Team Roles: A Review and Integration

Abstract

The concept of roles is ubiquitous in the social sciences, and a number of scholars have examined the operation of roles in task teams. In fact, this research has resulted in a seemingly unlimited number of roles that have been described as relevant to team performance. In this study, we attempt to integrate this research by deriving a model that describes three primary behavioral dimensions that underlie team role behavior: (a) dominance, (b) sociability, and (c) task orientation. Based on this model, we conduct a cluster analysis of the 154 team roles described in previous research. We identify 13 primary team role clusters, and discuss the implications of this approach for gaining further insight into team role structure and performance. We believe this is one step toward speaking a common language in discussing team roles.

Keywords

roles teamwork team

Roles are important in teams because they represent patterns of behavior that are interrelated with the activities of other team members in pursuit of the overall team goal. The term role comes from the French rôle, which originally referred to the rolls of paper on which actors’ parts were written (and often read to the actors if they forgot their lines). Thus, the theatrical analogy is quite apt—the term role came to refer to the theatrical part enacted by an actor in a play. However, in the theater, an actor adopts and sheds roles easily, whereas in real-world teams, a role is a more internalized, enduring set of behaviors. Accordingly, Stewart, Fulmer, and Barrick (2005) define a role as a set of behaviors that are repetitive activities characteristic of a person in a particular setting.

Sports teams often provide a clear illustration of how roles structure team interaction. In a basketball team, for example, the five primary roles include a point guard, a shooting guard, a center, a small forward, and a power forward. Although these roles may vary within any particular team, it is generally accepted that some variation of this basic structure underlies the nature of basketball team performance. Unfortunately, there is less consensus on the nature of the basic role structure of task teams. For example, some team role taxonomies have described primary task team roles in as few as two role dimensions (Bales & Slater, 1955) to as many as 27 (Benne & Sheats, 1948). Moreover, Gregory, Shimono, Burke, and Salas (2015) reviewed 23 existing team role taxonomies and identified a total of 164 different roles described by these taxonomies. This would lead one to conclude that there are a seemingly infinite number of team roles, or that different researchers are using different terms for similar role dimensions. We believe it is the latter.

In this article, we attempt to integrate existing team role taxonomies to derive a primary or core set of team roles. We do this by presenting a model that describes three primary behavioral dimensions that underlie team role behavior: (a) dominance, (b) sociability, and (c) task orientation. We conduct a cluster analysis of team roles identified in existing team role taxonomies to identify primary role clusters or dimensions. We describe the results of this analysis and the implications of this approach for further research.

Early Theoretical and Empirical Work

As Biddle (1986), Moxnes (1999), and others have noted, the concept of role is one of the most central in social science. Yet, it has no single scientific origin, nor a single core definition. Early proponents of the role concept include Ralph Linton (an anthropologist), George Herbert Mead (a social theorist), and Jacob Moreno (a psychologist), each with a different perspective on roles. In the audaciously titled book, “The Study of Man,” Linton (1936) described how the functioning of a group depends on the patterns of reciprocal behavior between individuals. Linton noted that when a person puts a given set of rights and responsibilities (which he terms a status or position, such as father or leader) into effect, he or she is performing a role. Thus, Linton described a role as a set of patterned behaviors that are inseparable from a position.

During the same time period as Linton, George Herbert Mead (1934) examined how social structure is organized and maintained, and emphasized the concept of role taking as a key to social control. In the process of role taking, the individual takes the perspective of a significant other and is able to view his or herself from that standpoint, and it is this reciprocal role taking that makes social activity possible (Heimer & Matsueda, 1994). Another contemporary, Jacob Moreno (1934), viewed role playing as a mechanism of role change, and emphasized the value of playing or practicing a role in a simulated psychodrama setting in order to gain a more in-depth understanding of the role. This work, couched within the psychodrama context, was influential but perhaps less impactful on mainstream group research.

The influential early work of Mead, Linton, and others lead to a proliferation of research on various aspects of group roles. This included work on role conflict (Kahn, Wolfe, Quinn, Snoek, & Rosenthal, 1964), role ambiguity (Rizzo, House, & Lirtzman, 1970), role strain (Beehr, 1976), role transitions (Nicholson, 1984), role overload (Kahn et al., 1964), role change (Turner, 1990), role efficacy (Beauchamp, Bray, Eys, & Carron, 2002), and person-role fit (DeRue & Morgeson, 2007),

Roles have been defined as institutionalized sets of expectations, relationships, and behaviors that ensure stable and predictable social interactions (Stryker & Statham, 1985). Kreps and Bosworth (1993) note that:

In any situation where people interact there is a tendency for their behaviors, sentiments, and motives to become differentiated into discrete entities called roles (Turner, 1980, p. 126). Once roles have become differentiated, the behavior, sentiments, and motives that appear subsequently in similar situations will tend to become patterned (i.e., they will be performed conventionally). (p. 436)

However, roles are not simply static sets of expectations and behaviors, but instead they change and adapt in response to role demands. In fact, two overarching goals of research on roles within teams are to account for both role stability and role change.

Early research efforts were also devoted to determining the basic role dimensions that define group interaction (see Hare, 1972). Stated simply, if we define roles broadly as coherent sets of expectations and behaviors, then what are the core roles that we observe in task teams? This fundamental question has been posed by various researchers in various guises: “What are the characteristics which can be evaluated by observing people interacting?” (Carter, 1954, p. 477), “What are the ‘functional role behaviors’ that emerge in small groups?” (Mudrack & Farrell, 1995, p. 542), and “What are ‘the roles members play in executing critical team functions?’” (Mumford, Van Iddekinge, Morgeson, & Campion, 2008, p. 250). In the following, we discuss attempts to develop a classification of team roles.

Team Role Taxonomies

Although Stewart et al. (2005) state that “A universally accepted taxonomy of team member roles does not exist” (p. 346), this is not necessarily the case. The fact is that, whereas there is no one way of characterizing team roles that is used to the exclusion of others, there are a number of different types of role taxonomies. A broad or higher order characterization of team roles may adopt a simple two-factor model, distinguishing roles as either primarily task oriented or primarily socioemotionally oriented (see Bales, 1950; Bales & Slater, 1955). More detailed team role taxonomies include Belbin (1993), who described nine team roles based on observations of management teams, and Benne and Sheats (1948), who defined team roles in terms of 27 discrete dimensions, including those of (a) initiator, (b) information giver, (c) coordinator, (d) harmonizer, (e) encourager, and so on.

There are several problems with these and other role taxonomies. First, there is a considerable divergence in terms of the team contexts examined. For example, Belbin’s classification was based on observations of managerial personnel who performed a 1-week computerized business game as part of a general management course (Dulewicz, 1995). One of the most heavily cited role taxonomies, proposed by Benne and Sheats (1948), was based on observations of sensitivity training groups, or t-groups, that were popular in the human potential movement of the 1960s, and whose primary purpose was interpersonal growth. In fact, Benne noted that these observations were never meant to apply to work groups, noting that it was a “mistake to think that a group that was specifically created for learning . . . about self . . . could become the prototype of a work group” (quoted in Freedman, 1996, p. 336). Thus, while Benne and Sheats’ work is often described as “the earliest and perhaps most pervasive” attempt at role classification (Mumford et al., 2008 p. 251), it is prudent to conclude that no one taxonomy constitutes a single gold standard.

Second, within the existing team role taxonomies, there is also considerable divergence in terms of the roles described. For example, Gregory et al. (2015) conducted a comprehensive inventory of existing team role taxonomies. To generate a list of task role taxonomies, Gregory et al. conducted a search of multiple databases (including PsycNET, Sociological Abstracts, EBSCOhost, Web of Science, and ProQuest) using the following search terms: team roles, group roles, social roles, and emotional roles paired with the term taxonomy. A series of additional searches were conducted using the keywords role sets, role strain, role transition, role differentiation, and role conflict paired with the terms teams, groups, and crew. Finally, a third search was conducted using the keywords using the terms team [group] role measurement, team [group] role assessment, team [group] role questionnaire, and related terms. Based on this search, 187 articles were deemed to be potentially relevant, from which a final set off 139 articles were selected, which produced 23 unique team role taxonomies (including Bales, 1950; Benne & Sheats, 1948; Helbin, 1981; Mathieu, Tannenbaum, Kukenberger, Donsbach, & Alliger, 2015; Mumford, Campion, & Morgeson, 2006; Stempfle, Hübner, & Badke-Schaub, 2001; and 17 others). Cumulatively, these 23 team role taxonomies rendered 164 role descriptors. Some roles were similar across some of the taxonomies. For example, Williams, Morgan, and Cameron (2011); Beck, Eng, and Brusa (1989); and Belbin (1993) identified leader, task leader, and chairman roles, respectively. However, many roles were unique to specific taxonomies. For example, the procedural technician role was only identified by Benne and Sheats (1948), the shaper role was only identified by Belbin (1993), and the contractor role was only identified by Mumford et al. (2006). As Neiman and Hughes (1951) observed some 60 years ago, “In surveying the literature, one is confronted with . . . a hopeless mass of different definitions, usages, and implications” (p. 142). It seems that little has changed.

The Nature and Structure of Team Roles

One question that group researchers have attempted to answer is what is the basic structure of roles in task teams? In other words, is there a primary set of role dimensions that are descriptive of role performance in task teams, and can we use this model of the nature and structure of team roles to better understand how role performance relates to team functioning and performance?

Given that group roles can be viewed as patterns of behavior (Mumford et al., 2008), Carter (1954, 1962) was one of the first researchers to attempt to account for the regularities of behavior that are observed in task groups. Couch and Carter (1952) observed small groups performing discussion, reasoning, and mechanical assembly tasks. They asked observers to rate their behavior on a number of variables, including aggressiveness, cooperativeness, leadership, task orientation, talkativeness, and other characteristics. A series of factor analyses found that, across differences in team tasks and group sizes, three primary factors emerged. They labeled these dimensions (a) Individual Prominence (behaviors associated with leadership, striving for recognition, and aggressiveness), (b) Sociability (behaviors related to positive social interaction, sociability, and maintenance of group relations), and (c) Group Goal Facilitation (behaviors related to achieving the group goal, cooperation, and attainment of the task). Carter (1954) concluded that “It seems apparent that the interaction behavior of individuals involved in small group situations can be adequately described by three factors” (p. 284). Couch (1960) replicated this research with a more comprehensive coding scheme incorporating 55 team behavior measures and found a similar factor structure, which he labeled (a) Interpersonal Dominance (prominence, activity, initiative), (b) Interpersonal Affect (positive affect toward others), and (c) Task Serious versus Social Expressivity (seriousness of purpose in working on the group task). Similar results were found by Mann (1961) and Borgatta (1963). Mudrack and Farrell (1995) also found that three primary factors emerged in examining behavior in small problem solving groups, which they labeled (a) Individual Roles (e.g., asserts authority), (b) Maintenance Roles (e.g., praises and encourages others), and (c) Task Roles (e.g., coordinates task activities, suggests task solutions).

Drawing on the previous work of Parsons, Bales, and Shils (1953) as well as Couch and Carter (1952), Bales (2001) argued that there are basic problems that must be addressed in all task groups (e.g., directing the group’s progress, motivating group members, problem solving), and that discrete roles develop to solve these problems. Bales described three functional problems that are faced by all groups: (a) activity or managing dominance or the exercise of power (i.e., some people dominate group activity to the exclusion of others), (b) likability or managing the maintenance of positive relations (i.e., positive relations must be maintained to support the continuance of the group), and (c) task ability or managing efforts to facilitate goal achievement (i.e., efforts must be marshaled to pursue the solution of group goals).

Thus, at a broad level, Bales argued that these problems (dominance, sociability, and task orientation) are essential to the effective functioning of all groups. Accordingly, Bales (see Bales, 1970; Bales & Cohen, 1979) proposed three primary dimensions to describe group behavior: Upward/Downward (dominance or submissiveness), Negative/Positive (friendly or unfriendly), and a third dimension labeled Forward/Backward that was defined as instrumentally controlled or task oriented versus emotionally expressive. Thus, Bales argued that regularities of group behavior emerge along these three primary dimensions.

Finally, Fiske, Cuddy, and Glick (2007) have described what they term as universal dimensions of social cognition. That is, when people attempt to interpret behavior or form impressions of others, they do so in terms of the basic dimensions of competence (industrious, intelligent, skillful) and warmth (sociable, good-natured). Fiske et al. also note a third dimension, Bales’ activity or dominance dimension, which they note is likely more relevant to actual interaction than to perceptions.

Following Carter (1954, 1962) and Bales (2001) et al., we believe that role behavior in groups can be described in terms of the three broad dimensions of (a) individual prominence, or Dominance, (b) Sociability, and (c) group goal facilitation, or Task Orientation. The dominance dimension captures the distinction between behavior that is dominant versus submissive, active versus passive, or seeking control versus deference. The sociability dimension captures the distinction between behavior that is sociable, friendly, and agreeable versus behavior that is withdrawn, unfriendly, and aloof. The task orientation dimension captures the distinction between behavior that is oriented toward the solution of task problems versus behavior that shirks or evades task responsibilities. Table 1 presents behavioral descriptors for these dimensions.

Table 1.

Behavioral Descriptors for Dominance, Sociability, and Task Orientation.

High dominance	Low dominance
Dominate, control, direct, influence, assert, take charge, lead, command, active	Defer, comply, follow, compliant, submissive, support, take orders, passive
High sociability	Low sociability
Friendly, interested in others, cordial, warm, gregarious, supportive	Withdrawn, aloof, avoids contact with others, prefers working alone, solitary
High task orientation	Low task orientation
Achievement-oriented, organized, reliable, dependable, conscientious, planful, responsible, serious	Careless, irresponsible, disordered, impulsive, spontaneous, untrustworthy, inactive, work-shy

We present a three-dimensional model (termed TRIAD [Tracking Roles In and Across Domains]) that comprises the dimensions of Dominance (high dominance vs. low dominance), Sociability (high sociability vs. low sociability), and Task Orientation (high task orientation vs. low task orientation). Figure 1 presents a three-dimensional representation of the TRIAD model. The horizontal (left to right) axis represents low versus high Task Orientation, the vertical (down to up) axis represents low versus high Sociability, and the third axis coming out from back to front represents low versus high Dominance. The assumption is that any role can be mapped onto this three-dimensional space. For example, we may expect that the Team Leader role would occupy a high Dominance, moderate Sociability, high Task Orientation space.

Figure 1.

The TRIAD model.

The Present Study

We believe that this model can be useful in several ways. First, it represents a seeming consensus on three primary behavioral dimensions that underlie team task behavior. Second, it provides a means by which to map team roles in order to determine core roles that occupy a similar conceptual space. In pursuit of this second goal, we developed a rating scale designed to tap each of the three dimensions of the TRIAD model. We utilized this scale to rate each of the 154 role descriptors identified by Gregory et al. (2015).¹ Cluster analysis is then used to determine the extent to which certain roles cluster in adjacent spaces in order to identify primary or core role categories.

Method

Cluster analysis is logically appropriate for suggesting underlying structures in data. We can take as an operating hypothesis that among the 154 distinct roles that have been drawn from the research literature, there are some roles that are similar to one another and which form a distinct subset, and that there are other roles that group together to form a subset unique from the first, and so on. That is, we assume that there is an underlying structure within the data that may be informative in terms of understanding the data. In the present case, the 154 roles can be sorted by similarity on the TRIAD dimensions of Dominance, Sociability, and Task Orientation. In performing a cluster analysis, the purpose is to join together objects into successively larger clusters according to their similarity, such that the clusters have maximum internal homogeneity and maximum between-cluster heterogeneity (Bortz, 2005). Or, as described by Bailey (1975), cluster analysis attempts to divide a set of objects into a smaller number of relatively homogeneous groups on the basis of their similarity over N variables. That is, cluster analysis is a means to draw boundaries in multidimensional space such that objects are grouped by their similarity on all variables considered simultaneously. Once significant clusters have been identified, this emergent structure may suggest an entirely new way of approaching the data. This approach has been applied in a variety of domains, including medical research (e.g., Moore et al., 2010), marketing research (Punj & Stewart, 1983), and psychological research (e.g., Kivlighan & Tibbits, 2012). For example, Meyer and Glenz (2013) describe a cluster-analysis-based faultline measure to quantify the subgroup structure of work teams.

Procedure

Two raters (doctoral-level researchers) independently rated each of the 154 role descriptions on the dimensions of dominance, sociability, and task orientation. The rating scale included three items describing each of these three dimensions. These items were chosen based on concurrence with Bales (2001), Couch (1960), and Couch and Carter (1952). Specifically, dominance was assessed by the following bipolar items: takes a leading role–takes a supporting role, directs activities–follows direction, and dominant behavior–passive behavior. Sociability was assessed by the following bipolar items: friendly–unfriendly, warm–cold, and interested in others–aloof. Finally, task orientation was assessed by the following bipolar items: hard-working–work-shy, focused on task–unfocused, and conscientious–careless. Each of the 154 role descriptions, from Gregory et al. (2015), was derived from the original source documents, similar to the following:

Encourager. Praises, agrees with and accepts the contribution of others. He indicates warmth and solidarity in his attitude toward other group members, offers commendation and praise and in various ways indicates understanding and acceptance of other points of view, ideas and suggestions. (Benne & Sheats, 1948)

Ratings were completed on a 7-point Likert-type scale. After scoring was completed, the scale was transformed from a range of 1 to 7 to a range of (−3) to (+3). This was done so that, for each item (such as dominance, for example), zero would then represent the midpoint of the scale, +3 would represent high dominance, and −3 would represent low dominance. This also allows scores to be more easily represented in a three-dimensional plane.

Interrater agreement was assessed by ICC(2) scores. Overall agreement was high with all items scoring over .85. Subsequently, the ratings of the two-raters were averaged to compute a single score for the nine total items. To obtain a single score for each facet, the three items representing each facet were averaged, thus creating an overall score for dominance, sociability, and task orientation. Cronbach’s alpha was used as justification for data aggregation. Alphas for dominance, sociability, and task orientation were .99, .99, and .99, respectively.

Results

First, we performed a hierarchical cluster analysis on the 154 roles extracted from the literature. Overall ratings for dominance, sociability, and task orientation were used as input variables in the analysis. The analysis using Ward’s method produced 13 clusters. The resultant dendrogram is represented in tabular form in Table 2. Table 2 shows the 13 clusters and the specific roles (i.e., the titles of the roles from the original sources) that comprise each cluster.

Table 2.

Role Clusters and Corresponding Role Descriptors.

Cluster 1Team leader	Cluster 2Task motivator	Cluster 3Power seeker	Cluster 4Critic	Cluster 5Attention seeker	Cluster 6Negative	Cluster 7Social	Cluster 8Coordinator	Cluster 9Follower	Cluster 10Teamwork support	Cluster 11Evaluator	Cluster 12Problem solver	Cluster 13Task completer
Chairman	Gives orientation	Ineffective Supporter	People diagnoser	Concealer	Blocker	People supporter	Conciliator	Listener	Information seeker	Ineffective collaborator	Orienter	Collaborator
Information giver	Manipulative/persuasive	Squelcher	Flaw-finder	Eager beaver	Shows tension	Ineffective challenger	Model member	Gatekeeper	Summarizer	Naysayer	Opinion seeker	Proceduralist
Explainer	Challenger	Dominator	Nonconformist	Playboy/playgirl	Pessimist	Mediator	Coordinator	Conformist	Conciliator	Evaluator	Opinion giver	Elaborator
Flight-operations	Talker	Challenger	Social pleader	Self-confessor	Debunker	Tension-reliever	Gate-keeper/expediter	Follower	Mediator	Monitor-evaluator	Information/opinion seeker	Mechanical/technical
Standard setter/ego ideal	Recognition seeker	Aggressive One	Brilliant one	Help-seeker	Bored one	Social	Social leadership	Follower	Ineffective knowledge contributor	Controller-inspector	Information/opinion giver	Intellectual/analytic
Organizer	Motivator	Shaper	Emotional one	Wannabe	Silent one	Personnel-support	Explorer-promoter	Assessor-developer	Recorder		Idea generator	Logical/precision
Leader	Energizer	Aggressor	Suspicious one	Spoiler		Encourager	Team worker	Communicator	Environmental-support		Gatekeeper	Scientific investigative
Envisioning leadership	Recognition seeker	Dominator	Blocker	Coat-tails		Harmonizer	Consul	Cooperator	Procedural technician		Imaginative/aesthetic	Production
Organizing leadership	Special interest pleader	Defiant Leader	Plant			Resource investigator		Agrees	Recorder		Evaluator-critic	Contributor
Spanning leadership	Energizer		Playboy			Helper of others			Group-observer/commentator		Compromiser	Completer-finisher
Thruster-organizer	Challenger		Creator-innovator			Upholder-maintainer			Questioner		Clarifier	Knowledge contributor
Concluder-producer	Calibrator		Scapegoat leader			Linker			Factual contributor		Idea creator
Task leader	Gives suggestion		Disagrees			Emotional leader			Agreeable enabler		Critical tester
Contractor			Shows antagonism			Communicator			Supportive worker		Critic
Contributor						Shows solidarity			Company worker		Coordinator
Initiator-contributor						Shows tension release			Reporter-adviser		Gives opinion
Creator									Completer		Asks for orientation
									Process observer		Asks for opinion
									Collaborator		Asks for suggestion

The hierarchical cluster analysis was supplemented by a two-step cluster analysis. Analogous to the hierarchical analysis, a 13-cluster solution provided the best fit for the data. The output from this analysis is provided in Table 3. Table 3 shows the rated levels of task orientation, sociability, and dominance for each cluster. These data points provide the necessary coordinates to map each cluster in a three-dimensional plane (see Figure 2). In this three-dimensional space, the midpoint of each dimension is zero, and the high end of the dimension is represented as +3, with the low end represented as −3. Thus, as shown in Figure 2, Cluster 5 is situated at the low end of the Task Orientation dimension (−2.46), and at the approximate midpoints of the Social dimension (0.00) and the Dominance dimension (0.50). For purposes of interpretation, we have chosen 2 to 3 to represent high scores on the scale, 0.75 to 1.9 to represent moderately high scores, 0.74 to −0.74 to represent average scores, −0.75 to −1.9 to represent moderately low scores, and −2 to −3 to represent low scores. Table 4 provides a summary description of each of the 13 role clusters.

Table 3.

Two-Step Cluster Analysis Results.

Cluster	1Team leader	2Task motivator	3Power seeker	4Critic	5Attention seeker	6Negative	7Social	8Coordinator	9Follower	10Teamwork support	11Evaluator	12Problem solver	13Task completer
Task	2.58	0.64	−0.43	−0.92	−2.46	−2.75	−0.03	1.69	0.56	2.24	2.30	1.28	2.64
Soc	0.04	−0.04	−2.43	−1.31	0.00	−2.22	2.84	2.15	1.24	0.11	−2.23	0.02	−0.08
Dom	2.35	1.96	2.13	−0.30	0.50	−2.22	−0.45	0.56	−2.39	−2.15	−0.10	−0.25	−0.56

Note. Task = task orientation; Soc = sociability; Dom = dominance.

Figure 2.

Three dimensional scatterplot of the 13 role clusters.

Table 4.

Summary Role Clusters.

Cluster	1Team leader	2Task motivator	3Power seeker	4Critic	5Attention seeker	6Negative	7Social	8Coordinator	9Follower	10Teamwork support	11Evaluator	12Problem solver	13Task completer
Description	Directs, organizes, and leads the team toward task accomplish-ment	Acts as a second in command or task manager to stimulate the group to action	Seeks power and recogni-tion through domin-ance	Adopts a critical, fault-finding, or devil’s advocate role	Seeks attention and resources but does little work	Displays negative affective behavior	Contributes to group harmony and maintenance of inter-personal relations	Coordinates task activities within the team and with those outside the team	Embraces followership activities of listening, agreeing, and cooperating	Supports team task accompli-shment	Inspects and evaluates team outputs	Involved in generation of ideas, problem solving.	Performs detail-oriented, routine activities related to complet-ing the task
Location
Task Orientation	High	Average	Average	Moderately low	Low	Low	Average	Moderately high	Average	High	High	Moderately high	High
Sociability	Average	Average	Low	Moderately low	Average	Low	High	High	Moderately high	Average	Low	Average	Average
Dominance	High	Moderately high	High	Average	Average	Low	Average	Average	Low	Low	Average	Average	Average
Behaviors	Guides, brings order, convenes the group, structures the task, commands and organizes	Prods the group to action, stimulates the group, energizes the group	Asserts authority or super-iority, interrupts and belittles others, seeks power	Disagrees, shows cynicism, judges others	Seeks attention without responsibility, seeks sympathy, expresses personal views	Shows tension, with-draws, com-plains	Relieves tension, mediates disagree-ments, praises others, shows warmth	Facilitates team functions, coordinates activities, facilitates participation	Participates in discussions, listens to others’ ideas, conforms to tasks, seeks cooperation	Pitches in to achieve team goals, implements plans, sums up and clarifies, records team activities	Analyses and evaluates propos-als, focuses on facts and figures	Asks for information, offers facts and opinion, clarifies inconsis-tencies, offers solutions	Adheres to proced-ures and deadlines, detail-oriented and orderly

The naming or labeling of clusters is a subjective process, similar to naming a factor in a factor analysis (see Kivlighan & Arseneau, 2009). However, representative names were selected for each cluster based on (a) the original labeling of the roles that comprise each cluster and (b) consideration of the themes and behavioral descriptors that described the original roles. The 13 derived role clusters include the following.

Cluster 1: Team Leader

Cluster 1 is characterized by high task orientation (2.58), average sociability (0.04), and high dominance (2.35). We define Cluster 1 as comprising the Team Leader role. This cluster is primarily defined by roles identified as leader, chairman, task leader, and organizing leadership. Descriptive behaviors (drawn from the role descriptions in the original team task role taxonomies) that define this role cluster include guiding and controlling, facilitates activities, brings order, convenes the group and acts as guide to task accomplishment, structures the task, organizes and coordinates, and commands.

Cluster 2: Task Motivator

Cluster 2 is characterized by average task orientation (0.64), average sociability (−0.04), and moderately high dominance (1.96). We define Cluster 2 as comprising the Task Motivator role. This cluster is defined by roles identified as energizer, challenger, motivator, and manipulative/persuasive. We view this cluster as representing a 2nd in command or task manager role that supports the team leader. Descriptive behaviors that define this role cluster include prods the group to action, stimulates the group to keep going, encourages the team, and energizes the team.

Cluster 3: Power Seeker

Cluster 3 is characterized by average task orientation (−0.43), low sociability (−2.43), and high dominance (2.13). We define Cluster 3 as comprising the Power Seeker role. This cluster is defined by roles identified as dominator, aggressor, and defiant leader. We view this cluster as representing a domineering or autocratic role that may have negative consequences for the team. Descriptive behaviors that define this role cluster include asserts authority or superiority, interrupts the contributions of others, belittles team members, attempts to seize control, opposes the leader, and is aggressive toward others.

Cluster 4: Critic

Cluster 4 is characterized by moderately low task orientation (−0.92), moderately low sociability (−1.31), and average dominance (−0.30). We define Cluster 4 as comprising the Critic role. This cluster is defined by roles identified as disagrees, shows antagonism, flaw-finder, and blocker. We view this cluster as representing a critical, fault-finding, or devil’s advocate type of role. Descriptive behaviors that define this role cluster include negativistic and stubborn, disagrees and opposes, shows cynicism, judges others, tends to go against the rest of the group, and may make others defensive. Note that the devil’s advocate role can be useful in decision making, but tends to be a negatively evaluated role.

Cluster 5: Attention Seeker

Cluster 5 is characterized by low task orientation (−2.46), average sociability (0.00), and average dominance (0.50). We define Cluster 5 as comprising the Attention Seeker role. This cluster is defined by roles identified as spoiler, coat-tails, concealer, and wannabe. We view this cluster as representing a primarily negative role in which the incumbent seeks attention from the group but shirks actual taskwork. Descriptive behaviors that define this role cluster include uses team to express personal views, seeks attention without responsibility, seeks attention and sympathy from group, expects others to do work, withholds information until it is personally advantageous, and leaves serious thinking to others.

Cluster 6: Negative

Cluster 6 is characterized by low task orientation (−2.75), low sociability (−2.22), and low dominance (−2.22). We define Cluster 6 as comprising the Negative role. This cluster is defined by roles identified as shows tension, pessimist, bored one, and silent one. We view this cluster as representing negative affective behavior. Descriptive behaviors that define this role cluster include shows tension, withdraws, has nothing to contribute, possesses a “can’t do” attitude, gripes and complains, erodes team spirit, and views effort as a waste of time.

Cluster 7: Social

Cluster 7 is characterized by average task orientation (−0.03), high sociability (2.84), and average dominance (−0.45). We define Cluster 7 as comprising the Social role. This cluster is defined by roles identified as harmonizer, helper of others, emotional leader, people supporter, and encourager. We view this cluster as representing positive social behavior, or behavior that contributes to group harmony and maintenance of socioemotional relations. Descriptive behaviors that define this role cluster include relieves tension and jokes, mediates disagreements, praises others, shows warmth and solidarity, promotes group cohesiveness, supportive of others, and maintains morale.

Cluster 8: Coordinator

Cluster 8 is characterized by moderately high task orientation (1.69), high sociability (2.15), and average dominance (0.56). We define Cluster 8 as comprising the Coordinator role. This cluster is defined by roles identified as coordinator, team worker, conciliator, gatekeeper, and consul. We view this cluster as representing coordinating task activities within the team and with those outside of the team. Descriptive behaviors that define this role cluster include facilitates team functions, coordinates activities of team members of subgroups, clarifies task relationships, keeps communication channels open, facilitates participation of others, and promotes ideas to others inside and outside the team.

Cluster 9: Follower

Cluster 9 is characterized by average task orientation (0.56), moderately high sociability (1.24), and low dominance (−2.39). We define Cluster 9 as comprising the Follower role. This cluster is defined by roles identified as follower, cooperator, agrees, listener, and communicator. We view this cluster as primarily comprised of taskwork behaviors of listening, agreeing, and cooperating. The Follower role is generally viewed as reciprocal to the Leader role. Descriptive behaviors that define this role cluster include participates in group discussion and decisions, builds on others’ ideas, effective listener and facilitator, conforms to assignments, seeks cooperation, and avoids disagreements.

Cluster 10: Teamwork Support

Cluster 10 is characterized by high task orientation (2.24), average sociability (0.11), and low dominance (−2.15). We define Cluster 10 as comprising the Teamwork Support role. This cluster is defined by roles identified as supportive worker, summarizer, recorder, process observer, company worker, and completer. We view this cluster as comprising behaviors that support team task accomplishment. Descriptive behaviors that define this role cluster include implements plans, does things for the group, makes records of group processes and output, takes initiative to ensure group’s success, puts information together, prepares for team meetings, sums up and clarifies, and pitches in to achieve team goals.

Cluster 11: Evaluator

Cluster 11 is characterized by high task orientation (2.30), low sociability (−2.23), and average dominance (−0.10). We define Cluster 11 as comprising the Evaluator role. This cluster is defined by roles identified as monitor-evaluator, controller-inspector, and evaluator. We view this cluster as comprising behaviors related to inspection and evaluation. Descriptive behaviors that define this role cluster include analyzes and evaluates proposals, is careful and meticulous, and focused on facts and figures of the task.

Cluster 12: Problem Solver

Cluster 12 is characterized by moderately high task orientation (1.28), average sociability (0.02), and average dominance (−0.25). We define Cluster 12 as comprising the Problem Solver role. This cluster is defined by roles identified as opinion seeker, opinion giver, clarifier, idea creator, information seeker, information giver, and critical tester. We view this cluster as comprised of behaviors related to idea production and evaluation, problem solving, and decision making. Descriptive behaviors that define this role cluster include orients the group to the task; asks for opinion; asks for information; asks for clarification; offers facts, information, and opinions; points out inconsistencies and clarifies them; synthesizes new ideas and comes up with solutions; and repeats, clarifies, and confirms.

Cluster 13: Task Completer

Cluster 13 is characterized by high task orientation (2.64), average sociability (−0.08), and average dominance (−0.56). We define Cluster 13 as comprising the Task Completer role. This cluster is defined by roles identified as completer-finisher, proceduralist, logical/precision, and production. We view this cluster as representing detail-oriented, routine behaviors oriented to completion of the task. Descriptive behaviors that define this role cluster include focused on details and deadlines, conscientious and orderly, adheres to responsibilities, focused on procedures, and performs explicit and routine tasks.

Discussion

This analysis integrates previous taxonomies within the team roles literature, provides further insight into the nature and structure of team roles, and provides some new directions for role research. Moreover, the role clusters identified resonate to the original roles literature. Consistent with the early work of Bales and colleagues (Bales, 1950; Bales & Slater, 1955), we capture the two primary roles of task leader (team leader role) and socioemotional leader (social role). Consistent with Benne and Sheats (1948), we capture the roles of energizer (task motivator role) and coordinator (coordinator role). Consistent with Belbin (1981, 1993), we capture the roles of completer-finisher (task completer role) and opinion giver (problem solver role). Furthermore, this classification captures both positive team roles as well as potentially negative team roles (power seeker, critic, attention seeker, and negative roles).

It is useful to discuss the relationship between team roles and team composition models. Team composition models describe how factors such as team roles may be configured to optimize team composition and, subsequently, team effectiveness. For example, Mathieu, Tannenbaum, Donsbach, and Alliger (2014) have presented an overview of four types of team composition models, offering different perspectives on how roles may affect team effectiveness. The first type of team composition model is a traditional personnel-position fit model that emphasizes the particular set of individual knowledge, skills, abilities and other characteristics (KSAOs) that contribute to successful performance in a specific position or role. The second type is a personnel model with teamwork considerations model that emphasizes team-generic competencies such as collective orientation (Driskell, Salas, & Hughes, 2010). The third type is a team profile model that considers team member’s KSAOs collectively in terms of how they are distributed in the team. The fourth type of team composition model is a relative contribution model that examines the relative or disproportional impact that some characteristics may have on team effectiveness, such as a negative or overly critical team member (Felps, Mitchell, & Byington, 2006).

Furthermore, Mathieu et al. (2014) propose that team composition can be more comprehensively examined by integrating these different team composition approaches. They provide the example in the context of replacing a team member. In this case, the examination of position-specific KSAOs would be consistent with a traditional person-fit model. The further examination of teamwork-oriented KSAOs would be consistent with a personnel with teamwork considerations model. The impact of the KSAOs of the team member replacement in terms of the distribution of KSAOs across the team would be consistent with a team profile model. Finally, the relative impact of the specific KSAOs subtracted and added to the team and the interrelationship with specific team roles would be consistent with a relative contribution model.

Moreover, this organizing framework suggests that we can examine the impact of team roles on team effectiveness in a similar manner. From a person-position fit perspective, we would focus on the individual team member and how he or she would fill specific roles to support team effectiveness. For example, we have proposed that the team leader role is comprised of behaviors related to high task orientation, average sociability, and high dominance. KSAOs related to this role may include comprehensive knowledge of the task, skills such as organizing and directing team members, as well as other characteristics such as initiative. Generally, a person who possesses these characteristics should be more adept at meeting team leader role demands.

Adopting a personnel model with teamwork considerations perspective, we would further examine the contributions of team competencies such as performance monitoring (Cannon-Bowers, Tannenbaum, Salas, & Volpe, 1995) or teamwork knowledge (Stevens & Campion, 1994). Although these types of team-generic skills are viewed as valuable for any team setting, we would propose that some team roles such as the social role may place a higher value on a competency such as interpersonal relations; whereas, another team role such as coordinator would place a higher value on a competency such as coordination.

A team profile perspective leads us to examine how team member’s characteristics collectively fit team requirements. For example, we would propose that effective teams require a balance within the three-dimensional space shown in Figure 2. That is, too much activity located in the leftmost dimension of the three-dimensional model reflects a team that is not task oriented, whereas too little activity in the uppermost dimension of the model suggests that team maintenance functions may be compromised. (Although it is difficult to represent on the printed page, the 3-D scattergram illustrated in Figure 2 can be rotated for visualization on each of its three axes.) An optimal team profile would involve team members enacting role activity that spreads over the high task orientation, high dominance, and high sociability spaces.

On the other hand, we may consider that a team profile in which multiple team members attempt to perform the same role may impair team effectiveness. Too many people performing a single role may be not only redundant but lead to inefficiencies and conflict. Dynes (1986) noted that, under emergency conditions, you may have more people trying to assume relevant roles than there are relevant roles to fill. Furthermore, in an examination of group performance in polar winter-overs, Johnson, Boster, and Palinkas (2003) found that role collision may occur when multiple individuals in a group attempt to perform roles which overlap.

Finally, a relative contribution perspective allows us to examine whether some roles may be more influential than others. For example, Mann (1961) and others have suggested that for teams performing a task that is primarily social, interpersonal, or emotionally nuanced, team maintenance activities may be more relevant; whereas, for teams performing a primarily instrumental task, task-oriented activities may be more relevant. Or, we may examine the interplay among roles. For example, the social role may be disproportionally important in a situation in which the task leader is especially dominant. Team roles that are negatively valenced (in the leftmost space in Figure 2) such as the critic, attention seeker, or negative roles may disproportionally affect team effectiveness in that negative emotionality can be contagious and spread throughout the team (Barsade, 2002).

In brief, the team composition models described by Mathieu et al. (2014) provide a perspective for examining the various ways in which team roles may affect team effectiveness. Whereas traditional person-fit models have been adopted to study team roles (Burke et al., 2017), team profile perspectives can suggest more complex analyses of how various roles may supplement or complement one another.

Research Implications

There are several implications for further research that can be drawn from the present study. Foremost, research is needed to empirically test and instantiate the model presented. First, can an independent set of raters obtain similar results to that of the expert raters in the present study? Second, can observers use the model presented in situ to distinguish among role behaviors in real-world groups? Third, we offer suggestions for the use of this model for examining personality-role fit, the impact of contextual factors such time or temporal dynamics, and effects of the type of task on team role requirements.

Early in the past century, Linton (1936, p. 476) observed that “It is vitally necessary for the functioning of society that the personalities of its members be at least superficially adapted to their statuses” (or roles). Research on personality-role fit is still sparse (DeRue & Morgeson, 2007). Personality-role fit refers to the compatibility between a team member’s personality characteristics and his or her role within the team. In terms of personality/role fit, we can imagine that those role clusters that are located in the high social space in Figure 2 (Cluster 7, Social; Cluster 8, Coordinator; Cluster 9, Follower; and Cluster 10, Teamwork Support) would all have a strong link to the big five extraversion or sociability personality dimensions. That is, a person scoring low on trait sociability is probably ill-suited to roles that fall within the upper zone of this three-dimensional space.

In fact, we may hypothesize linkages between personality and each of the broad TRIAD dimensions (see Driskell, Goodwin, Salas, & O’Shea, 2006; Stewart et al., 2005). For example, dominance may be predicted by the big five factor of extraversion (particularly the subfacets of dominance and ambition). Sociability may be predicted by the factors of agreeableness and low neuroticism, and cognate subfacets of sociability, warmth, and expressiveness. Task orientation may be predicted by the factor of conscientiousness and cognate subfacets of achievement-orientation and dependability. Empirical research is needed to examine these relationships.

Further research can utilize this model to examine role performance under various contexts or conditions. Note that there are no preferential spaces per se in the three-dimensional model—that is, a role that enacts low task orientation, low sociability, and high dominance (i.e., the power seeker role) may not be highly functional or supportive of team effectiveness under normal task conditions, but may perhaps be more functional under emergency conditions when authority needs to be asserted. Furthermore, a specific role, such as completer, may be more or less functional over time, such that the role may be prominent at Time A, but less so at Time B. In brief, teams and team roles are dynamic. Team performance conditions may change over time, team members may change over time, and the task itself may require different types of activities as it evolves over time. In fact, the importance of examining how some roles become more relevant or less relevant over time has been noted by Mann (1961) and others but remains an underresearched area of group dynamics.

As Driskell and Salas (2013) have noted, “it is important to note that almost any overall statement regarding teams must be qualified by factors such as the type of task” (p. 745). Holland (1966) has presented a RIASEC classification of task environments based on the activities that describe each category, including realistic, investigative, artistic, social, enterprising, and conventional tasks (see also Driskell, Hogan, & Salas, 1987). The visualization of role dimensions in the TRIAD model presented in Figure 2 suggests potential linkages between roles and types of task environments. For example, realistic tasks are production tasks that typically involve little interaction with others but may require the production and completion of activities. Realistic tasks may be most dependent on effective role enactment in the low dominance, low sociability, high task orientation space (i.e., problem solver, completer). Social tasks require personal interaction with others and may involve assisting or supporting other team members and behaviors related to maintaining team harmony. Social tasks may be most dependent on effective role enactment in the low dominance, high sociability, average task orientation space (i.e., social, coordinator). Further research is needed to elaborate and test these assumptions.

Applications

The examination of team role is relevant to various types of applied organizational activities, including the selection of team members, the removal of team members, and the replacement of team members (see Mathieu et al., 2014). In selecting team members, one should be cognizant of the type of team, the type of task, the type of roles to be filled, and the characteristics (KSAOs) of those who may potentially fill those roles. In general, the optimal strategy is to select the individuals that have the optimal KSAOs to match specific team roles, although there are certainly elaborations on this rule. For example, for a small team (two to three members) it may be advantageous to select generalists (that have moderate levels of task orientation, dominance, and sociability) who would be able to fulfill an idealized team role. Larger teams would provide the affordance to select more specialists, that is, persons who could fill separate task leader, social, problem solver, and other roles. Caution should be exercised in selecting too many high dominance task leaders or too many highly sociable members. At least one person fulfilling these roles may be ideal, but there may be diminishing returns with multiple team members attempting to lead or multiple team members engaging in social activity.

Our model suggests that the removal of some negative team roles may be beneficial to the team. For example, the power seeker role is characterized by very high dominant behavior and very low social behavior, which may represent an overbearing team member with little concern for others. The attention seeker and negative roles are characterized by very low task orientation, which suggests that persons fulfilling these roles contribute little to overall team goals.

In replacing team members because of turnover or attrition, the TRIAD model provides a means to examine role balance in the team. The 3-D representation of the 13 role clusters within the TRIAD conceptual space (Figure 2) provides insight into the topography of role performance. For example, the examination of Figure 2 reveals that some role clusters are relatively close in distance. Thus, we see that Cluster 10 (teamwork support) and Cluster 12 (problem solver) are close and thus somewhat related (suggesting, for example, that these roles could perhaps be performed by a single person). Figure 2 also provides insight on what role clusters are dissimilar and spatially distant. Thus, Cluster 7 (social) and Cluster 11 (evaluator) are spatially separated (and suggests that these two roles are more disparate and would be more difficult to perform by a single person). Furthermore, in terms of role shifts or role replacement, this model also provides potentially useful insight on how to replace or change a specific role incumbent (e.g., ideally, with the role incumbent that is spatially nearest to that location or cluster).

Team training interventions may also serve to support effective role performance. For example, cross-training of team members may enhance team member’s understanding of their own and other team members’ roles (Cannon-Bowers & Salas, 1998; Marks, Sabella, Burke, & Zaccaro, 2002). Ellis, Bell, Ployhart, Hollenbeck, and Ilgen (2005) found that generic teamwork skills training enhanced overall team performance, and that this training was more valuable for some roles than others. Other team training interventions that may support effective role performance include team building interventions (Salas, Rozell, Mullen, & Driskell, 1999) and facilitating knowledge building across team members (Rentsch, Delise, Salas, & Letsky, 2010).

One further applied research task is to identify the competencies, or KSAOs that underlie effective role performance for a given task. For example, the leader role requires both task-specific taskwork skills as well as role-relevant teamwork skills. That is, an effective leader possesses skills that are relevant to the specific task at hand (taskwork skills) as well as skills that are required to support effective teamwork (teamwork skills; Morgan, Glickman, Woodard, Blaiwes, & Salas, 1986; Stevens & Campion, 1994). In the model presented, we have tended to emphasize teamwork skills that are relevant across tasks. For example, leader role activities include organizing, explaining, giving information, and leadership behaviors (Figure 2). These types of teamwork behaviors can be trained generically. However, taskwork competencies (such as knowledge of the task requirements) are tied to the specific task. In an applied setting, it is necessary to define both the taskwork competencies as well as the teamwork competencies that are required to perform a given role.

Conclusion

The present research is an attempt to integrate a large and disparate number of team role descriptions within the research literature and derive a smaller, core set of team roles, using a cluster analytic approach. To accomplish this task, we first proposed a model that describes three primary behavioral dimensions that underlie team role behavior: (a) dominance, (b) sociability, and (c) task orientation. We use this three-dimensional model to map team roles in order to determine “core” roles that occupy a similar conceptual space. We believe that the value of this approach is twofold. First, we have attempted to reduce as complex data set into a smaller, homogeneous set of categories that represent a theoretically based and quantitatively derived set of 13 core team roles. We believe the resultant team role classification provides a reasonable approximation of the nature and structure of team roles. Second, the TRIAD roles model presented suggests a number of potentially valuable, yet untested, avenues to examine the linkage between role performance and different types of team tasks, team member personality, and team performance over time. Although, at this point, these links are only speculative, we believe that further research will yield greater insight into team role structure and performance.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by funding from the National Aeronautics and Space Administration (Grant NNX09AK48G) and the National Science Biomedical Research Institute (NCC-9-58-401/NBPF03402).

Notes

Author Biographies

Tripp Driskell is a research scientist at Florida Maxima Corporation, USA. He received his PhD in applied experimental and human factors psychology from the University of Central Florida.

James E. Driskell, PhD, is president and senior scientist at Florida Maxima Corporation, USA. His research examines team interaction in high demand environments.

C. Shawn Burke is a professor (research) at the University of Central Florida, USA. Her expertise includes team adaptability, team training, measurement/evaluation, and team effectiveness. All of the above work is conducted with an interest in team leadership and the training of teams operating in complex environments.

Eduardo Salas, PhD, is the professor and Allyn R. & Gladys M. Cline chair of psychology at Rice University. He received the 2016 American Psychological Association Award for outstanding lifetime contributions to psychology.

References

Bailey

K. D.

(1975). Cluster analysis. In Heise

D. L.

(Ed.), Sociological methodology (pp. 59-128). San Francisco, CA: Jossey-Bass. doi:10.2307/270894.

Bales

R. F.

(1950). A set of categories for the analysis of small group interaction. American Sociological Review, 15, 257-263. doi:10.2307/2086790

Bales

R. F.

(1970). Personality and interpersonal behavior. New York, NY: Holt, Rinehart and Winston.

Bales

R. F.

(2001). Social interaction systems: Theory and measurement. New Brunswick, NJ: Transaction Publishers.

Bales

R. F.

Slater

P. E.

(1955). Role differentiation in small decision-making groups. In Parsons

Bales

R. F.

(Eds.), Family, socialization and interaction process (pp. 259-306). Glencoe, IL: The Free Press.

Barsade

S. G.

(2002). The ripple effect: Emotional contagion and its influence on group behavior. Administrative Science Quarterly, 47, 644-675. doi:10.2307/3094912

Beauchamp

M. R.

Bray

S. R.

Eys

M. A.

Carron

A. V.

(2002). Role ambiguity, role efficacy, and role performance: Multidimensional and mediational relationships within interdependent sport teams. Group Dynamics: Theory, Research, and Practice, 6, 229-242. doi:10.1037/1089-2699.6.3.229

Beck

A. P.

Eng

A. M.

Brusa

J. A.

(1989). The evolution of leadership during group development. Group, 13, 155-164. doi:10.1007/BF01473148

Beehr

T. A.

(1976). Perceived situational moderators of the relationship between subjective role ambiguity and role strain. Journal of Applied Psychology, 61, 35-40. doi:10.1037/0021-9010.61.1.35

10.

Belbin

R. M.

(1981). Management teams. Why they succeed or fail. London, England: Butterworth-Heinemann.

11.

Belbin

R. M.

(1993). Team roles at work. Oxford, UK: Butterworth-Heinemann.

12.

Benne

K. D.

Sheats

(1948). Functional roles of group members. Journal of Social Issues, 4, 41-49. doi:10.1111/j.1540-4560.1948.tb01783.x

13.

Biddle

B. J.

(1986). Recent development in role theory. Annual Review of Sociology, 12, 67-92. doi:10.1146/annurev.so.12.080186.000435

14.

Borgatta

E. F.

(1963). A new systematic interaction observation system: Behavior scores system (BSs System). Journal of Psychological Studies, 14, 24-44.

15.

Bortz

(2005). Statistik für Human- und Sozialwissenschaftler [Statistics for the human and social sciences] (6th ed.). Heidelberg, Germany: Springer.

16.

Burke

C. S.

Driskell

J. E.

Howell

Marlow

Driskell

Salas

(2017, January). Team roles revisited. Presented at the 2017 Human Research Program Investigators Workshop (NASA), Galveston, TX.

17.

Cannon-Bowers

J. A.

Salas

(1998). Team performance and training in complex environments: Recent findings from applied research. Current Directions in Psychological Science, 7, 83-87. doi:10.1111/1467-8721.ep10773005

18.

Cannon-Bowers

J. A.

Tannenbaum

S. I.

Salas

Volpe

C. E.

(1995). Defining team competencies and establishing team training requirements. In Guzzo

Salas

(Eds.), Team effectiveness and decision making in organizations (pp. 333-380). San Francisco, CA: Jossey-Bass.

19.

Carter

L. F.

(1954). Evaluating the performance of individuals as members of small groups. Personnel Psychology, 7, 477-484. doi:10.1111/j.1744-6570.1954.tb01049.x

20.

Carter

L. F.

(1962). Recording and evaluating the performance of individuals as members of small groups. In Hare

A. P.

Borgatta

E. F.

Bales

R. F.

(Eds.), Small groups: Studies in social interaction (pp. 492-497). New York, NY: Knopf.

21.

Couch

A. S.

(1960). Psychological determinants of interpersonal behavior (Doctoral dissertation). Harvard University, Cambridge, MA.

22.

Couch

A. S.

Carter

L. F.

(1952, March). A factorial study of the rated behavior of group members. Paper presented at the meeting of the Eastern Psychological Association, Atlantic City, NJ.

23.

DeRue

D. S.

Morgeson

F. P.

(2007). Stability and change in person-team and person-role fit over time: The effects of growth satisfaction, performance, and general self-efficacy. Journal of Applied Psychology, 92, 1242-1253. doi:10.1037/0021-9010.92.5.1242

24.

Driskell

J. E.

Goodwin

G. F.

Salas

O’Shea

P. G.

(2006). What makes a good team player? Personality and team effectiveness. Group Dynamics, 10, 249-271. doi:10.1037/1089-2699.10.4.249

25.

Driskell

J. E.

Hogan

Salas

(1987). Personality and group performance. In Hendrick

(Ed.), Group process and intergroup relations (pp. 91-112). Thousand Oaks, CA: SAGE.

26.

Driskell

J. E.

Salas

(2013). Personality and work teams. In Christiansen

N. D.

Tett

R. P.

(Eds.), Handbook of personality at work (pp. 744-771). New York, NY: Routledge.

27.

Driskell

J. E.

Salas

Hughes

(2010). Collective orientation and team performance: Development of an individual differences measure. Human Factors, 52, 16-328. doi:10.1177/0018720809359522

28.

Dulewicz

(1995). A validation of Belbin’s team roles from 16PF and OPQ using bosses’ ratings of competence. Journal of Occupational and Organizational Psychology, 68, 81-99. doi:10.1111/j.2044-8325.1995.tb00574.x

29.

Dynes

R. R.

(1986). The concept of role in disaster research. Newark: Disaster Research Center, University of Delaware.

30.

Ellis

A. P.

Bell

B. S.

Ployhart

R. E.

Hollenbeck

J. R.

Ilgen

D. R.

(2005). An evaluation of generic teamwork skills training with action teams: Effects on cognitive and skill-based outcomes. Personnel Psychology, 58, 641-672. doi:10.1111/j.1744-6570.2005.00617.x

31.

Felps

Mitchell

T. R.

Byington

(2006). How, when, and why bad apples spoil the barrel: Negative group members and dysfunctional groups. Research in Organizational Behavior, 27, 175-222. doi:10.1016/S0191-3085(06)27005-9

32.

Fiske

S. T.

Cuddy

A. J.

Glick

(2007). Universal dimensions of social cognition: Warmth and competence. Trends in Cognitive Sciences, 11, 77-83. doi:10.1016/j.tics.2006.11.005

33.

Freedman

A. M.

(1996). The values and legacy of the founders of NTL: An interview with Ken Benne. Journal of Applied Behavioral Science, 32, 332-344. doi:10.1177/0021886396323007

34.

Gregory

Shimono

Burke

Salas

(2015). Development of a literature-driven, integrative team role taxonomy. Unpublished manuscript.

35.

Hare

A. P.

(1972). Four dimensions of interpersonal behavior. Psychological Reports, 30, 499-512. doi:10.2466/pr0.1972.30.2.499

36.

Heimer

Matsueda

R. L.

(1994). Role-taking, role commitment, and delinquency: A theory of differential social control. American Sociological Review, 59, 365-390. doi:10.2307/2095939

37.

Holland

J. L.

(1966). The psychology of vocational choice. Waltham, MA: Blaisdell.

38.

Johnson

J. C.

Boster

J. S.

Palinkas

L. A.

(2003). Social roles and the evolution of networks in extreme and isolated environments. Journal of Mathematical Sociology, 27, 89-121. doi:10.1080/00222500305890

39.

Kahn

R. L.

Wolfe

D. M.

Quinn

R. P.

Snoek

J. D.

Rosenthal

R. A.

(1964). Organizational stress: Studies in role conflict and ambiguity. New York, NY: John Wiley.

40.

Kivlighan

D. M.

Jr. Arseneau

J. R.

(2009). A typology of critical incidents in intergroup dialogue groups. Group Dynamics: Theory, Research, and Practice, 13, 89-102. doi:10.1037/a0014757

41.

Kivlighan

D. M.

Jr. Tibbits

B. M.

(2012). Silence is mean and other misconceptions of group counseling trainees: Identifying errors of commission and omission in trainees’ knowledge structures. Group Dynamics: Theory, Research, and Practice, 16, 14-34. doi:10.1037/a0026558

42.

Kreps

G. A.

Bosworth

S. L.

(1993). Disaster, organizing, and role enactment: A structural approach. American Journal of Sociology, 99, 428-463. doi:10.1086/230270

43.

Linton

(1936). The study of man: An introduction. New York, NY: Appleton-Century-Crofts.

44.

Mann

R. D.

(1961). Dimensions of individual performance in small groups under task and social-emotional conditions. Journal of Abnormal and Social Psychology, 62, 674-682. doi:10.1037/h0041180

45.

Marks

M. A.

Sabella

M. J.

Burke

C. S.

Zaccaro

S. J.

(2002). The impact of cross-training on team effectiveness. Journal of Applied Psychology, 87, 3-13. doi:10.1037/0021-9010.87.1.3

46.

Mathieu

J. E.

Tannenbaum

S. I.

Donsbach

J. S.

Alliger

G. M.

(2014). A review and integration of team composition models: Moving toward a dynamic and temporal framework. Journal of Management, 40, 130-160. doi:10.1177/0149206313503014

47.

Mathieu

J. E.

Tannenbaum

S. I.

Kukenberger

M. R.

Donsbach

J. S.

Alliger

G. M.

(2015). Team role experience and orientation: A measure and tests of construct validity. Group & Organization Management, 40, 6-34. doi:10.1177/1368430201004002005

48.

Mead

G. H.

(1934). Mind, self and society. Chicago, IL: The University of Chicago Press.

49.

Meyer

Glenz

(2013). Team faultline measures a computational comparison and a new approach to multiple subgroups. Organizational Research Methods, 16, 393-424. doi:10.1177/1094428113484970

50.

Moore

W. C.

Meyers

D. A.

Wenzel

S. E.

Teague

W. G.

. . . Bleecker

E. R.

(2010). Identification of asthma phenotypes using cluster analysis in the severe asthma research program. American Journal of Respiratory and Critical Care Medicine, 181, 315-323. doi:10.1164/rccm.200906-0896OC

51.

Moreno

(1934). Who shall survive? A new approach to the problem of human interrelations. Washington, DC: Nervous and Mental Disease.

52.

Morgan

B. B.

Jr. Glickman

A. S.

Woodard

E. A.

Blaiwes

A. S.

Salas

(1986). Measurement of team behaviors in a Navy environment (NTSC Tech. Rep. No. 86-014). Orlando, FL: Naval Training Systems Center.

53.

Moxnes

(1999). Understanding roles: A psychodynamic model for role differentiation in groups. Group Dynamics: Theory, Research, and Practice, 3, 99-113. doi:10.1037/1089-2699.3.2.99

54.

Mudrack

P. E.

Farrell

G. M.

(1995). An examination of functional role behavior and its consequences for individuals in group settings. Small Group Research, 26, 542-571. doi:10.1177/1046496495264005

55.

Mumford

T. V.

Campion

M. A.

Morgeson

F. P.

(2006). Situational judgment in work teams: A team role typology. In Weekley

J. A.

Ployhart

R. E.

(Eds.), Situational judgment tests: Theory, measurement, & application (pp. 319-343). Mahwah, NJ: Lawrence Erlbaum.

56.

Mumford

T. V.

Van Iddekinge

C. H.

Morgeson

F. P.

Campion

M. A.

(2008). The team role test: Development and validation of a team role knowledge situational judgment test. Journal of Applied Psychology, 93, 250-267. doi:10.1037/1089-2699.3.2.99

57.

Neiman

L. J.

Hughes

J. W.

(1951). Problem of the concept of role: A re-survey of the literature. Social Forces, 30, 141-149. doi:10.2307/2571625

58.

Nicholson

(1984). A theory of work role transitions. Administrative Science Quarterly, 29, 172-191. doi:10.2307/2393172

59.

Parsons

Bales

R. F.

Shils

E. A.

(Eds.) (1953). Working papers in the theory of action. Glencoe, IL: The Free Press.

60.

Punj

Stewart

D. W.

(1983). Cluster analysis in marketing research: Review and suggestions for application. Journal of Marketing Research, 20, 134-148. doi:10.2307/3151680

61.

Rentsch

J. R.

Delise

L. A.

Salas

Letsky

M. P.

(2010). Facilitating knowledge building in teams: Can a new team training strategy help? Small Group Research, 41, 505-523. doi:10.1177/1046496410369563

62.

Rizzo

J. R.

House

R. J.

Lirtzman

S. I.

(1970). Role conflict and ambiguity in complex organizations. Administrative Science Quarterly, 15, 150-163. doi:10.2307/2391486

63.

Salas

Rozell

Mullen

Driskell

J. E.

(1999). The effect of team building on performance: An integration. Small Group Research, 30, 309-329. doi:10.1177/104649649903000303

64.

Stempfle

Hübner

Badke-Schaub

(2001). A functional theory of task role distribution in work groups. Group Processes & Intergroup Relations, 4, 138-159. doi:10.1177/1368430201004002005

65.

Stevens

M. J.

Campion

M. A.

(1994). The knowledge, skill, and ability requirements for teamwork: Implications for human resource management. Journal of Management, 20, 503-530. doi:10.1016/0149-2063(94)90025-6

66.

Stewart

G. L.

Fulmer

I. S.

Barrick

M. R.

(2005). An exploration of member roles as a multilevel linking mechanism for individual traits and team outcomes. Personnel Psychology, 58, 343-365. doi:10.1111/j.1744-6570.2005.00480.x

67.

Stryker

Statham

(1985). Symbolic interaction and role theory. In Lindzey

Aronson

(Eds.), Handbook of social psychology (Vol. 1, pp. 311-378). New York, NY: Random House.

68.

Turner

R. H.

(1980). Strategy for developing an integrated role theory. Humboldt Journal of Social Relations, 7, 123-139.

69.

Turner

R. H.

(1990). Role change. Annual Review of Sociology, 16, 87-110. doi:10.1146/annurev.so.16.080190.000511

70.

Williams

K. C.

Morgan

Cameron

B. A.

(2011). How do students define their roles and responsibilities in online learning group projects? Distance Education, 32, 63-76. doi:10.1080/01587919.2011.565498