Committee Conclusion: Research has identified a number of individual-differences attributes that are broadly predictive of success in a team environment. There has also been progress in identifying attributes that when aggregated across team members (e.g., mean level of cognitive ability, minimum agreeableness), are predictive of team effectiveness. More research is needed to expand and amplify this work in the context of potential utility in military accession. The committee concludes that the teamwork knowledge, skills, abilities, and other characteristics (KSAO) domain merits inclusion in a program of basic research with the long-term goal of improving the Army’s enlisted accession system.
The small unit has always been critical to an army’s success. The U.S. Army’s selection of soldiers for assignment into a particular team, squad, and platoon is the basis for much of the soldier’s military experience and achievement. There are thousands of military units that serve a wide variety of functions such as combat, medical, aviation, rescue, and support (Dyer et al., 1980). Furthermore, Essens and colleagues (2005) argue that more specialized units will be needed to meet new demands as the Army is tasked to add political and social objectives to more traditional military missions. Today’s soldiers are challenged to work in multinational coalitions, joint forces operations, and ad hoc teams with nonroutine tasks. This chapter examines current theory and research on teams, which the committee applies to the Army’s organizational level of the small unit, and proposes future research directions that are likely to enhance the unit’s collective capacity to perform.
This chapter uses Kozlowski and Ilgen’s (2006, p. 79) definition of a team:
A team can be defined as (a) two or more individuals who (b) socially interact (face-to-face or, increasingly, virtually); (c) possess one or more common goals; (d) are brought together to perform organizationally relevant tasks; (e) exhibit interdependencies with respect to workflow, goals, and outcomes; (f) have different roles and responsibilities; and (g) are together embedded in an encompassing organizational system, with boundaries and linkages to the broader system context and task environment.
About 60 years of research on teams have yielded a significant literature on team processes and team effectiveness (Kozlowski and Ilgen, 2006). This research builds on the small-group literature founded in social psychology and extends McGrath’s (1964) Input-Process-Output (I-P-O) heuristic to examine what factors shape team processes, how they interact in efforts to reach team goals, and the types of outcomes these interactions or team processes produce. The I-P-O model generally describes inputs (I) as factors at the individual level (e.g., team member personality), team level (e.g., task structure), and organizational/environmental level (e.g., organizational design). These factors are antecedents that enable, inhibit, or enhance team member interactions. Team processes (P) are generally acknowledged as critical mediators between inputs and team outcomes. They involve interpersonal processes as teams cycle through transition phases (e.g., plans for action) and action phases (Marks et al., 2001). Thus, efforts to improve the selection of potential team members and the composition of teams should focus on how individual differences, considered independently or in combination, relate to team processes and their outcomes. Finally, outcomes (O) are results of team processes that include performance outcomes (e.g., team effectiveness, team efficiency) as well as behavioral (e.g., absenteeism, turnover) and affective outcomes (e.g., team member commitment, team viability) (Cohen and Bailey, 1997; Mathieu et al., 2008). Variations and extensions of the I-P-O model abound, with different emphases on temporal dynamics (Marks et al., 2001), multilevel aspects of I-P-O (Kozlowski and Klein, 2000), and emergent states that serve as additional mediators between inputs and outcomes (Ilgen et al., 2005).
This chapter identifies future research needs that can improve the Army’s collective capacity to perform. Specifically, it focuses on how selection and classification of entry-level enlisted soldiers can improve unit performance and mission success. The I-P-O model will serve as a loose framework to identify future research objectives. Starting with the end goal, the committee first discusses team outcomes to define the criteria domain for selection and classification. Next, we examine team processes and emergent states as more proximal criteria of collective capacity. Finally, we
examine how future research on individual-level inputs to teams might help understand who is best suited for teamwork and how individuals might be better classified into specific Army small units, including teams, squads, and platoons.
Mathieu and Gilson (2012) noted that there has been relatively little research on team outcomes. Defining and measuring team outcomes have been challenging because they are often tied to specific team tasks and organizational conditions. These idiosyncratic measures can limit the generalizability of the research. A review of work team research (Sundstrom et al., 2000) found a wide variety of outcome constructs (e.g., productivity, communication, satisfaction, accidents, prosocial behavior) as well as measures to represent team outcomes, (e.g., objective measures of quantity and quality of team output; aggregated measures of individual ratings on team satisfaction and motivation; and managerial or customer ratings of team overall performance). Production teams were more likely to have objective measures of team performance, whereas service teams were more likely to have subjective self-ratings of team outcomes.
Despite the wide variety of team outcomes, Mathieu and Gilson (2012) identified two general forms. Tangible outcomes are directly related to team goals and include criteria tapping productivity, efficiency and quality, or composites of these outcomes. In contrast, influences on team members are outcomes that include team-level emergent states (e.g., unit cohesiveness) as well as individual-level outcomes tapping attitudes, behaviors, reactions, and individual development related to teamwork. More research attention has been paid to tangible outcomes at the individual role, team, and organization levels (Mathieu et al., 2008); however, team member reactions are also important because they are likely to drive future team interactions and team viability (Hackman, 1990; Mathieu et al., 2008).
Other challenges to defining and measuring these outcomes focus on temporal dynamics and the multilevel nature of team outcomes. Mathieu and colleagues (2008) noted that teams vary on how long it may take to develop stable outcomes. Teams do different things at different times (Marks et al., 2001) and evolve over the course of the team’s developmental stages (Tuckman, 1965). LePine (2003) examined how team-level averages of member cognitive ability, achievement, dependability, and openness to change predicted team performance before and after an unforeseen change in the task. None of these team composites predicted routine performance before the change, but significant relationships were found between these predictors, mediated by role structure adaptability, and team performance in a changing context. Thus, research gaps exist in understanding not only
what defines team effectiveness but also in understanding when these outcomes should be measured and over what period of time.
Teams vary on how individual team members’ actions are combined into a team-level outcome (Kozlowski et al., in press). Combinations range from composition models (e.g., individual team member errors are aggregated to represent team errors) to compilation models (e.g., differences in knowledge expertise across team members yield new insights on task goals). In composition models, constructs at different levels are isomorphic, amenable to simple aggregations based on sums or means, and empirically supported by indices of within-group agreement (e.g., rwg; as defined by Bliese, 2000). The performance of an army fire team (e.g., small team of riflemen firing at multiple targets) may be a composition construct if construed as the sum of targets that are hit by any team member. Individual hits and team hits share the same form and function.
In contrast, compilation models involve constructs in a common domain, but differ in their emergence. For example, the performance of an artillery team may be a compilation construct if construed as the number of targets that are hit by the team. In this team context, individual performance differs across team members and must be highly coordinated for team performance to emerge. The fire direction officer checks the target location, a technical expert ensures all equipment is ready, two fire direction specialists coordinate horizontal and vertical operations, and other team members may drive vehicles, operate the radio, chart data, etc.1 Thus, the form of a team outcome can influence how it is measured and how individual-level actions and characteristics may be combined to understand the team outcome.
Attempts to use individual-level characteristics to directly predict team-level outcomes are likely to result in cross-level and misspecification fallacies (Ployhart and Schneider, 2002). Given the multilevel nature of individuals and teams, selecting individuals to maximize team outcomes can be achieved in two basic ways (Ployhart and Schneider, 2002, 2005). In the first selection model, individual-level KSAOs are used to predict individual-level outcomes that are related to team effectiveness. This is the traditional selection model in human resources management, incorporating team-relevant criteria such as individual-level reactions, attitudes, behavior, and personal development (Mathieu and Gilson, 2012). This approach also requires theoretical and empirical links between individual-level criteria and team-level outcomes. These links are obvious in composition models
1 This example derives from a paper prepared by Captain Andrew Miller, former U.S. Army, for the National Research Council’s Committee on the Context of Military Environments: Social and Organizational Factors. The paper is available by request from the public access file of that committee.
but pose research and measurement challenges for compilation models (Kozlowski and Klein, 2000). An example of this approach is a study by Morgeson and colleagues (2005) that found conscientiousness, extraversion, and agreeableness to be significantly related to supervisor ratings of an individual’s contextual (team) performance.
The second selection model aggregates individual-level KSAOs to form team-level measures to directly predict team-level outcomes. Chan (1998) described five types of models to combine lower-level data to represent higher-level phenomena: additive (e.g., mean), direct consensus (e.g., within-group consensus of individual perceptions), referent-shift (e.g., within-group consensus of individual perceptions of the team), dispersion (e.g., variance), and process (e.g., focus on process or change). These types designate different functional relationships of the bottom-up process of outcome emergence. With regard to personnel selection, all types may be used to represent team-level outcomes, but only additive and dispersion models are used to represent team-level predictors.
In addition to combinations of individual-level data, team-level constructs may be represented by a single score from one team member. Minimum or maximum scores within a team can describe situations where one individual has a great effect on the entire group (e.g., one bad apple spoils the whole barrel or one brilliant mind carries the whole team) (Day et al., 2004). An example of this approach is a study by Mohammed and Angell (2003) that found student team variability on extraversion was positively correlated with team presentation grades (team oral performance).
These two selection models have direct implications for selection system design. The traditional individual-level approach can be used to measure KSAOs at pre-accession to predict individual-level outcomes related to team performance. These KSAOs can also be used in a multilevel approach to aid classification of individuals, post-accession, into teams whose team profiles are most likely to enhance team effectiveness.
In addition to their influence on team outcomes as described above, team processes may also serve as more proximal criteria for selection purposes. Instead of focusing on how individuals contribute to team performance, one can examine relationships between individual KSAOs on specific teamwork behaviors such as coordination, communication, and conflict resolution. In this approach, team processes and emergent states of teams are presented as critical mediators between individual characteristics and team outcomes.
Team outcomes represent bottom-line or distal criteria for selection systems. However, the I-P-O model identifies several processes and emergent
states that mediate relationships between individual KSAOs and ultimate tangible outcomes of teams. Consequently, they can serve as selection criteria because many team processes and emergent states have been shown to predict team effectiveness (Mathieu et al., 2008). Marks and colleagues (2001, p. 237) define team processes as “members’ interdependent acts that convert inputs to outcomes through cognitive, verbal, and behavioral activities directed toward organizing taskwork to achieve collective goals.” Cycles of I-P-O episodes involve transition and action phases, with previous episodes influencing subsequent episodes. Transition phases include team processes related to planning and evaluation activities as a team plans activities for goal accomplishment or reviews an action for lessons learned. In contrast, action phases include processes related to individual and coordinated behaviors that are directly tied to goal attainment. Finally, interpersonal processes occur through planning and action phases, focusing on motivation, conflict, and affect management.
As an example of a model of team processes, Marks and colleagues (2001) describe 10 team processes that are nested under the rubrics of transition, action, and interpersonal processes as follows:
- a team’s mission analysis;
- goal specification (prioritization of goals and identified sub-goals); and
- strategy formulation and alternative action plans.
- monitoring goal progress;
- monitoring environment and resources;
- monitoring team members and providing back-up support if needed; and
- coordination of individual tasks in an efficient sequence.
- managing conflict within the team;
- managing motivation for taskwork; and
- managing emotions of team members.
A meta-analysis examining relationships between team processes and two outcomes, team performance and team member satisfaction, showed all team processes were positively related to both team outcomes (LePine et al., 2008). Furthermore, there was some evidence that these relationships were moderated by team size and task interdependence, so that stronger relationships were found for larger teams and teams with high task interdependence. However, the meta-analysis was limited by its domain of a
small number of studies that predominantly used paper-and-pencil surveys for data collection.
Marks and colleagues (2001, p. 237) described emergent states as “cognitive, motivational, and affective states of teams, as opposed to the nature of their member interaction.” These states develop from dynamic interactions of I-P-O factors within a team context, emerging after early team experiences and changing over time with subsequent experiences. Emergent states may be viewed as team outcomes or as mediators between team inputs and tangible team outcomes (Mathieu et al., 2008). Some states, such as team cognition and team cohesion, may initially be influenced by surface-level individual characteristics (e.g., race or sex), but deeper forms of individual differences (e.g., personality or knowledge) may be better predictors in more mature teams (Kozlowski and Chao, 2012). Like team processes, the emergent states can serve as proximal criteria for team selection and classification decisions.
New technologies should be explored to better assess teamwork behaviors beyond paper-and-pencil measures. For example, current research with sociometric badges (a wearable electronic device about the size of an ID card that measures patterns of behavior) allows researchers to collect real-time data in social networks (Hollingshead and Poole, 2012; Kozlowski, in press; Pentland, 2010). These technologies can capture team interactions with several biomarkers (e.g., physical activity, identity, vocal intensity, heart rate, physical proximity with other team members), tracking who interacts with whom, when, and for how long (Kozlowski, in press). Badges can record not only what was said but also physiological data that may be able to capture qualitative metrics of the conversations (e.g., changes in heart rate during interaction with a particular team member may be interpreted as stress or anxiety). Technological improvements on the capabilities of these badges to record team interactions should increase the reliability and validity of these team-behavior measures.
In another example, computational models can be used to examine the correlation of a wide variety of initial team characteristics with emerging outcomes (McGrath et al., 2000). They simulate teams by specifying mathematical equations (e.g., logical if-then statements) to describe team interactions from one point in time to the next (Kozlowski et al., 2013). Studies using computational modeling can avoid typical constraints of experimental methods such as limited sample sizes, fatigue effects, and restriction of range in subject characteristics. Computational models have been used to identify possible effects of various individual member learning rates on team learning (Kozlowski et al., 2014). These results were combined with experiments on human teams to validate metrics for the emergence of team knowledge. Computational models have also been used to compare different social network measures on leadership influence (Braun et al., 2014).
This method enabled Braun and colleagues to examine over 500,000 simulated teams in a wide range of conditions. Their results showed that social network metrics that were most commonly used in experimental designs (e.g., reciprocity and centralization metrics) were not the best predictors of leadership influence. Although there was no single best network metric for all leadership outcomes, indirect network metrics (e.g., betweenness and closeness metrics) were better predictors of leadership influence (Braun et al., 2014). Given the practical constraints of team research, computational modeling may prove useful in the study of several individual-level KSAOs and their simultaneous effects on team behaviors and emergent team states.
As noted above, there has been more research on team processes than on team outcomes (Mathieu and Gilson, 2012). Thus, there is a need to develop a better understanding of, and new metrics to operationalize, team outcomes and effectiveness. In addition, new technologies should be explored to better assess teamwork behaviors beyond paper-and-pencil measures (see objective A in the research recommendation at the end of this chapter). To identify additional areas for future research, the committee next reviews what is known about individual-level predictors of team processes and outcomes.
Early research on military teams identified two primary skill tracks necessary for effective team performance (Glickman et al., 1987; Morgan et al., 1986; see also Shuffler et al., 2012 for a review of teams in military environments2). Taskwork requires specific job-related knowledge, skills, and abilities that are directly tied to performance demands of the job. They are bound by job requirements at the individual level. Traditional selection systems focus on assessing how well candidates can perform individual taskwork. In contrast, teamwork addresses the coordinated efforts of team members as they work together to accomplish individual and team goals. Teams that clearly understand each member’s role, communicate well, and have members who support each other are likely to be more effective than teams without good teamwork. Thus, proficiency in both taskwork and teamwork, operating together, is necessary for teams to be efficient and effective. Although the importance of teamwork has been widely recognized
2 Note that Glickman and colleagues (1987) and Morgan and colleagues (1986) both considered Navy teams, whereas Shuffler and colleagues (2012) provide a review of research relevant to teams across military environments and services.
in the team’s literature, research on selection for teamwork is not well developed (Mohammed et al., 2010; Tannenbaum et al., 2012).
Employee selection at the individual level generally includes examination of job analyses, identification of KSAO predictors, defining criteria, and measurement issues regarding the reliability and validity of specific selection procedures. Selection for team members may be viewed with parallel features at multiple levels. Although teams are defined by the interactions of team members, some taskwork may be performed at the individual level (Arthur et al., 2005). Thus, taskwork can be a blend of individual and coordinated work efforts. Job analysis generally defines taskwork at the individual level, whereas team task analysis examines the criticality of team task interdependence from both taskwork and teamwork perspectives (Bowers et al., 1994; Mohammed et al., 2010). Recent work has identified team-relatedness, the extent to which team members must interact in order to maximize team effectiveness, and team workflow among team members as important components of team task analysis (Arthur et al., 2012). However, this area is relatively undeveloped, and more research is needed to define team tasks (Allen and West, 2005).
Reviews on team selection have generally identified demographics (e.g., race), task-related KSAOs (e.g., experience), and psychological individual differences (e.g., cognitive ability, personality) as predictors of team selection and classification (Allen and West, 2005; Mohammed et al., 2010; Morgeson et al., 2012). Current research on team composition (the process by which individuals are selected for and assigned to a team, including consideration of potential and existing team members’ individual characteristics) has examined the diversity of team members on a wide range of individual characteristics. Reviews examining surface-level or demographic diversity found null (Horwitz and Horwitz, 2007) or negative (Mannix and Neale, 2005) relationships between this type of team composition and team outcomes. Extensions of this line of research have examined multiple characteristics that may define subgroups, such as group fault lines (Lau and Murnighan, 1998), with stronger fault lines related to team conflict (Lau and Murnighan, 2005). However, a meta-analysis found the negative effects of team demographic diversity on team performance diminished over time (Bell et al., 2007).
In contrast, reviews of team composition based on deeper-level individual differences such as general intelligence, personality, or values showed significant effects on team performance (Bell, 2007; Stewart, 2006). Team-level operationalizations of team composition were generally means on a single individual difference, although other measures (e.g., maximum score for disjunctive tasks, minimum score for conjunctive tasks) have been used with less frequency. Most robust was the finding that team means on general mental ability were positively related to team performance (Bell, 2007;
Stewart, 2006). Bell (2007) also found team means on conscientiousness, openness to experience, collectivism, and preference for teamwork were positively related to team performance in field studies but not in laboratory studies. In addition, team minimum values on agreeableness were stronger predictors of team performance than other operationalizations of a team agreeableness composite. Bell called for future research to explore additional operationalizations (e.g., proportion of team with high conscientiousness), as well as possible combinations (e.g., mean and maximum) to better represent team composition.
More recently, Carton and Cummings (2013) examined the number of subgroups based on social identities (i.e., surface-level) and knowledge (i.e., deeper-level). They found that having just two subgroups based on social identities had a more negative impact on team performance than having either no subgroups or more than two subgroups. However, having more knowledge-based subgroups generally had a positive effect on team performance. Furthermore, teams performed better when identity-based subgroups were not balanced in size but performed better when knowledge-based subgroups were balanced. Thus, conclusions about the effects of team composition on team performance must take into account the individual-level variables that are used to compose the team as well as the number and relative sizes of subgroups. The meta-analyses also show that relationships between team composition and team performance are moderated by factors such as research setting (field or laboratory studies), operationalization of team-level variables (Bell, 2007), and task type (Stewart, 2006). Furthermore, most of the current research examines only one or a few individual differences. Future research is needed to identify individual and team cognitions, affect/motivation, and behaviors that are linked to successful team outcomes and effectiveness. A team task analysis that, for example, identifies critical generic KSAOs, such as core teamwork skills, could be useful for initial selection (pre-accession). It could also aid in classifying individuals, post-accession, with respect to the more-contingent teamwork competencies identified as important for the proficiency of high-value teams. Essential to this research area is developing methods of team task analysis (see objective B of the research recommendation at the end of this chapter).
In addition to effective team composition, successful teams can be described as having individuals who are experienced and skilled in teamwork. Mohammed and colleagues (2010) described teamwork skills, such as interpersonal skills and communication skills, as core teamwork competencies that can be measured at the individual level and used to aid team selection. These skills would be valuable to all types of teams, making them generic predictors of teamwork. Often assessed by a work sample or paper-and-pencil test, specific teamwork skills like adaptability skills (Salas et al., 2007), interpersonal skills (Morgeson et al., 2005), and communication
skills (Bowers et al., 2000; Smith-Jentsch et al., 1996) have been shown to be valid predictors of team outcomes.
Stevens and Campion (1994) developed a Teamwork Knowledge, Skills, and Ability (KSA) test to help select individuals who are suited to teamwork; however, results on the validation of this test as a selection tool are mixed (Allen and West, 2005). Self-reported KSAs of teamwork were significantly related to individual performance (McClough and Rogelberg, 2003), but mean Teamwork KSA scores were not related to team performance (Miller, 2001). Furthermore, Teamwork KSA scores were found to be significantly correlated with cognitive ability (Stevens and Campion, 1999), potentially limiting the utility of this predictor if added to an existing test battery that already includes general mental abilities (Miller, 2001).
This report describes a number of individual differences that may be predictors of an individual’s ability to work in teams. For example, an individual’s inhibitory control capacity (see Chapter 2), cognitive biases (see Chapter 3), and emotional regulation (see Chapter 6) may be related to how well he or she adapts when team members engage in potentially stressful interactions. A soldier who is capable of controlling emotional and behavioral impulses may be more likely to work well in a team context. Conversely, a soldier who is low in emotional regulation may be likely to disrupt or distract the team from accomplishing a mission.
Individual assessments may be combined to help predict a number of outcomes related to long-term team performance and satisfaction. For selection and placement decisions, it is likely that there are no simple rules to find the best individuals for a particular team. Individual characteristics related to teamwork and taskwork may provide supplementary fit to a team (e.g., all team members are similarly conscientious and responsive to one another); or the characteristics may provide complementary fit (e.g., one team member’s expertise fills a team’s need for that knowledge).
In addition to core teamwork competencies that apply to all teams, contingent teamwork competencies are sensitive to team tasks, structures, and environmental conditions that may change relationships between predictors and outcomes (Mohammed et al., 2010). A meta-analysis of studies examining person-group fit found this individual-level construct to be significantly correlated with individual-level outcomes such as job satisfaction, organizational commitment, and intentions to quit (Kristof-Brown et al., 2005). Unfortunately, the small number of studies in this meta-analysis (n’s ranged from 4 to 12, depending on outcome) did not permit any investigation of possible moderators to these relationships. Perceptions of fit based on goals, values, and/or personality are likely to be influenced by the existing team and organization environments that a newly selected team member joins. For example, team size will influence the division of labor and coordination demands (Steiner, 1972). An individual’s perceptions of
fit within a team can be shaped by these contextual features. As stated in objective C of the research recommendation at the end of this chapter, future research is needed to identify optimal within-individual profiles that are linked to team effectiveness. This research should also consider types of team structures, tasks, and environmental conditions that moderate relationships between profile attributes and their combined influence on team processes and outcomes.
In addition to potential moderators that can change predictor-outcome relationships, it is possible for post-selection experiences to change the predictive power of individual profiles. Pre-selection experiences help shape an applicant’s task-related and team-related knowledge and skills, and they can be used as predictors of future team outcomes. Similarly, post-selection experiences help shape an employee’s task-related and team-related knowledge and skills, potentially mitigating the utility of a selection measure. Training programs and team experiences can increase an individual’s capacity toward effective teamwork. For example, Chen and colleagues (2004) found a course on teamwork significantly improved Teamwork KSA scores and observer ratings of teamwork competencies for college students. Thus, training may compensate for low pre-selection scores on this predictor. In another example, team leadership may require adjustments when a team encounters an extreme context that puts members in harm’s way (Rumsey, 2013; Yammarino et al., 2010). As a team assesses a situation, plans for action, and executes those plans, the team learns how it impacts the environment. In turn, these lessons influence subsequent teamwork (Burke et al., 2006). Future research should investigate the effects of teamwork training and team experiences on the predictive power of individual-differences measures (see objective D in the research recommendation at the end of this chapter).
Research Gaps and Future Directions
Teams are critical units for military performance. Improving team performance can be aided by selecting individuals who are most capable of teamwork and composing teams with individuals who have compatible KSAOs. The criterion domain of team effectiveness has received relatively little research attention compared to research on team inputs and team processes. A better understanding of team outcomes, both tangible performance metrics and influences on team members, is needed to validate team selection methods. Well-defined team outcomes can also inform more-thorough descriptions and analyses of team tasks. Despite repeated calls for
more research on team task analysis, few researchers have answered this call (Allen and West, 2005).
Team selection can use traditional selection models, finding individual-level predictors (e.g., cognitive ability) to predict individual-level outcomes (e.g., individual performance). Team selection can also take a multilevel perspective, examining links between individual- and team-level predictors and their relationships to individual- and team-level criteria. Some important team characteristics (e.g., team cohesiveness, team diversity) have no individual-level equivalents, so multilevel perspectives are better able to assess a team’s collective capacity to perform. Indeed, Ployhart and Schneider (2005) argued that some desired organizational characteristics (e.g., workforce diversity) may not be achieved if selection systems only focus on maximizing individual performance on a single job. Research on team-level composites of predictors in team selection can help identify new predictors at the individual level and how they might be best combined to measure team composition.
It is important to recognize the limitations of team selection. The benefits of a valid selection system may be nullified if team members fail to cooperate with one another (Schneider et al., 2000). Good selection can identify those individuals who are most likely to succeed in teams; however, the actual interactions of team members in a specific context would be more directly responsible for team outcomes (Hackman and Katz, 2010). What happens post-accession—how individuals are trained, equipped, organized, socialized, led, and rewarded—will also be important predictors of how team members interact and perform, but this topic was beyond the scope of this study. Likewise, the composition or assembly of individuals into teams brings together a wide variety of individual characteristics, team task designs, and contextual features that can critically affect team performance. Furthermore, different team members can assume “leadership” roles as individuals mature, members are reassigned, and time changes role demands (Contractor et al., 2012). Teamwork behaviors may also be influenced by such negative factors as stereotypes or implicit bias, but the committee judges that the utility of using such predictors for selection purposes would be limited due to potential mitigating effects of contextual features such as good leadership and clear tasks or of surface-level diversity whereby the effects of race or gender differences dissipate over time as team members get to know one another (Harrison et al., 2002). Lastly, the committee notes that a collective capacity to perform can be defined by larger units than teams as defined in this chapter. The Army’s squads, platoons, companies, battalions, brigades, divisions, corps, and even field army extend the “collective capacity” to many levels. Since this chapter is focused on individual selection, it makes sense to confine future research at individual and team levels. However, as Ployhart and Schneider (2002) noted, some higher-level
goals may require lower-level goals to be suboptimized in order to accommodate performance requirements across all levels.
Relative to individual personnel selection, research on team selection is in a developmental stage. More research is needed to identify KSAOs that are required for specific team taskwork as well as generic teamwork. More research is needed to define and measure team effectiveness as teams develop, evolve, and change. More research is needed to identify individual-level predictors, how they are combined into individual profiles, and how they are combined into team composites. Together, these research directions can maximize the potential for individuals to work effectively in dynamic teams.
The U.S. Army Research Institute for the Behavioral and Social Sciences should support research on individual- and team-level knowledge, skills, abilities, and other characteristics that influence the collective capacity to perform. Future research should include the following objectives:
- Develop a better understanding of, and new metrics to operationalize, team outcomes and effectiveness. In addition, new technologies should be explored to better assess teamwork behaviors beyond paper-and-pencil measures.
- Identify individual and team cognitions, affect/motivation, and behaviors that are linked to successful team outcomes and effectiveness. Essential to this is developing methods of team task analysis.
- Identify optimal within-individual profiles that are linked to team effectiveness. This research should also consider types of team structures, tasks, and environmental conditions that moderate relationships between profile attributes and their combined influence on team processes and outcomes.
- Investigate the effects of teamwork training and team experiences on the predictive power of individual-differences measures.
Allen, N.J., and M.A. West. (2005). Selection for teams. In A. Evers, N. Anderson, and O. Voskuijl, Eds., The Blackwell Handbook of Personnel Selection (pp. 476–494). Malden, MA: Wiley-Blackwell.
Arthur, W., Jr., B.D. Edwards, S.T. Bell, A.J. Villado, and W. Bennett, Jr. (2005). Team task analysis: Identifying tasks and jobs that are team-based. Human Factors, 47(3):654–669.
Arthur, W., Jr., R.M. Glaze, A. Bhupatkar, A.J. Villado, W. Bennett, Jr., and L.J. Rowe. (2012). Team task analysis: Differentiating between tasks using team relatedness and team workflow as metrics of team task interdependence. Human Factors, 54(2):277–295.
Bell, S.T. (2007). Deep-level composition variables as predictors of team performance: A meta-analysis. Journal of Applied Psychology, 92(3):595–615.
Bell, S.T., A.J. Villado, M.A. Lukasik, A. Briggs, and L. Belau. (2007). Revisiting Team Demographic Diversity and Performance Relationship: A Meta-Analysis. Paper presented at the Annual Conference of the Society for Industrial and Organizational Psychology, New York, NY.
Bliese, P.D. (2000). Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. In K.J. Klein and S.W.J. Kozlowski, Eds., Multilevel Theory, Research, and Methods in Organizations: Foundations, Extensions, and New Directions (pp. 349–381). San Francisco, CA: Jossey-Bass.
Bowers, C.A., D.P. Baker, and E. Salas. (1994). Measuring the importance of teamwork: The reliability and validity of job/task analysis indices of team-training design. Military Psychology, 6(4):206–214.
Bowers, C.A., J.A. Pharmer, and E. Salas. (2000). When member homogeneity is needed in work teams: A meta-analysis. Small Group Research, 31(3):305–327.
Braun, M.T., L.A. DeChurch, and D.R. Carter. (2014). Measuring Shared Leadership: A Computational Modeling Study. Paper presented at the 29th Annual Conference of the Society for Industrial and Organizational Psychology, Honolulu, HI.
Burke, C.S., K.C. Stagl, E. Salas, L. Pierce, and D. Kendall. (2006). Understanding team adaptation: A conceptual analysis and model. Journal of Applied Psychology, 91(6):1,189–1,207.
Carton, A.M., and J.N. Cummings. (2013). The impact of subgroup type and subgroup configurational properties on work team performance. Journal of Applied Psychology, 98(5):732–758.
Chan, D. (1998). Functional relations among constructs in the same content domain at different levels of analysis: A typology of composition models. Journal of Applied Psychology, 83(2):234–246.
Chen, G., L.M. Donahue, and R.J. Klimoski. (2004). Training undergraduates to work in organizational teams. Academy of Management Learning and Education, 3(1):27–40.
Cohen, S.G., and D.E. Bailey. (1997). What makes teams work: Group effectiveness research from the shop floor to the executive suite. Journal of Management, 23(3):239–290.
Contractor, N.S., L.A. DeChurch, J. Carson, D.R. Carter, and B. Keegan. (2012). The topology of collective leadership. The Leadership Quarterly, 23(6):994–1,011.
Day, E.A., A. Winfred, B. Miyashiro, B.D. Edwards, T.C. Tubre, and A.H. Tubre. (2004). Criterion-related validity of statistical operationalizations of group general cognitive ability as a function of task type: Comparing the mean, maximum, and minimum. Journal of Applied Psychology, 34(7):1,521–1,549.
Dyer, J.L., T.R. Tremble, Jr., and D.L. Finley. (1980). The Structural, Training, and Operational Characteristics of Army Teams (Technical Report 507). Alexandria, VA: U.S. Army Research Institute for the Behavioral and Social Sciences.
Essens, P., A. Vogelaar, J. Mylie, C. Blendell, C. Paris, S. Halpin, and J. Baranski. (2005). Military Command Team Effectiveness: Model and Instrument for Assessment and Improvement (Research and Technology Organisation Technical Report TR-HFM-087). NATO. Available: http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA437898 [February 2015].
Glickman, A.S., S. Zimmer, R.C. Montero, P.J. Guerette, W.J. Campbell, B.B. Morgan, and E. Salas. (1987). The Evolution of Team Skills: An Empirical Assessment with Implications for Training (Technical Report NTSC 87-016). Orlando, FL: Naval Training Systems Center, Human Factors Division.
Hackman, J.R., Ed. (1990). Groups that Work (and Those that Don’t): Creating Conditions for Effective Teamwork. San Francisco, CA: Jossey-Bass.
Hackman, J.R., and N. Katz. (2010). Group behavior and performance. In S.T. Fiske, G.T. Gilbert, G. Lindzey, and A.E. Jongsma, Eds., Handbook of Social Psychology (Vol. 5, pp. 1,208–1,251). New York: Wiley & Sons.
Harrison, D.A., K.H. Price, J.H. Gavin, and A.T. Florey. (2002). Time, teams, and task performance: Changing effects of surface- and deep-level diversity on group functioning. Academy of Management Journal, 45(5):1,029–1,045.
Hollingshead, A.B., and M.S. Poole. (2012). Research Methods for Studying Groups and Teams: A Guide to Approaches, Tools, and Technologies. New York: Routledge.
Horwitz, S.K., and I.B. Horwitz. (2007). The effects of team diversity of team outcomes: A meta-analytic review of team demography. Journal of Management, 33(6):987–1,015.
Ilgen, D.R., J.R. Hollenbeck, M. Johnson, and D. Jundt. (2005). Teams in organizations: From input-process-output models to IMOI models. Annual Review of Psychology, 56:517–543.
Kozlowski, S.W.J. (in press). Advancing research on team process dynamics: Theoretical, methodological, and measurement consideration. Organizational Psychology Review. Available: http://iopsych.msu.edu/koz/Recent%20Pubs/Kozlowski%20%28in%20press%29%20-%20Team%20Dynamics.pdf [February 2015].
Kozlowski, S.W.J., and G.T. Chao. (2012). The dynamics of emergence: Cognition and cohesion in work teams. Managerial and Decision Economics, 33(5-6):335–354.
Kozlowski, S.W.J., and D.R. Ilgen. (2006). Enhancing the effectiveness of work groups and teams. Psychological Science, 7(3):77–124.
Kozlowski, S.W.J., and K.J. Klein. (2000). A multilevel approach to theory and research in organizations: Contextual, temporal, and emergent processes. In K.J. Klein and S.W.J. Kozlowski, Eds., Multilevel Theory, Research, and Methods in Organizations (pp. 3–90). San Francisco, CA: Jossey-Bass.
Kozlowski, S.W.J., G.T. Chao, J.A. Grand, M.T. Braun, and G. Kuljanin. (2013). Examining the dynamics of multilevel emergence: Learning and knowledge building in teams. Organizational Research Methods, 16(4):581–615.
Kozlowski, S.W.J., G.T. Chao, J.A. Grand, M.T. Braun, G. Kuljanin, D. Pickhardt, and S. Mak. (2014). Team Macrocognition: Integrating Computational Modeling and Research Methods. Paper presented at the 29th Annual Conference of the Society for Industrial and Organizational Psychology, Honolulu, HI.
Kozlowski, S.W.J., J.A. Grand, S.K. Baard, and M. Pearce. (in press). Teams teamwork, and team effectiveness: Implications for human systems integration. In D. Boehm-Davis, F. Durso, and J. Lee, Eds., The Handbook of Human Systems Integration. Washington, DC: American Psychological Association.
Kristof-Brown, A.L., R.D. Zimmerman, and E.C. Johnson. (2005). Consequences of individuals’ fit at work: A meta-analysis of person-job, person-organization, person-group, and person-supervisor fit. Personnel Psychology, 58(2):281–342.
Lau, D.C., and J.K. Murnighan. (1998). Demographic diversity and faultlines: The compositional dynamics of organizational groups. Academy of Management Review, 23(2): 25–34.
Lau, D.C., and J.K. Murnighan. (2005). Interactions within groups and subgroups: The effects of demographic faultlines. Academy of Management Journal, 48(4):645–659.
LePine, J.A. (2003). Team adaptation and postchange performance: Effects of team composition in terms of members’ cognitive ability and personality. Journal of Applied Psychology, 88(1):2–39.
LePine, J.A., R.F. Piccolo, C.L. Jackson, J.E. Mathieu, and J.R. Saul. (2008). A meta-analysis of teamwork processes: Tests of a multidimensional model and relationships with team effectiveness criteria. Personnel Psychology, 61(2):273–307.
Mannix, E., and M.A. Neale. (2005). What differences make a difference? Psychological Science in the Public Interest, 6(2):31–55.
Marks, M.A., J.E. Mathieu, and S.J. Zaccaro. (2001). A temporally based framework and taxonomy of team processes. Academy of Management Review, 26(3):356–376.
Mathieu, J.E., and L.L. Gilson. (2012). Criteria issues and team effectiveness. In S.W.J. Kozlowski, Ed., The Oxford Handbook of Organizational Psychology (pp. 910–932). New York: Oxford University Press.
Mathieu, J., M.T. Maynard, T. Rapp, and L. Gilson. (2008). Team effectiveness 1997–2007: A review of recent advancements and a glimpse into the future. Journal of Management, 34(3):410–476.
McClough, A.C., and S.G. Rogelberg. (2003). Selection in teams: An exploration of the teamwork knowledge, skills, and ability test. International Journal of Selection and Assessment, 11(1):56–66.
McGrath, J.E. (1964). Social Psychology: A Brief Introduction. New York: Holt, Rinehart, and Winston.
McGrath, J.E., H. Arrow, and J.L. Berdahl. (2000). The study of groups: Past, present, and future. Personality and Social Psychology Review, 4(1):95–105.
Miller, D.L. (2001). Reexamining teamwork KSAs and team performance. Small Group Research, 32(6):745–766.
Mohammed, S., and L.C. Angell. (2003). Personality heterogeneity in teams: Which differences make a difference for team performance? Small Group Research, 34(6):651–677.
Mohammed, S., J. Cannon-Bowers, and F.S. Chuen. (2010). Selection for team membership: A contingency and multilevel perspective. In J.L. Farr and N.T. Tippins, Eds., Handbook of Employee Selection (pp. 801–822). New York: Routledge.
Morgan, B.B., A.S. Glickman, E.A. Woodard, A.S. Blaiwes, E. Salas, W.J. Campbell, D.L. Miller, R.C. Montero, and S. Zimmer. (1986). Measurement of Team Behaviors in a Navy Training Environment (NTSC Technical Report No. TR-86-014). Orlando, FL. Human Factors Division, Naval Training Systems Command, Department of the Navy. Available: http://www.dtic.mil/dtic/tr/fulltext/u2/a185237.pdf [January 2015].
Morgeson, F.P., M.H. Reider, and M.A. Campion. (2005). Selecting individuals in team settings: The importance of social skills, personality characteristics, and teamwork knowledge. Personnel Psychology, 58(3):83–611.
Morgeson, F.P., S.E. Humphrey, and M.C. Reeder. (2012). Team selection. In N. Schmitt, Ed., The Oxford Handbook of Personnel Assessment and Selection (pp. 832–848). New York: Oxford University Press.
Pentland, A.S. (2010). To signal is human: Real-time data mining unmasks the power of imitation, kith and charisma in our face-to-face social networks. American Scientist, 98:203–211.
Ployhart, R.E., and B. Schneider. (2002). A multi-level perspective on personnel selection research and practice: Implications for selection system design, assessment, and construct validation. In F.J. Yammarino and F. Dansereau, Eds., Research in Multi-level Issues, Vol. 1, (pp. 95–140). Oxford, UK: Elsevier Science.
Ployhart, R.E., and B. Schneider. (2005). Multilevel selection and prediction: Theories, methods, and models. In A. Evers, N. Anderson, and O. Voskuijl, Eds., The Blackwell Handbook of Personnel Selection (pp. 495–516). Malden, MA: Blackwell.
Rumsey, M. (2013). Military understanding of teams. In G.B. Graen and J.A. Graen, Eds., Management of Team Leadership in Extreme Context: Defending Our Homeland, Protecting Our First Responders. Charlotte, NC: Information Age.
Salas, E., D.R. Nichols, and J.E. Driskell. (2007). Testing three team training strategies in intact teams: A meta-analysis. Small Group Research, 38(4):471–488.
Schneider, B., D.B. Smith, and W.P. Sipe. (2000). Personnel selection psychology: Multi-level considerations. In K.J. Klein and S.W.J. Kozlowski, Eds., Multi-level Theory, Research, and Methods in Organizations: Foundations, Extensions, and New Directions (pp. 91–120). San Francisco, CA: Jossey-Bass.
Shuffler, M.L., D. Pavlas, and E. Salas. (2012). Teams in the military. In J.H. Laurence and M.D. Matthews, Eds., The Oxford Handbook of Military Psychology (pp. 282–310). New York: Oxford University Press.
Smith-Jentsch, K.A., E. Salas, and D.P. Baker. (1996). Training team performance-related assertiveness. Personnel Psychology, 49(4):909–936.
Steiner, I.D. (1972). Group Processes and Productivity. New York: Academic Press.
Stevens, M.J., and M.A. Campion. (1994). The knowledge, skill, and ability requirements for teamwork: Implications for human resource management. Journal of Management, 20(2):503–530.
Stevens, M.J., and M.A. Campion. (1999). Staffing work teams: Development and validation of a selection test for teamwork settings. Journal of Management, 25(2):207–228.
Stewart, G.L. (2006). A meta-analytic review of relationships between team design features and team performance. Journal of Management, 32(1):29–54.
Sundstrom, E., M. McIntyre, T. Halfhill, and H. Richards. (2000). Work groups: From the Hawthorne studies to work teams of the 1990s and beyond. Group Dynamics: Theory, Research, and Practice, 4(1):44–67.
Tannenbaum, S.I., J.E. Mathieu, E. Salas, and D. Cohen. (2012). Teams are changing: Are research and practice evolving fast enough? Industrial and Organizational Psychology: Perspectives on Science and Practice, 5(1):2–24.
Tuckman, B.W. (1965). Developmental sequence in small groups. Psychological Bulletin, 63:384–399.
Yammarino, F.J., M.D. Mumford, M.S. Connelly, and S.D. Dionne. (2010). Leadership and team dynamics for dangerous military contexts. Military Psychology, 22(Suppl 1):S15–S41.