Content uploaded by Matthew S. O'Connell
Author content
All content in this area was uploaded by Matthew S. O'Connell on Feb 23, 2015
Content may be subject to copyright.
Leadership and Work Teams 1
Leadership and Semi-Autonomous Work Team Performance:
A Field Study
Matthew S. O'Connell
Select International, Inc.
San Diego, CA
Dennis Doverspike and Alana Blumental
Psychology Department
University of Akron
Published in Group and Organization Management, 27, 14-49
We gratefully acknowledge the input and assistance of Robert Lord, Kevin Klinvex, Greg Smith,
Lee Ashton, Rosanna Miguel, David Bernal, the late Ralph Alexander, and the three anonymous
reviewers.
Correspondence concerning this article should be addressed to Dennis Doverspike, Psychology
Department, University of Akron, Akron, Ohio, 44325, ddoverspike@uakron.edu, 330.972.8372.
Leadership and Work Teams 2
Leadership and Semi-Autonomous Work Team Performance:
A Field Study
Abstract
The impact of leadership on group performance was investigated in the context of a semi-
autonomous work team environment. The relationship between group performance and team
leadership, as exhibited by the designated, internal leader, was moderated by team size. Team
leadership, as rated by team members, was related to the managers’ ratings of quality and overall
performance for smaller teams, but not for larger teams. In addition, the managers’ ratings of
team performance were not significantly correlated with ratings of team performance obtained
from the team members.
Leadership and Work Teams 3
Leadership and Semi-Autonomous Work Team Performance:
A Field Study
Self-directed work teams can be defined based on the presence of a number of
characteristics such as team members taking responsibility for the quality of the work process,
the sharing of management and leadership functions, and involvement in the training, hiring and
disciplining of group members (Wellins, Byham, & Dixon, 1994; Wellins, Byham, & Wilson,
1991). While a substantial body of empirical and anecdotal evidence on the design and
functioning of self-directed teams has accumulated in professional journals and the applied press
(c.f., Hackman, 1987; Pearce & Ravlin, 1987; Sundstrom, Demeuse, & Futrell, 1990; Wellins et
al., 1991, 1994; Wilson, George, Wellins, & Byham, 1994), there is very little published
empirical research concerned with leadership and self-directed work team performance.
Although the concept of "leadership in self-directed work teams" would appear to be an
oxymoron, self-directed teams do appear to retain some form of assigned administrator,
coordinator, or supervisor (Stewart & Manz, 1995).
In the laboratory, it is possible to create relatively pure versions of the self-directed work
group. However, in industrial settings, self-directed work groups must operate within the
organizational hierarchy. In response, self-directed work groups in organizational contexts often
have a designated leader. Most writers in the area of self-directed teams, indicate that although
the role of the leader is dramatically changed in such teams, becoming more of a coordinator than
a supervisor, most self-directed teams do maintain some form of team leadership (Jessup, 1990;
Stewart & Manz, 1995; Wellins et al., 1991). Jessup (1990) discussed three roles that leaders,
Leadership and Work Teams 4
whether they are called team leaders, team coordinators, or coaches, served in team-based
organizations. The three roles were those of the administrator, coach, and advisor. Echols and
Mitchell (1990) suggested that the role of the leader appeared to have shifted from one of formal
authority in terms of decision making, goal setting, directing floor activity, and troubleshooting,
to one that focused primarily on coaching and training team members, as well as directing special
projects, ensuring adequate resources, and monitoring overall performance. Manz and Sims
(1987) proposed that the role of effective leaders was to teach team members how to perform
most of the functions typically reserved for management, such as setting their own goals,
gathering performance data, and evaluating their own performance.
The question that might be asked then is whether this shift means that the extensive body
of research on effective group leadership is outdated and meaningless for self-directed teams.
That is, does the appointed leader's performance of traditional leadership behaviors make a
difference in work teams which are self-directed by design but retain a designated leader.
Substantial research evidence exists indicating that leadership is related to group performance in
traditional groups (c.f., Day & Lord, 1988; Fiedler & Garcia, 1987; Misumi, 1985; Weiss &
Friedrichs, 1986). However, almost all of this research was conducted in situations involving
traditional leader-group member relations. With the growing popularity of self-directed teams
and employee empowerment, the question of the impact of designated team leadership on self-
directed team performance is a salient issue for leadership researchers, as well as practicing
managers and consultants.
The purpose of the current research was to study the effect of a designated, internal leader
on performance by a type of self-directed work group, the semi-autonomous team. The semi-
Leadership and Work Teams 5
autonomous team is defined here as a self-directed work group that had a designated, internal
leader along with a “star” governance structure. As the name “star” suggests, responsibilities in
the semi-autonomous work group can be thought of as organized according to a star diagram,
with each team member accepting responsibility for one function or point on a star (e.g.,
monitoring safety, coordinating time off, training, maintaining equipment, etc.). The designated,
internal team leader coordinated the activities of the team members (Stewart & Manz, 1995;
Wellins et al., 1994) and represented the team in meetings with managers and other team leaders.
The organization in which this research was conducted had been committed to a self-
directed team environment and had semi-autonomous teams in place for the previous three and a
half years. The semi-autonomous teams included in this study were performing a production task
in a large, manufacturing plant. Although it was the organization's expressed desire to dissipate
much of the authority traditionally associated with supervisors, it was the company's policy to
designate a leader for each team. It was this appointment of a leader for the semi-autonomous
work team which provided the opportunity to conduct research on the role of leadership in this
unique environment.
Relevant Research on Leadership in Self-Directed Teams
According to Wellins et al. (1991), a self-directed work team “is an intact group of
employees who are responsible for a ‘whole’ work process or segment that delivers a product or
service to an internal or external customer” (p. 3). In addition, Wellins, Wilson, Katz, Laughlin,
& Day (1990) also identified several factors which often characterized self-directed teams,
including: (a) Sharing various management and leadership functions, (b) planning, controlling,
and improving their own work processes, (c) setting goals and inspecting work, (d) creating their
Leadership and Work Teams 6
own schedules, (e) ordering materials, keeping inventories, and working with suppliers, (f)
acquiring training as needed, (g) hiring and disciplining members of the team, and (h) taking
responsibility for the quality of their products or services. Of course, the degree to which self-
directed teams carry out these responsibilities may vary widely, with very few teams actually
carrying out all of these responsibilities. Results from a national survey of over 500 organizations
using self-directed work teams indicated that for more than 50% of the organizations surveyed,
responsibility for many of these functions was either shared between the team leader and the
team or solely the responsibility of the team leader (Wellins et al., 1990). Thus, it seems clear
that although these teams hold a greater degree of responsibility for their own self-management
than do more traditional teams, they often still have a designated leader and team leaders may
still play a significant leadership role.
There are some recent studies which look at leadership in less traditional types of teams.
A study from the field of sport psychology (Chelladurai, 1984), investigated the relationship
between coaching behaviors and win/loss percentage for college basketball teams. Team member
evaluations of the leadership behaviors of their coach from 23 teams showed that leader behavior
significantly predicted the team's win/loss record. In a baseball study, Jacobs and Singell (1993)
found that superior managers either improved individual performance or had excellent tactical
skills. In an extension of earlier work on strategy-style matching theory (Gupta, 1984; 1988),
Slater (1989) investigated the influence of managerial style on Strategic Business Unit (SBU)
performance. Using self-reports from 72 general managers, the study found that managers who
used a thinking mode of decision making tended to have higher SBU performance. In a study
which was very similar to the current one in the overall design, Hater and Bass (1988)
Leadership and Work Teams 7
investigated the impact of two different leadership styles, transformational and transactional, on
team performance. In the study, the authors used subordinate ratings of their manager’s behavior,
the effectiveness of the manager, the effectiveness of the work group, and the work group’s
satisfaction with their group. They also collected ratings of the manager's and work group's
effectiveness from the managers’ superiors. Hater & Bass (1988) found that transformational
leadership ratings added significantly to the prediction of worker satisfaction and work-group
performance.
Turning to research which was focused directly on leadership in self-directed work teams,
a model and typology for understanding leadership in self-managed work team was developed by
Stewart and Manz (1995). Their typology was based upon two, independent dimensions - leader
involvement and leader power orientation. The poles for the leader involvement dimension were
“active” and “passive.” The poles for the leader power orientation dimension were “autocratic”
and “democratic.” The result was a four cell taxonomy where the four different types were
labeled “overpowering leadership,” “powerless leadership,” “power building leadership,” and
“empowered leadership.” Based on this typology, various predictions or hypotheses were then
offered concerning the relationship between leader behaviors and group performance in self-
managed work teams.
Manz and Sims (1987) developed a questionnaire measure for answering questions
regarding the effect of external leadership on self-managed work teams and validated this
instrument with employees of a manufacturing plant. For external leaders, the performance of
self-managing leadership behaviors (i.e., facilitating self-observation, self-evaluation, and self-
Leadership and Work Teams 8
reinforcement) was positively correlated with perceived leadership effectiveness. As compared to
internal leaders, external leaders encouraged the development of self-management by the team.
The Manz and Sims’ (1987) Self-Management Leadership Questionnaire was also
employed in a field study conducted by Cohen, Chang, and Ledford (1997). In their study, which
involved external leaders, the relationship between outcomes and self-managed leadership was
similar for self-managed work teams and traditional teams. There was a low but significant
correlation between group effectiveness, as rated by the team members, and leadership behaviors
as rated by the team members.
In a recent study on empowered teams, Kirkman and Rosen (1999) found that external
leader behavior was signficantly related to team member ratings of empowerment. Kirkman and
Rosen’s definition of external leadership was similar to that employed by Manz and associates
(Manz and Sims, 1987, 1989; Stewart and Manz, 1995) in that it emphasized the facilitation of
self-management. External leader behavior, as rated by the external leader, was also significantly
correlated with team productivity, as rated by the external leader.
The results of previous studies, taken as a whole, indicate that in intact, “real world”
teams, leadership is significantly and positively related to team performance. In self-directed
teams, the leadership behaviors of external leaders, especially their encouragement of group self-
management, is significantly and positively related to group outcome variables including
performance.
Current Study
The current study was designed to assess the impact of internal leadership, by a
designated leader, on semi-autonomous work group performance. This study extends earlier
Leadership and Work Teams 9
research in at least three ways. First, it is an empirical study directly aimed at assessing the
relationship between leadership behaviors as performed by a designated, internal leader and
group performance in a real world, self-directed work team environment. As compared to the
self-management leader behaviors assessed by Manz and associates (Cohen et al., 1997; Manz
and Sims, 1987, 1989; Stewart and Manz, 1995), the leadership behaviors included in this study
could be considered to be more traditional in that they assessed actions taken by the leader to
maintain group performance and encourage group solidarity (Misumi, 1985).
Second, the current study gathered group performance ratings from upper-level managers
and leader behavior ratings from team members. This was done in order to limit potential
inferential biases due to collecting ratings from the same source. There is a substantial body of
research, beginning with the work of Calder (1977), which indicates that the perception of
leadership effectiveness can often best be described as a causal attributional process through
which observers attempt to explain events or behaviors that are believed to be representative of
leadership (Lord, Binning, Rush, & Thomas, 1978; Lord & Maher, 1991; Mitchell, Larson, &
Green, 1977; Rush & Lord, 1981). Research in this area indicates that observers tend to rate a
target individual on various leadership behaviors, or qualities, based more on the performance of
the leader's work team than on objective leader behaviors (Lord, Foti, & Phillips, 1982). Thus, it
is possible that the use of same source to provide ratings of leader effectiveness, or leader
behaviors, and team performance may be confounded by such inferential processes (Lord &
Maher, 1991). Thus, it is critical, in our opinion, that studies of this type obtain independent
ratings of group processes and group performance, otherwise, a clear possibility exists that any
relationship which is found could simply be the result of rater artifacts.
Leadership and Work Teams 10
Third, this study also investigated the effect of group size on the relationship between
leadership and work group performance in a field setting. As is the case with traditional group
structures, one might expect that various situational or contextual factors might moderate the
effects of leadership. Although the result of previous studies of the effects of group size have
been inconsistent (Cummings, Huber, & Arendt, 1974; Gist, Locke, & Taylor, 1987; Hackman &
Vidmar, 1990), based on Latané's (1981) social impact theory, we hypothesized that group size
might moderate the relationship between leadership behavior and group performance. One of the
three psychosocial principles that Latané (1981) discussed was what he referred to as "division of
impact." According to this principle, the extent to which a given source, the leader, has influence
over a target, the team members, will vary inversely as a function of the number of targets or
team members.
Method
Participants - Team Leaders
Data on 102 team leaders from a large manufacturing plant were gathered. Each of the
team leaders were in charge of one semi-autonomous work group which was composed of from
four to thirteen team members. Team leaders had to have been in that position with the current
team for at least 6 consecutive months in order to be included in the study. Team leaders were
selected for their positions by the company based upon test results from an assessment center
plus input from managers and team members.
Team Members’ Ratings of Leadership and Group Performance
For each of the 102 teams, 2 team members were chosen at random to provide a
subordinate rating of their team leader's performance. In order to provide ratings of leadership,
Leadership and Work Teams 11
team members had to have been in their current team for at least 6 consecutive months. Thus,
team leaders and members providing ratings had been working together on a daily basis for at
least six months.
The form used by team members to rate team leaders was designed based on job analysis
information and was designed to tap certain key dimensions using behavioral examples
representative of each dimension. Using as a starting point the job analysis, several existing
leadership questionnaires, and Misumi’s (1985) theory of leadership, three I/O psychologists
generated a series of behavioral statements that pertained to each dimension. After generating an
initial pool of behavioral statements, the statements were then reviewed with two managers at the
participating organization for accuracy and representativeness for the position. The final
instrument was a 21-item appraisal form covering 7 dimensions of leadership performance
including: team leadership, teamwork, problem identification, problem solution, work standards,
initiative, and safety. Items were rated using a 7-point Likert scale and an average or overall
rating of leadership performance was calculated. In order to gain an understanding of team
member's perceptions of team performance, team members were also asked to rate their work
team with regard to productivity, quality, and overall performance.
Ratings from the two team members from each team were averaged to create a single
team member rating. Team leadership in this study was operationalized as the sum of the 21-item
leadership scale ( = .89). Based on the raters as items method for calculating the reliability of
member ratings (Cohen et al., 1997), the interrater coefficient alpha was .55. The ratings by team
members of productivity (interrater coefficient alpha equaled .36), quality (interrater coefficient
Leadership and Work Teams 12
alpha equaled .35) and overall performance (interrater coefficient alpha equaled .48) were all
based on responses to one item by each of the two team members.
Managers’ Ratings of Team Performance
Forty-seven managers provided ratings of their respective team's performance. Managers
were asked to rate team productivity, quality, and overall team performance using a 7 point
Likert-type rating scale. Most of the 47 managers rated more than one team. The result of a one-
way ANOVA designed to test for a manager effect was non-significant (F
(46, 55)
=1.43, > .05).
Procedure
Performance ratings for team leaders and perceptions of work team performance were
gathered from a random sample of team members. Team members completed the rating forms
during a proctored administration of the instrument which included rater training. They were told
that this was part of a research project, that their responses would be totally confidential and that
the results would have no impact on the person they were rating. To assist in promoting a spirit
of confidentiality, team members placed their completed forms, along with other forms
associated with the study, into a large envelope which was then mailed to the first author for
scoring. Ratings of team performance from managers were obtained by sending the
questionnaires through the organization's mail system. Instructions similar to those provided to
the team members during training were furnished to the managers (i.e., confidentiality, rating
errors, and the use of the ratings).
Results
Descriptive statistics and the correlations between variables are presented in Table 1.
Inspection of Table 1 reveals that none of the three measures of team performance, as rated by
Leadership and Work Teams 13
managers, were significantly correlated with team leadership as rated by team members.
However, that team member ratings of leadership were significantly correlated with team
member ratings of productivity, quality, and overall performance. In addition, the managers’
ratings of team productivity, team quality, and overall performance were not correlated
significantly with the ratings of the same team characteristics by the team members.
It should be noted that rather than being unexpected, the disagreement found between
rating sources is probably the norm (Murphy & Cleveland, 1995). Studies on the relationship
between various ratings sources generally show low levels of agreement (Bass & Yammarino,
1991; Cohen et al., 1997; Furnham & Stringfield, 1994; Hoffman, Nathan, & Holden, 1991;
Murphy & Cleveland, 1995; Williams & Levy, 1992).
Insert Table 1 and Table 2 about here.
In order to test for the potential moderating effect of team size, a series of regression
analyses were conducted and the results are summarized in Table 2. The results of the analyses
indicated that team size moderated the relationship between team leadership and the managers’
ratings of team performance for both quality and overall performance. In order to illustrate the
nature of the interaction, the correlation between team leadership ratings and the team
performance ratings made by the managers was calculated separately for small groups, less than
8 members, and large groups, more than 8 members. For productivity, the correlation between
team leadership ratings and managers’ ratings of team performance was r=.20 for small groups
and r=-.03 for large groups. For quality, the correlation between team leadership ratings and
Leadership and Work Teams 14
managers’ ratings of team performance was r=.31 for small groups and r=-.12 for large groups.
For overall performance, the correlation between team leadership ratings and managers’ ratings
of team performance was r=.42 for small groups and r=-.03 for large groups. Thus, there was a
strong positive relationship between team leadership and manager rated-group performance for
smaller teams, especially those smaller than 8 members, whereas the relationship begins to
decrease as the size of the team increased.
Discussion
Team members' ratings of leadership as exhibited by the internal, designated leader were
not significantly correlated with the managers’ rating of self-directed work team performance.
However, the relationship between leadership and manager rated-team performance was
moderated by team size. Team leadership was significantly related to the managers’ ratings of
quality and overall performance ratings for teams with fewer members, but not for those with
larger numbers of team members. This differential strength of impact for leadership depending
on the size of the team is consistent with the principles of social impact theory (Latané, 1981)
and is also consistent with a "contextualist" perspective of leadership. The contextualist
perspective argues that the impact of a leader is constrained by organizational factors, such as the
organizational culture, the rules, formal authority structure, etc., which serve to diffuse the
possible impact of leadership. These factors are similar to what Fiedler and his colleagues refer
to as "multiple screens" that serve as potential barriers through which leadership must pass (c.f.
Fiedler & Leister, 1977). The size of the team may serve to diffuse or dilute the impact of the
team leader on the performance of the group. Thus, the findings from this study, that the
relationship between team leadership ratings and the managers ratings of team performance tend
Leadership and Work Teams 15
to be higher in smaller teams, is quite consistent with Latané’s (1981) principle of division of
impact and also the general contextualist approach.
While managers’ ratings of team performance were not significantly correlated with team
members’ ratings of team performance, team members' ratings of team performance were
correlated with their ratings of leader performance. If we believe for some reason that team
members were simply more accurate observers of team performance, then it is possible that this
simply reflected the actual relationship between performance and leadership. However, team
members were not given formal feedback on performance, but rather were given data that had
been aggregated across teams. An equally likely explanation for the correlation between team
members’ ratings of leadership and performance was that team member ratings were
substantially influenced by some type of biased, inference process (Lord & Maher, 1990); that is,
if a team member had an overall favorable impression of their group, they would be more likely
to rate both their team and their leader more positively. This explanation would be consistent
with various information processing approaches to performance appraisal and leadership ratings
(Lord et al., 1982). It was to avoid just this type of possible contamination when obtaining ratings
from the same source, as well as any other ensuing artifacts, that the decision was made to obtain
ratings from the managers.
As can be seen in Table 1, none of the ratings of team performance from managers were
significantly related to corresponding ratings from team members. One explanation for the
differential perceptions may be that team members observe or perceive different aspects of team
leader performance than do managers (Murphy & Cleveland, 1995). Team members may have
only been responding to internal aspects of team performance, and the internal leader’s role in
Leadership and Work Teams 16
that team performance, while the managers were also familiar with the performance of the
internal team leaders when asked to perform functions outside of their own team. It should be
noted, however, that the external responsibilities of the internal leaders were limited and
primarily involved representing the group in meetings with other team leaders and with the
managers. The agreement between team members was also fairly low, although within in the
range of reliabilities usually found in the literature, which suggests that team members may also
have been observing or reporting different aspects of the leader’s behavior. It is also possible that
in this case the managers had information on relative team performance which was not available
to the team members. In addition, both from the perspective of multiple levels of analysis theory
(c.f. Avolio, O’Connell, Maritz, & Kennedy, 1999; Bass & Yammarino, 1991) and stakeholder
theory (Donaldson & Preston, 1995) perspectives, we might expect to see inconsistencies in
ratings of performance from different levels or constituencies within the organization. Thus, the
lack of agreement between sources is a finding which should be of interest not only to those
interested in leadership perceptions in self-directed work teams, but also to those interested in
multi-source performance appraisals such as 360's.
A separate explanation could be that the managers themselves were not in a good position
to rate the performance of the team. They may have been unable to accurately rate the
performance of the team because of the continuous nature of the production process in the plant.
Although work teams were broken into functional areas along segments of the production
process, the organization itself only maintained performance information for groupings of teams;
it was for this reason that we were unable to get objective information on group performance.
The organization only collected hard data on the quality and production of the primary processes
Leadership and Work Teams 17
and not the individual teams involved in the process. Thus, it might have been quite difficult for
managers to actually differentiate, in a reliable manner, the performance of a single team,
although it would seem that it would be even more difficult for team members to have access to
this type of information and to make accurate ratings of the differential performance of teams..
Self-directed work teams and the leadership within the team can take on a number of
different forms within various organizations (Wellins et al., 1991). The organization in which
this study was conducted had decided to shift responsibility for making decisions and other
assignments and to dissipate much of the formal authority traditionally associated with
supervisors, or team leaders, to team members. Given this shift in responsibility, it is conceivable
that the role of internal team leader would have less direct impact on team performance than
would an external leader or a supervisor in a traditional organization .
According to literature surrounding the changing face of supervision and work team
design, however, the role that the team leader plays, is still a significant one (Echols & Mitchell,
1990; Jessup, 1990; Manz & Sims, 1987; 1989). The role, however, appears to be shifting from
one of formal authority in terms of decision making, goal setting, directing floor activity, and
troubleshooting, to one that focuses primarily on coaching and training team members, as well as
directing special projects, ensuring adequate resources, and monitoring overall performance
(Echols & Mitchell, 1990). Given this shift in role, the impact of team leadership may be more of
an indirect one, which manifests itself in better intra-team self-management, and more team
member-initiated troubleshooting, and less in externally noticeable shifts in productivity. This
indirect leadership function would make the role of the team leader more analogous, in some
ways, to the role of a coach on a professional sports team, than for a traditional supervisor. If
Leadership and Work Teams 18
team development is seen as an evolutionary instead of a static phenomenon, then the impact of a
team leader is likely to take months, or years, to actually manifest itself. Further, if a team leader
is truly effective in creating and maintaining a self-directed team, then the value and impact of
his/her role becomes difficult to assess behaviorally and must in a large sense be inferred from
the functioning of the group, both in terms of performance as well as in terms of team climate
and the degree of empowerment among team members.
The results of the present study provide several potentially useful insights into the role of
leadership in semi-autonomous work teams. A central finding from this study is that the direct
relationship between internal team leadership and semi-autonomous work team performance
does not appear to be as strong as might be anticipated. Although potentially disturbing to
leadership theorists, this finding is probably good news to designers of self-directed teams. It
suggests that these teams may in fact be self-governing and that the role of the internal leader is
much more indirect. Although it was impossible to evaluate in the current study, it is quite
possible that the mark of a truly effective team leader is to be more of a coach than a leader and
to empower the team to make their decisions and govern themselves (Cohen et al., 1997; Manz &
Sims, 1987, 1989). This type of behavior would be more indirect and would not likely show up
on traditional leadership behavior questionnaires, such as the one used in this study. This, in fact,
is just the role posited by proponents of self-directed teams (Echols & Mitchell, 1990; Jessup,
1990; Wellins et al., 1991). This has interesting implications for the measurement and evaluation
of leadership in teams. Future research needs to identify the types of behaviors that really do
create effective self-directed teams (Cohen et al., 1997; Manz & Sims, 1987, 1989). In addition,
future research should look at additional measures of team performance in order to gain a better
Leadership and Work Teams 19
understanding of the type of impact that internal team leaders may actually have on the group.
These might include measures of team climate, satisfaction, expansion of team member skills,
time spent in meetings, number of suggestions generated and/or implemented, etc.
The results of the study indicated that internal team leader behaviors are related to team
performance, but only on those teams with a small number of team members. This could be
caused by several factors. The first might have to do with the sheer number of direct interactions
that team members have with their internal team leader. On a smaller team those interactions
would likely be more frequent whereas on larger teams, the sheer number of opportunities that a
given team member would have to interact with the internal team leader would likely be much
less. Thus, the perceived impact of the internal leader may be less on larger teams, even though
the actual impact of that leader is the same. Another possible explanation could be that larger
teams force an internal leader to delegate a larger amount of responsibility to team members in
order to accomplish all of the tasks of the team. This, in turn, may lead other team members to
take on more leadership responsibilities and thereby, reduce the perceived impact of the team
leader. Further research should address this issue of differential impact of team leadership as it
relates to size, to determine if these are actually the reasons or if there are reasons for the
decreased impact of team leaders on larger teams.
Conclusion
Leadership and Work Teams 20
The impact of leadership on group performance was investigated using semi-autonomous
work groups. The relationship between team leadership and group performance as rated by the
managers was moderated by team size, a finding consistent with the principles of social impact
theory (Latané, 1981) and also the "contextualist" perspective of leadership. Team leadership, as
rated by team members, was related to managers ratings of quality and overall performance for
teams with fewer members, but not for those with larger numbers of members. In a finding of
relevance to those interested in multi-source performance assessments, the managers’ rating of
team performance was not significantly correlated with ratings of team performance obtained
from the team members. Consistent with various information processing theories of ratings, the
team members’ assessment of group and leadership performance were significantly correlated
with each other. Although it was not feasible to evaluate in the current study, it is quite possible
that the mark of a truly effective internal team leader is to be more of a coach and to empower the
team to make its own decisions and govern itself.
Leadership and Work Teams 21
References
Avolio, B. J., O’Connell, M. S., Maritz, D., & Kennedy, C. (1999, April). Assessment of
transformational leadership potential. Paper presented at the Fourteenth Annual Conference of
the Society for Industrial and Organizational Psychology, Atlanta, Ga.
Bass, B., & Yammarino, F. (1991). Congruence of self and others’ leadership ratings of
naval officers for understanding successful performance. Applied Psychology: An international
review, 40, 437-454.
Calder, B. J. (1977). An attribution theory of leadership. In B. M. Staw & G. R. Salancik
(Eds.), New directions in organizational behavior. Chicago, IL: St. Clair Press. (pp. 179-204).
Chelladuria, P. (1984). Discrepancy between preferences and perceptions of leadership
behavior and satisfaction of athletes in varying sports. Journal of Sport Psychology, 6, 27-41.
Cohen, S. G., Chang, L., & Ledford, G. (1997). A hierarchical construct of self-
management leadership and its relationship to quality of work life and perceived work group
effectiveness. Personnel Psychology, 50, 275-308.
Cummings, L. L., Huber, G. P., & Arendt, E. (1974). Effects of size and spatial
arrangements on group decision making. Academy of Management Journal, 17, 460-475.
Day, D. V., & Lord, R. G. (1988). Executive leadership and organizational performance:
Suggestions for a new theory and methodology. Journal of Management, 14: 453-464.
Donaldson, T., & Preston, L. (1995). The stakeholder theory of the corporation:
Concepts, evidence, implications. Academy of Management Review, 20, 65-91.
Echols, D., & Mitchell, R. (1990). Champion or victim? The supervisor's new role in a
team-based work system. Paper presented at Seventh Annual Fall Forum, Association for Quality
Leadership and Work Teams 22
and Participation, Denver, CO.
Fiedler, F. E., & Garcia, J. E. (1987). New approaches to effective leadership: Cognitive
resources and organizational performance. New York: Wiley.
Fiedler, F. E., & Leister, A. F. (1977). Leader intelligence and task performance: A test of
a multiple screen model. Organizational Behavior and Human Performance, 20, 1-14.
Furnham, A., & Stringfield, A. (1994). congruence of self and subordinate ratings of
managerial practices as a correlate of supervisor evaluation. Journal of Occupational and
Organizational Psychology, 67, 57-67.
Gist, M. E., Locke, E. A. & Taylor, M. S. (1987). Organizationa1 behavior: Group
structure, process, and effectiveness. Journal of Management, 13, 237-257.
Gupta, A. K. (1984). Contingency linkages between strategy and general manager
characteristics: A conceptual examination. Academy of Management Review, 9, 399-412.
Gupta, A. K. (1988). Contingency perspectives on strategic leadership: Current
knowledge and future research directions. In Hambrick, D.C. (Ed.), The executive effect:
Concepts and methods for studying top managers (pp. 147-178). Greenwich, CT: JAI Press.
Hackman, J. R. (1987). The design of work teams. In J. Lorsch (ed.), Handbook of
Organizational Behavior. Englewood Cliffs, N.J.: Prentice-Hall.
Hackman, J. R., & Vidmar, N. (1970). Effects of size and task type on group
performance and member relations. Sociometry, 33, 37-54.
Hater, J. J., & Bass, B. M. (1988). Superiors' evaluations and subordinates' perceptions of
transformational and transactional leadership. Journal of Applied Psychology, 73, 695-702.
Hoffman, C. C., Nathan, B. R., & Holden, L. M. (1991). a comparison of validation
Leadership and Work Teams 23
criteria: Objective versus subjective performance measures and self- versus supervisor ratings.
Personnel Psychology, 44, 601-619.
Jacobs, D., & Singell, L. (1993). Leadership and organizational performance: Isolating
links between managers and collective success. Social Science Research, 22, 165-189.
Jessup, H. R. (1990, November). New roles in team leadership. Training and
Development Journal, 44, 79-83.
Kirkman, B. L. & Rosen, B. (1999). Beyond self-management: Antecedents and
consequences of team empowerment. Academy of Management Journal, 42, 58-74.
Latané, B. (1981). The psychology of social impact. American Psychologist, 36, 343-356.
Lord, R. G., Binning, J. F., Rush, M. C., & Thomas, J. C. (1978). The effects of
performance cues and leader behavior on questionnaire ratings of leadership behavior.
Organizational Behavior and Human Performance, 21, 27-39.
Lord, R. G., Foti, R. J., & Phillips, J. S. (1982). A theory of leadership categorization. In
J. G. Hunt, U. Sekaran, & C. Schriescheim (Eds.), Leadership: Beyond established views. (pp.
104-121). Carbondale, IL: Southern Illinois University Press.
Lord, R. G., & Maher, K. J. (1990). Alternative information-processing models and their
implications for theory, research, and practice. Academy of Management Review, 15, 9-28.
Manz, C. C. & Sims, H. P., Jr., (1987). Leading workers to lead themselves: The external
leadership of self-managed work teams. Administrative Science Quarterly, 32, 106-128.
Manz, C. C., & Sims, H. P., Jr., (1989). Super leadership: Leading others to lead
themselves. Englewood Cliffs, N.J.: Prentice-Hall.
Misumi, J. (1985). The behavioral science of leadership: An interdisciplinary Japanese
Leadership and Work Teams 24
research program. Ann Arbor: The University of Michigan Press.
Mitchell, T. R., Larson, J. R., & Green, S. G. (1977). Leader behavior, situational
moderators, and group performance: An attributional process. Organizational Behavior and
Human Performance, 18, 254-268.
Murphy, K. & Cleveland, J. N. (1995). Understanding performance appraisal: Social,
organizational and goal-based perspectives. Thousand Oaks, CA: Sage.
Pearce, J. A., & Ravlin, E. C. (1987). The design and activation of self-regulating work
groups. Human Relations, 40, 751-782.
Phillips, J. S., & Lord, R. G. (1981). Causal attributions and perceptions of leadership.
Organizational Behavior and Human Performance. 28, 143-163.
Slater, S. F. (1989). The influence of managerial style on business unit performance.
Journal of Management, 15, 441-455.
Stewart, G. L. & Manz, C. C. (1995). Leadership for self-managing work teams: A
typology and integrative model. Human Relations, 48, 747-770.
Sundstrom, E., Demeuse, K., & Futrell, D. (1990). Work teams: Applications and
effectiveness. American Psychologist, 45, 120-133.
Weiss, M. R., & Friedrichs, W. D. (1986). The influence of leader behaviors, coach
attributes, and institutional variables on performance and satisfaction of collegiate basketball
teams. Journal of Sport Psychology, 8, 332-346.
Wellins, R. S., Byham, W. C., & Dixon, G. R. (1994). Inside teams: How 20 world-class
organizations are winning through teamwork. San Francisco, CA: Jossey-Bass.
Wellins, R. S., Byham, W. C, & Wilson, J. M. (1991). Empowered teams: Creating self-
Leadership and Work Teams 25
directed work groups that improve quality, productivity, and participation. San Francisco: Jossey-
Bass.
Wellins, R. S., Wilson, J., Katz, A. J., Laughlin, P., and Day, C. R., Jr. (1990). Self-
directed teams: A study of current practice. Survey report. Pittsburgh, PA.: Development
Dimensions International, Association for Quality and Participation, and Industry Week.
Williams, J., & Levy, P. (1992). The effect of perceived system knowledge on the
agreement between self-ratings and supervisor ratings. Personnel Psychology, 45, 835-847.
Wilson, J.M., George, J., Wellins, R. S., & Byham, W. C. (1994). Leadership trapeze:
Strategies for leadership in team-based organizations. Jossey-Bass: San Francisco.
Yammarino, F. J., & Bass, B. M. (1991). Person and situation views of leadership: A
multiple levels of analysis approach. Leadership Quarterly, 2, 121-139.
Leadership and Work Teams 26
Table 1. Descriptive Statistics and Correlations for All Variables.
Correlations
Variable
Mean
SD
1
2
3
4
5
6
7
Team Member Ratings of:
1. Team Productivity
5.71
1.25
2. Team Quality
5.53
1.07
.63*
3. Overall Performance
5.51
1.16
.70*
.79*
4. Team Leadership
4.46
1.33
.33*
.46*
.30*
Manager Ratings of:
5. Team Productivity
4.86
1.19
.17
.10
-.03
.04
6. Team Quality
4.85
1.22
.09
.02
-.10
.10
.61*
7. Overall Performance
4.68
1.09
.18
.08
.01
.11
.79*
.79*
Team Size
8. Team Size
8.55
2.73
-.15
-.12
.02
-.04
.01
-.14
.00
Note: All data was based on an N of 102 teams.
*p < .05
Leadership and Work Teams 27
Table 2. Moderated regression analysis: Team size as a potential moderator of the relationship
between leadership ratings for the designated, internal team leader (as rated by team members) and
the managers’ ratings of team productivity, quality and overall team performance.
Independent Variable
R
2
R
2
b
F
Manager-Rated Productivity
Team Size
.00
.19
.04
Leadership Ratings
.00
.00
.18
.00
Team Size X Leadership
.03
.03
-.19
2.88
Manager-Rated Quality
Team Size
.00
.22
.15
Leadership Ratings
.00
.00
.21
.17
Team Size X Leadership
.06
.05
-.23
4.59*
Manager-Rated Overall Performance
Team Size
.00
.27
.11
Leadership Ratings
.27
Leadership and Work Teams 28
.01
.00
.35
Team Size X Leadership
.08
.07
-.27
6.25**
* p < .05
**p < .01
Note: The F test was computed to test the significance of the increment in R-squared obtained at
the step the variable was added to the model, it does not refer to the significance of that term in
the final model. The b is the standardized partial regression coefficient associated with that term
in the final model.
A preview of this full-text is provided by SAGE Publications Inc.
Content available from Group & Organization Management
This content is subject to copyright.
