Conflict Management, Efficacy, and Performance

in Organizational Teams

Steve Alper
Covenant Behavioral Health, Milwaukee, WI
(414)327-9750, (414)327-7436 (FAX)
[email protected] (work)

Dean Tjosvold
Department of Management, Lingnan University,
Tuen Mun, Hong Kong
and
Kenneth S. Law
Hong Kong University of Science and Technology
February, 2000

The authors thank Geoff Maruyama, David W. Johnson, and other members of
Steve Alper dissertation committee for their support. They appreciate the financial
support of the Research Grants Council of Hong Kong, RGC grant project No:
LC890/96H to the second author thank Eleanor MacDonald and Michelle Berneror
their valuable contributions.

Abstract
The study empirically links conflict management literature with research on
efficacy and organizational teams. Sixty-one self-managing teams with 489 employees
were recruited from the production department of a leading electronic manufacturer.
Structural equation analysis supports the model that a cooperative instead of competitive
approach to conflict leads to conflict efficacy that in turn results in effective performance
as measured by managers. Findings suggest how organizational teams can be prepared to
make use of their autonomy to deal with problems and conflicts so that they are
productive.

Conflict Management, Efficacy, and Performance

in Organizational Teams
Teams are popular means used worldwide to improve quality, reduce costs, and
develop new products to help organizations cope with the highly competitive marketplace
and restrictive government funding (Jassawalla & Sashittal, 1999; Pfeffer & Veiga, 1999;
Salem & Banner, 1992). However, these teams, as they confront many issues and
divisions, must be able to manage conflict to be successful (Ilgen, 1999; Neck & Manz,
1994; Sims, 1995). Teams that are confident they can deal with their conflicts are likely to
work productively; teams that doubt their conflict management abilities may become
demoralized and ineffectual. This study empirically relates conflict management research
with the efficacy and teamwork literatures. It suggests that the concept of efficacy is useful
for understanding the impact of different approaches to conflict on the effectiveness of
organizational teams. The major hypothesis is that groups that rely on cooperative
approaches to managing conflict develop efficacy that they can deal with their conflicts;
this efficacy in turn results in effective team performance.
Conflict in organizational teams
Although organizational research on groups and conflict have proceeded
somewhat independently (Hackman, 1990; Rubin, Pruitt, & Kim, 1994; Thomas, 1992;
van de Vliert & Kabanoff, 1990; Weldon, Jehn, & Pradhan, 1991), recent studies
emphasize the critical role of conflict in groups (Amason, 1996; Bettenhausen, 1991;
Bettenhausen & Murnighan, 1991; Jehn, 1997, 1995: Nemeth & Owens, 1996). Groups
must contend, among other issues, with conflicts over effective and fair distribution of
work and rewards, social loafing, and the best ways to accomplish their goals (Wageman,
1995). Groups provide an interpersonal context in which conflicts occur and attempts to
manage them are made.

In traditional hierarchical organizations, employees are expected to inform their

managers and supervisors of problems and conflicts and abide by their decisions. In
organizations that use teams, especially self-managing and other forms of empowered
teams, employees are supposed to resolve problems and conflicts themselves (Cohen &
Ledford, 1994; Kirkman & Rosen, 1999; Spreitzer, 1995, 1996; Spreitzer, Kizilos, &
Nason, 1997). Self-managing team theorists have proposed that employees, as they are
closer to the source of errors and variances in production, are better situated to correct
them (Pasmore, Francis, Haldeman, & Shani, 1982). They are trained in quality
management and given the power to halt or speed up production. Their participation in
resolving these issues is expected to increase "ownership" of problems and more
commitment to implement their solutions that in turn results in improved productivity,
product quality, and work life (Herbst, 1974; Pasmore et al, 1982; Weisbord, 1987).
Self-managing proponents and sociotechnical theorists have argued for the
minimum critical specification principle: employees will work more effectively when they
are in control of their own internal functioning and work coordination without external
supervision (Herbst, 1974; Trist, 1977). Variances, problems, and frustrations do not
disappear in self-managing but are dealt with directly by employees in their teams.
Although developing conflict management capabilities would then appear to be
useful for all organizational teams, they may be particularly critical for employees in
empowered and self-managing work teams. They must resolve issues around their
personalities, work roles and habits, production procedures, the quality of work,
scheduling, and the best ways to complete the work. They also have conflicts with area
managers and with other teams in the organizations. Effective conflict resolution is needed
for employees to manage their internal functioning successfully and to make decisions to
which they are committed (Tjosvold, 1987).
Conflict management ideas may contribute to theorizing on organizational teams
and suggest the conditions under which these teams are productive. Studies have not

provided much support for team-building interventions focused generally on relationships

(Salas, Rozell, Driskell, & Mullen, 1999). Previous studies have suggested poorly
managed conflict increases the stress and strain for managers and supervisors involved in
the change to self-managing (Wall, Kemp, Jackson, & Clegg, 1986; Walton, 1982). There
is, however, little direct empirical support for the value of conflict management for
empowered organizational teams.
This study explores the dynamics by which managing conflict can contribute to the
effective performance of teams. In particular, it examines how conflict efficacy might
mediate between conflict approach and team performance.
Conflict management, efficacy and team effectiveness
Efficacy is defined as the confidence that one can use one's capabilities to execute a
course of action that will result in performance (Lee & Bobko, 1994). Research has shown
that individuals who believe they can perform needed actions exert effort and are
productive; those with little efficacy are unproductive and fail to take the initiative to
contribute to the organization (Bandura, 1993; Wood, Bandura, & Bailey, 1990). Bandura
(1982) argued that efficacy can also occur at the group level and suggested developing
task specific measures. Group efficacy may have important effects on team performance
(Gibson, 1999).We argue that conflict management is a central task for members of teams,
especially empowered, self-managing teams. Teams may come to much different
conclusions about their efficacy in handling conflict situations. As a result, we follow
Bandura argument and define conflict efficacy as the team beliefs that it can deal with
issues to manage the team conflicts productively.
Because conflict is so central to organizational groups, conflict efficacy may
contribute significantly to the team overall performance. With low levels of conflict
efficacy, teams become demoralized because they doubt that they will combine their ideas
and pool their resources to solve problems. They are then unable to perform effectively
(Guzzo, Yost, Campbell, & Shea, 1993).

This study uses Deutsch's (1990, 1980, 1973) theory of cooperation and
competition to identify major approaches to managing conflict. He defined conflict as
incompatible activities, where one person is interfering, obstructing, or in other ways
making the behavior of another less effective. He argued that whether conflict is handled
cooperatively or competitively affects the dynamics and outcomes of conflict. Protagonists
can emphasize their cooperative goals; as one moves toward goal attainment, others also
move toward goal achievement. They tend to view a conflict as a mutual problem that
needs common consideration and solution. Within this context, protagonists are confident
that others will reciprocate and work for mutually beneficial solutions. They understand
that they can pursue their interests as they pursue the interests of others. These
expectations lead to full exchange of diverse ideas and perspectives that are combined into
effective, mutually advantageous solutions. Experiences of confirming these positive
expectations and engaging in flexible, mutually beneficial conflict management processes
strengthen the efficacy among group members that they can handle their conflicts
effectively.
Protagonists can also emphasize their competitive interests; as one succeeds the
other moves away from goal attainment. They tend to view the conflict as a win-lose
struggle; if the other wins, they lose. This social context induces the expectations that
others will fail to reciprocate and indeed will obstruct one own efforts as they pursue
their incompatible interests. These doubts lead to biased communication and inflexibility
and results in deadlock or an imposition of a solution by the more powerful protagonist.
Confirming these suspicions and experiences in competitive conflict management induce
skepticism that the team can deal effectively with conflict.
Social psychological research has documented that whether protagonists
emphasize cooperative or competitive goals very much alters the dynamics and outcomes
of conflict (Deutsch, 1990, 1980, Tjosvold, 1998; Tjosvold, Leung, & Johnson, in press).
A great deal of research has developed our understanding of the impact of cooperative and

competitive goal interdependence on relationships more generally (Johnson & Johnson,

1993; Johnson, Maruyama, Johnson, Nelson & Skon, 1981). Recent studies have extended
the cooperative-competitive conflict approach to organizational settings (Barker, Tjosvold,
& Andrews, 1988; Tjosvold, 1999; Tjosvold, Dann, & Wong, 1992; Tjosvold, Morishima,
& Belsheim, 1999).
This study tests the applicability of Deutsch conflict framework to the important
issue of organizational teams. Specifically, cooperative conflict is expected to induce high
levels of conflict efficacy; competitive conflict management induces low levels of conflict
efficacy. Teams with conflict efficacy believe that they can work together effectively
resulting in team productivity. These proposed relations are summarized in the following
three hypotheses:
H1a. Teams that rely on a cooperative approach to conflict develop feelings of efficacy
that they can deal with their conflicts.
H1b. Teams that rely on a competitive approach to conflict develop low efficacy that
they can deal with their conflicts.
H2 Teams that develop perceptions of high conflict efficacy will be more effective than
those with low perceived conflict efficacy.
The above discussions and hypotheses together suggest that conflict approaches
affect conflict efficacy which in turn impacts team effectiveness. Figure 1 pictures the
theorized relationships.
-------------------------------------Insert Figure 1 about here
--------------------------------------

Method
Measures

A leading manufacturer of portable and stationary electric generators sets and

related switchgear and controls, small gasoline engines, and alternating current generators
agreed to participate in the study. Self-directing teams, which had developed five years
before data were collected, were recruited from its production department. Nearly all
respondents had been in their teams for over six months. The company is in the Midwest
and top and middle management supported the study. Only employees who volunteered
completed the survey. Eleven employees did not agree. The survey was completed during
work time and took about 20 minutes. The initial sample consisted of 67 teams with 538
employees.
Conflict Approaches
Scales for cooperative and competitive approaches to conflict were developed
from a series of experimental studies (Tjosvold, 1985) and from a questionnaire study on
project managers (Barker, et al, 1986). The five cooperative approach (COOP) items
measured the emphasis on mutual goals, understanding everyone's views, orientation
toward joint benefit, and incorporating several positions to find a solution good for all. A
sample item for the cooperative approach scale is e seek a solution that will be good for
the whole team. Subjects were asked to rate on a 7-point scale (1=strongly agree,
7=strongly disagree) their degree of agreement to the five statements. The competitive
approach scale (COMP) had five items with similar anchors to measure the assumption
that the conflict was a win-lose situation, and the use of pressure and intimidation to get
others to conform to one's view. A sample item is ndividual team members treat conflict
as a win-lose contest.
Conflict Efficacy
The conflict efficacy (CE) scale was a new 6-item scale developed for this study
which measured the beliefs team members have that their team could successfully manage
different conflict situations. In discussions with employees at the factory, they indicated the
most common and difficult conflict situations; the most frequently mentioned conflict

situations were included in this scale. In addition, items concerning work quality and work
productivity were included because they are considered important reasons why work
teams are implemented. Subjects were asked about their degree of agreement with the six
statements on a 7-point scale (1=strongly agree, 7=strongly disagree). The six items are
listed in the Appendix.
Team Effectiveness Measures
As with other work team research (Cohen & Ledford, 1994; Goodman, Devadas,
& Griffith-Hughson, 1988), obtaining objective work outcome measures proved
impossible despite the willingness of the organization to provide them. The company did
not collect team level productivity data. Quality data were unreliable because some quality
inspectors did not report defects; some teams performed much more complex functions
and comparison of reported defects could be easily misinterpreted. On-time-delivery
measures were abandoned because often they were due to factors external to the team.
Therefore, we used manager ratings of team performance as the effectiveness measure.
Proposing that there is no strictly objective measure of performance in organizations,
Pritchard (1992) argued that ratings can measure the extent users of the team outputs find
them productive. In addition, these managers should be knowledgeable about the group
performance (Hackman, (1987).
Supervisors and team leaders completed an 18 item team effectiveness scale
(RATE) developed for this study. The items involve productivity, quality, and cost savings
because these are central reasons why self-directed work teams are initiated (Appendix).
The managers completed this scale four weeks after the employees questionnaire was
administered. For most of the teams, either the supervisor of the team or the team leader
rated the performance of the whole team. There were, however, some teams where two
to four supervisors rated team performance. When there were more than one rater, the
average across all raters was used as the manager rating of team performance.
Analysis

Team members provided ratings of the whole team on the conflict approaches used
by their team members as well as their perception of conflict efficacy of the team. Team
managers provided a single rating for the effectiveness of the whole team. Since the
hypotheses are developed at the team level, individual ratings on cooperative (COOP) and
competitive (COMP) conflict approach and conflict efficacy (CE) were aggregated to the
team level.
Data aggregation
Aggregating individual rating to the team level is logically justified because all
three variables (COOP, COMP and CE) are directed to the characteristics of the work
team. We still tested whether the ratings of group members are reasonably homogeneous
before the data were aggregated to the group level, using the James, Demaree, and Wolf
(1984) procedure to estimate the inter-ratter reliability of members within each team for
the variables of COOP, COMP and CE. James et al. rWG(J) index was used as an estimate
of inter-rater reliability because each of the four variables were measured by multiple
items. Two indicators showed that the ratings among members in each group were quite
homogeneous. First, the median rWG(J) for the four variables across the 67 teams were .
87, .79 and .93 respectively. Second, George and Bettenhausen (1990) argued that rWG(J)
which was greater than or equal to .70 could be considered as indicators of good
agreement within group. Of the 67 teams, the percentages of teams with rWG(J) greater than
or equal to .70 across the three variables were 84%, 73% and 93% respectively. We
therefore concluded that the within-team ratings were homogeneous enough to be
aggregated to the team level. Individual team members ratings were aggregated to the
team level and the data merged with supervisory ratings of team performance (RATE).
The final sample size of the merged data file was 61 teams with 489 team members.
Correlations among the three predictor variables and the outcome variable at the team
level are shown in Table 1.

------------------------------Insert Table 1 about here

------------------------------Conflict efficacy as a full or partial mediator
Because we do not have specific hypothesis on whether conflict efficacy is a full or
partial mediator, we tested the base-line path model (M1) shown in Figure 1 against two
alternative partial mediating models (M2 and M3). In addition to the base-line model (M1),
M2 has two additional direct paths from COOP and COMP to team managers rating of
team performance. These two paths are added because we do not have concrete
theoretical support that conflict efficacy will fully mediate the conflict approach-team
performance relationship. It is possible, for example, that COOP may have some direct
effects on team performance on top of its mediating effect through conflict efficacy.
The second alternative model (M3) tested whether the relational order between
conflict efficacy (CE) and conflict approach (COOP and COMP) should be reversed. The
theory of cooperation and competition suggested the hypothesis that cooperative conflict
approach leads to a sense of conflict efficacy which in turn affects team performance and a
competitive conflict approach decreases the team conflict efficacy that leads to poor
team performance. However, given the seemingly high correlation between conflict
efficacy and conflict approach and that they are measured from the same source (i.e. team
members), we also tested the alternative model that conflict efficacy is antecedent to
conflict approach which, in turn, leads to team performance. This alternative model is
labeled M3.
The series of path analyses were conducted using LISREL 8.12a (Jeskog &
Sbom, 1993). We used three indicators to judge if the observed data fit into our
hypothesized models: An overall chi-square measure and its associated degrees of
freedom, the comparative fit index (CFI) and the Tucker-Lewis index (TLI, Tucker and
Lewis, 1973). Bentler and Bonnett (1980) suggest that the CFI should be above .90 for

sufficient fit. The TLI compares the relative improvement in fit for the proposed model
over a strict null model of complete independence among the various items. In contrast to
the CFI, the TLI appears to be relatively robust across model characteristics (Wheaton,
1987) and by small or large sample sizes (Marsh, Balla & McDonald, 1988). Results of
the nested model analyses are shown in Table 2.
------------------------------Insert Table 2 about here
-------------------------------

Results
Table 1 indicates the mean, standard deviation, reliabilities, and correlations among
the constructs in this study. The correlations indicated that the cooperative approach to
conflict was positively related to conflict efficacy (r=.78 , p<.01). In contrast, the
competitive approach to conflict was negatively related to conflict efficacy (r=-.61,
p<.01). The competitive approach to conflict was negatively related to managers rating of
team effectiveness (r=-.27, p<.01).
Table 2 shows the results of the path analysis. As model M2 has a path joining
each and every variable, it is fully saturated with zero degrees of freedom and model Chisquare. Model M1 is, however, nested within M2 with the two paths from COOP and
COMP to RATE (team managers ratings of team performance) dropped. We can,
therefore, compare the model Chi-square between M1 and M2. Table 2 shows that M1 has
a model Chi-square of 1.47 and two degrees of freedom. The change in model Chi-square
between M1 and M2 is not statistically different. As a result, we conclude that the two
direct paths from COOP and COMP to RATE are unnecessary. Conflict efficacy will fully
mediate the conflict approach-team performance relationship.
Table 2 also shows the results of reversing the ordering of the variables in the
model. Specifically, when CE is modeled as the antecedent of COOP and COMP which in

turn affect RATE (M3), the model Chi-square is 2.35 with 2 degrees of freedom. As
predicted CE has positive effects on COOP and negative effects on COMP. However,
both the two paths from COOP and COMP to RATE are insignificant. In contrast, when
CE is modeled as the mediator between conflict approach and team performance, all the
paths are significant as predicted. Although the two models are not nested within each
other, M1 has a relatively lower model Chi-square than M3. Both the CFI and TLI show
almost perfect model fit. Based on these observations, we conclude that more evidence
supports our original model that conflict efficacy fully mediates the conflict approach and
team performance relationship.
The path coefficient estimates of the final mediating model (M1) are shown in
Figure 2. Consistent with the theorizing, results of the structural equation analysis suggest
that a cooperative approach results in effective outcomes. Cooperative approach had a
highly significant effect on conflict efficacy (b=.72, p<.01); competitive approach had a
significant negative impact on conflict efficacy (b=-.26, p<.01). The results provide good
support for hypothesis 1.
-------------------------------------Insert Figure 2 about here
-------------------------------------Conflict efficacy was in turn found to have a significant effect on team managers
rating of team effectiveness (b=.29, p<.05). Hypothesis 2 is also strongly supported.
Results overall provide good support for the hypotheses and the cooperative and
competitive theory of conflict management. Teams that handle conflicts cooperatively
tended to generate strong conflict efficacy. Teams with high conflict efficacy also have
superior performance.

Discussion

This study empirically links conflict management and efficacy research with the
literature on organizational teams. Empowered organizational teams can be effective but
they have a great many issues and conflicts to manage (Barker, 1993; Cohen & Ledford,
1994). The results of this study, including the path analyses, suggest that how team
members manage their conflicts can affect not only their sense of efficacy in dealing with
conflicts but their overall team performance. Specifically, managing conflict for mutual
benefit was found to predict to the extent team members believed they could handle
various conflicts and to their supervisor conclusions about their team effectiveness.
Results of the correlational analyses support the reasoning that competitive conflict
has a largely negative impact on conflict efficacy. Teams that relied on competitive conflict
were found to exhibit low levels of conflict efficacy and reduced group performance. It
should be noted, however, that in the path analysis, competitive conflict did not
significantly lead to low conflict efficacy.
Findings provide further support for the utility of the Deutsch perspective on
conflict management for understanding important organizational issues. Deutsch's theory
of cooperative and competitive conflict has been developed mostly by social psychologists,
often in laboratory studies. Some recent studies have used interviews to test the
generalizability of the Deutsch theory to organizations (Tjosvold, 1999; Tjosvold, et al,
1999). This study supplements these studies by using questionnaires with a large sample
and the team as the unit of analysis to suggest how and by what approach managing
conflict can contribute to effective organizational teams. The study makes methodological
contributions to previous research in that it allowed independent measures of conflict
approaches and effectiveness. Team members rated their conflict management approaches
and managers rated team productivity.
Results further support current evidence and theorizing on the benefits of conflict
for groups and organizations (De Dreu & van de Vliert, 1997). Recent studies have found
that task types of conflicts compared to relationship ones are more useful for groups

(Jehn, 1997, 1995). It may be that task conflicts are more easily discussed cooperatively
whereas relationship ones become competitive. Future research is needed to clarify this
speculation.
This study contributes to efficacy research by examining a particular kind of
collective efficacy (Bandura, 1982, 1993; Gibson, 1999; Parker, 1993; Wood, et al, 1990).
Efficacy research has focused on the consequences of self and collective efficacy on
performance and other outcomes (Geringer & Frayne, 1993; Wood, et al, 1990). Less
research has identified the conditions under which people feel efficacious (Major,
Cozzarelli, Sciacchitano, Cooper, Testa, & Mueller, 1990). The study analyses supports
the argument that cooperative conflict experiences are important antecedents to conflict
efficacy and as well as for team effectiveness. In a team setting, knowing that group
members tend to manage conflict cooperatively can strengthen conflict efficacy and team
productivity.
Limitations
The results of this study are, of course, limited by the sample and operations.
Cooperative conflict management may be particularly useful for self-managing teams, but
other kinds of organization teams must also manage conflict. The data are self-reported
and subject to biases, and may not accurately describe the situation and dynamics,
although recent research suggests that self-reported data are not as limited as commonly
expected, that people often accurately perceive their social environment, and that,
therefore, self-reported data are useful for understanding people psychological
experience (Balzer & Sulsky, 1992; Crampton & Wagner, 1994; Funder, Kolar, &
Blackman, 1995; Murphy, Jako, & Anhalt, 1992; Spector, 1994). These data are also
correlational and do not provide direct evidence of causal links between conflict
approaches, conflict efficacy, and effectiveness. In addition, team members supplied most
of the measures. However, studies suggest that common method variance may not be as
much of an artifact as often assumed (Avolio, Yammarino, & Bass, 1991; Spector, 1987).

The limitations of this study should be considered in the context of previous research,
which provide experimental support with behavioral measures for the major findings of
this study.
Practical Implications
If successfully replicated, this study has potentially significant implications for
teams in organizations. This study helps to specify the group processes critical for team
effectiveness (Campion, Medsker, & Higgs, 1993; Manz & Sims, 1987; Salas, Rozell,
Driskell, & Mullen, 1999). Developing effective ways of managing conflict, and, in
particular, cooperative approaches, results suggest, could be useful for helping teams feel
they can deal with conflict so that they are able to take advantage of their autonomy and
opportunities to contribute successfully to the organization.
Previous research provides guidance for fostering cooperative conflict approaches.
To the extent that teams are committed to cooperative goals as well as discuss their views
open-mindedly they are more able to deal with their conflicts constructively (Tjosvold &
Tjosvold, 1995, 1994). They recognize that they want to resolve the conflict for mutual
benefit. They realize that their goal is to help each other get what each other really needs
and values, and not to try to win or to outdo each other.

Organizational teams cannot be expected automatically to feel empowered and

confident (Kirkman & Rosen, 1999; Spreitzer, Kizilos, & Nason, 1997). Indeed, without
the ability to discuss their problems and manage their conflicts they may well feel
demoralized. Organizational teams are alternative ways of dealing with critical issues; they
do not by themselves improve quality, reduce costs, and develop new products.
Empowering is not simply giving teams the autonomy to be self-directing. These teams
also need the abilities to manage conflict. Organizational teams that rely on cooperative
approaches to conflict would appear to be good candidates for making use of their
autonomy to work effectively for themselves and the organization.

Appendix
Conflict Efficacy
I believe that our team will manage the following conflicts in an effective manner:
1. among team members concerning personality differences.
2. among team members concerning work habits.
3. among team members concerning safety issues
4. among team members concerning work roles.
5. among team members concerning scheduling.
6. among team members concerning the best way to get a project done.
Manager Rating of Team Effective Performance
1. Team members actively engage in reviewing their work so that they can improve it.
2. Team members come up with ideas on how to reduce costs.
3. Team members work effectively.
4. Team members have to redo their work because of sloppy workmanship.
5. Team members have successfully implemented ideas to reduce costs.
6. Team members put considerable effort into their jobs.
7. Team members are concerned about the quality of their work.
8. Team members are wasteful in how they use their work materials.
9. Team members meet or exceed their productivity requirements.
10. Team members are committed to producing quality work.
11. Team members take good care of their tools and machinery.
12. Team members do their part to ensure that their products will be delivered on time.
13. Team members comes up with ideas on how to produce higher-quality work.
14. Team members take preventive action so that machinery and tools will not be
damaged.
15. Team members search for ways to be more productive.

16. Team members have successfully implemented ideas to come up with higher quality
work.
17. Team members do not abuse their sick leave policy.
18. Team members have successfully implemented plans to be more productive.

References
Amason,A.C.(1996).Distinguishingtheeffectsoffunctionalanddysfunctionalconflict
onstrategicdecisionmaking:Resolvingaparadoxfortopmanagementteams.
AcademyofManagementJournal,39,123148.
Avolio,B.J.,Yammarino,F.J.&Bass,B.M.(1991).Identifyingcommonmethod
variancewithdatacollectedfromasinglesource.JournalofManagement,17,
571587.
Balzer,W.K.&Sulsky,L.M.(1992).Haloandperformanceappraisalresearch:A
criticalexamination.JournalofAppliedPsychology,77,975985.
Bandura,A.(1982).Selfefficacymechanisminhumanagency.AmericanPsychologist,
37,122147.
Bandura,A.(1993).Perceivedselfefficacyincognitivedevelopmentandfunctioning.
EducationalPsychologist,28,117148.
Barker,J,Tjosvold,D.,&Andrews,I.R.(1988).Conflictapproachesofeffectiveand
ineffectivemanagers:Afieldstudyinamatrixorganization.Journalof
ManagementStudies,25,167178.
Barker,J.R.(1993).Tighteningtheironcage:Concertivecontrolinselfmanagingteams.
AdministrativeScienceQuarterly,38,408437.
Bentler,P.M.,&Bonnett,D.G.(1980).Significancetestsandgoodnessoffitinthe
analysisofcovariancestructure.PsychologicalBulletin,88:588606.

Bettenhausen,K.L.&Murnighan,J.K.(1991).Thedevelopmentofanintragroupnorm
andtheeffectsofinterpersonalandstructuralchallenges.AdministrativeScience
Quarterly,36,2035.
Bettenhausen,K.L.(1991).Fiveyearsofgroupsresearch:Whatwehavelearnedand
whatneedstobeaddressed.JournalofManagement,17,345381.
Champion,M.A.,Medsker,G.J.Higgs,A.C.(1993).Relationsbetweenworkgroup
characteristicsandeffectiveness:Implicationsfordesigningeffectivework
groups.PersonnelPsychology,46,823850.
Cohen,S.G.&Ledford,G.E.,Jr.(1994).Theeffectivenessofselfmanagingteams:A
quasiexperiment.HumanRelations,47,1343.
DeDreu,C.andvandeVliert,E.(Eds.)(1997).UsingConflictinOrganizations.
BeverlyHills,CA:Sage.
Deutsch,M.(1973).Theresolutionofconflict.NewHaven,CT:YaleUniversityPress.
Deutsch,M.(1980).Fiftyyearsofconflict.InL.Festinger(ed.).,Retrospectionson
socialpsychology.(pp.4677)NewYork:OxfordUniversityPress.
Deutsch,M.(1990).Sixtyyearsofconflict.TheInternationalJournalofConflict
Management,1,237263.
Funder,D.C.,Kolar,D.C.,&Blackman,M.C.(1995).JournalofPersonalityandSocial
Psychology,69,656672.
George,J.M.,&Bettenhausen,K.(1990).Understandingprosaicbehavior,sales
performance,andturnover:Agrouplevelanalysisinaservicecontext.Journalof
AppliedPsychology,75,698709.

Geringer,J.M.&Frayne,C.A.(1993).Selfefficacy,outcomeexpectancyand
performanceofinternationaljointventuregeneralmanagers.CanadianJournalof
AdministrativeSciences,10,322333.
Gibson,C.B.(1999).Dotheydowhattheybelievetheycan?Groupefficacyandgroup
effectivenessacrosstasksandcultures.AcademyofManagementJournal,42,
138152.
Goodman,P.S.,Devadas,R.&GriffithHughson,T.L.(1988).Groupsandproductivity:
Analyzingtheeffectivenessofselfmanagingteams.InJ.P.Campbell&R.J.
Campbell(Eds.),Productivityinorganizations:NewPerspectivesfromindustrial
andorganizationalpsychology.SanFrancisco:JosseyBass,295327.
Guzzo,R.A.,Yost,P.R.,Campbell,R.J.&Shea,G.P.(1993).Potencyingroups:
Articulatingaconstruct.BritishJournalofSocialPsychology,32,87106.
Hackman,J.R.(1987).Thedesignofworkteams.InJ.Lorsch(ed.),Handbookof
OrganizationalBehavior.EnglewoodCliffs:PrenticeHall,pp.315342.
Hackman,J.R.(1990).Groupsthatwork(andthosethatdon't).SanFrancisco:Jossey
Bass.
Ilgen,D.R.(1999).Teamsembeddedinorganizations.AmericanPsychologist,54,129
139.
James,L.R.,Demaree,R.G.,&Wolf,G.(1984).Estimatingwithingroupinterrater
reliabilitywithandwithoutresponsebias.JournalofAppliedPsychology,69,
8598.
Jassawalla,A.R.&Sashittal,H.C.(1999).Buildingcollaborativecrossfunctionalnew
productteams.AcademyofManagementExecutive,13,5063.

Jehn,K.A.(1997).Aqualitativeanalysisofconflicttypesanddimensionsin
organizationalgroups.AdministrativeScienceQuarterly,42,530557.
Jehn,K.A.(1995).Amultimethodexaminationofthebenefitsanddetrimentsof
intragroupconflict.AdministrativeScienceQuarterly,40,256282.
Johnson,D.W.&Johnson,R.T.(1989).Cooperationandcompetition:Theoryand
research.Edina,MN:InteractionBookCompany.
Johnson,D.W.,G.Maruyama,R.T.Johnson,D.Nelson,&S.Skon.(1981).Effectsof
cooperative,competitive,andindividualisticgoalstructuresonachievement:A
metaanalysis.PsychologicalBulletin,89,4762.
Jeskog,K.G.&Sbom,D.(1993).LISREL:Structuralequationmodelingwith
SIMPLUScommandlanguage.Hillsdale,NJ:ErlbaumandScientificSoftware
International.
Kirkman,B.L.&Rosen,B.(1999).Beyondselfmanagement:Antecedentsand
consequencesofteamempowerment.AcademyofManagementJournal,42,58
74.
Lee,C.&Bobko,P.(1994).Selfefficacybeliefs:Comparisonoffivemeasures.Journal
ofAppliedPsychology,79,364369.
Major,B.,Cozzarelli,C.,Sciacchitano,A.M.,Cooper,M.L.,Testa,M.&Mueller,P.M.
(1990).Perceivedsocialsupport,selfefficacy,andadjustmenttoabortion.
JournalofPersonalityandSocialPsychology,59,452463.
Manz,C.C.&Sims,H.P.,Jr.(1987).Leadingworkerstoleadthemselves:Theexternal
leadershipofselfmanagingworkteams.AdministrativeScienceQuarterly,32,
106129.

Marsh,H.W.,Balla,J.R.,&McDonald,R.P.1988.Goodnessoffitindexesin
confirmatoryfactoranalysis:Theeffectofsamplesize.PsychologicalBulletin,
103:391410.
Murphy,K.R.,Jako,R.A.&Anhalt,R.L.(1992).Natureandconsequencesofhalo
error:Acriticalanalysis.JournalofAppliedPsychology,78,218229.
Neck,C.P.&Manz,C.C.(1994).Fromgroupthinktoteamthink:Towardthecreationof
constructivethoughtpatternsinselfmanagingworkteams.HumanRelations,47,
929952.
Nemeth,C.&Owens,P.(1996).Makinggroupsmoreeffective:Thevalueofminority
dissent.In.M.A.West(Ed.),HandbookofWorkGroupPsychology,pp.125
141.Wiley:Chichester.
Parker,L.E.(1993).Whentofixitandwhentoleave:Relationshipsamongperceived
control,selfefficacy,dissent,andexit.JournalofAppliedPsychology,78,949
959.
Pasmore,W.,Francis,C.Haldeman,J.&Shani,A.(1982).Sociotechnicalsystems:A
NorthAmericanreflectiononempiricalstudiesoftheSeventies.Human
Relations,35,11791204.
Pfeffer,J.&Veiga,J.F.(1999).Puttingpeoplefirstfororganizationalsuccess.The
AcademyofManagementExecutive,13,3748.
Pritchard,D.(1992).Organizationalproductivity.InM.D.Dunnette&L.M.Hough
(eds.),HandbookofIndustrialandOrganizationalPsychology,Vol.3.PaloAlto:
ConsultingPsychologistsPress,pp.443471.

Rubin,J.Z.,Pruitt,D.G.,&Kim,S.H.(1994).Socialconflict:Escalation,stalemate,
andsettlement.(2nded.).NewYork:McGrawHill.
Salas,E.,Rozell,D.,Driskell,J.E.,&Mullen,B.(1999).Theeffectofteambuildingon
performance:Anintegration.SmallGroupResearch,30,309329.
Salem,M.A.&Banner,D.K.(1992).Selfmanagingworkteams:Aninternational
perspective.Leadership&OrganizationDevelopmentJournal,13,38.
Sims,H.P.Jr.(1995).ChallengestoimplementingselfmanagingteamsPart2.Journal
forQuality&Participation,18,2431.
Spector,P.E.(1987).Methodvarianceasanartifactinselfreportedaffectand
perceptionsatwork:Mythorsignificantproblem.JournalofAppliedPsychology,
72,438443.
Spector,P.E.(1994).UsingselfreportquestionnairesinOBresearch:Acommentonthe
useofacontroversialmethod.JournalofOrganizationalBehavior,15,385392.
Spreitzer,G.M.(1995).Psychologicalempowermentintheworkplace:Dimensions,
measurement,andvalidation.AcademyofManagementJournal,38,14421465.
Spreitzer,G.M.(1996).Socialstructurecharacteristicsofpsychologicalempowerment.
AcademyofManagementJournal,39,483504.
Spreitzer,G.M.,Kizilos,M.A.&Nason,S.W.(1997).Adimensionalanalysisofthe
relationshipbetweenpsychologicalempowermentandeffectiveness,satisfaction,
andstrain.JournalofManagement,23,679704.

Thomas,K.W.(1992).Conflictandnegotiationprocessesinorganizations.InM.D.
Dunnette&L.M.Hough(eds.).HandbookofIndustrialandOrganizational
Psychology,3,(PaloAlto,CA:ConsultingPsychologistPress.)651717.
Tjosvold,D.(1982).Effectsoftheapproachtocontroversyonsuperiors'incorporationof
subordinates'informationindecisionmaking.JournalofAppliedPsychology,
67,189193.
Tjosvold,D.(1985).Implicationsofcontroversyresearchformanagement.Journalof
Management,11,2137.
Tjosvold,D.(1987).Participation:Acloselookatitsdynamics.JournalofManagement,
13,739750.
Tjosvold,D.(1987).Participation:Acloselookatitsdynamics.JournalofManagement,
13,739750.
Tjosvold,D.(1990).Makingatechnologicalinnovationwork:Collaborationtosolve
problems.HumanRelations.43,11171131.
Tjosvold,D.(1998).Thecooperativeandcompetitivegoalapproachtoconflict:
Accomplishmentsandchallenges.AppliedPsychology:AnInternationalReview,.
Tjosvold,D.(1999).BridgingEastandWesttodevelopnewproductsandtrust:
InterdependenceandinteractionbetweenaHongKongparentandNorth
Americansubsidiary.InternationalJournalofInnovationManagement,3,233
252.
Tjosvold,D.Dann,V.&Wong,C.L.(1992).Managingconflictbetweendepartmentsto
servecustomers.HumanRelations,45,10351054.

Tjosvold,D.Leung,K.&Johnson,D.W.(inpress).Cooperativeandcompetitive
conflictinChina.InM.DeutschandP.T.Coleman(Eds.)HandbookofConflict
Resolution:TheoryandPractice.SanFrancisco:JosseyBass.
Tjosvold,D.Morishima,M.&JBelsheim,J.A.(1999).Complainthandlingontheshop
floor:Cooperativerelationshipsandopenmindedstrategies.InternationalJournal
ofConflictManagement,10,4568,1999.
Tjosvold,D.&Tjosvold,M.M.(1994).Cooperation,competition,andconstructive
controversy:Knowledgetoempowerselfmanagingteams.InM.M.Beyerlein
andD.AJohnson(eds.),Advancesininterdisciplinarystudiesofworkteams.
Vol.1.Greenwich,CT:JAIPress.119144.
Tjosvold,D.&Tjosvold,M.M.(1995).Crossfunctionalteamwork:Thechallengeof
involvingprofessionals.InM.M.Beyerlein,D.AJohnson,andS.T.Beyerlein,
(eds.),Advancesininterdisciplinarystudiesofworkteams.Vol.2.Greenwich,
CT:JAIPress,134.
Trist,E.(1977).Collaborationinworksettings:Apersonalperspective.Journalof
AppliedBehavioralSciences,13,268278.
Tucker,L.R.,&Lewis,C.(1973).Thereliabilitycoefficientformaximumlikelihood
factoranalysis.Psychometrika,38:110.
Tucker,L.R.,&Lewis,C.1973.Thereliabilitycoefficientformaximumlikelihood
factoranalysis.Psychometrika,38:110.
VandeVliert,E.&Kabanoff,B.(1990).Towardtheorybasedmeasuresofconflict
management.AcademyofManagementJournal,33,199209.

Wageman,R.(1995).Interdependenceandgroupeffectiveness.AdministrativeScience
Quarterly,40,145180.
Wall,T.D.,Kemp,N.J.,Jackson,P.R.,&Clegg,C.W.(1986).Outcomesof
autonomousworkgroups:Alongtermfieldexperiment.Academyof
ManagementJournal,29,280304.
Walton,R.E.(1982).TheTopekaworksystem:Optimisticvisions,pessimistic
hypotheses,andreality.InR.Zager&M.P.Rosnow(Eds.),Theinnovative
organization:Productivityprogramsinaction.Elmsford,NY:PergamonPress.
Weldon,E.,Jehn,K.A.,&Pradhan,P.(1991).Processesthatmediatetherelationship
betweenagroupgoalandimprovedgroupperformance.JournalofPersonality
andSocialPsychology,61,555569.
Wheaton,B.1987.Assessmentoffitinoveridentifiedmodelswithlatentvariables.
SociologicalMethodsandResearch,16:118154.
Wood,R.,Bandura,A.&Bailey,T.(1990).Mechanismsgoverningorganizational
performanceincomplexdecisionmakingenvironments.OrganizationalBehavior
andHumanDecisionProcesses,46,181201.

Table 1. Correlations Among the Variables

Variable

Mean SD

COOP

COMP

CE

1. Cooperative

2.95

.58

(.92)

2. Competitive

4.20

.56

-.55**

(.88)

3. Conflict Efficacy

3.06

.63

.78**

-.61**

(.92)

5. Manager Rating of
Performance

2.83

.73

.22

-.27*

.25*

Note:
(1) *p<..05; **p<.01
(2) Values in bracket are reliability (coefficient alpha) estimates.
(3) N = 61.

RATE

(.94)

Table 2. Results of the Nested Model Analyses of the measurement models 12

Model M1
Path
Path
coefficient
COOP -> CE
.72**
COMP -> CE
-.26**
CE -> RATE
.29*
Model
d.f.
CFI
TLI

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

1.47
2
1.00
1.02

Model M3
Path
Path
coefficient
CE -> COOP
.72**
CE -> COMP
-.52**
COOP -> RATE
.12
COMP -> RATE
-.28
2
2.35
Model
d.f.
2
CFI
1.00
TLI
1.00

Figure 1. The Proposed Model of Conflict Dynamics and Outcomes

C o n flic t re s o lu tio n

In te r a c tio n s

O u tc o m e s

C o o p e rative
C o n flic t
E ffic a c y

S u p e rvis o ry
E ffe c tive n e s s
ra tin g s

C o m p e titive

C o n flic t re s o lu tio n

In te r a c tio n s

O u tc o m e s

C o o p e rative
C o n flic t
E ffic a c y

C o m p e titive

S u p e rvis o ry
E ffe c tive n e s s
ratin g s

Figure 2. Path Estimates of the Final Model of Conflict Dynamics and Outcomes

Conflict resolution

Interactions

Outcomes

Cooperative
.72**

Conflict
Efficacy

-.26**
Competitive

Note: **p<.01; *p<.05; ns p>.10

.29*

Supervisory
Effectiveness
ratings