562000

research-article2014
GOMXXX10.1177/1059601114562000Group & Organization ManagementMathieu et al.

Article
Group & Organization Management
2015, Vol. 40(1) 6­–34
Team Role Experience © The Author(s) 2014
Reprints and permissions:
and Orientation: A sagepub.com/journalsPermissions.nav
DOI: 10.1177/1059601114562000
Measure and Tests of gom.sagepub.com

Construct Validity

John E. Mathieu1, Scott I. Tannenbaum2,

Michael R. Kukenberger3, Jamie S. Donsbach2,
and George M. Alliger2

Abstract
We review and synthesize previous team research and suggest that individuals’
previous experiences and orientations combine to yield predispositions to
occupy six different team roles, which we refer to as Team Role Experience
and Orientation (TREO) dimensions. We report the development of a
survey measure of TREO dimensions and establish its content validity using
a sample of subject matter experts’ item classifications. Furthermore, we
provide evidence that TREO dimensions are distinguishable from, but related
to, measures of the “Big 5” personality constructs. We also illustrate the
temporal stability of the measures. Moreover, we test the predictive validity
of TREO scores as related to peer ratings of members’ behaviors during
team activities. We discuss future theoretical and research implications of
TREO dimensions, and potential future applications of the measure.

Keywords
group or team composition or diversity, personality, roles or role behavior,
teams or teamwork

1University of Connecticut, Storrs, USA

2The Group for Organizational Effectiveness, Albany, NY, USA
3University of New Hampshire, Durham, USA

Corresponding Author:
John E. Mathieu, School of Business, University of Connecticut, 2100 Hillside Road, Unit
1041MG, Storrs, CT 06269-1041, USA.
Email: Jmathieu@business.uconn.edu
Mathieu et al. 7

Teams are widely recognized as the basic building blocks of most modern-
day organizations (Ilgen, Hollenbeck, Johnson, & Jundt, 2005; Kozlowski &
Ilgen, 2006; Mathieu, Maynard, Rapp, & Gilson, 2008). Team-based designs
enable organizations to quickly align their human resources with the multi-
tude of changing work demands and competitive pressures. Enhancing team
effectiveness offers a powerful means by which organizations can gain and
maintain competitive advantage. Team effectiveness can be driven by a num-
ber of factors such as a supportive organizational environment, team-oriented
external leadership, design features, dynamic processes and emergent states,
and a host of other variables (Mathieu et al., 2008). However, research and
practice have suggested that the best teams are well designed up-front. Teams
that have an optimal mix of members’ knowledge, skills, abilities, and other
characteristics (KSAOs) are better positioned to work well together and to
perform effectively than are teams composed of a less-optimal combination
of members (Bell, 2007; Ilgen, 1999). In short, team composition serves as
the foundation upon which other team factors are built, and represents a key
enabling feature of teams. Moreover, understanding how a team is “com-
posed” can provide insights for targeted team development activities.
Numerous characteristics have been used to index team composition,
including personality, functional expertise, competencies, goal orientations,
teamwork orientations, and a host of other attributes (Klimoski & Zukin,
1999; Mathieu, Tannenbaum, Donsbach, & Alliger, 2014). Importantly, these
individual attributes motivate and enable individuals to occupy different
team roles (Stewart, Fulmer, & Barrick, 2005). A role is generally defined as
a cluster of related and goal-directed behaviors taken on by a person within a
specific situation (Stewart, Manz, & Sims, 1999). Teams rely on different
members to fulfill different critical needs such as organizing work, maintain-
ing group harmony, and aligning their efforts with those of others in an orga-
nization (Aritzeta, Swailes, & Senior, 2007; Stewart et al., 2005). Accordingly,
both research and practice will benefit from a greater understanding of indi-
vidual differences that are associated with team role fulfillment, and from
tools to assess those differences.
Although the existing literature has advanced numerous team role taxono-
mies, we believe that there have been a number of major barriers to a produc-
tive discourse on the topic. Team roles are largely considered a critical part of
effective teaming (Ancona & Caldwell, 1988; Belbin, 1993) and viewed as a
basic feature of work teams (Sundstrom, De Meuse, & Futrell, 1990). Yet, the
topic has received relatively little empirical attention beyond the introduction
of team role classifications. Although a number of taxonomies of team roles
exist in the literature (Bales, 1950; Belbin, 1981, 1993; Benne & Sheats, 1948;
Parker, 1994; Humphrey, Morgeson, & Mannor, 2009), work in this area is
8 Group & Organization Management 40(1)

stifled due to a lack of integration and the absence of publicly available and
construct valid measures of role propensities. Accordingly, we have several
goals for this article. First, we briefly review and synthesize previous work that
has focused on team roles. From this, we distill six different roles that have
potential widespread applicability. Second, we argue that individuals’ previous
experiences, on one hand, and personal orientations and predilections, on the
other hand, combine to yield predispositions to occupy these different types of
team roles. We suggest that this synthesis of preferences and experiences rep-
resents individual differences that are predictive of individuals’ future role-
related behaviors in teams. We refer to these individual differences as Team
Role Experience and Orientation (TREO) dimensions. Third, we describe the
potential relations between TREO dimensions and other variables in a nomo-
logical network, and suggest that TREO dimensions are distinguishable from,
but related to, measures of traditional “Big 5” personality dimensions.
Fourth, we describe the development and validation of a survey measure of
TREO dimensions. This effort includes qualitative grounding and content
validity work by subject matter experts (SMEs) to generate and classify survey
items. It also involves gathering data from two development samples, evaluat-
ing and refining the construct validity of the items, and then administering the
revised measures to two validation samples. Fifth, we evaluate the discriminant
validity of the TREO measures from measures of Big 5 personality dimensions,
and test their pattern of relations using meta-analyses. Sixth, we demonstrate
the stability of TREO scores over time using a fifth sample. And finally, we test
the predictive validity of TREO measures as related to peers’ ratings of role-
related behaviors exhibited over the course of a 2-month long team business
simulation exercise. We conclude with a discussion of our findings in terms of
the nomological network associated with team’s role-related behaviors and
potential applications of TREO measures in the future.
In sum, the primary contributions of this work are threefold. We synthe-
size the diverse prior team role taxonomies and offer a six-dimensional
framework designed to balance comprehensiveness with parsimony. Second,
we develop measures of those six dimensions and provide evidence of their
psychometric properties, construct validity, and predictive validity. The items
are presented in the appendix and are freely available for research purposes.
Finally, we outline an agenda for future research incorporating the six-dimen-
sional role taxonomy.

Team Role Theories

Roles are often considered to be one of the fundamental and defining features
of both organizations (Ilgen & Hollenbeck, 1991; Sluss, van Dick, &
Mathieu et al. 9

Thompson, 2011) and teams (Hackman, 1990). Aritzeta et al. (2007) noted
that there are two heritages in the team role literature. One approach, which
we could term role as position, equates roles with expected behavior associ-
ated with the particular position that a team member occupies (e.g., Katz &
Kahn, 1978). Essentially, this view focuses on the characteristics and
demands of jobs and how they give rise to certain expected role behaviors of
occupants. A second approach, “role as person,” suggests that roles can be
defined as a combination of the values, attitudes, and behaviors of individu-
als who occupy particular locations in a social network. From this perspec-
tive, roles emerge from a combination of members’ natural inclinations or
preferences, as well as the social-psychological dynamics of the group (Ilgen
& Hollenbeck, 1991). We adopt this latter approach because we are interested
in developing indices of individual differences that may predispose people to
fulfill particular roles in teams.
Early research attempted to outline typologies of roles that team members
take when interacting with others (e.g., Bales, 1950; Benne & Sheats, 1948).
Benne and Sheats (1948) examined small discussion groups that were
engaged in problem-solving activities. They observed the appearance of two
types of roles: Task and Maintenance. Behaviors such as facilitating and
coordinating group activities, suggesting new ideas and ways of solving
problems fell under the “task role.” In contrast, behaviors that encouraged
praising, agreeing, and accepting the contribution of others within the group
were considered part of the “maintenance role.” Bales (1950, 1970) built on
this team role research by analyzing the interactions between members of
small groups and categorized their behaviors into task-oriented and socio-
emotional categories. Task roles are held by those group members who
engage in behaviors that are designed to facilitate task completion. Task
activities involve behaviors such as delegation and the coordination of group
communication. Socio-emotional leaders are those group members who
engage in behaviors that are designed to facilitate positive relations within
the group. Socio-emotional activities involve behaviors such as encourage-
ment and compromising. Interestingly, Bales and colleagues argued that task
and relationship roles emerge as a result of one another in that when task
leaders work toward the completion of goals, they engage in behaviors that
do not promote positive group relations, such as criticizing and giving orders.
As such, most successful groups would have some members specializing in
task activities and other members specializing in socio-emotional activities to
balance the group dynamics. These early studies centered on the individual’s
behaviors within a group and the classification of these behaviors into broader
roles. Benne and Sheats (1948) and Bales (1950) provided insight into two
critical functions that roles are instrumental in performing: task execution
10 Group & Organization Management 40(1)

and the encouragement of strong interpersonal relations. Another primary

contribution made by these researchers is that their work demonstrated the
utility of using role constructs to cluster team members’ behaviors.
The interest in team roles gained momentum in the 1980s with the publi-
cation of Belbin’s (1981) work on successful management teams. Belbin’s
(1981) theory advanced eight distinct team role types: (a) idea generator, (b)
resource investigator, (c) chairman, (d) shaper, (e) monitor evaluator, (f) team
worker, (g) company worker, and (h) completer–finisher. In later editions, he
changed various names (i.e., chairman to coordinator, company worker to
implementer) and introduced a new role called specialist. Belbin (1981,
1993) examined management teams playing executive simulations (e.g.,
computerized management and business exercises) during training courses
where team performance was measured in terms of winning or losing.
Belbin’s team role model has been associated with both team behaviors
and performance (for a recent review, see Aritzeta et al., 2007). Belbin and
others have argued that successful teams that have all nine roles present,
although individuals may display multiple roles, will be more balanced,
which will result in higher levels of success. Although a number of studies
have reported nonsignificant results (Anderson & Sleap, 2004; Jackson,
2002; Partington & Harris, 1999; Sommerville & Dalziel, 1998; Rushmer,
1996), other research has yielded positive relationships between such balance
and management styles (Lessem & Baruch, 2000), team performance
(Aritzeta & Ayestaran, 2003; Senior, 1998), cognitive styles (Aritzeta, Senior,
& Swailes, 2005), and to the exercise of power and control (Fisher, Hunter, &
Macrosson, 2001). Specifically related to team performance, in a small sam-
ple (n = 11), Senior (1997) provides some support for the link Belbin makes
between team role balance and team performance. Aritzeta and Ayestaran
(2003) demonstrate positive evidence for the team role balance argument as
the majority of their mostly female work teams were balanced and showed
high performance. However, using the same criteria, Park and Bang (2002)
found that less than 5% of their largely male work teams were balanced, and
did not find evidence to support that team role balance results in higher per-
formance. Gender differences between both studies could explain differences
observed with respect to team role balance and some of the inconsistent
results across Belbin’s team role studies.
In addition to the possible gender basis, the lack of consistent results
across studies may be related to the type of team being assessed and the origi-
nal sample and context. Belbin’s work focused extensively on management
teams, and while certainly of interest, others have argued that top manage-
ment teams are qualitatively different from other types of teams (Hollenbeck,
Beersma, & Schouten, 2012). Hambrick (1994) went so far as to refer to them
Mathieu et al. 11

as top management groups. Suffice it to say that a taxonomy of management

team roles may have limited applicability to other types of teams.
Furthermore, the psychometric properties of the scale may contribute to
the inconsistent results and, importantly, question the validity of the scale.
The team role model has predominately been measured through the Team
Role Self Perception Inventory (TRSPI-8R; TRSPI-9R), in some cases along-
side an Observer Assessment Sheet (OAS) where teammates could rate each
other. As originally conceived and implemented, Belbin’s inventory utilizes
ipsative responses, which have demonstrated weak internal consistencies
(Furnham, Steele, & Pendleton, 1993) and yield a contaminated covariance
matrix, undermining efforts to evaluate its factor structure (Cheung & Chan,
2002). Given the popularity of the Belbin role theory, many studies have
attempted to validate the scale (cf. Aritzeta et al., 2007; Broucek & Randell,
1996; Dierendonck & Groen, 2011; Fisher et al., 2001; Furnham et al., 1993;
Senior, 1998; Smith, Polglase, & Parry, 2012). In their review of the research
examining Belbin’s scale, Aritzeta et al. (2007) notes the following:

Having reviewed psychometric studies it is clear that neither the eight role nor
the nine role version of the TRSPI has unequivocal psychometric support and
most of the studies show low or at best average effect sizes indicating only
partial psychometric support. (p. 110)

This lack of psychometric support is rather surprising given the widespread

use and popularity of the Belbin scale.
Margerison and McCann (1985) developed a model containing eight roles
located on four dimensions, which the researchers referred to as (a) relation-
ships, (b) information, (c) decision making, and (d) organization (for a
detailed review of Margerison and McCann’s work, see Sadler-Smith, 2001).
McCann and Margerison (1989) also developed a team role measure that has
eight types: (a) explorer–promoter, (b) assessor–developer, (c) thruster–orga-
nizer, (d) concluder–producer, (e) checker–inspector, (f) upholder–main-
tainer, (g) reporter–advisor, and (h) creator–innovator. While recognizing
team behavior as being related to the task and the sociological factors (includ-
ing the roles that people play, and the norms and values operating), Margerison
and McCann argue that individual preferences that people bring to teamwork
are key psychological factors influencing individual and team performance.
Like Belbin, they suggest that a team should have a balance of required
behaviors and role preferences. Although this measure has also received
practitioner attention, there seems not to be rigorous empirical evidence to
support the balance hypotheses related to these roles. This measure is heavily
influenced by the Jungian theories underlying the Myers-Briggs test (Myers,
12 Group & Organization Management 40(1)

McCaulley, & Most, 1985). It not only has norms but also has evidence of
internal reliability and concurrent validity; however, there appears to be little
or no evidence of the factorial structure of the measure (to confirm the clas-
sification or taxonomic scheme), or any evidence of the predictive or con-
struct validity of the test.
Parker (1994, 1996) developed a set of team player styles, which fall into
four types: (a) contributor, (b) collaborator, (c) communicator, and (d) chal-
lenger. In contrast to Belbin’s emphasis on preferred team roles, Parker
focused on personal style—a function or process may be performed in a vari-
ety of ways depending on the team members’ personal style. In her review of
team roles, Sadler-Smith (2000) note that according to Parker, an individual
who approaches a function such as planning with a Contributor style, may
adopt a tactical, statistical, specific, measurable, and conservative approach,
whereas an individual with a Collaborator style facing the same process
would use a strategic, visionary, open, and involving approach. While
Margerison and McCann’s and Belbin’s team roles are based on Jungian tra-
dition (Jung, 1923), Parker argued that his style-based team roles provide
more latitude for different members taking different team roles and feedback
for improvement. There is little research exploring the applicability or valid-
ity of Parker’s work, with the exception of Kirnan and Woodruff’s (1994)
study that found evidence of construct reliability and validity of peer to
self-ratings.
Barry (1991) classified leadership behavior approaches within self-man-
aged teams. He described the following four types of leadership roles: (a)
envisioning leadership (innovative and vision fostering), (b) organizing lead-
ership (giving orders on missions), (c) spanning leadership (facilitating the
activities that connect the team to the organization), and (d) social leadership
(developing and maintaining the psychology and sociability of the team).
DuBrin (1995) advanced 10 team roles: (a) knowledge contributor, (b) pro-
cess observer, (c) collaborator, (d) people supporter, (e) challenger, (f) lis-
tener, (g) summarizer, (h) conciliator, (i) mediator, and (j) gate keeper. No
validation evidence has been provided in support of Barry’s or DuBrin’s
largely practitioner-based leadership roles.
Mumford, Van Iddekinge, Morgeson, and Campion (2006) attempted to
address the fragmentation in the team role literature by sorting existing roles
into categories they suggested captured the essence of each role. Their pro-
cess fit 120 roles into 10 unique team knowledge roles that they in turn clus-
tered into three broader categories of task, social, and boundary-spanning
roles. Task roles share the common function of carrying out the work that
constitutes the team’s objective and included (a) contractor, (b) creator, (c)
contributor, (d) completer, and (e) critic. Social roles involve maintaining the
Mathieu et al. 13

social environment in which teams function and included sub-dimensions of

(a) communicator, (b) cooperator, and (c) calibrator. The communicator role,
for example, encompasses behaviors that create a social environment that is
positive, open, and conducive to collaboration. One of the key theoretical
contributions of Mumford et al. (2006) is that they included Ancona’s
(Ancona & Caldwell, 1988, 1992) work by integrating roles that interact both
internally and externally to the team into their typology. Boundary-spanning
roles address members’ effort to align their team’s actions with those of indi-
viduals outside of the team and include a counselor and a coordinator sub-
role. T. V. Mumford, Van Iddekinge, Morgeson, and Campion (2008) then
developed situational judgment measures of members’ team role knowledge
and provided evidence of their validity. However, it is not clear the extent to
which having knowledge of certain team roles relates to whether individuals
will occupy or perform such roles in practice. In addition, Mumford created
role-category level variables by collapsing roles into three categories: task,
social, and boundary-spanning roles, which raises questions regarding the
validity of their 10 specific roles.

TREO Classification and Dimensions

We believe that there is a long and valuable history exploring team roles. Yet,
in a recent review of the literature, Aritzeta and colleagues (2007) questioned
the existence of the team roles: “…we raise the question about the real exis-
tence of nine well-differentiated team roles and whether these team roles, in
fact, are better differentiated using some other grouping suggested in the lit-
erature” (p. 105). Accordingly, in Table 1, we list and compare the types of
roles that have been discussed by the different authors noted above and offer
a synthesis in terms of six unifying themes or orientations. We submit that an
individual’s role propensities or predispositions are likely to be a function of
their previous experiences and orientations. This view rests on the assump-
tion that past behavior is a good indicator of future behavior (cf. Brown,
1978; Mumford & Owens, 1984). Work history measures also tend to yield
consistent factor structures over time and across samples (cf. Davis, 1984;
Neiner & Owens, 1982). By “orientations,” we mean individual, natural
inclinations that may well reflect personality or other individual differences
(Stewart et al., 2005). All else being equal, we would anticipate these orienta-
tions or preferences to guide individuals’ behaviors. However, we recognize
that different people may have different opportunities to occupy various team
roles as a function of the positions to which they have been assigned and of
the mix of other team members’ and their respective inclinations. Thus, we
believe that individuals’ predispositions to occupy various roles in future
14
Table 1. Previous Literature Integrated Within the TREO.
Roles studied within previous team role research

McCann and
TREO Benne and Sheats Belbin Ancona and Margerison Barry DuBrin Parker Mumford et al.
roles (1948) Bales (1950) (1981, 1993) Caldwell (1988) (1989) (1991) (1995) (1994) (2006)

Organizer Coordinator; Gives and asks Coordinator, Assessor Organizing Collaborator, Collaborator Contractor,
group observer; for orientation shaper thruster summarizer calibrator
gate keeper opinion,
suggestion
Doer Procedural, Completer– Concluder, Contributor Completer,
technician, finisher, controller, creator,
recorder specialist, reporter contributor
implementer
Challenger Evaluator-critic Disagrees Monitor, Challenger Challenger Critic
evaluator
Innovator Plant/idea Creator Envisioning Knowledge Innovator
generator contributor
Team Encourager, Shows solidarity, Team/company Upholder, People Communicator,
Builder follower, tension release, worker social supporter, cooperator
compromiser, agrees listener
harmonizer
Connector Resource Task coordinator, Explorer Spanning Collector,
investigator scout, consult
ambassador,
guard

Note. TREO = Team Role Experience and Orientation.

Mathieu et al. 15

teams will be a joint function of what they have done in past teams (past
experience), as well as roles that they would like to occupy (behavioral pre-
dispositions). Taking both an orientation and experience approach allows us
to consider both work history and natural inclinations providing a more holis-
tic method to measuring individual team roles. We believe that the result will
increase the likelihood of clearly validated team roles.
Table 1 lists the roles identified by previous theorists clustered into six
TREO categories. As illustrated, the various sources list up to 10 specific
types of roles, yet empirical evidence has yet to demonstrate their discrimi-
nant validity (see Aritzeta et al., 2007, for a detailed review). In short, we
doubt whether there are 10+ empirically distinguishable team roles that
members may occupy. Alternatively, many approaches have reduced their
role profiles to two or three clusters (e.g., task vs. maintenance) that may
obscure important differences. We endeavored to strike a balance between
comprehensiveness and inclusion versus discriminant validity and parsi-
mony, and propose a six-dimensional framework. Admittedly, this synthesis
represents our beliefs about the number and nature of important team roles,
based on our collective experience working with hundreds of different teams
across a wide variety of settings. Yet, this framework is also grounded in the
extant literature as detailed below. Table 2 lists the TREO dimensions and
their associated definitions. Notably, the TREO dimensions are designed to
represent individuals’ general propensities to fulfill different team roles inde-
pendent of any particular team context. In other words, we envision these
dimensions as reflecting individual differences in propensities, not situated
preferences or self-reports of behavior in a given context.
We distilled two team roles that are primarily directed at getting task work
accomplished. First, we define “Organizer” as someone who acts to structure
what the team is doing. An Organizer also keeps track of accomplishments
and how the team is progressing relative to team and individual goals and
timelines. This dimension is prominent in Benne and Sheats’ (1948) coordi-
nator and gate keeper dimensions, Barry’s (1991) organizer, and DuBrin’s
(1995) and Parker’s (1994) collaborator facet. Team success rests not only on
the extent to which members can process information and coordinate their
actions but also on the sheer fact that work gets accomplished. In other words,
getting work done is a prime role feature in teams. This may be manifested
differently in different teams working on different tasks, but the common ele-
ment is that some members need to devote themselves to getting task work
done. Accordingly, we define the second task work oriented dimension as
“Doer” as someone who willingly takes on work and gets things done. A
Doer can be counted on to complete work, meet deadlines, and take on tasks
to ensure the team’s success. This theme was evident in Benne and Sheats’s
16 Group & Organization Management 40(1)

Table 2. Team Role Definitions.

Role Definition
Organizer Someone who acts to structure what the team is doing. An
Organizer also keeps track of accomplishments and how the
team is progressing relative to goals and timelines.
Doer Someone who willingly takes on work and gets things done.
A “Doer” can be counted on to complete work, meet
deadlines, and take on tasks to ensure the team’s success.
Challenger Someone who will push the team to explore all aspects
of a situation and to consider alternative assumptions,
explanations, and solutions. A Challenger often asks “why”
and is comfortable debating and critiquing.
Innovator Someone who regularly generates new and creative ideas,
strategies, and approaches for how the team can handle
various situations and challenges. An Innovator often offers
original and imaginative suggestions.
Team Builder Someone who helps establish norms, supports decisions, and
maintains a positive work atmosphere within the team. A
Team Builder calms members when they are stressed, and
motivates them when they are down.
Connector Someone who helps bridge and connect the team with
people, groups, or other stakeholders outside of the team.
Connectors ensure good working relationships between
the team and “outsiders,” whereas Team Builders work to
ensure good relationship within the team.

(1948) cluster of procedural, technician, and recorder dimensions; McCann

and Margerison’s (1989) controller, concluder, and reporter bundle; and
Mumford et al.’s (2006) and Belbin’s (1993) completer–finisher themes.
Based on our review, we also distilled two socio-emotional types of roles.
Effectively managing members’ interpersonal processes is critical to team
success (Marks et al., 2001), and teams are social mechanisms for integrating
the expertise and views held by diverse individuals (Larson & LaFasto,
1989). Team members often need to socially integrate opposing views and
complex expertise. Accordingly, we define “Team Builders” as people who
help to establish norms, support decisions, and maintain a positive work
atmosphere within the team. This is a common theme in previous taxonomies
and can been seen in Benne and Sheats’s (1948) encourager, follower, com-
promiser, and harmonizer set; Bales’s (1950) show solidarity, tension release,
and agree dimensions; and DuBrin’s (1995) people supporter and listener
bracket. Whereas Team Builders focus their attention within the team, in
Mathieu et al. 17

contrast, it can also be important to have one or more members coordinating

team actions with external constituencies. We define a “Connector” as some-
one who helps bridge and connect the team with people, groups, or other
stakeholders outside of the team. Connectors ensure good working relation-
ships between the team and outsiders. This external linking role was the focal
feature of Ancona and Caldwell’s (1988) work, but can also be seen in
Belbin’s (1993) resource investigator, McCann & Margerison’s (1989)
explorer, Barry’s (1991) spanning, and Mumford et al.’s (2006) collector and
consult dimensions.
Finally, we also identified two destabilizing or change-oriented team
roles. Various authors have described a challenger (e.g., DuBrin, 1995;
Parker, 1994), critic (Benne & Sheats, 1948; Mumford et al., 2006), or dis-
agrees (Bales, 1950) team role. This type of behavior offers value by mixing
things up and getting members to consider alternative conceptions or work
modes and helps them to prevent premature closure on decisions. Accordingly,
we define a “Challenger” as someone who will push the team to explore all
aspects of a situation and to consider alternative assumptions, explanations,
and solutions. A Challenger often asks “why” and is comfortable debating
and critiquing ideas. Whereas Challengers question what is being done, we
define “Innovators” as people who regularly generate new and creative ideas,
strategies, and approaches for how the team can handle various situations and
challenges. An Innovator often offers original and imaginative suggestions.
This theme was articulated by Belbin’s (1993) plant/idea generator, McCann
and Margerison’s (1989) creator, Barry’s (1991) envisioning, DuBrin’s
(1995) knowledge contributor, and Parker’s (1994) innovator dimensions.

Psychometric and Nomological Network Analyses

The online supplement to this article provides extensive details concerning
the development of Likert-type scales for measuring these six TREO dimen-
sions. Table 3 summarizes the samples that we used and the purpose(s) for
each. We first drafted preliminary items for each of the six TREO dimen-
sions. Within each role, we generated two sub-sets of items: role orientations
and specific behavioral experiences. We then had those items classified into
the six substantive areas by 12 SMEs. From this classification, we identified
48 items, 4 orientations, and 4 behavioral experiences for each dimension
that were consistently classified into their respective TREO categories. We
then administered the 48 TREO items, along with measures of the Big 5 per-
sonality dimensions, to 317 Military officers and 266 upper-level undergrad-
uate business students. Analyses using those data, with a few exceptions,
confirmed the internal consistencies of the different scales, their convergent
18 Group & Organization Management 40(1)

Table 3. Summary of Study Samples and Their Use.

Sample population Primary purpose(s)

1. Academic (n = 6) and practitioner Classify 105 draft survey items into six
(n = 6) subject matter experts TREO dimensions
2. Military officers (n = 317) Evaluate TREO scale reliabilities
3. Large university capstone business and factor structure, and their
students (n = 266) discriminant validity from measures of
Big 5 personality constructs. Revise
items.
4. Military officers (n = 341) Evaluate revised TREO scale reliabilities
5. Large university capstone business and factor structure, and their
students (n = 515) discriminant validity from measures of
Big 5 personality constructs.
6. Small college business students Evaluate the test–retest stability of the
(n = 172) in a strategy course TREO measures
7. Large university capstone business Evaluate the predictive validity of
students (n = 225) playing complex TREO scale as related to peer ratings
simulation of members’ behaviors during a 10-
week business simulation

Note. TREO = Team Role Experience and Orientation.

validity across the orientation and behavioral experience subscales, and their
discriminant validity from measures of the Big 5 personality scales.
We made some minor wording changes to some items, and then adminis-
tered the TREO and Big 5 items to another sample of 341 military officers and
a second sample of 515 upper-level undergraduate business students for cross-
validation purposes. Again, analyses of these data confirmed the internal con-
sistencies of the different scales, their convergent validity across the orientation
and behavioral experience subscales, and their discriminant validity from mea-
sures of the Big 5 personality scales. We further generated hypothesized pat-
terns of correlations between the TREO and Big 5 measures. Then, using “bare
bones” meta-analyses (Hunter & Schmidt, 2004), we synthesized the results of
the four samples and concluded that the TREO scales, by and large, exhibited
a pattern of correlations with the Big 5 measures that were consistent with the
anticipated nomological network. Finally, we used a fifth sample of 172 upper-
level business students from a small college to evaluate the stability of TREO
scores over a 2.5-month long period. Each of the dimensions exhibited signifi-
cant test–retest correlations (mean r = .52, p < .001).
In sum, the supplement to this article details how the 48-item TREO mea-
surement tool (shown in the appendix) was developed and its psychometric
Mathieu et al. 19

properties. The separate orientation and behavioral experience facets of each

dimension provided unique variance and information, while also demonstrat-
ing significant convergent validity. The TREO measures are distinguishable
from, yet generally showed the expected pattern of correlations with, mea-
sures of the Big 5 personality dimensions. And the TREO measures illus-
trated significant and reasonably high test–retest reliabilities over a 2.5-month
long period. Nevertheless, it is also important to test whether the self-report
TREO predisposition measures actually predict the extent to which individu-
als enact these different team role behaviors in group settings. The following
investigation was focused on that very question.

Method
Sample
A sample of 225 students who were enrolled in seven capstone business strat-
egy courses taught at a large northeastern public university were recruited for
this study. Besides typical class activities, the students worked in four- to six-
person teams on a business strategy simulation for 10 weeks. We collected
TREO survey responses and demographics from students using an online sur-
vey before they began the simulation in their classes. Reliabilities for the
TREO scales were all above .80 and are detailed in Table 4. We collected peer
ratings of members’ TREO-related behaviors exhibited during the simulation
3 months later using another online survey. We had complete information for
196 students who were members of 66 teams that competed in the simulation.
This sample was 64% men with an average age of 21 (SD = 5.1), and 87%
reported their ethnicity as White with the remainder distributed across other
categories. They represented a spectrum of business majors (Accounting =
20%, Finance = 29%, Marketing = 14%, Management = 19%, other = 18%)
and had an average overall grade point average of 3.51 (SD = .27).

Simulation
The “StratSim Management” business strategy simulation is based on the
automobile industry and students function as the top management teams of
virtual firms competing with one another (S. W. James & Deighan, 2008).
The simulation calls for teams to make strategic decisions over time concern-
ing a number of business functions (i.e., operations, marketing, finance, and
R&D) for their firms for each weekly round. Students’ grades were based, in
part, on how well their team performed, as measured by a variety of objective
indices (e.g., the change in firm stock price).
20 Group & Organization Management 40(1)

Table 4. Predictive Validity of TREO Scales as Related to Peer Ratings of Team

Contributions.
Peer ratings of TREO exhibited behaviors

TREO scales Organizer Doer Challenger Innovator Team Builder Connector Overall

Organizer .33** .15* .09 .09 .23** .12 .27**

Doer .20** .18** .07 .14* .15* .06 .17*
Challenger .12 .04 .16* .19* .10 .08 .11
Innovator .03 .06 .12 .18* .14 .05 .07
Team Builder .21** .11 .00 .10 .24** .17* .14*
Connector .10 .09 .06 .16* .16* .10 .10

TREO scale M 3.80 4.26 3.73 3.81 4.04 3.60

Scale SD .54 .41 .47 .52 .46 .59
Alpha .85 .83 .88 .84 .82 .85
Peer rating M 3.81 4.04 3.81 3.78 3.78 3.63 4.57
Peer rating SD .79 .72 .65 .68 .68 .83 .59
Median rwg .83 .83 .75 .75 .75 .65 .88

Note. n = 196. TREO = Team Role Experience and Orientation.

*p < .05. **p < .01.

The simulation lasted 10 weeks, with each week representing one virtual
year of operation. During each period, teams used a decision interface to
gather information about topics including virtual macroeconomic parameters,
general market conditions, competitive landscape, and a rich array of demo-
graphic information on customer segments and their preferences. They made
a wide variety of decisions concerning matters such as product attributes (e.g.,
safety, capacity, economy, interior design); operations (e.g., manufacturing
capacity); sales and marketing (e.g., pricing, targeted or generic advertising,
geographic dealership coverage); financial decisions (e.g., cash flow, issuing
or buying back bonds, loans, and common stock); as well as other strategic
decisions such as launching new models or discontinuing existing models,
bidding for business-to-business contracts, and licensing to and from competi-
tors. Team decisions were uploaded and processed by the software in a
dynamic and competitive simulation environment, which incorporated all
decisions of all teams in any given industry, thus leading to considerable vari-
ance between industry-level outcomes, such as competitiveness, innovative-
ness, rivalry, and so on. The simulation software then provided outcome
figures for macroeconomy, customer preferences, industry, and each firm’s
performance. Detailed feedback reports were made available for each com-
pany, along with nonproprietary information on competitors, as well as the
industry and the macroeconomic environment as a whole, per demand of
Mathieu et al. 21

teams within the simulation interface. Notably, members were free to structure
themselves and develop various roles and modes of operation as they desired.

Peer Ratings of TREO Behaviors

At the completion of the simulation, we collected self and peer evaluations of
members’ TREO-related behaviors exhibited during the simulation. Students
were assured that their ratings would remain confidential. Although they did
not translate directly to other students’ grades, these ratings were used as
input to instructors who assigned class participation grades for the course. We
used just the average peer evaluations as our criteria, and study participants
received ratings from 1 to 5 peers (median = 4 peer ratings).
Students rated each of their team members on the six TREO dimensions
and provided an overall rating. It is important to emphasize that these ratings
are about the extent to which each member exhibited TREO behaviors during
their group activities over the course of the simulation. For the TREO behav-
ior ratings, we provided the definitions of each dimension (see Table 2) and
asked members to “provide ratings on six scales concerning the extent to
which your teammates’ contributed effectively to each of the roles.” Responses
were made on the following 5-point scale: 5 = To a very great extent: This
person exhibited a great deal of this behavior; 4 = To a great extent: This
person exhibited more than an average amount of this behavior; 3 = To a
moderate extent: This person exhibited about an average amount of this
behavior; 2 = To a slight extent: This person exhibited some, but less than
average, amount of this behavior; and 1 = Not at all: This person did not
exhibit any of this behavior. We used the following item for the overall rat-
ing: “For this rating, think about the total contributions that each person made
to your team effort. This need not be simply an average of the other ratings.”
We instructed participants to select from the following the most appropriate
score for each person: Poor: We would have been better off without this per-
son. She or he was disruptive or simply didn’t show up or do anything for the
group (score 1); Fair: She or he really didn’t pull her or his weight (e.g.,
didn’t do a fair share of the work, missed meetings, etc.) and expected others
to do the work (2); Average: a good team player (3); Good: She or he was a
great help to others and did at least her or his share of the work (4); or
Excellent: She or he was absolutely critical to our team. She or he did sub-
stantially more than most people and helped everyone understand and accom-
plish what we were doing (5). We used L. R. James, Demaree, and Wolf’s
(1984) single item rwg agreement index to assess inter-rater agreement on
each of these ratings. As reported in Table 4, median rwg values were .75 or
higher for all dimensions except for Connector, which had an rwg = .65.
22 Group & Organization Management 40(1)

Results
Table 3 presents correlations between participants’ TREO responses and their
peers’ ratings of the extent to which they exhibited TREO-related behaviors
over the course of the 10-week simulation. Predictive validities are itali-
cized in Table 4 and show significant correlations for all dimensions
(Organizer r = .33, p < .01; Doer r = .18, p < .01; Challenger r = .16, p < .05;
Innovator r = .18, p < .05; Team Builder r = .24, p < .01) except Connector
(r = .10, ns). Moreover, the corresponding predictive validity coefficients
were the highest correlations per peer rating for Organizer, Doer, Challenger,
and Team Builder dimensions.
Among the non-predictive validity correlations that were evident, TREO
Doer (r = .20, p < .01) and Team Builder (r = .21, p < .01) scores correlated
significantly with peers’ ratings of members’ Organizer behaviors, whereas
TREO Organizer (r = .23, p < .01) and Doer (r = .15, p < .05) scores also
correlated with peers’ ratings of members’ team building behaviors.
Interestingly, TREO Doer (r = .14, p < .05), Challenger (r = .19, p < .05), and
Connector (r = .16, p < .05) scores all correlated significantly with members’
ratings of Innovator type behaviors. In terms of predicting peers’ overall
evaluations of members’ contributions to the team, the TREO dimensions of
Organizer (r = .27, p < .01), Doer (r = .17, p < .05), and Team Builder (r =
.24, p < .05) all evidenced significant correlations.
In sum, the results from this sample are encouraging. First, peers evidenced
sufficient agreement on the extent to which students exhibited TREO aligned
behaviors while working together on a business simulation. Second, TREO
scale responses demonstrated significant predictive validity with their corre-
sponding peer ratings of their behaviors 3 months later. Third, the exception to
this pattern concerned the Connector dimension for which peers had markedly
lower inter-rater agreement and the predictive validity was not significant.
This is not particularly surprising, however, as the business simulation
afforded virtually no opportunities for boundary-spanning activities to mani-
fest. Finally, at least in terms of this business simulation context, the bundle of
high Organizer, Doer, and Team Builder TREO scores appears to be associ-
ated with high ratings of individuals’ subsequent team contributions.

Discussion
We had several goals for this article. First, we offered a review and synthesis
of the team role literature. We argued that better incorporating individuals’
potential to occupy various team roles holds great promise for advancing our
understanding of team composition. We distilled six integrative types of team
Mathieu et al. 23

roles and developed an accompanying TREO survey measure that considers

both individuals’ previous team’s role-related experiences, as well as their
preferences to engage in certain types of role-related behaviors.
SMEs worked to establish the content validity of the items per dimension.
We demonstrated the convergent and discriminant validity of TREO mea-
sure, detailed their correlations with measures of individuals’ Big 5 personal-
ity constructs, and illustrated their stability over time. Moreover, we
demonstrated the predictive validity of the TREO scale as related to peers’
ratings of team members’ role-related behaviors. Below we consider the
implications of these findings in terms of future theory and research incorpo-
rating individuals’ predispositions to fulfill different team roles in the larger
team composition domain. We also consider the use of the TREO measure in
future work, and note some limitations of our investigations. We conclude
with recommendations for practice.

Role Theory
Team role theory dates back to at least the 1940s and Kurt Lewin’s Research
Center for Group Dynamics. In the years since then, numerous taxonomies
have been advanced, lists of team roles have proliferated, and there have been
limited attempts at integration. In areas where integration has occurred (e.g.,
Belbin’s framework), the construct validity of measurement tools has been
questioned (e.g., Aritzeta et al., 2007; Broucek & Randell, 1996; Dierendonck
& Groen, 2011; Furnham et al., 1993) and freely available measures are lack-
ing. We offer an integration of this literature in terms of six team role dimen-
sions, and have developed, validated, and made freely available an instrument
(TREO) to measure individuals’ propensities to occupy the roles for research
purposes.
We should note that the intercorrelations of the six TREO dimensions
were fairly high, averaging approximately r = .70, p < .001, across our item
development and validation samples. We hasten to add, however, that those
correlations are among latent variables that, in effect, are corrected for mea-
surement attenuation. In other words, latent variable correlations are always
higher than observed correlations if the measurement scales have less than
perfect reliabilities. Therefore, although the high inter-dimensional correla-
tions are still a source of concern, they are not as problematic as they appear
using conventional thresholds. The confirmatory factor analysis results con-
sistently illustrated that the TREO dimensions were empirically distinguish-
able. Moreover, the TREO scale correlations with the Big 5 personality
variables differed across dimensions, indicating that they are not redundant
(see supplement Table S8). That said, it does appear that there could be a
24 Group & Organization Management 40(1)

higher order latent structure underlying the six dimensions: (a) a task-ori-
ented factor (i.e., Organizer and Doer), (b) a change-oriented factor (i.e.,
Challenger and Innovator), and (c) a socio-emotional or linkage dimension
(i.e., Team Builder and Connector). Whether there is value in differentiating
the six TREO dimensions or operating at a higher level of abstraction remains
a question for future research. We suspect that the relative value of either
approach would hinge on the purpose of any particular investigation. But
what is clear from our and previous research is that it is not likely the case
that there are 10 or more clearly distinguishable team role dimensions.
However, there is evidence to support the six TREO dimensions.
We restricted our work to examining TREO as a measure of individual
differences and differentiated it from measures of other individual differ-
ences. Certainly, TREO can be used to predict the likelihood that individuals
will take on certain type of roles and exhibit certain types of behaviors in
teams. But the TREO measure is also perhaps valuable in terms of assessing
and testing team compilation models. For example, the popular Belbin (1993)
framework advocates composing a team to have an ideal “balance” of styles.
Individuals are categorized as one of nine types on the basis of their or
observers’ ratings, and presumably teams are more effective to the extent that
they have a greater rather than lesser variety of team roles present. The
empirical evidence on the value of such balance is inconsistent (cf. Senior,
1997; Smith et al., 2012), and there are shortcomings in the way in which the
framework is implemented (e.g., classifying individuals ipsatively into one
and only one type, equating different varieties of balance, etc.). Nevertheless,
the general idea that different member combinations or profiles may be more
or less advantageous in different situations warrants far more scrutiny
(Mathieu et al., 2014).
We advocate using individuals’ TREO scores in a more holistic fashion than
merely to categorize members into types. For example, perhaps a team might
benefit from having at least one member who is high on Organizer scores, one
who is high on team building, and at least two who have high Doer scores. In
this situation, there may be two different individuals who fulfill the Organizer
and team building needs, or there may be a single person who fulfills both
needs for a team. Naturally, this begs the question of whether it is preferable for
one person to be occupying multiple team roles, or whether it is better to spread
the fulfillment of different roles across members. This represents an interesting
question for future theory, research, and application.
Roles may be naturally related to certain positions (Ilgen & Hollenbeck,
1991; Sluss et al., 2011). For example, an individual may occupy a for-
mal leadership position whereby he or she would be expected to organize
work, provide individuals with performance-related feedback, and take
Mathieu et al. 25

responsibility for team functioning. However, those functions may well be

performed by others in the group and are not necessarily limited to a person
who occupies a particular position in the team (Manz & Sims, 1980). In other
words, roles are emergent phenomena and are not tied inextricably to formal
positions. Nevertheless, future scholars might consider the value of different
member TREO profiles in team situations where individuals are fixed into
certain positions (i.e., where there are substantial “role as position” by “role
as person” interactions). For example, what are the implications of a surgeon,
head nurse, or anesthesiologist scoring high on team building? What are the
implications of the most junior member of a team versus a designated leader
scoring the highest on the TREO Challenger dimension? In short, the team
role area is ripe for development. We believe that optimal team compositions
are not likely to be adequately captured by a simple average or variety (i.e.,
variance) index; rather, we believe that configural or compilation (Kozlowski
& Ilgen, 2006) combinations are likely to better capture the complexities of
team memberships. What such combinations are and whether they are a prod-
uct of situational demands, members’ work histories, or the extent to which
teams have rigid or loosely defined positions (if positions at all) are all inter-
esting questions worthy of pursuit. The TREO measure will help to enable
those inquiries.

Limitations
We developed the TREO measures using SME input and then explored dif-
ferent facets of its construct validity using seven different samples. Although
leveraging multiple samples is certainly advantageous, the fact that we sam-
pled military officers and business students does raise questions about the
generalizability of our findings. Additional evidence of the construct validity
of the TREO measures using different populations is warranted. Moreover,
the relative importance of TREO versus other individual differences is wor-
thy of investigation. For example, are individuals with certain TREO profiles
seen as more valuable team members than individuals who possess task-spe-
cific knowledge, general intelligence, or experience in a particular domain?
Do TREO profiles interact with any of these other KSAOs as related to
important individual or team outcomes? Additional research along these lines
would be welcome.
The TREO dimensions did evidence relatively high intercorrelations.
Notably, items are implicitly worded in terms of positive attributes—about
one’s propensity to fulfill important team roles. Thus, dimension correlations
may be inflated to the extent that respondents exhibited any social desirabil-
ity bias. Incorporating measures that assess such bias would be a welcome
26 Group & Organization Management 40(1)

addition in future investigations. It may also be valuable to develop force-

choice or other ipsative method of measurements that would yield sharper
TREO profiles, although such techniques introduce different psychometric
models and considerations. In a related vein, we have discussed TREO
dimensions as though greater amounts of each are always advantageous.
However, there may well be value in examining whether there are tipping
points or thresholds beyond which a positive attribute turns into a negative
one (see Smith et al., 2012). For example, whereas a propensity to organize
team activities is generally seen as a positive influence, at some point, such
structuring may lapse into being overly controlling and stifling. Some chal-
lenging behavior is healthy for teams and prevents them from prematurely
closing discussion or failing to consider alternatives. Yet too much challeng-
ing is likely to breed contempt and be viewed as antagonistic. There may also
be problematic combinations or profiles. For example, someone who is
exceedingly high on organizing and doing, yet very low on team building,
may be perceived as an overly dominating force in a team. Or, someone who
scores very high on challenging and connecting may be perceived as a nega-
tive element who is not invested in the team. In other words, tipping points,
acidic members, and team profiles all warrant future investigation.

Applications
Improving our understanding of team role propensities and being able to
measure them in a reliable manner can have several practical applications
starting with initial team formation through ongoing team, leader, and per-
sonal development. When composing a team, knowledge of candidates’ team
role propensities could be used to avoid configurations that are more likely to
result in team dysfunction. For example, Belbin (1993) recommends com-
posing teams to maximize members’ role differentiation. However, optimal
(and suboptimal) team configurations may well be more complex than sim-
ply establishing diversity. For example, a team that is low on Doers or has no
one who is an Organizer is apt to struggle, as is a team with too many or too
few Challengers. Research is needed to better understand the types of profiles
that are likely to be problematic, but a psychometrically sound measure is the
starting point for such research.
Once a team is formed, understanding its members’ TREO profile may
provide insights as to the utility of other team interventions. Use of the TREO
tool can provide a common language for discussing team members’ prefer-
ences and inclinations, enhancing awareness, and perhaps enabling the team
to better anticipate each other’s reactions in various situations. In effect,
knowing each other’s predispositions may enhance the effectiveness of team
Mathieu et al. 27

interventions such as charters, team training and development, debriefs, and

leadership strategies (Mathieu et al., 2014).
Finally, the TREO can be used during leader training and as a self-aware-
ness tool. For example, leaders could be taught about the various team role
propensities and how best to lead individuals and teams with certain team
role profiles. Similarly, an individual who completes a tool such as the TREO
may learn to better understand their natural inclinations in team settings,
increase their self-awareness, and better recognize when they need to operate
in a manner that is not their natural “default” mode. Naturally, the applica-
tions described here for team composition and development should be tested
in future research. All such applications are predicated on having a well-val-
idated measure such as TREO readily available.

Appendix
TREO Scale Items
Based on my prior experiences, as a member of different teams . . .

1. I learn how to get outside resources that our team needs to be

successful.
2. I’m comfortable being critical of my teammates.
3. I like it when we keep busy and get things done.
4. I like to challenge peoples’ assumptions.
5. I like to be the one that sorts out the details of a team project.
6. I often volunteer new ideas and suggestions without being asked my
opinion.
7. I can calm people down and get them focused on the task when things
get stressful.
8. I like to be the one who decides who will do which tasks on a team.
9. I am the one who questions why we are doing things in a certain way.
10. Sometimes, I just voice a different opinion to keep my team thinking
about what we should be doing.
11. I’m always ready to support a good suggestion in the common interest
of the team.
12. People usually look to me when something needs to be done in the
team.
13. I like to try out new ideas and approaches.
14. I question what my team should be doing to get the job done.
15. I can be counted on to follow through on any tasks which I’ve been
assigned.
28 Group & Organization Management 40(1)

16. I can be counted on when a task needs to be done.

17. I keep my team on pace and aware of deadlines.
18. I make sure that my teammates are clear about their responsibilities.
19. I’m comfortable dealing with interpersonal conflicts and helping peo-
ple work through them.
20. I enjoy coordinating team efforts with people or groups outside of the
team.
21. My primary focus is on getting my assignments done for the team.
22. I can be counted on to spread ideas between my team and people out-
side of my team.
23. I’m comfortable being the spokesperson for a team.
24. I am the one who steps up and does whatever is necessary to make the
team successful.
25. I’m often the first to volunteer for a difficult or unpopular assignment
if that is what the team needs.
26. I like to be the one who keeps track of how well my team is doing.
27. I am usually the one who suggests a new idea or direction when the
team gets stuck on something.
28. I bring a sense of organization to any job a team undertakes.
29. I get bored when we do the same task the same way every time.
30. I structure team activities.
31. I discover and connect with people who can help my team succeed.
32. I’m not afraid to question my teammates’ authority.
33. I’m known for thinking creatively and “outside the box.”
34. I typically find out what is going on outside my team and share that
with my teammates.
35. I like coming up with new ways that our team can accomplish our
tasks.
36. I usually suggest the appropriate steps that my team should follow to
get something done.
37. I like helping different kinds of people work effectively together.
38. I’m comfortable producing and sharing new ideas with my team.
39. I often work to maintain good working relationships within my team.
40. It bothers me when I see teammates getting frustrated or depressed.
41. I’m always committed to my team tasks.
42. I often point out the potential risks or hazards of a team plan or course
of action.
43. I help people move beyond their disagreements and find common
ground.
44. My teammates often view my suggestions as creative or innovative.
45. I often serve as a liaison between my team and outside groups.
Mathieu et al. 29

46. I promote my team’s mission and goals with other teams or units.
47. I can typically provide a strong rationale to refute ideas that I believe
are unsound.
48. I encourage my teammates when I know they have a difficult assign-
ment or challenge.

Item Mapping

Organizer: 5O, 8O, 17E, 18E, 26O, 28O, 30E, 36E.

Doer: 3O, 12E, 15E, 16O, 21E, 24E, 25O, 41O.
Challenger: 2O, 4O, 9E, 10E, 14E, 32O, 42E, 47O.
Innovator: 6E, 13O, 27E, 29O, 33E, 35O, 38O, 44E.
Team Builder: 7E, 11O, 19O, 37O, 39E, 40O, 43E, 48E.
Connector: 1E, 20O, 22O, 23O, 31E, 34O, 45E, 46E.

Note. E = experience subscale; O = orientation subscale.

©2008-2012. The Group for Organizational Effectiveness, Inc. Permission is granted to use
the TREO for research purposes. All other uses require permission from The Group for
Organizational Effectiveness (www.groupoe.com).

Authors’ Note
All opinions expressed herein are strictly those of the authors and not necessarily
those of the sponsoring organizations.

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: The work was supported by Grant
NNX11AR22G from the National Aeronautics and Space Administration (NASA)
and Contract W91WAW-08-C-0021 from the Army Research Institute awarded to the
authors.

References
Ancona, D. G., & Caldwell, D. F. (1988). Beyond task and maintenance: Defining
external functions in groups. Group & Organizational Studies, 13, 468-494.
Ancona, D. G., & Caldwell, D. F. (1992). Bridging the boundary: External activity
and performance in organizational teams. Administrative Science Quarterly, 37,
634-665.
30 Group & Organization Management 40(1)

Anderson, N., & Sleap, S. (2004). An evaluation of gender differences on the

Belbin Team Role Self Perception Inventory. Journal of Occupational and
Organizational Psychology, 77, 429-437.
Aritzeta, A., & Ayestaran, S. (2003). Applicability of Belbin’s team roles theory:
A longitudinal and comparative study with work teams. General and Applied
Psychology, 56, 61-75.
Aritzeta, A., Senior, B., & Swailes, S. (2005). Team role preference and cognitive
styles: A convergent validity study. Small Group Research, 36, 404-436.
Aritzeta, A., Swailes, S., & Senior, B. (2007). Belbin’s team role model: Development,
validity and applications for team building. Journal of Management Studies, 44,
96-118.
Bales, R. F. (1950). Interactive process analysis: A method for the study of small
groups. Cambridge, MA: Addison-Wesley.
Bales, R. (1970). Personality and interpersonal behaviour. New York: Holt, Rinehart
and Winston.
Barry, D. (1991). Managing the bossless team: Lessons in distributed leadership.
Organizational Dynamics, 20, 31-47.
Belbin, R. M. (1981). Management teams: Why they succeed or fail. London, England:
Butterworth-Heinemann.
Belbin, R. M. (1993). Team roles at work. London, England: Butterworth-
Heinemann.
Bell, S. T. (2007). Deep-level composition variables as predictors of team perfor-
mance: A meta-analysis. Journal of Applied Psychology, 92, 595-615.
Benne, K. D., & Sheats, P. (1948). Functional roles of group members. Journal of
Social Issues, 4, 41-49.
Broucek, W. G., & Randell, G. (1996). An assessment of the construct validity of
the Belbin Self-Perception Inventory and observer’s assessment from the per-
spective of the five-factor model. Journal of Occupational and Organizational
Psychology, 69, 389-405.
Brown, S. H. (1978). Long-term validity of a personal history item scoring procedure.
Journal of Applied Psychology, 63, 673-676.
Cheung, M. W., & Chan, W. (2002). Reducing uniform response bias with ipsative
measurement in multiple-group confirmatory factor analysis. Structural Equation
Modeling: A Multidisciplinary Journal, 9, 55-77.
Davis, K. R. (1984). A longitudinal analysis of biographical sub-groups using Owen’s
development-integrative model. Personnel Psychology, 37, 1-14.
Dierendonck, D. V., & Groen, R. (2011). Belbin revisited: A multitrait-multi-
method investigation of a team role instrument. European Journal of Work &
Organizational Psychology, 20, 345-366.
DuBrin, A. J. (1995). The breakthrough team player. New York, NY: American
Management Association.
Fisher, S. G., Hunter, T. A., & Macrosson, W. D. K. (2001). A validation study of
Belbin’s team roles. European Journal of Work & Organizational Psychology,
10, 121-144.
Mathieu et al. 31

Furnham, A., Steele, H., & Pendleton, D. (1993). A psychometric assessment of

the Belbin Team-Role Self-Perception Inventory. Journal of Occupational and
Organizational Psychology, 66, 245-257.
Hackman, J. R. (1990). Groups that work and those that don’t. San Francisco, CA:
Jossey-Bass.
Hambrick, D. C. (1994). Top management groups: A conceptual integration and
reconsideration of the “team” label. Research in Organizational Behavior, 16,
171-214.
Hollenbeck, J. R., Beersma, B., & Schouten, M. E. (2012). Beyond team types and
taxonomies: A dimensional scaling conceptualization for team description.
Academy of Management Review, 37, 82-106.
Humphrey, S. E., Morgeson, F. P., & Mannor, M. J. (2009). Developing a theory
of the strategic core of teams: A role composition model of team performance.
Journal of Applied Psychology, 94, 48-61.
Hunter, J. E., & Schmidt, F. L. (2004). Methods of meta-analysis: Correcting error
and bias in research findings (2nd ed.). Thousand Oaks, CA: SAGE.
Ilgen, D. R. (1999). Teams embedded in organizations: Some implications. American
Psychologist, 54, 129-139.
Ilgen, D. R., & Hollenbeck, J. R. (1991). The structure of work: Job design and roles.
In M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organiza-
tional psychology (Vol. 2, pp. 165-208). Palo Alto, CA: Consulting Psychologists
Press.
Ilgen, D. R., Hollenbeck, J. R., Johnson, M., & Jundt, D. (2005). Teams in organiza-
tions: From I-P-O models to IMOI models. Annual Review of Psychology, 56,
517-543.
Jackson, C. (2002). Predicting team performance from a learning process model.
Journal of Managerial Psychology, 17, 6-13.
James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-group interrater
reliability with and without response bias. Journal of Applied Psychology, 69,
85-98.
James, S. W., & Deighan, M. (2008). StratSim: Administrators guide. Charlottesville,
VA: Interpretive Software.
Jung, C. G. (1923). Psychological types. London, England: Routledge and Kegan
Paul.
Katz, D., & Kahn, R. L. (1978). The social psychology of organizations (2nd ed.).
New York, NY: John Wiley.
Kirnan, J. P., & Woodruff, D. (1994). Reliability and validity estimates of the
Parker Team Player Survey. Educational and Psychological Measurement, 54,
1030-1037.
Klimoski, R. J., & Zukin, L. B. (1999). Selection and staffing for team effective-
ness. In E. Sundstrom (Ed.), Supporting work team effectiveness (pp. 63-94). San
Francisco, CA: Jossey-Bass.
Kozlowski, S. W. J., & Ilgen, D. R. (2006). Enhancing the effectiveness of work
groups and teams. Psychological Science, 7, 77-124.
32 Group & Organization Management 40(1)

Larson, C. E., & LaFasto, F. M. (1989). Teamwork: What must go right/wrong can go
wrong. Newburg Park, CA: SAGE.
Lessem, R., & Baruck, Y. (2000). Testing the SMT and Belbin inventories in top
management teams. Leadership and Organization Development Journal, 34,
75-83.
Manz, C. C., & Sims, H. P. J. (1980). Self-management as a substitute for leader-
ship: A social learning theory perspective. Academy of Management Review, 5,
361-367.
Margerison, M., & McCann, D. (1985). How to lead a winning team. Bradford, UK:
MCB University Press.
Marks, M. A., Mathieu, J. E., & Zaccaro, S. J. (2001). A temporally based framework
and taxonomy of team processes. Academy of Management Review, 26, 356-376.
Mathieu, J. E., Maynard, M. T., Rapp, T., & Gilson, L. (2008). Team effectiveness
1997-2007: A review of recent advancements and a glimpse into the future.
Journal of Management, 34, 410-476.
Mathieu, J. E., Tannenbaum, S. I., Donsbach, J. S., & Alliger, G. M. (2014). A Review
and Intergration of Team Composition Models: Moving Towards a Dynamic and
Temporal Framework. Journal of Management, 40, 130-160.
McCann, D., & Margerison, C. (1989). Managing high-performance teams. Training
and Development Journal, 43, 52-60.
Mumford, M. D., & Owens, W. A. (1984). Individuality in a developmental con-
text: Some empirical and theoretical considerations. Human Development, 27,
84-108.
Mumford, T. V., Campion, M. A., & Morgeson, F. P. (2006). Situational judg-
ment in work teams: A team role typology. In J. A. Weekley & R. E. Ployhart
(Eds.), Situational judgment tests: Theory, measurement, and application (pp.
319– 343). Mahwah, NJ: Erlbaum.
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 situ-
ational judgment test. Journal of Applied Psychology, 93, 250-267.
Myers, I. B., McCaulley, M. H., & Most, R. (1985). Manual: A guide to the devel-
opment and use of the Myers-Briggs type indicator. Palo Alto, CA: Consulting
Psychologists Press.
Neiner, A. G., & Owens, W. A. (1982). Relationships between two sets of biodata
with 7 years separation. Journal of Applied Psychology, 67, 146-150.
Park, W. W., & Bang, H. (2002). Team role balance and team performance. Paper
presented at the Biannual Conference on Belbin’s Team Roles, Cambridge,
UK.July.
Parker, G. M. (1994). Cross-functional teams. San Francisco, CA: Jossey-Bass.
Parker, G. M. (1996). Team players and teamwork. San Francisco, CA: Jossey-Bass.
Partington, D., & Harris, H. (1999). Team role balance and team performance: An
empirical study. Journal of Management Development, 18, 694-705.
Ronnie, L., & Yehuda, B. (2000). Testing the SMT and Belbin inventories in top man-
agement teams. Leadership & Organization Development Journal, 21, 75-84.
Mathieu et al. 33

Rushmer, R. (1996). Is Belbin’s shaper really TMS’s thruster-organizer? An empiri-

cal investigation into the correspondence between the Belbin and TMS team role
models. Leadership & Organizational Development Journal, 17, 20-26.
Sadler-Smith, E. (2001). Self-perception of team-roles: Some of the implications for
business and management. In R. Riding & S. G. Rayner (Eds.), International per-
spectives on individual differences (Vol. 2, pp. 101-130). Westport, CT: Ablex.
Senior, B. (1997). Team roles and team performance: Is there “really” a link? Journal
of Occupational and Organizational Psychology, 70, 241-258.
Senior, B. (1998). An empirically-based assessment of Belbin’s team roles. Human
Resource Management Journal, 8, 54-60.
Sluss, D. M., van Dick, R., & Thompson, B. S. (2011). Role theory in organizations:
A relational perspective. In S. Zedeck (Ed.), APA handbook of industrial and
organizational psychology (Vol. 1, pp. 505-534). Washington, DC: American
Psychological Association.
Smith, M., Polglase, G., & Parry, C. (2012). Construction of student groups using
Belbin: Supporting group work in environmental management. Journal of
Geography in Higher Education, 36, 585-601.
Sommerville, J., & Dalziel, S. (1998). Project teambuilding—The applicability of
Belbin’s Team-Role Self-Perception Inventory. International Journal of Project
Management, 16, 165-171.
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.
Stewart, G. L., Manz, C. C., & Sims, H. P. J. (1999). Team work and group dynamics.
New York, NY: Wiley.
Sundstrom, E., De Meuse, K. P., & Futrell, D. (1990). Work teams: Applications and
effectiveness. American Psychologist, 45, 120-133.

Author Biographies
John E. Mathieu is a Professor of Management at the University of Connecticut, and
holds the Friar Chair in Leadership and Teams at UConn. His primary areas of interest
include models of team and multi-team effectiveness, leadership, training effective-
ness, and cross-level models of organizational behavior.
Scott Tannenbaum is President of the Group for Organizational Effectiveness
(gOE), a New York-based consulting and research firm. Formerly a tenured professor,
his research has been cited over 8000 times and he is a Fellow of SIOP and APS.
Michael R. Kukenberger is an assistant professor in the Department of Management
at the Peter T. Paul School of Business and Economics at University of New
Hampshire. He received his Ph.D. in Organizational Behavior at the University of
Connecticut and his primary areas of research interest include shared and team leader-
ship, team effectiveness, team and individual learning and multilevel theory
building.
34 Group & Organization Management 40(1)

Jamie S. Donsbach is a Senior Consultant with The Group for Organizational

Effectiveness, Inc., a consulting and research firm. Throughout her career, Dr.
Donsbach has supported corporate and public sector customers with a variety of orga-
nizational needs such as enhancing team effectiveness, training, identifying compe-
tency gaps, managing performance, and developing/validating assessment tools. Dr.
Donsbach received her doctorate in Industrial/Organizational Psychology from the
University at Albany, State University of New York.
George M. Alliger, PhD, is VP of Solutions for the Group for Organizational
Effectiveness. Dr. Alliger has over many years conducted research in and written
about job analysis, training, and assessment in organizations. He recently co-edited
the Handbook of Work Analysis.