw
een
hyb
rid
team
s an
d
face
-to
-fac
e te
ams
was
not
sig
nifi
can
t). M
oreo
ver,
the
rela
tion
ship
bet
wee
n
com
mu
nic
atio
n q
ual
ity
and
per
form
ance
see
ms
to b
e st
ron
ger
th
an t
he
rela
tion
ship
bet
wee
n c
omm
un
icat
ion
freq
uen
cy a
nd
per
form
ance
(H7)
.
The
rela
tion
ship
bet
wee
n c
omm
un
icat
ion
an
d te
am p
erfo
rman
ce d
oes
not
dep
end
on
lead
ersh
ip s
tyle
(sh
ared
vs
hie
rarc
hic
al le
ader
ship
, H4)
, tas
k in
terd
epen
den
ce (H
5), t
ask
typ
e (c
ogn
itiv
e-b
ased
vs
acti
on-b
ased
, H6)
, co
nte
nt
of c
omm
un
icat
ion
(tas
k-or
ien
ted
vs
per
son
al c
omm
un
icat
ion
, H8)
.
H1:
Р =
.31
H2:
ß =
0.3
H
3:Р
= .1
0 (
virt
ual
team
s)
Р =
.29
(hyb
rid
team
s)
Р =
.32
(fac
e-to
-fac
e
team
s)
H4
(not
su
pp
orte
d):
Р =
.27
(sh
ared
lead
ersh
ip)
Р =
.33
(hie
rarc
hic
al le
ader
ship
) H
5 (n
ot s
up
por
ted
): Р
= .2
7 (h
igh
ly
ind
epen
den
t ta
sks)
Р
= .3
9 (lo
w in
dep
end
ent
task
s)
H6
(not
su
pp
orte
d):
Р=
.30
(cog
nit
ive-
bas
ed
task
s)
Р =
.26
(act
ion
-bas
ed
task
s)
H7:
Р =
.36
(qu
alit
y)
Р =
.19
(fre
qu
ency
) H
8 (n
ot s
up
por
ted
): Р
= .2
2 (p
erso
nal
com
mu
nic
atio
n)
Р=
0.3
5 (t
ask-
rela
ted
co
mm
un
icat
ion
)
No
seri
ous
limit
atio
ns
C
Mat
hie
u
2015
met
a-an
alys
is
of lo
ng
itu
din
al
stu
die
s (a
nd
tw
o ad
dit
ion
al
sin
gle
lo
ng
itu
din
al
stu
die
s)
k=15
(N =
737
te
ams)
vari
ous
Coh
esio
n a
nd
per
form
ance
wer
e re
late
d p
osit
ivel
y an
d re
cip
roca
lly o
ver
tim
e (w
hile
con
trol
ling
for
pre
viou
s p
erfo
rman
ce).
How
ever
, on
ave
rag
e, t
he
coh
esio
n >
per
form
ance
rela
tion
ship
exc
eed
ed t
he
per
form
ance
> c
ohes
ion
rela
tion
ship
. Mor
eove
r, th
e co
hes
ion
> p
erfo
rman
ce
rela
tion
ship
gre
w s
tron
ger
ove
r ti
me
wh
erea
s th
e p
erfo
rman
ce >
coh
esio
n
rela
tion
ship
did
not
.
Res
ult
s su
gg
est
that
it t
akes
tim
e fo
r te
am c
ohes
ion
, as
an e
mer
gen
t st
ate,
to d
evel
op a
nd
sol
idif
y b
efor
e it
beg
ins
to re
late
sig
nifi
can
tly
to la
ter
per
form
ance
. Fol
low
ing
th
is lo
gic
, on
e m
igh
t co
ncl
ud
e th
at it
wou
ld b
e b
enefi
cial
to t
ry a
nd
acc
eler
ate
the
pro
cess
by
eng
agin
g in
team
bu
ildin
g,
char
terin
g e
xerc
ises
, an
d o
ther
act
ivit
ies
that
are
des
ign
ed to
en
han
ce te
am
mor
ale
and
coh
esio
n.
Р=
.27
(T1)
to .3
5 (T
2)n
o se
riou
s lim
itat
ion
sB
44
McE
wan
, 20
17
met
a-an
alys
is
of c
ontr
olle
d
bef
ore-
and
-af
ter
stu
die
s
Sam
ple
si
ze: 7
2 st
ud
ies,
84
39
par
tici
pan
ts
Team
s in
d
iffe
ren
t se
ttin
gs
(lab
orat
ory,
ac
adem
ia,
dif
fere
nt
sect
ors:
h
ealt
hca
re,
avia
tion
, in
du
stry
)
1. Te
amw
ork
trai
nin
g h
as a
med
ium
-to-
larg
e p
osit
ive
effe
ct o
n te
amw
ork
beh
avio
urs
.
2.
Team
wor
k tr
ain
ing
has
a la
rge
pos
itiv
e ef
fect
on
team
per
form
ance
.
Mod
erat
ors
for
effe
ct o
n te
amw
ork
beh
avio
urs
:
3.
Team
con
text
: str
ong
est
effe
ct in
avi
atio
n te
ams,
follo
wed
by
mili
tary
te
ams,
follo
wed
by
hea
lth
care
an
d la
b e
xper
imen
ts, t
hen
ind
ust
ry, a
nd
fi
nal
ly a
cad
emia
.
4.
Team
ten
ure
: tea
mw
ork
trai
nin
g h
as a
larg
er p
osit
ive
effe
ct o
n te
amw
ork
beh
avio
urs
in n
ewly
-for
med
team
s th
an in
exi
stin
g (i
nta
ct) t
eam
s.
5.
Trai
nin
g m
eth
od: s
imu
lati
on-b
ased
tra
inin
gs
has
a la
rge
pos
itiv
e ef
fect
, te
am re
view
s a
med
ium
-to-
larg
e ef
fect
, wor
ksh
op a
med
ium
effe
ct, a
nd
d
idac
tic
edu
cati
on a
sm
all e
ffect
.
6.
The
nu
mb
er o
f dim
ensi
ons
of te
amb
uild
ing
tar
get
ed b
y th
e tr
ain
ing
: ta
rget
ing
3 d
imen
sion
s h
as t
he
stro
ng
est
effe
ct, f
ollo
wed
by
two,
fou
r, an
d o
ne.
7.
The
dim
ensi
ons
targ
eted
by
the
team
bu
ildin
g: p
rep
arat
ion
had
th
e st
ron
ges
t ef
fect
, fol
low
ed b
y in
terp
erso
nal
dyn
amic
s, re
flect
ion
, an
d
exec
uti
on.
Mod
erat
ors
for
effe
ct o
n te
am p
erfo
rman
ce:
8.
Team
con
text
: str
ong
est
effe
ct in
team
s in
ind
ust
ry, f
ollo
wed
by
hea
lth
care
, mili
tary
, avi
atio
n, l
ab e
xper
imen
t, an
d a
cad
emia
.
9.
Team
ten
ure
: tea
mw
ork
trai
nin
g h
as a
larg
er p
osit
ive
effe
ct o
n te
am
per
form
ance
for
exis
tin
g (i
nta
ct) t
eam
s th
an fo
r n
ewly
-for
med
team
s.
10.
Trai
nin
g m
eth
od: t
eam
revi
ews
hav
e a
med
ium
-to-
larg
e ef
fect
, si
mu
lati
on-b
ased
tra
inin
gs
and
wor
ksh
ops
hav
e a
med
ium
effe
ct, a
nd
d
idac
tic
edu
cati
on a
sm
all-
to-m
ediu
m e
ffect
.
11.
The
nu
mb
er o
f dim
ensi
ons
of te
amb
uild
ing
tar
get
ed b
y th
e tr
ain
ing
: ta
rget
ing
4 d
imen
sion
s h
as t
he
stro
ng
est
effe
ct, f
ollo
wed
by
two,
on
e an
d t
hre
e.
The
dim
ensi
ons
targ
eted
by
the
team
bu
ildin
g: p
rep
arat
ion
had
th
e st
ron
ges
t ef
fect
, fol
low
ed b
y in
terp
erso
nal
dyn
amic
s, re
flect
ion
, an
d e
xecu
tion
.
1. d
=.68
(ou
tlie
rs
rem
oved
: d=.
55)
2.
d=.
92 (o
utl
iers
re
mov
ed: d
=.58
)
3.
?
4.
exis
tin
g te
ams:
d=.
33; n
ew
team
s: d
=.67
5.
sim
ula
tion
-bas
ed t
rain
ing
: d
=.78
; tea
m re
view
s: d
=.64
; w
orks
hop
s: d
=.50
; did
acti
c ed
uca
tion
: d
=.19
6.
thre
e d
imen
sion
s:
d=.
98; t
wo
dim
ensi
ons:
d=.
65;
fou
r d
imen
sion
s: d
=.57
; on
e d
imen
sion
: d=.
00
5
7.
Pre
par
atio
n:
d=.
75; i
nte
rper
son
al
dyn
amic
s: d
=.69
; refl
ecti
on:
d=.
65; e
xecu
tion
: d=.
64
8.
?
9.
exis
tin
g te
ams:
d=.
99; n
ew
team
s: d
=.54
10.
team
revi
ews:
d=.
69; s
imu
lati
on-
bas
ed t
rain
ing
: d=.
57;
wor
ksh
ops:
d=.
55; d
idac
tic
edu
cati
on:
d=.
41
11.
d b
etw
een
.4
6 an
d .6
7
d b
etw
een
.5
2 an
d .6
0
no
seri
ous
limit
atio
ns
A
45
McL
arn
on,
2019
RC
T
n =
13,2
24
(1,8
39 te
ams)
ind
ivid
ual
s w
ho
wer
e p
arti
cip
atin
g in
th
e X-
Cu
ltu
re
con
sult
ing
co
mp
etit
ion
Res
ult
s su
pp
orte
d a
str
ong
er in
dire
ct e
ffec
t b
etw
een
com
mu
nic
atio
n
freq
uen
cy a
nd
per
form
ance
, via
pro
cess
coo
rdin
atio
n, w
hen
glo
bal
vir
tual
te
am m
emb
ers
gav
e an
d re
ceiv
ed w
eekl
y (p
eer)
feed
bac
k.u
ncl
ear
arti
fici
al
sett
ing
an
d
task
sA
Mes
mer
-M
agn
us,
20
09
Met
a-an
alys
is,
incl
ud
es R
CTs
k =
72
(4,7
95 g
rou
ps,
n =
17,2
79)
vari
ous
Info
rmat
ion
sh
arin
g p
osit
ivel
y p
red
icte
d te
am p
erfo
rman
ce a
cros
s al
l lev
els
of m
oder
ator
s. H
owev
er, I
S u
niq
uen
ess
pre
dic
ts te
am p
erfo
rman
ce m
ore
stro
ng
ly t
han
IS o
pen
nes
s. In
ad
dit
ion
, it
was
fou
nd
th
at te
ams
shar
e m
ore
info
rmat
ion
wh
erei
n (a
) tas
k d
emon
stra
bili
ty is
hig
h (s
olve
vs
jud
ge)
, b)
dis
cuss
ion
str
uct
ure
is h
igh
(fre
efor
m v
s h
igh
ly fo
cuse
d),
and
(c) m
emb
ers
are
mor
e co
oper
ativ
e d
urin
g d
iscu
ssio
ns.
IS >
per
f: Р
= .4
2
IS u
n >
per
f: Р
= .5
0
IS o
p >
per
f: Р
= .3
2
task
dem
> IS
: Р =
.45
dis
c st
ruct
> IS
: Р =
.41
coop
dis
c >
IS: Р
= .5
7
no
seri
ous
limit
atio
ns
AA
Mes
mer
M
agn
us,
20
17
met
a-an
alys
is
k =
28
(res
ult
s fr
om
4,9
43
team
s / 1
9,57
5 in
div
idu
als)
.
vari
ous
Res
ult
s sh
ow c
onsi
sten
t ef
fect
s fo
r te
am c
ogn
itio
n in
team
pro
cess
an
d
per
form
ance
. How
ever
, wh
erea
s or
igin
ally
com
pila
tion
al c
ogn
itio
n (T
MS)
was
m
ore
stro
ng
ly re
late
d to
bot
h te
am p
roce
ss a
nd
team
per
form
ance
th
an
was
com
pos
itio
nal
cog
nit
ion
(SM
M),
in t
he
up
dat
ed d
atab
ase,
com
pila
tion
al
cog
nit
ion
(TM
S) is
mor
e st
ron
gly
rela
ted
to te
am p
roce
ss a
nd
com
pos
itio
nal
co
gn
itio
n (S
MM
) is
mor
e st
ron
gly
rela
ted
to te
am p
erfo
rman
ce.
In e
ssen
ce, t
his
up
dat
ed fi
nd
ing
su
gg
ests
th
at k
now
ing
wh
o kn
ows
wh
at
(TM
S) is
mor
e im
por
tan
t to
pre
dic
tin
g e
ffec
tive
an
d e
ffici
ent
team
pro
cess
, w
hile
hav
ing
a s
har
ed u
nd
erst
and
ing
of t
he
pro
ble
m, t
ask,
or
team
(e.g
. SM
Ms)
is m
ore
influ
enti
al in
pre
dic
tin
g t
he
exte
nt
to w
hic
h a
team
will
be
succ
essf
ul.
over
all
Р =
.36
com
pos
itio
nal
(SM
M)
Р =
.39
com
pila
tion
al (T
MS)
Р =
.29
des
ign
of
the
incl
ud
ed
stu
die
s n
ot
spec
ified
C
46
Pee
tes,
20
06
Met
a-an
alys
is
k =
10
Pro
fess
ion
al
and
stu
den
t te
ams
Met
a-an
alys
is o
n t
he
rela
tion
ship
bet
wee
n te
am c
omp
osit
ion
in te
rms
of t
he
Big
-Fiv
e p
erso
nal
ity
trai
ts (t
rait
ele
vati
on a
nd
var
iab
ility
) an
d
team
per
form
ance
. Th
e h
igh
er t
he
aver
age
leve
l of a
gre
eab
len
ess
and
co
nsc
ien
tiou
snes
s w
ith
in te
ams,
an
d t
he
mor
e si
mila
r te
am m
emb
ers
are
wit
h re
spec
t to
ag
reea
ble
nes
s an
d c
onsc
ien
tiou
s- n
ess,
th
e b
ette
r th
eir
team
p
erfo
rms.
H1a
Ele
vati
on o
f ext
rave
rsio
n is
NO
T re
late
d to
team
per
form
ance
. H
1b V
aria
bili
ty in
ext
rave
rsio
n is
NO
T p
osit
ivel
y re
late
d to
team
per
form
ance
.
H2a
Ele
vati
on o
f ag
reea
ble
nes
s is
pos
itiv
ely
rela
ted
to te
am p
erfo
rman
ce.
H2b
Var
iab
ility
in a
gre
eab
len
ess
is n
egat
ivel
y re
late
d to
team
per
form
ance
.
H3a
Ele
vati
on o
f con
scie
nti
ousn
ess
is p
osit
ivel
y re
late
d to
team
per
form
ance
.
H3b
Var
iab
ility
in c
onsc
ien
tiou
snes
s is
neg
ativ
ely
rela
ted
to te
am
per
form
ance
.
H4
a E
leva
tion
of e
mot
ion
al s
tab
ility
is N
OT
rela
ted
to te
am p
erfo
rman
ce.
H4
b V
aria
bili
ty in
em
otio
nal
sta
bili
ty is
NO
T re
late
d to
team
per
form
ance
.
H5a
Ele
vati
on o
f op
enn
ess
to e
xper
ien
ce is
NO
T re
late
d to
team
per
form
ance
.
H5b
Var
iab
ility
in o
pen
nes
s to
exp
erie
nce
is n
ot re
late
d to
team
per
form
ance
.
Mod
erat
ion
by
typ
e of
team
was
test
ed fo
r p
rofe
ssio
nal
team
s ve
rsu
s st
ud
ent
team
s. M
oder
atio
n re
sult
s fo
r ag
reea
ble
nes
s an
d c
onsc
ien
tiou
snes
s w
ere
in
line
wit
h t
he
tota
l sam
ple
resu
lts.
How
ever
, stu
den
t an
d p
rofe
ssio
nal
team
s d
iffe
red
in e
ffec
ts fo
r em
otio
nal
sta
bili
ty a
nd
op
enn
ess
to e
xper
ien
ce.
1a Р
= 0
.04
1b Р
= 0
.05
(ns)
2a Р
= 0
.24
2b Р
= =
-0
.12
3a Р
= 0
.20
3b Р
= -
0.2
4
4a
Р =
0.0
4
4b
Р =
0.0
2
5a Р
= 0
.03
5b Р
= -
0.0
1
Des
ign
of
incl
ud
ed
stu
die
s n
ot
rep
orte
d
Smal
l n
um
ber
of
corr
elat
ion
s
C
47
Pre
wet
t,
200
9
Met
a-an
alys
is
k =
70N
ot re
por
ted
Stu
dy
exam
ined
rela
tion
s b
etw
een
team
per
son
alit
y an
d te
am p
erfo
rman
ce
con
sid
erin
g a
.o. t
he
choi
ce o
f cri
teri
on (b
ehav
iou
ral v
s. o
utc
ome)
an
d m
eth
od
of a
gg
reg
atio
n (m
ean
, min
imu
m, m
axim
um
an
d v
aria
nce
).
1. C
onsc
ien
tiou
snes
s ac
ts a
s a
sup
ple
men
tary
tra
it, s
uch
th
at (A
) mea
n
and
(B) m
inim
um
met
hod
s of
ag
gre
gat
ing
Con
scie
nti
ousn
ess
pos
itiv
ely
rela
tes
to p
erfo
rman
ce. (
C) V
aria
nce
in C
onsc
ien
tiou
snes
s is
not
n
egat
ivel
y re
late
d to
per
form
ance
.
2.
Ext
rove
rsio
n a
cts
as a
com
ple
men
tary
tra
it (A
) min
imu
m s
core
s d
id n
ot
neg
ativ
ely
rela
te to
per
form
ance
(B) v
aria
nce
in E
xtro
vers
ion
pos
itiv
ely
pre
dic
ts te
am p
erfo
rman
ce.
3.
Ag
reea
ble
nes
s ac
ts a
s a
sup
ple
men
tary
tra
it, s
uch
th
at (A
) mea
n a
nd
(B
) min
imu
m m
eth
ods
of a
gg
reg
atio
n p
osit
ivel
y re
late
to p
erfo
rman
ce,
and
(C) v
aria
nce
in A
gre
eab
len
ess
neg
ativ
ely
rela
tes
to p
erfo
rman
ce.
4.
Em
otio
nal
Sta
bili
ty d
id n
ot d
emon
stra
te s
up
ple
men
tary
ch
arac
teris
tics
, su
ch t
hat
(A) m
ean
an
d (B
) min
imu
m m
eth
ods
of a
gg
reg
atio
n w
ill
pos
itiv
ely
rela
te to
per
form
ance
, an
d (C
) var
ian
ce m
eth
ods
will
neg
ativ
ely
rela
te to
per
form
ance
.
5.
Team
Ext
rove
rsio
n re
late
s m
ore
stro
ng
ly to
(A) t
eam
beh
avio
r th
an to
(B
) tea
m o
utc
ome
mea
sure
s.
6.
Team
Ag
reea
ble
nes
s re
late
s m
ore
stro
ng
ly to
(A) t
eam
beh
avio
r th
an to
(B
) tea
m o
utc
ome
mea
sure
s.
7.
Team
Em
otio
nal
Sta
bili
ty re
late
s m
ore
stro
ng
ly to
(A) t
eam
beh
avio
r th
an
to (B
) tea
m o
utc
ome
mea
sure
s.
1a r
= .1
3 1b
r =
.13
1c r
= -.
06
(ns)
2a r
= .0
3 (n
s)
2b r
= .0
6
3a r
= .1
0
3b r
= .1
0
3c r
= -.
07
4a
r =
.08
4b
r =
.06
(ns)
4
c r
= -.0
3 (n
s)
5a r
= .2
0
5b r
= .0
6 (n
s)
6a r
= .2
0
6b r
= .0
8
7a r
= .1
7 7b
r =
.05
(ns)
Des
ign
of
incl
ud
ed
stu
die
s n
ot
rep
orte
d
Larg
e n
um
ber
of
hyp
oth
eses
C
Rap
p,
2014
lon
git
ud
inal
st
ud
y,
n =
153
team
s
sale
s te
ams
in
a in
a m
ediu
m-
size
d h
igh
te
chn
olog
y fi
rm
Team
goa
l mon
itor
ing
mod
erat
es t
he
rela
tion
ship
bet
wee
n te
am e
ffica
cy
and
team
per
form
ance
, su
ch t
hat
:
(a) t
her
e is
an
inve
rted
U-s
hap
ed re
lati
onsh
ip b
etw
een
th
ese
vari
able
s am
ong
te
ams
that
en
gag
e in
low
leve
ls o
f mon
itor
ing
an
d
(b) a
pos
itiv
e re
lati
onsh
ip in
team
s th
at e
ng
age
in h
igh
leve
ls o
f mon
itor
ing
.
ß =
.37
no
seri
ous
limit
atio
ns
C
48
Rob
erts
on,
2013
cros
s-se
ctio
nal
st
ud
y n
= 3
83
vari
ous
Tru
st in
team
mat
es p
red
icte
d t
ran
sact
ive
mem
ory.
Tru
st in
man
agem
ent
did
N
OT
pre
dic
t tr
ansa
ctiv
e m
emor
y.
tru
st te
amm
ates
> T
MS:
ß
= .4
6 tr
ust
man
agem
ent
> TM
S:
ß =
.08
no
seri
ous
limit
atio
ns
D
Sala
s,
200
8
met
a-an
alys
is
of c
ross
-se
ctio
nal
st
ud
ies
k =
45
(2
,650
team
s)
Mos
t su
bje
cts
eith
er in
m
ilita
ry d
omai
n
or s
tud
ents
in
lab
set
tin
gs
Inte
rven
tion
: tea
m t
rain
ing
: a s
et o
f too
ls a
nd
met
hod
s th
at, i
n c
omb
inat
ion
w
ith
req
uire
d [t
eam
-bas
ed] c
omp
eten
cies
an
d t
rain
ing
ob
ject
ives
, for
m a
n
inst
ruct
ion
al s
trat
egy.
Tas
k fo
cuse
d te
am t
rain
ing
en
able
s te
am m
emb
ers
to b
ecom
e aw
are
of, l
earn
ab
out,
and
pra
ctic
e re
qu
isit
e te
am c
omp
eten
cies
(i.
e., K
SAs)
an
d p
erfo
rman
ce p
roce
sses
wh
ile re
ceiv
ing
feed
bac
k on
th
eir
per
form
ance
.
1. Te
am t
rain
ing
has
a m
oder
ate
pos
itiv
e ef
fect
on
team
ou
tcom
es.
2.
Team
tra
inin
g h
as a
med
ium
-to-
larg
e p
osit
ive
effe
ct o
n c
ogn
itiv
e an
d
pro
cess
ou
tcom
es, a
nd
a m
ediu
m p
osit
ive
effe
ct o
n a
ffect
ive
outc
omes
.
3.
Team
tra
inin
g h
as a
mod
erat
e-to
-hig
h p
osit
ive
effe
ct o
n te
am
per
form
ance
.
Mod
erat
ors
of t
he
effe
ct o
f tea
m t
rain
ing
on
team
per
form
ance
:
4.
The
con
ten
t of
th
e tr
ain
ing
(tas
kwor
k, te
amw
ork,
bot
h) r
esu
lts
in li
ttle
d
iffer
ence
s in
th
e ef
fect
siz
e es
tim
ates
.
5.
The
stab
ility
of t
he
team
: th
e ef
fect
in in
tact
(exi
stin
g) t
eam
s is
hig
her
th
an in
ad
-hoc
team
s.
The
size
of t
he
team
: th
e ef
fect
is g
reat
est
in la
rge
team
s, fo
llow
ed b
y sm
all
team
s an
d t
hen
by
med
ium
team
s.
1. Р
=.34
2.
cog
nit
ive
ou
tcom
es: Р
=.4
2; p
roce
ss
outc
omes
: Р=.
44
; affe
ctiv
e
outc
omes
: Р=.
35
3.
Р=.
39
4.
task
wor
k: Р
=.35
; tea
mw
ork:
Р
=.38
; b
oth
: Р=.
40.
5.
exis
tin
g te
ams:
Р
=.54
; ad
-hoc
te
ams:
Р=.
38
larg
e te
ams:
Р=.
50;
smal
l tea
ms:
Р=.
39; m
ediu
m
team
s: Р
=.34
no
seri
ous
limit
atio
ns
C
49
San
tos,
20
15
Lon
git
ud
inal
st
ud
y
N =
67
team
s,
314
ind
ivid
ual
s
Com
pan
y m
anag
ers,
u
niv
ersi
ty
stu
den
ts
Team
lear
nin
g p
roce
sses
do
not
au
tom
atic
ally
lead
to p
erfo
rman
ce
imp
rove
men
t. In
ord
er to
ach
ieve
an
incr
ease
in te
am p
erfo
rman
ce o
ver
tim
e,
team
s n
eed
to c
omp
lem
ent
thei
r te
am le
arn
ing
beh
avio
urs
wit
h s
har
ed t
ask
and
tem
por
al m
enta
l mod
els.
1) T
he
exte
nt
to w
hic
h t
he
team
mem
ber
s en
gag
e in
team
lear
nin
g p
roce
sses
is
not
pos
itiv
ely
rela
ted
to te
am p
erfo
rman
ce im
pro
vem
ent.
2a) T
he
rela
tion
ship
bet
wee
n te
am le
arn
ing
pro
cess
es a
nd
team
p
erfo
rman
ce im
pro
vem
ent
is m
oder
ated
by
task
men
tal m
odel
sim
ilari
ty
in s
uch
a w
ay t
hat
wh
en te
am m
emb
ers
hav
e a
sim
ilar
task
men
tal m
odel
, th
e re
lati
onsh
ip w
ill b
e m
ore
pos
itiv
e th
an w
hen
th
ey d
o n
ot h
ave
a si
mila
r m
enta
l mod
el.
2b) T
he
rela
tion
ship
bet
wee
n te
am le
arn
ing
pro
cess
es a
nd
team
p
erfo
rman
ce im
pro
vem
ent
is n
ot m
oder
ated
by
team
men
tal m
odel
si
mila
rity
in s
uch
a w
ay t
hat
wh
en te
am m
emb
ers
hav
e a
sim
ilar
men
tal
mod
el t
he
rela
tion
ship
will
be
mor
e p
osit
ive
than
wh
en t
hey
do
not
hav
e a
sim
ilar
men
tal m
odel
.
2c) T
he
rela
tion
ship
bet
wee
n te
am le
arn
ing
pro
cess
es a
nd
team
p
erfo
rman
ce im
pro
vem
ent
is m
oder
ated
by
tem
por
al m
enta
l mod
el
sim
ilari
ty in
su
ch a
way
th
at w
hen
team
mem
ber
s h
ave
a si
mila
r te
mp
oral
m
enta
l mod
el t
he
rela
tion
ship
will
be
mor
e p
osit
ive
than
wh
en t
hey
do
not
h
ave
a si
mila
r te
mp
oral
men
tal m
odel
.
no
effe
ct s
izes
p
rovi
ded
Con
cern
s a
sim
ula
tion
, p
artl
y w
ith
st
ud
ents
C
Sch
ipp
ers,
20
13
Lon
git
ud
inal
st
ud
y
N =
73
team
s (g
rou
ps
of 3
st
ud
ents
)
Bu
sin
ess
stu
den
ts
1) P
rior
team
per
form
ance
mod
erat
es t
he
rela
tion
ship
bet
wee
n te
am
refle
xivi
ty a
nd
fin
al te
am p
erfo
rman
ce s
uch
th
at te
am re
flexi
vity
will
be
pos
itiv
ely
rela
ted
to fi
nal
team
per
form
ance
for
a) te
ams
wit
h re
lati
vely
low
p
rior
per
form
ance
an
d u
nre
late
d to
fin
al te
am p
erfo
rman
ce fo
r b
) tea
ms
wit
h
rela
tive
ly h
igh
pri
or p
erfo
rman
ce.
2) P
rior
team
per
form
ance
mod
erat
es t
he
rela
tion
ship
bet
wee
n te
am
refle
xivi
ty a
nd
team
lear
nin
g s
uch
th
at a
) tea
m re
flexi
vity
will
be
pos
itiv
ely
rela
ted
to te
am le
arn
ing
for
team
s w
ith
rela
tive
ly lo
w p
rior
per
form
ance
an
d
b) u
nre
late
d to
fin
al te
am p
erfo
rman
ce fo
r te
ams
wit
h re
lati
vely
hig
h p
rior
p
erfo
rman
ce.
3) T
eam
lear
nin
g m
edia
tes
the
inte
ract
ion
bet
wee
n te
am re
flexi
vity
an
d p
rior
te
am p
erfo
rman
ce o
n fi
nal
team
per
form
ance
.
1a ß
= .2
5
1b ß
= -.
16 (n
s)
2a ß
= .4
9
2b ß
= .0
9 (n
s)
No
maj
or
wea
knes
ses
C
50
Siva
sub
ram
- an
iam
, 20
12
Met
a-an
alys
is
of 3
8 st
ud
ies
Org
anis
atio
nal
se
ttin
g, t
eam
s fr
om d
iffe
ren
t in
du
stri
es (e
.g.,
hig
h-t
ech
, m
anu
fact
urin
g,
soft
war
e,
elec
tron
ics,
h
ealt
hca
re)
Team
inp
uts
: tea
m le
ader
ship
(H4)
, tea
m a
bili
ty (H
3) a
nd
team
ten
ure
(H1)
ar
e p
osit
ivel
y re
late
d to
NP
D* t
eam
per
form
ance
. Fu
nct
ion
al d
iver
sity
(H2)
is
un
rela
ted
to N
PD
ou
tcom
es.
Team
pro
cess
var
iab
les:
In
tern
al (H
5) a
nd
ext
ern
al (H
6) c
omm
un
icat
ion
, g
rou
p c
ohes
iven
ess
(H7)
goa
l cla
rity
(H8)
wer
e fo
un
d to
be
pos
itiv
ely
asso
ciat
ed w
ith
NP
D o
utc
omes
.
Ad
dit
ion
ally
, NP
D e
ffec
tive
nes
s se
ems
to b
e p
red
icte
d b
y te
am le
ader
ship
, in
tern
al a
nd
ext
ern
al c
omm
un
icat
ion
, an
d g
rou
p c
ohes
iven
ess
(H9,
H10
). N
PD
ef
fici
ency
was
pre
dic
ted
by
all b
ut
team
siz
e an
d te
am te
nu
re (H
9).
* NP
D =
new
pro
du
ct d
evel
opm
ent.
* tea
m le
ader
ship
= t
he
exte
nt
to w
hic
h t
he
team
lead
er is
ch
aris
mat
ic a
nd
tr
ansf
orm
atio
nal
, an
d u
tilis
es a
sty
le c
har
acte
rise
d a
s b
ein
g p
arti
cip
ativ
e,
emp
ower
ing
, fac
ilita
tive
, an
d c
omm
un
icat
ive.
H1:
r =
.28
H2:
r =
.0
2
H3:
r =
.29
H4
: r =
.4
4
H5:
r =
.31
H6:
r =
.18
H7:
r =
.20
H8
: r =
.50
H9,
H10
:
R2
= .4
5
R2
= .3
4
Des
ign
of
the
incl
ud
ed
stu
die
s
not
rep
orte
d
C
Sola
nsk
y,
2011
Stu
dy
1 Lo
ng
itu
din
al
stu
dy
N1 =
20
team
s (8
6 st
ud
ents
)
Stu
dy
2 N
2a =
126
(10
team
s)
N2b
= 5
8 (1
2 te
ams)
1) S
tud
ent
team
s, U
S
2) W
orki
ng
te
ams,
US,
co
nst
ruct
ion
, ed
uca
tion
1. Te
am id
enti
fica
tion
is p
osit
ivel
y re
late
d to
per
form
ance
.St
ud
y 1 r
= .6
3
Stu
dy
2 r
= .5
8
no
seri
ous
wea
knes
ses
1) C
2) D
Sola
nsk
y,
2019
Stu
dy
1 Q
uas
i exp
st
ud
y N
= 8
6 (2
0 te
ams)
Stu
dy
2 C
ross
-sec
t
N =
126
(10
team
s)
1) S
tud
ents
, US
2) W
orki
ng
te
ams,
US,
co
nst
ruct
ion
1. C
olle
ctiv
e m
ind
sco
res
incr
ease
ove
r ti
me.
2.
Col
lect
ive
min
d s
core
s ar
e p
osit
ivel
y as
soci
ated
wit
h te
am p
erfo
rman
ce.
Not
e: c
olle
ctiv
e m
ind
col
lect
ive
min
d re
fers
to a
team
or
an o
rgan
isat
ion
th
at
acts
inte
llig
entl
y as
a c
olle
ctio
n o
f in
div
idu
als.
Stu
dy
1 1)
Not
rep
orte
d
2) ß
= .6
8
Stu
dy
2 2)
ß =
0.8
0
Un
clea
r re
por
tin
g o
f re
sult
s
1) C
2) D
51
Stew
art,
20
06
Met
a-an
alys
is
of 9
3 st
ud
ies
Team
s p
erfo
rmin
g
real
-life
tas
ks
in a
nat
ura
l se
ttin
g
(man
agem
ent,
pro
du
ctio
n a
nd
p
roje
ct te
ams)
The
stu
dy
exam
ines
th
e re
lati
onsh
ips
bet
wee
n te
am d
esig
n fe
atu
res
(gro
up
co
mp
osit
ion
, tas
k d
esig
n a
nd
lead
ersh
ip) a
nd
team
per
form
ance
an
d re
por
ts
the
follo
win
g re
sult
s:
1. A
gg
reg
ated
mea
sure
s of
ind
ivid
ual
ab
ility
an
d d
isp
osit
ion
cor
rela
te
pos
itiv
ely
wit
h te
am p
erfo
rman
ce.
2.
Team
mem
ber
het
erog
enei
ty a
nd
per
form
ance
cor
rela
te n
ear
zero
, bu
t th
e ef
fect
var
ies
som
ewh
at b
y ty
pe
of te
am (p
roje
ct, p
rod
uct
ion
an
d
man
agem
ent)
.
3.
Pro
ject
an
d m
anag
emen
t te
ams
hav
e sl
igh
tly
hig
her
per
form
ance
wh
en
they
incl
ud
e m
ore
mem
ber
s.
4.
Team
-leve
l tas
k m
ean
ing
fuln
ess
exh
ibit
s a
mod
est
bu
t in
con
sist
ent
rela
tion
ship
wit
h p
erfo
rman
ce.
5.
Incr
ease
d a
uto
nom
y an
d in
tra-
team
coo
rdin
atio
n c
orre
spon
d w
ith
h
igh
er p
erfo
rman
ce, b
ut
the
effe
ct v
arie
s d
epen
din
g o
n t
ask
typ
e (t
eam
s en
gag
ed in
ph
ysic
al v
s kn
owle
dg
e w
ork)
.
6.
Lead
ersh
ip, p
arti
cula
rly t
ran
sfor
mat
ion
al a
nd
em
pow
erin
g le
ader
ship
, im
pro
ves
team
per
form
ance
.
1. Р
= .2
2
2. Р
= -.
04
(gen
eral
) Р
= .0
4 (p
roje
ct te
ams)
Р
= -.
07
(pro
du
ctio
n te
ams)
Р
= -.
03
(man
agem
ent
te
ams)
3. Р
= .0
4
4. Р
= .1
6
5. A
uto
nom
y:
Р =
.25
(gen
eral
) Р
= .3
6 (p
hys
ical
wor
k)
Р =
.26
(kn
owle
dg
e w
ork)
in
tra-
team
coo
rdin
atio
n:
Р =
.25
(gen
eral
) Р
= .1
2 (p
hys
ical
wor
k)
Р =
.29
(kn
owle
dg
e w
ork)
6. Р
= .2
6 (t
ran
sfor
mat
ion
al
lead
ersh
ip)
Р =
.33
(em
pow
erin
g le
ader
ship
)
Des
ign
of
the
incl
ud
ed
stu
die
s n
ot
rep
orte
d
C
52
Svya
nte
c,
1999
met
a an
alys
is
of b
efor
e-af
ter
stu
die
s an
d
con
trol
led
b
efor
e-af
ter
stu
die
s
n =
11 s
tud
ies
Em
plo
yees
in
bu
sin
ess
or
gov
ern
men
t
sett
ing
s
Mod
erat
ors
of te
am b
uild
ing
eff
ect
on te
am p
rod
uct
ivit
y:
Fact
ors
bef
ore
team
bu
ildin
g
1. Th
e in
itia
tor
of t
he
team
bu
ildin
g: w
hen
th
e in
itia
tor
is e
xter
nal
(vs.
in
tern
al to
th
e te
am),
the
pos
itiv
e ef
fect
is s
tron
ger
.
2.
The
rati
onal
e fo
r th
e te
amb
uild
ing
: wh
en t
he
acti
on is
cor
rect
ive
(vs.
p
reve
nti
ve),
the
effe
ct is
str
ong
er.
3.
The
exp
ecta
tion
s fr
om te
am b
uild
ing
ben
efits
: wh
en t
he
exp
ecta
tion
is
to c
han
ge
bot
h, t
he
effe
ct is
str
ong
est,
follo
wed
by
exp
ecta
tion
s to
im
pro
ve p
erfo
rman
ce, a
nd
last
ly b
y ex
pec
tati
ons
to im
pro
ve a
ttit
ud
es
4.
The
invo
lvem
ent
of t
he
gro
up
in p
lan
nin
g t
he
team
bu
ildin
g: w
hen
th
e g
rou
p is
invo
lved
, th
e ef
fect
is w
eake
r th
an w
hen
it is
n’t
invo
lved
.
Fact
ors
du
rin
g t
eam
bu
ildin
g
5.
Focu
s of
team
bu
ildin
g: a
mix
ed fo
cus
on g
oal s
etti
ng
an
d in
terp
erso
nal
re
lati
ons
has
th
e h
igh
est
effe
ct, f
ollo
wed
by
only
goa
l set
tin
g, f
ollo
wed
b
y on
ly in
terp
erso
nal
rela
tion
ship
s.Th
e p
rese
nce
of o
ther
inte
rven
tion
s to
get
her
wit
h te
am b
uild
ing
: in
th
e p
rese
nce
of o
ther
inte
rven
tion
s, t
he
effe
ct o
f tea
m b
uild
ing
is s
tron
ger
th
an w
hen
it is
alo
ne.
6.
Wh
o m
anag
ed t
he
inte
rven
tion
(s):
for
team
bu
ildin
gs
wit
h a
n e
xter
nal
an
d a
n in
tern
al c
onsu
ltan
t th
e ef
fect
is s
tron
ges
t, fo
llow
ed b
y on
ly
exte
rnal
con
sult
ant,
follo
wed
last
ly b
y on
ly in
tern
al c
onsu
ltan
t.
7.
The
focu
s of
th
e te
amb
uild
ing
: an
intr
agro
up
focu
s h
as a
str
ong
er e
ffect
th
an a
n in
div
idu
al fo
cus.
Org
anis
atio
nal
su
pp
ort
fact
ors
8.
Sup
ervi
sory
su
pp
ort
for
the
team
bu
ildin
g: w
hen
su
pp
ort
is p
rese
nt,
the
effe
ct is
str
ong
er.
9.
Sup
por
t fo
r ch
ang
e ef
fort
s: w
hen
su
pp
ort
from
hig
her
leve
ls is
pre
sen
t, th
e ef
fect
is s
tron
ges
t, fo
llow
ed b
y or
gan
isat
ion
s w
ith
no
evid
ence
for
sup
por
t, an
d fo
llow
ed la
stly
by
sup
por
t fr
om d
iffer
ent
leve
ls.
10.
Org
anis
atio
nal
ch
arac
teris
tics
.
11.
Size
of o
rgan
isat
ion
: sm
all o
rgan
isat
ion
s h
ave
the
stro
ng
est
effe
ct,
follo
wed
by
larg
e or
gan
isat
ion
s, fo
llow
ed b
y m
ediu
m o
nes
.
12.
Typ
e of
org
anis
atio
n: i
n in
du
stria
l/man
ufa
ctu
ring
org
anis
atio
ns,
th
e ef
fect
is
str
ong
er t
han
in g
over
nm
ent
org
anis
atio
ns.
13.
Team
’s re
spon
sib
ility
for
own
per
form
ance
: wh
en t
he
team
is s
olel
y re
spon
sib
le fo
r ow
n p
erfo
rman
ce, t
he
effe
ct is
str
ong
er t
han
wh
en t
he
team
dep
end
s on
oth
er e
lem
ents
in t
he
org
anis
atio
n.
Man
agem
ent
styl
e: in
team
s w
ith
a p
arti
cip
ativ
e m
anag
emen
t st
yle,
th
e ef
fect
was
str
ong
er t
han
in te
ams
wit
h a
n a
uto
crat
ic m
anag
emen
t st
yle.
1. In
tern
al in
itia
tor:
d=.
43; e
xter
nal
in
itia
tor:
d=.
78
2.
Pre
ven
tive
act
ion
: d=.
69;
corr
ecti
ve a
ctio
n: d
=.86
)
3.
Exp
ect
to c
han
ge
atti
tud
es &
p
erfo
rman
ce: d
=.86
; exp
ect
to
chan
ge
per
form
ance
: d=.
79;
exp
ect
to c
han
ge
atti
tud
es:
d=.
23
4.
Gro
up
invo
lved
: d=.
50; g
rou
p n
ot
invo
lved
: d=1
.07
5.
Inte
rper
son
al fo
cus:
d=.
58; g
oal
sett
ing
focu
s: d
=.62
; mix
ed
focu
s:
d=.
79
6.
On
ly te
amb
uild
ing
: d
=.53
; oth
er in
terv
enti
ons:
d=.
82
7.
On
ly in
tern
al c
onsu
ltan
t: d
=.35
; on
ly e
xter
nal
con
sult
ant:
d=.
74;
bot
h in
tern
al a
nd
ext
ern
al:
d=1
.75
8.
Ind
ivid
ual
focu
s: d
=.4
8;
intr
agro
up
focu
s: d
=.79
9.
Sup
por
t p
rese
nt:
d=1
.02;
su
pp
ort
mis
sin
g: d
=.49
10.
Sup
por
t fr
om h
igh
er le
vels
: d
=.90
; no
ev
iden
ce o
f su
pp
ort:
d=.
64;
sup
por
t fr
om d
iffer
ent
leve
ls:
d=.
50
11.
(Sm
all o
rgan
isat
ion
s: d
=.80
; m
ediu
m o
rg: d
=.43
; lar
ge
org
: d
=.56
)
12.
(Ind
ust
rial/m
an.
d=.
89; g
over
nm
ent:
d=.
21)
13.
Res
pon
sib
le fo
r p
erfo
rman
ce:
d=.
92;
inte
rdep
end
ent:
d=.
76
Par
tici
pat
ive
styl
e:
d=2
.62;
au
tocr
atic
: d=.
18
larg
e n
um
ber
of
rela
tion
ship
s te
sted
A
53
Svya
nte
c,
1999
met
a an
alys
is
of b
efor
e-af
ter
stu
die
s an
d
con
trol
led
b
efor
e-af
ter
stu
die
s
n =
11 s
tud
ies
Em
plo
yees
in
bu
sin
ess
or
gov
ern
men
t
sett
ing
s
Mod
erat
ors
of te
am b
uild
ing
eff
ect
on te
am p
rod
uct
ivit
y:
Fact
ors
bef
ore
team
bu
ildin
g
1. Th
e in
itia
tor
of t
he
team
bu
ildin
g: w
hen
th
e in
itia
tor
is e
xter
nal
(vs.
in
tern
al to
th
e te
am),
the
pos
itiv
e ef
fect
is s
tron
ger
.
2.
The
rati
onal
e fo
r th
e te
amb
uild
ing
: wh
en t
he
acti
on is
cor
rect
ive
(vs.
p
reve
nti
ve),
the
effe
ct is
str
ong
er.
3.
The
exp
ecta
tion
s fr
om te
am b
uild
ing
ben
efits
: wh
en t
he
exp
ecta
tion
is
to c
han
ge
bot
h, t
he
effe
ct is
str
ong
est,
follo
wed
by
exp
ecta
tion
s to
im
pro
ve p
erfo
rman
ce, a
nd
last
ly b
y ex
pec
tati
ons
to im
pro
ve a
ttit
ud
es
4.
The
invo
lvem
ent
of t
he
gro
up
in p
lan
nin
g t
he
team
bu
ildin
g: w
hen
th
e g
rou
p is
invo
lved
, th
e ef
fect
is w
eake
r th
an w
hen
it is
n’t
invo
lved
.
Fact
ors
du
rin
g t
eam
bu
ildin
g
5.
Focu
s of
team
bu
ildin
g: a
mix
ed fo
cus
on g
oal s
etti
ng
an
d in
terp
erso
nal
re
lati
ons
has
th
e h
igh
est
effe
ct, f
ollo
wed
by
only
goa
l set
tin
g, f
ollo
wed
b
y on
ly in
terp
erso
nal
rela
tion
ship
s.Th
e p
rese
nce
of o
ther
inte
rven
tion
s to
get
her
wit
h te
am b
uild
ing
: in
th
e p
rese
nce
of o
ther
inte
rven
tion
s, t
he
effe
ct o
f tea
m b
uild
ing
is s
tron
ger
th
an w
hen
it is
alo
ne.
6.
Wh
o m
anag
ed t
he
inte
rven
tion
(s):
for
team
bu
ildin
gs
wit
h a
n e
xter
nal
an
d a
n in
tern
al c
onsu
ltan
t th
e ef
fect
is s
tron
ges
t, fo
llow
ed b
y on
ly
exte
rnal
con
sult
ant,
follo
wed
last
ly b
y on
ly in
tern
al c
onsu
ltan
t.
7.
The
focu
s of
th
e te
amb
uild
ing
: an
intr
agro
up
focu
s h
as a
str
ong
er e
ffect
th
an a
n in
div
idu
al fo
cus.
Org
anis
atio
nal
su
pp
ort
fact
ors
8.
Sup
ervi
sory
su
pp
ort
for
the
team
bu
ildin
g: w
hen
su
pp
ort
is p
rese
nt,
the
effe
ct is
str
ong
er.
9.
Sup
por
t fo
r ch
ang
e ef
fort
s: w
hen
su
pp
ort
from
hig
her
leve
ls is
pre
sen
t, th
e ef
fect
is s
tron
ges
t, fo
llow
ed b
y or
gan
isat
ion
s w
ith
no
evid
ence
for
sup
por
t, an
d fo
llow
ed la
stly
by
sup
por
t fr
om d
iffer
ent
leve
ls.
10.
Org
anis
atio
nal
ch
arac
teris
tics
.
11.
Size
of o
rgan
isat
ion
: sm
all o
rgan
isat
ion
s h
ave
the
stro
ng
est
effe
ct,
follo
wed
by
larg
e or
gan
isat
ion
s, fo
llow
ed b
y m
ediu
m o
nes
.
12.
Typ
e of
org
anis
atio
n: i
n in
du
stria
l/man
ufa
ctu
ring
org
anis
atio
ns,
th
e ef
fect
is
str
ong
er t
han
in g
over
nm
ent
org
anis
atio
ns.
13.
Team
’s re
spon
sib
ility
for
own
per
form
ance
: wh
en t
he
team
is s
olel
y re
spon
sib
le fo
r ow
n p
erfo
rman
ce, t
he
effe
ct is
str
ong
er t
han
wh
en t
he
team
dep
end
s on
oth
er e
lem
ents
in t
he
org
anis
atio
n.
Man
agem
ent
styl
e: in
team
s w
ith
a p
arti
cip
ativ
e m
anag
emen
t st
yle,
th
e ef
fect
was
str
ong
er t
han
in te
ams
wit
h a
n a
uto
crat
ic m
anag
emen
t st
yle.
1. In
tern
al in
itia
tor:
d=.
43; e
xter
nal
in
itia
tor:
d=.
78
2.
Pre
ven
tive
act
ion
: d=.
69;
corr
ecti
ve a
ctio
n: d
=.86
)
3.
Exp
ect
to c
han
ge
atti
tud
es &
p
erfo
rman
ce: d
=.86
; exp
ect
to
chan
ge
per
form
ance
: d=.
79;
exp
ect
to c
han
ge
atti
tud
es:
d=.
23
4.
Gro
up
invo
lved
: d=.
50; g
rou
p n
ot
invo
lved
: d=1
.07
5.
Inte
rper
son
al fo
cus:
d=.
58; g
oal
sett
ing
focu
s: d
=.62
; mix
ed
focu
s:
d=.
79
6.
On
ly te
amb
uild
ing
: d
=.53
; oth
er in
terv
enti
ons:
d=.
82
7.
On
ly in
tern
al c
onsu
ltan
t: d
=.35
; on
ly e
xter
nal
con
sult
ant:
d=.
74;
bot
h in
tern
al a
nd
ext
ern
al:
d=1
.75
8.
Ind
ivid
ual
focu
s: d
=.4
8;
intr
agro
up
focu
s: d
=.79
9.
Sup
por
t p
rese
nt:
d=1
.02;
su
pp
ort
mis
sin
g: d
=.49
10.
Sup
por
t fr
om h
igh
er le
vels
: d
=.90
; no
ev
iden
ce o
f su
pp
ort:
d=.
64;
sup
por
t fr
om d
iffer
ent
leve
ls:
d=.
50
11.
(Sm
all o
rgan
isat
ion
s: d
=.80
; m
ediu
m o
rg: d
=.43
; lar
ge
org
: d
=.56
)
12.
(Ind
ust
rial/m
an.
d=.
89; g
over
nm
ent:
d=.
21)
13.
Res
pon
sib
le fo
r p
erfo
rman
ce:
d=.
92;
inte
rdep
end
ent:
d=.
76
Par
tici
pat
ive
styl
e:
d=2
.62;
au
tocr
atic
: d=.
18
larg
e n
um
ber
of
rela
tion
ship
s te
sted
A
Tan
gh
e,
2010
Stu
dy
1 C
ross
-se
ctio
nal
N
= 7
1 tea
ms
Stu
dy
2 Sc
enar
io
exp
erim
ent
2x2
des
ign
N
= 12
1
1. E
mp
loye
es,
mai
nly
ser
vice
or
gan
isat
ion
s
2. S
tud
ents
Team
iden
tifi
cati
on le
ads
gro
up
mem
ber
s to
aff
ecti
vely
con
verg
e to
th
eir
fello
w g
rou
p m
emb
ers
and
th
at t
his
aff
ecti
ve c
onve
rgen
ce, i
n tu
rn, e
xpla
ins
sub
seq
uen
t te
am-o
rien
ted
att
itu
des
:
1. Th
e h
igh
er t
he
gro
up
iden
tifi
cati
on is
, th
e st
ron
ger
th
e af
fect
ive
con
verg
ence
am
ong
team
mem
ber
s.
2.
Pos
itiv
e g
rou
p a
ffect
ive
(PA
) ton
e is
pos
itiv
ely
asso
ciat
ed w
ith
team
ef
fect
iven
ess
and
th
is e
ffect
will
be
stro
ng
er fo
r h
igh
er le
vels
of g
rou
p
iden
tifi
cati
on.
3.
Neg
ativ
e g
rou
p a
ffect
ive
(NA
) ton
e is
not
neg
ativ
ely
asso
ciat
ed w
ith
team
ef
fect
iven
ess
and
th
is e
ffect
will
be
stro
ng
er fo
r h
igh
er le
vels
of g
rou
p
iden
tifi
cati
on.
See
arti
cle
Mea
sure
g
rou
p a
ffec
t st
ud
y 2
som
ewh
at
un
clea
r
1) D
2) B
Tan
nen
bau
m,
2013
met
a-an
alys
is
of b
efor
e-an
d-
afte
r st
ud
ies
n =
31
stu
die
s, 2
,136
par
tici
pan
ts
vari
ous
Inte
rven
tion
: deb
rief
s. D
ebri
efs
lead
ind
ivid
ual
s or
team
s th
rou
gh
a s
erie
s of
q
ues
tion
s th
at a
llow
par
tici
pan
ts to
refle
ct o
n a
rece
nt
exp
erie
nce
, con
stru
ct
thei
r ow
n m
ean
ing
from
th
eir
acti
ons,
an
d u
nco
ver
less
ons
lear
ned
in a
non
-p
un
itiv
e en
viro
nm
ent.
A d
ebri
ef h
as fo
ur
elem
ents
: (1)
par
tici
pan
ts re
flect
on
sp
ecifi
c ev
ents
; (2)
par
tici
pan
ts a
re a
ctiv
ely
invo
lved
in s
elf-
dis
cove
ry; (
3)
the
envi
ron
men
t is
non
jud
gm
enta
l an
d fo
cuse
d o
n le
arn
ing
; (4)
bot
h t
he
par
tici
pan
ts a
nd
at
leas
t on
e ot
her
ext
ern
al s
ourc
e g
ive
inp
ut
reg
ard
ing
th
e ev
ents
un
der
revi
ew.
1. D
ebrie
fs h
ave
a m
ediu
m-t
o-la
rge
pos
itiv
e ef
fect
on
team
per
form
ance
.
2.
Deb
riefs
focu
sed
on
team
-leve
l im
pro
vem
ent
hav
e a
larg
er e
ffect
on
te
am p
erfo
rman
ce t
han
on
ind
ivid
ual
per
form
ance
.
Mod
erat
ors
of t
he
effe
ct o
f deb
rief
s on
team
per
form
ance
:
3.
Faci
litat
ion
in d
ebrie
fs: f
acili
tate
d d
ebrie
fs h
ave
a g
reat
er e
ffect
on
team
p
erfo
rman
ce t
han
non
-fac
ilita
ted
deb
riefs
.
4.
The
deg
ree
of s
tru
ctu
re o
f deb
riefs
: hig
hly
str
uct
ure
d d
ebrie
fs h
ave
a st
ron
ger
effe
ct t
han
mod
erat
ely
stru
ctu
red
deb
riefs
.
d=.
67
team
per
form
ance
: d=1
.2; i
nd
ivid
ual
p
erfo
rman
ce: d
=.41
faci
litat
ed d
ebri
efs:
d=.
75; n
on-
faci
litat
ed d
ebri
efs:
d=.
25
hig
hly
str
uct
ure
d d
ebri
efs:
d=.
69;
mod
erat
ely-
stru
ctu
red
deb
rief
s:
d=.
54
no
seri
ous
wea
knes
ses
B
54
Tekl
eab
, 20
16
lon
git
ud
inal
st
ud
y
n =
227
(4
5 te
ams)
emp
loye
es
pu
rsu
ing
a
gra
du
ate
deg
ree
at a
larg
e M
idw
este
rn
un
iver
sity
.
1. Th
ere
is N
O c
urv
e lin
ear
rela
tion
ship
bet
wee
n fu
nct
ion
al d
iver
sity
an
d
team
coh
esio
n.
2.
Beh
avio
ura
l in
teg
rati
on p
osit
ivel
y in
fluen
ces
team
coh
esio
n.
3.
A h
igh
leve
l of b
ehav
iou
ral i
nte
gra
tion
att
enu
ates
th
e n
egat
ive
imp
act
of
fun
ctio
nal
div
ersi
ty o
n te
am c
ohes
ion
su
ch t
hat
th
e re
lati
onsh
ip w
ill b
e n
egat
ive
only
un
der
a lo
w le
vel o
f beh
avio
ura
l in
teg
rati
on.
4.
The
rela
tion
ship
bet
wee
n te
am c
ohes
ion
an
d o
bje
ctiv
e te
am
per
form
ance
is m
edia
ted
by
team
lear
nin
g.
Not
e: B
ehav
iou
ral i
nte
gra
tion
is a
met
a-co
nst
ruct
, wh
ich
incl
ud
es t
he
team
’s
info
rmat
ion
exc
han
ge,
col
lab
orat
ive
beh
avio
r, an
d jo
int
dec
isio
n m
akin
g.
1. ß
= −
.12 n
s
2.
ß =
.73
3.
ß =
−.4
8
4.
coh
> le
arn
ing
ß =
.45
lear
nin
g >
per
form
ance
ß
= .6
6
Con
cern
s a
cap
ston
e si
mu
lati
on
C
Turn
er,
2014
Met
a-an
alys
is,
incl
ud
es R
CTs
k
= 18
(7
68 te
ams,
n
= 13
,491
)
vari
ous
The
prim
ary
focu
s of
th
is m
eta-
anal
ysis
is to
iden
tify
wh
ich
mea
sure
of t
he
six
team
cog
nit
ion
con
stru
cts
(sh
ared
men
tal m
odel
s, S
MM
; tea
m m
enta
l m
odel
s, T
MM
; in
form
atio
n s
har
ing
, IS;
tra
nsa
ctiv
e m
emor
y sy
stem
s, T
MS;
co
gn
itiv
e co
nse
nsu
s, C
C; g
rou
p le
arn
ing
, GL)
pro
du
ced
th
e b
est
per
form
ance
ou
tcom
e re
sult
s.
The
one
team
cog
nit
ion
con
stru
ct t
hat
sto
od o
ut
was
th
at o
f IS,
wit
h
stat
isti
cal fi
nd
ing
s g
reat
er t
han
th
e co
nst
ruct
s of
TM
M, G
L, a
nd
TM
S. T
he
two
shar
ed c
ogn
itiv
e co
nst
ruct
s th
at w
ere
not
sta
tist
ical
ly d
iffe
ren
t fr
om IS
w
ere
SMM
an
d C
C; n
eith
er o
f th
ese
con
stru
cts
was
fou
nd
to b
e si
gn
ifica
ntl
y d
iffe
ren
t fr
om T
MM
, GL,
or
TMS.
Not
e: s
ee d
iscu
ssio
n &
con
clu
sion
.
SMM
r =
.39
TMM
r =
.19
ns
IS r
= .5
1
TMS
r =
.30
CC
= .4
2
GL
= .15
ns
no
seri
ous
limit
atio
ns
AA
Van
der
Veg
t, 20
10
Lon
git
ud
inal
st
ud
y
N =
47
team
s (a
vera
ge
team
si
ze =
10)
Tru
ck
man
ufa
ctu
ring
1. Tu
rnov
er is
neg
ativ
ely
asso
ciat
ed w
ith
soc
ial i
nte
gra
tion
wit
hin
sel
f-m
anag
ing
wor
k te
ams.
2.
Turn
over
is n
egat
ivel
y as
soci
ated
wit
h te
am le
arn
ing
beh
avio
r w
ith
in s
elf-
m
anag
ing
wor
k te
ams
bu
t is
not
inve
rse
U-s
hap
ed.
3.
Team
tu
rnov
er is
neg
ativ
ely
asso
ciat
ed w
ith
tas
k fle
xib
ility
in s
elf-
man
agin
g w
ork
team
s.
4.
Soci
al in
teg
rati
on d
oes
not
par
tial
ly m
edia
te a
gen
eral
ly n
egat
ive
rela
tion
ship
bet
wee
n te
am t
urn
over
an
d e
ffect
iven
ess
in s
elf-
man
agin
g
wor
k te
ams.
5.
Team
lear
nin
g b
ehav
ior
par
tial
ly m
edia
tes
a g
ener
ally
neg
ativ
e re
lati
onsh
ip b
etw
een
team
tu
rnov
er a
nd
effe
ctiv
enes
s in
sel
f-m
anag
ing
w
ork
team
s.
6.
Task
flex
ibili
ty p
arti
ally
med
iate
s a
gen
eral
ly n
egat
ive
rela
tion
ship
b
etw
een
team
tu
rnov
er a
nd
effe
ctiv
enes
s in
sel
f-m
anag
ing
wor
k te
ams.
7.
Team
tu
rnov
er is
neg
ativ
ely
rela
te to
team
effe
ctiv
enes
s.
1. ß
= -.
32
2.
ß =
-.41
3.
ß =
-.37
4.
ß =
-.36
No
full
use
of
val
idat
ed
scal
esC
55
Tekl
eab
, 20
16
lon
git
ud
inal
st
ud
y
n =
227
(4
5 te
ams)
emp
loye
es
pu
rsu
ing
a
gra
du
ate
deg
ree
at a
larg
e M
idw
este
rn
un
iver
sity
.
1. Th
ere
is N
O c
urv
e lin
ear
rela
tion
ship
bet
wee
n fu
nct
ion
al d
iver
sity
an
d
team
coh
esio
n.
2.
Beh
avio
ura
l in
teg
rati
on p
osit
ivel
y in
fluen
ces
team
coh
esio
n.
3.
A h
igh
leve
l of b
ehav
iou
ral i
nte
gra
tion
att
enu
ates
th
e n
egat
ive
imp
act
of
fun
ctio
nal
div
ersi
ty o
n te
am c
ohes
ion
su
ch t
hat
th
e re
lati
onsh
ip w
ill b
e n
egat
ive
only
un
der
a lo
w le
vel o
f beh
avio
ura
l in
teg
rati
on.
4.
The
rela
tion
ship
bet
wee
n te
am c
ohes
ion
an
d o
bje
ctiv
e te
am
per
form
ance
is m
edia
ted
by
team
lear
nin
g.
Not
e: B
ehav
iou
ral i
nte
gra
tion
is a
met
a-co
nst
ruct
, wh
ich
incl
ud
es t
he
team
’s
info
rmat
ion
exc
han
ge,
col
lab
orat
ive
beh
avio
r, an
d jo
int
dec
isio
n m
akin
g.
1. ß
= −
.12 n
s
2.
ß =
.73
3.
ß =
−.4
8
4.
coh
> le
arn
ing
ß =
.45
lear
nin
g >
per
form
ance
ß
= .6
6
Con
cern
s a
cap
ston
e si
mu
lati
on
C
Turn
er,
2014
Met
a-an
alys
is,
incl
ud
es R
CTs
k
= 18
(7
68 te
ams,
n
= 13
,491
)
vari
ous
The
prim
ary
focu
s of
th
is m
eta-
anal
ysis
is to
iden
tify
wh
ich
mea
sure
of t
he
six
team
cog
nit
ion
con
stru
cts
(sh
ared
men
tal m
odel
s, S
MM
; tea
m m
enta
l m
odel
s, T
MM
; in
form
atio
n s
har
ing
, IS;
tra
nsa
ctiv
e m
emor
y sy
stem
s, T
MS;
co
gn
itiv
e co
nse
nsu
s, C
C; g
rou
p le
arn
ing
, GL)
pro
du
ced
th
e b
est
per
form
ance
ou
tcom
e re
sult
s.
The
one
team
cog
nit
ion
con
stru
ct t
hat
sto
od o
ut
was
th
at o
f IS,
wit
h
stat
isti
cal fi
nd
ing
s g
reat
er t
han
th
e co
nst
ruct
s of
TM
M, G
L, a
nd
TM
S. T
he
two
shar
ed c
ogn
itiv
e co
nst
ruct
s th
at w
ere
not
sta
tist
ical
ly d
iffe
ren
t fr
om IS
w
ere
SMM
an
d C
C; n
eith
er o
f th
ese
con
stru
cts
was
fou
nd
to b
e si
gn
ifica
ntl
y d
iffe
ren
t fr
om T
MM
, GL,
or
TMS.
Not
e: s
ee d
iscu
ssio
n &
con
clu
sion
.
SMM
r =
.39
TMM
r =
.19
ns
IS r
= .5
1
TMS
r =
.30
CC
= .4
2
GL
= .15
ns
no
seri
ous
limit
atio
ns
AA
Van
der
Veg
t, 20
10
Lon
git
ud
inal
st
ud
y
N =
47
team
s (a
vera
ge
team
si
ze =
10)
Tru
ck
man
ufa
ctu
ring
1. Tu
rnov
er is
neg
ativ
ely
asso
ciat
ed w
ith
soc
ial i
nte
gra
tion
wit
hin
sel
f-m
anag
ing
wor
k te
ams.
2.
Turn
over
is n
egat
ivel
y as
soci
ated
wit
h te
am le
arn
ing
beh
avio
r w
ith
in s
elf-
m
anag
ing
wor
k te
ams
bu
t is
not
inve
rse
U-s
hap
ed.
3.
Team
tu
rnov
er is
neg
ativ
ely
asso
ciat
ed w
ith
tas
k fle
xib
ility
in s
elf-
man
agin
g w
ork
team
s.
4.
Soci
al in
teg
rati
on d
oes
not
par
tial
ly m
edia
te a
gen
eral
ly n
egat
ive
rela
tion
ship
bet
wee
n te
am t
urn
over
an
d e
ffect
iven
ess
in s
elf-
man
agin
g
wor
k te
ams.
5.
Team
lear
nin
g b
ehav
ior
par
tial
ly m
edia
tes
a g
ener
ally
neg
ativ
e re
lati
onsh
ip b
etw
een
team
tu
rnov
er a
nd
effe
ctiv
enes
s in
sel
f-m
anag
ing
w
ork
team
s.
6.
Task
flex
ibili
ty p
arti
ally
med
iate
s a
gen
eral
ly n
egat
ive
rela
tion
ship
b
etw
een
team
tu
rnov
er a
nd
effe
ctiv
enes
s in
sel
f-m
anag
ing
wor
k te
ams.
7.
Team
tu
rnov
er is
neg
ativ
ely
rela
te to
team
effe
ctiv
enes
s.
1. ß
= -.
32
2.
ß =
-.41
3.
ß =
-.37
4.
ß =
-.36
No
full
use
of
val
idat
ed
scal
esC
Wag
ner
, 20
12
Ran
dom
ised
ex
per
imen
t N
= 2
06
(82
team
s)
Stu
den
ts, U
S
1. Fo
r w
ork
con
sist
ing
of a
com
bin
atio
n o
f in
div
idu
aliz
ed a
nd
in
terd
epen
den
t ta
sks,
team
mem
ber
per
form
ance
is h
igh
er fo
r in
div
idu
als
pos
sess
ing
a m
ix o
f in
div
idu
alis
m a
nd
col
lect
ivis
m t
han
for
ind
ivid
ual
s w
ho
lack
sim
ilar
het
erog
enei
ty.
2.
The
rela
tion
ship
bet
wee
n in
trap
erso
nal
het
erog
enei
ty in
ind
ivid
ual
ism
–co
llect
ivis
m a
nd
per
form
ance
pre
dic
ted
in H
1 is
mod
erat
ed b
y st
ruct
ura
l in
terd
epen
den
ce s
uch
th
at t
he
effe
ct is
str
ong
er u
nd
er c
ond
itio
ns
of lo
ose
stru
ctu
ral i
nte
rdep
end
ence
th
an u
nd
er c
ond
itio
ns
of t
igh
t st
ruct
ura
l in
terd
epen
den
ce.
See
arti
cle
no
seri
ous
wea
knes
ses
A
Wan
g,
2019
m
eta-
anal
ysis
k
= 47
(2
,832
team
s)va
riou
s
1a) S
urf
ace‐
leve
l div
ersi
ty in
cu
ltu
rally
div
erse
team
s is
not
rela
ted
to te
am
crea
tivi
ty/in
nov
atio
n, (
2a) i
rres
pec
tive
wh
eth
er t
he
team
is c
ollo
cate
d o
r n
on-c
ollo
cate
d, l
evel
of t
ask
ind
epen
den
cy (3
a), t
ask
com
ple
xity
(4a)
, or
task
in
telle
ctiv
enes
s (j
ud
gm
enta
l vs
inte
llect
ive
task
s) (5
a).
1b)…
wh
erea
s d
eep
‐lev
el d
iver
sity
in c
ult
ura
lly d
iver
se te
ams
is s
omew
hat
p
osit
ivel
y re
late
d to
team
cre
ativ
ity/
inn
ovat
ion
. 2) T
his
rela
tion
ship
is
stro
ng
er fo
r co
lloca
ted
team
s th
an fo
r n
on-c
ollo
cate
d te
ams,
3) s
tron
ger
for
inte
rdep
end
ent
task
s th
an fo
r in
dep
end
ent
task
s, 5
)str
ong
er fo
r in
telle
ctiv
e ta
sks
than
for
jud
gm
enta
l tas
ks. T
asks
com
ple
xity
an
d in
telle
ctiv
enes
s d
id n
ot
mod
erat
e th
e ef
fect
.
All
ES
very
sm
all
1. a)
ns
and
mos
tly
clos
e to
zer
o b
) Р =
.16?
2.
col:
Р =
.18
non
col
: Р =
.02
ns
3.
inte
r: Р
= .1
9 in
dep
: Р =
-.10
ns
Des
ign
of
incl
ud
ed
stu
die
s n
ot
rep
orte
d
Eff
ect
size
s re
por
ted
in
text
do
not
co
rres
pon
d
wit
h t
he
ES
in t
able
2
C
Web
ber
, 20
01
met
a-an
alys
is
k =
24
(45
corr
elat
ion
s)
Res
ult
s sh
owed
th
at jo
b-r
elat
ed d
iver
sity
has
NO
rela
tion
ship
wit
h c
ohes
ion
or
per
form
ance
.cl
ose
to z
ero
and
ns
Des
ign
of
incl
ud
ed
stu
die
s n
ot
rep
orte
d
C
56
Web
ber
, 20
08
Lon
git
ud
inal
st
ud
y
n =
69
(9 te
ams)
Un
der
gra
du
ate
stu
den
ts,
Can
adia
n
Un
iver
sity
Not
e: d
irect
eff
ect
size
s in
corp
orat
ed in
De
Jon
g 2
016
.
1. E
arly
tru
st e
mer
ges
as
a on
e-d
imen
sion
al fa
ctor
ear
ly in
th
e lif
e sp
an
of a
team
.
2.
Cog
nit
ive
and
affe
ctiv
e tr
ust
em
erg
e as
sep
arat
e co
mp
onen
ts o
ver
tim
e.
3.
Fam
iliar
ity
wit
h te
am m
emb
ers
is p
osit
ivel
y re
late
d to
ear
ly t
rust
.
4.
Inte
ract
ion
fre
qu
ency
is N
OT
rela
ted
to (a
ffect
ive
or c
ogn
itiv
e) t
rust
.
5.
Mon
itorin
g b
ehav
iou
rs b
y te
am m
emb
ers
(eg
tra
ckin
g t
he
wor
k of
ot
her
s, c
reat
ing
bac
kup
pla
ns,
or
wor
kin
g a
rou
nd
team
mem
ber
s to
get
ta
sks
don
e) s
omew
hat
neg
ativ
ely
affe
cts
the
dev
elop
men
t of
cog
nit
ive
and
affe
ctiv
e tr
ust
in te
ams
(eve
n a
fter
con
trol
ling
for
fam
iliar
ity
and
ea
rly t
rust
).
3) R
2 =
.06,
ß =
.24
4) A
ff: R
2= .0
1, ß
= .0
9 C
ogn
: R2=
.00
, ß =
.03,
5) s
mal
lar
tifi
cial
se
ttin
gC
Wee
r, 20
16
lon
git
ud
inal
st
ud
y w
ith
3
mea
sure
men
t p
oin
ts o
ver
4
,5 y
ears
n =
71
4 te
ams
Em
plo
yees
of
a la
rge,
m
ult
inat
ion
al
tech
nol
ogy-
dri
ven
firm
.
Inte
rven
tion
: coa
chin
g b
y te
am le
ader
: fac
ilita
tive
coa
chin
g (p
rovi
din
g
gu
idan
ce b
y al
ign
ing
team
mem
ber
asp
irat
ion
s w
ith
org
anis
atio
nal
goa
ls
and
faci
litat
e th
e ac
hie
vem
ent
of b
oth
ind
ivid
ual
an
d te
am o
bje
ctiv
es) a
nd
p
ress
ure
-bas
ed c
oach
ing
(pro
vid
ing
dire
ctio
n b
y ap
ply
ing
ext
ensi
ve p
ress
ure
to
get
resu
lts.
Thes
e m
anag
ers
com
mu
nic
ate
exp
ecta
tion
s b
y b
ecom
ing
vis
ibly
up
set
and
co
mp
lain
ing
vig
orou
sly
if g
oals
are
not
met
, an
d m
ay c
hal
len
ge
emp
loye
es
to im
pro
ve b
y re
prim
and
ing
poo
r p
erfo
rman
ce a
nd
/or
pu
blic
ly c
riti
cizi
ng
m
ista
kes.
).
Team
lead
er’s
faci
litat
ive
coac
hin
g p
red
icts
team
eff
ecti
ven
ess
(as
rate
d b
y th
e te
am le
ader
).
Team
lead
er’s
pre
ssu
re-b
ased
coa
chin
g n
egat
ivel
y p
red
icts
team
ef
fect
iven
ess.
Low
eff
ect
size
s
r=.14
at
tim
e 1;
r=.0
at
tim
e 2;
r=.
02
and
ß=.
02
at t
ime
3.
r =-
.05
at t
ime
1; r
=-.11
at
tim
e 2;
r
=-.0
6 an
d ß
=-.2
4 a
t ti
me
3
no
seri
ous
wea
knes
ses
B
Wild
man
, 20
16
Syst
emat
ic
revi
ew o
f 31
cro
ss-
sect
ion
al
(28)
an
d
lon
git
ud
inal
(3
) stu
die
s.
Wor
k te
ams,
d
iffe
ren
t in
du
stri
es
Con
cern
ing
sp
ecifi
c te
am le
arn
ing
beh
avio
urs
, sh
arin
g, t
eam
refle
ctio
n, a
nd
te
am a
ctiv
ity
ten
d to
hav
e th
e st
ron
ges
t im
pac
t on
team
s’ e
ng
agem
ent
in
inn
ovat
ion
dev
elop
men
t. Le
arn
ing
an
d in
nov
atio
n d
evel
opm
ent
are
mu
tual
ly
dep
end
ent
asp
ects
of t
eam
wor
k an
d t
hat
fost
erin
g o
ne
asp
ect
will
als
o b
e b
enefi
cial
for
the
oth
er.
Un
clea
r
It s
eem
s th
at
the
auth
ors
pre
ten
d to
es
tim
ate
cau
sal
infe
ren
ce
bas
ing
on
co
rrel
atio
nal
stud
ies.
B
57
Wild
man
, 20
18
Cro
ss-
sect
ion
al
stu
dy
N =
117
team
s (5
93
resp
ond
ents
)
Voc
atio
nal
ed
uca
tors
in
voca
tion
al
colle
ges
Team
lear
nin
g b
ehav
iou
rs (T
LBs)
, esp
ecia
lly te
am re
flexi
vity
and
bou
nd
ary
span
nin
g, r
elat
e p
osit
ivel
y to
inn
ovat
ive
wor
k b
ehav
iou
r (IW
B).
Furt
her
mor
e,
team
str
uct
ure
, tas
k in
terd
epen
den
ce a
nd
gro
up
pot
ency
rela
te p
osit
ivel
y to
TL
Bs.
TLB
s ca
n b
e fo
ster
ed b
y es
tab
lish
ing
th
ese
team
lear
nin
g c
ond
itio
ns.
1. a)
Kn
owle
dg
e sh
arin
g is
not
rela
ted
to IW
B.
b) T
eam
refle
xivi
ty re
late
s p
osit
ivel
y to
IWB
. c)
Bou
nd
ary
span
nin
g re
late
s p
osit
ivel
y to
IWB
. d
) Sto
rag
e an
d re
trie
val i
s n
ot re
late
d to
IWB
.
2.
Team
str
uct
ure
rela
tes
pos
itiv
ely
to (a
) kn
owle
dg
e sh
arin
g, (
b) t
eam
re
flexi
vity
, (c)
bou
nd
ary
span
nin
g, (
d) s
tora
ge
and
retr
ieva
l.
3.
Task
inte
rdep
end
ence
rela
tes
pos
itiv
ely
to (a
) kn
owle
dg
e sh
arin
g, (
b)
team
refle
xivi
ty, (
c) b
oun
dar
y sp
ann
ing
. It
doe
s n
ot p
osit
ivel
y re
late
to (d
) st
orag
e an
d re
trie
val.
4.
Gro
up
pot
ency
rela
tes
pos
itiv
ely
to (b
) tea
m re
flexi
vity
, bu
t n
ot to
(a)
know
led
ge
shar
ing
, (c)
bou
nd
ary
span
nin
g, (
d) s
tora
ge
and
retr
ieva
l.
1a n
s 1b
ß =
0.5
4
1c ß
= 0
.61
1d n
s
2a ß
= 0
.73
2b ß
= 0
.44
2c
ß =
0.4
9 2d
ß =
0.2
9
3a ß
= 0
.24
3b
ß =
0.2
5 3c
ß =
0.2
6 3d
ns
4a
ns
4b
ß =
.28
4c
ns
4d
ns
no
seri
ous
limit
atio
ns
D
Zhou
, 20
15
met
a-an
alys
is
k =
31en
trep
ren
euri
al
team
s
1. N
o co
ncl
usi
on c
ould
be
dra
wn
reg
ard
ing
th
e re
lati
onsh
ip b
etw
een
ag
e d
iver
sity
an
d e
ntr
epre
neu
rial t
eam
per
form
ance
du
e to
inco
nsi
sten
t re
sear
ch re
sult
s.
2.
No
con
clu
sion
cou
ld b
e d
raw
n re
gar
din
g t
he
rela
tion
ship
bet
wee
n ra
ce
div
ersi
ty a
nd
en
trep
ren
euria
l tea
m p
erfo
rman
ce d
ue
to in
con
sist
ent
rese
arch
resu
lts.
3.
The
asso
ciat
ion
bet
wee
n e
du
cati
onal
-leve
l div
ersi
ty a
nd
en
trep
ren
euria
l te
am p
erfo
rman
ce re
mai
ns
inco
ncl
usi
ve.
4.
Ove
rall,
th
e u
sual
ly h
ypot
hes
ized
pos
itiv
e re
lati
onsh
ip b
etw
een
ed
uca
tion
al-b
ackg
rou
nd
div
ersi
ty a
nd
en
trep
ren
euria
l tea
m
per
form
ance
was
not
su
pp
orte
d b
y em
piri
cal s
tud
ies.
5.
Em
piri
cal fi
nd
ing
s re
gar
din
g re
lati
onsh
ips
bet
wee
n fu
nct
ion
al d
iver
sity
an
d e
ntr
epre
neu
rial t
eam
per
form
ance
may
be
des
crib
ed a
s d
iver
gen
t an
d in
con
sist
ent.
6.
Alt
hou
gh
info
rmat
ion
al d
iver
sity
has
oft
en b
een
ass
um
ed to
be
ben
efici
al fo
r fi
rm-le
vel a
s w
ell a
s te
am-le
vel e
ntr
epre
neu
rial t
eam
p
erfo
rman
ce, h
ere
agai
n e
mp
irica
l evi
den
ce is
inco
nsi
sten
t an
d t
her
efor
e in
con
clu
sive
.
ns
Des
ign
of
incl
ud
ed
stu
die
s n
ot
rep
orte
dC
58
Excluded studies
Author & year Reason for exclusion
Barczak, 2010
Cross-sectional study, partly focusses on EI, confirms findings from recent MAs.
Beal, 2003
Included in Chioccio 2009.
Budworth, 2011
Concerns the assessment of a specific training (employee selection methods).
Buljac, 2010
Literature review, narrative summary of findings (including of individual studies included).
Buvik, 2016
Cross-sectional study, small sample, Norwegian construction teams, confirms findings from recent MAs.
Carboni, 2013
Cross-sectional study, sales teams. Results indicated that individuals close to the core of a team outperformed more peripheral individuals, but only to the extent that teams were high-performing or had been together longer as a team.
Chen, 2018
Limited generalizability (undergraduate students in Singapore). Experiment comparing different methods of team formation: (1) random assignment; (2) self-selection; and (3) algorithm assignment designed to maximize skill complementarity. The study found that self-selection creates high-performing teams.
Cheng, 2016
Study with several weaknesses, artificial setting and artificial tasks, findings are inconclusive.
Chi, 2012
Paper cannot be retrieved.
Curseau, 2014
Concerns the effect of individual goals rather than group goals.
DeChurch, 2010
Incorporated in Mesmer-Magnus, 2017.
de Pillis, 2015
Concerns student teams in an educational setting.
Devine, 2001
Focusses only on differences between field and lab studies.
Ehrhardt, 2014
Cross-sectional study, mainly confirms findings from meta-analyses.
Eisele, 2013
Assessment of feedback sessions based on the Team Diagnostic Survey, lacks detailed information regarding the effect.
Eisele, 2015
The study doesn’t examine an intervention. It is focused only on the predictive validity of the Team Diagnostic Survey.
Espinosa, 2015
Concerns dyadic teams.
Farh, 2017
Concerns TMX, findings rather hard to apply: strong TMX produces obligations to utilize resources provided by one’s teammates, and these obligations enhance performance when (a) teammates provide resources of high quality or (b) the quality of resources available from individuals outside of the TMX relationship (i.e., the leader) are low, purportedly because TMX-based obligations protect individuals from over-utilizing low-quality resources from the leader.
Gaggioli, 2015
Longitudinal study, concerns social network indices and experience of flow.
59
Gilley, 2010
The study doesn’t examine an intervention. It is focused only on the managerial skills that predict the manager’s ability to facilitate and build teams.
Hasa, 2019
Limited generalizability (startups in India). The findings highlight how prior social connections, which are often a source of knowledge and influence, can limit new interactions and thus the ability of organisations to leverage peer effects to improve the performance of their members.
Harty, 2016
No estimate of effect sizes is given, only significance levels. Furthermore, data is analysed at individual level, and the groups might not be teams.
Huang, 2013
Cross-sectional study, mainly confirms findings from meta-analyses (Bachrach 2019, Mesmer-Magnus 2017).
Jarrett, 2016
The study is included in the McEwan (2017) meta-analysis, with another reference (Jarrett et al. 2012) – this is a dissertation that the article from 2016 was based on (the same sample & data).
Kaymak, 2011
The effect of positive past experiences working on group cohesion and group performance was only indirectly measured (SEM).
Kuipers, 2009
The study doesn’t examine an intervention. It examines whether three team processes predict.
Lee, 2013
Limited generalizability (operational service teams in Hong Kong en Macao. In this study, operational service teams’ attribute patterns and their associated performance levels were examined using a configuration approach.
Liu, 2011
Cross-sectional study, Taiwanese companies, confirms findings from recent MA’s.
Marques-Quinteiro, 2019
Cross-setional study, confitms findings from recent MA’s.
Matta, 2018
The meta-analysis doesn’t focus on an intervention, but on leader-member exchange (team leader’s behaviour).
McHaney, 2018
Off topic, study about whether groups with prior history of interaction outperform individuals in deception detection. Results indicated that groups which exhibited higher levels of relational links, that is, established groups, were more accurate in deception detection than ad hoc groups.
McNeese, 2017
Not relevant, laboratory-based study of collocated student teams undertaking information retrieval tasks.
Mell, 2014
Effect sizes are incorporated in Bachrach 2019. In addition, the central hypothesis ('Teams with a centralized TMS structure perform better than teams with a decentralized TMS structure when there is a disconnected distribution of interdependent task information, but not when there is a connected distribution of interdependent task information') is too detailed and too academic for this REA.
Meneghel, 2016
Cross-sectional study.
Mertins, 2015
Simulation game, no hypotheses, methods section rather unclear.
Moser, 2019
Cross-sectional study, confirms findings from recent MAs.
Muhlberger, 2015
The study doesn’t measure team effectiveness, but the effectiveness of several types of coaching (including group coaching) interventions for individual goal attainment.
Mullen, 1994
Included in Chioccio 2009.
Naidoo, 2011
(included in Matta, 2018) The study doesn’t focus on an intervention, but on leader-member exchange (team leader’s behaviour).
Nielsen, 2010
Concerns training of managers rather than teams.
60
Nielsen, 2012
The study doesn’t measure team effectiveness, but individual-level variables such as autonomy, social support, job satisfaction, and affective well-being.
Nielsen, 2017
Only marginally relevant, most of the findings are incorporated in recent MAs (eg McEwan, 2017).
Nisula, 2016
Cross-sectional study, focusses on individual creative self-efficacy and several other constructs measured at the individual level.
Nouri, 2013
Nice study, but unfortunately concerns dyads.
Oertel, 2015
Partly incorporated in Bachrach 2019. In addition, hypotheses tested mostly descriptive and fairly self-evident (eg “During the formation phase of teams, knowledge-based team learning behaviours (storing and retrieving task- and teamwork-relevant knowledge) are positively related to the emergence of transactive memory).
O’ Leary, 1994
Included in Kleingeld, 2011.
Pierro, 2015
Cross-sectional study on the relationship between need for cognitive closure and employee performance.
Rapp, 2007
Included in Delise 2010, McEwan 2017, and Salas 2008.
Revilla, 2012
Cross-sectional study, very small sample, confirms findings from recent MAs.
Rico, 2011
Examined the effects of person-focused organisational citizenship behaviours on the performance of teams characterized by different levels of virtuality and task interdependence.
Rodriguez-Sanchez, 2016
Longitudinal study, outcomes are incorporated in recent MAs.
Rousseau, 2013
Cross-sectional study on team coaching by team leaders, uses a self-constructed 5-item questionnaire that takes a very broad approach to coaching (eg: Our team leader points out the areas we need to improve; Our team leader suggests means to improve our performance).
Salanova, 2015
Off topic, study about collective efficacy and collective flow. No performance measure.
Salas, 1999
Rather old MA of cross-sectional studies, outcome is partly refuted by more recent, high quality MAs (eg Klein, 2009).
Shazi, 2015
Cross-sectional study, small sample, confirms findings from recent MAs.
Staats, 2012
Off topic. Study about ‘team scaling fallacy’. As team size increases, people increasingly underestimate the number of labor hours required to complete projects.
Sonnentag, 2010
Focusses on individual performance (findings suggest that team members can improve their individual performance when engaging in teamwork processes that are relevant for the team as a whole).
Stevens, 2003
Participants were all female and from sports teams.
Tanghe, 2010
Experimental lab study with students, very artificial. Concerns the effect of team members’ affective state on propensity to trust. People who are less trusting will show more cooperative behaviours when confronted with group members displaying high activation affective states than when confronted with group members displaying low activation affective states.
Tindale, 2012
Unclear framework, hypotheses and methods, limited generalizability.
Troster, 2014
Concerns only self-managed teams and the effect of nationality. Outcomes are potency’’ (the team’s confidence in its ability to perform)and its performance as rated by expert judges.
61
Unger-Aviram, 2015
Lab setting and artificial task (bridge planning task), focusses mainly on the effect of goal orientation on adaption to change.
van der Haar, 2015
Concerns on-scene-command teams that coordinate the interdisciplinary aid efforts of fire departments, the police, and disaster medicine in case of natural or man-made emergencies, such as floods, fire breakouts, or car accidents.
Van Mierlo, 2010
Lab setting and artificial task (tower building task), very academic, implications for practice somewhat unclear.
Vora, 2012
Longitudinal study, simulation with undergraduate students, effects unclear.
Woehr, 2013
Concerns an RCT with students in an artificial setting and an artificial task (building a replica of a real bridge, using 33 plastic pipes of three different sizes and 20 rubber bands). Note: Results indicated that value diversity among team members had NO significant impact on task performance.
Wu, 2016
Cross-sectional study, student population (undergraduates), confirms findings from recent MAs.
Xu, 2019
Longitudinal study, only marginally relevant to the REA question.
Yee-Young, 2015
Concerns whether social category (gender and age) and informational diversity (education and work experience) in work teams may affect a team’s perceived fit, which in turn may influence leader-rated group performance.
Zhang, 2015
Cross-sectional study of Chinese teams. The statistical technique (support vector machine) that is used to build the model is rather unclear.
Zhang, 2016
Population concerns Chinese employees and students. However, in China group members may differ from their American and European counterparts in terms of group diversity effects. In addition, most beta’s found were practically irrelevant.
Zhu, 2018
Longitudinal survey, confirms findings from recent MAs (Bachrach 2019, Mesmer-Magnus 2017).
62
REA Antecedents of workplace incivility a summary of research literature - January 2020 REA Antecedents of workplace incivility a summary of research literature - January 2020
The ACT Government acknowledges and thanks Novartis for allowing the content of their REA to be reproduced and redesigned by ACT Health.
Any enquiries in relation to the content of this REA should be directed to CEBMa through their website: www.cebma.org
A partnership between the ACT Government through the ACT public health system and the ANU Research School of Management.