Self-management teams are a relatively new type of teams which are frequently
used in modern organizations. Self-management teams have several defining characteristics.
They are given relatively whole work tasks and are allowed increased autonomy
and control over their work (Hackman, 1986; Manz,
1992). In addition, the members of such teams are responsible for many traditional
management functions, such as assigning members to various tasks, solving within-team
quality and interpersonal problems and conducting team meetings (Lawler,
1986). Leaders in self-management teams usually dont try to control
the team; their primary function is to enable self-management.
Many benefits have been attributed to self-managing teams, including increased
productivity, better quality work and improved quality of work life for employees,
as well as decreased absenteeism and turnover (Cohen and
Ledford, 1994; Manz, 1992). Despite many advantage
of self-management teams (Manz and Sims, 1987), sometimes
they may fail. Self-managing weaken the influences of authority out-of-team
and increase uncertainty. For example, if team conflict occurred in self-management
team, it must resolve in the team and by the team members themselves. Team would
disintegrate if the level of conflict out of control.
Team conflict means serious disagreement and argument among team members about
important issues related to the team goals (Jehn, 1995;
Simons and Peterson, 2000). Any divergence or differences
between team members may lead to team conflict, such as different interests,
various values and cultures, or diversity in personality traits.
Academic perspective about team conflict developed in three stages (Robbins,
2004). In the first stage, scholars regarded conflict as irrational, harmful,
destructive monster, which should be eliminated in organizations or in teams.
In the second stage, people realized that conflict is inevitable and it should
be fully understood and accepted. Managers should find some valid ways to depress
the negative effects of team conflict. In the last stage, constructive aspect
of team conflict was recognized by academic community, people believed that
certain level of team conflict can motivate the potential of team.
Team conflict is surely an amazing concept not only in academic circle but
also in business world. If conflict is assumed to be harmful and should be eliminated,
team may be in harmony without viability; if conflict is supposed to be helpful
and should be tolerated, team may fall into irrational debate and personal attack.
Nelson (1989) suggested that the thresholds should
be identified at which conflict becomes problematic in a team.
Some scholars speculated that different types of team conflict may induce different
outcomes; they try to make a classification system of team conflict to predict
the effect of conflict. Literatures indicated that there are three categories
of team conflict in organizations: Relationship conflict, task conflict and
process conflict (Jehn, 1994; Jehn
and Mannix, 2001). Relationship conflict is defined as perceived disagreement
among team members regarding issues not related to the tasks or goals (e.g.,
fights about different political opinions or value differences); it is likely
to limit the information processing of team members and time or energy is devoted
to discussing, resolving, or ignoring the conflict instead of working on the
task (Jehn and Bendersky, 2003). Task conflict is defined
as perceived disagreement among team members regarding issues of specific task
the team is working on and may in contrast, increase divergent opinion about
task which ultimately may improve team performance. Process conflict refers
to perceived disagreement among team members regarding issues of means or methods
by which team goal would be attained (e.g., assignment of task, distribution
of responsibility and power, etc.,).
A lot of empirical studies indicated that relationship conflict is detrimental
(Roseman et al., 1994) but task conflict is
beneficial. But some theoretician doubted the generalization of this conclusion;
they argued that beneficial conflict (e.g., task conflict) may be not really
helpful sometimes. For example, a comment that express substantial standpoint
may be interpreted as personal attack when violent debate about task occurred
among team members.
The effect of team conflict can not be identified by characters or types of
team conflict. Team contextual variables may influence the effect of team conflict
significantly. It is assumed that team trust is an important factor that influences
the relationship between team conflict and team effectiveness.
Team trust is a shared believed that the team is safe for interpersonal risk
taking (Edmondson, 1999). Team members showed
their perspectives directly without any worry about jeer or rebuke from others
in the team. Edmondson et al. (2001) examined
several hospitals implementing new cardiac surgery technology. In hospital teams
whose team trusts are low, they are less likely to engage in risk taking and
they exhibit more behaviors consistent with the status quo. Team trust is a
part of team climates, in which critical comment may be regarded as impersonal
suggestion in conflict context. So, team trust may reduce the negative effect
of team conflict.
Team effectiveness is a variable to describe the outcome of team process. Gladstein,
(1984) and Hackman (1987) believed that there are
3 dimensions belonging to team effectiveness: (1) Team performance, (e.g., team
efficiency, productivity, quality). (2) Team satisfaction, (e.g., subjective
well-being or happiness of team members). (3) Team viability (Gladstein, 1984;
Hackman, 1987). In practice, team effectiveness is broadly
defined and accessed in various way. It therefore lacks precision of a theoretical
construct; a wise researcher must look to its specification for particular types
of teams to determine its grounded meaning (Goodman et
al., 1987). Because of large diversity of team tasks in different self-management
teams and too much conceptual overlap between team viability and team satisfaction,
team effectiveness in this research is defined as two variables: team learning
and team satisfaction.
Team learning is defined as a process of using information to guide behavior
in such a way as to promote ongoing adaptation. It is a kind of knowledge management
activity, such as seeking for feedback, asking for help, discussing error, experimenting
and monitoring the results of action (Altschuler and Armstrong,
1996). Team learning promotes performance obviously and it is beneficial
activity that organizations might want to support.
Two hypotheses were developed as below:
||Hypothesis 1:Team conflict is negative correlated with
team learning and team satisfaction
||Hypothesis 2: Team trust is mediate variable between team conflict
and team satisfaction, team trust is also mediate variable between team
conflict and team satisfaction
Instruments: In this research, 5 questionnaires are adopted as instruments.
Team conflict questionnaire developed by Jehn and Mannix
(2001) was used to assess team trust conflict level. The questionnaire is
consisted of 9 items. All types of team conflicts are all measured in this questionnaire;
they are relationship conflict (3 items), task conflict (3 items) and process
conflict (3 items).
Cronbach α of the questionnaire is 0.88. Likert 7 point scale is adopted.
Team trust questionnaire developed by Altschuler and Armstrong
(1996) was used to access team trust. The questionnaire consists 8 items
and adopts Likert 7 points scale, Cronbach α is 0.60.
Team learning questionnaire. Level of team learning behavior was measured by
team learning questionnaire developed by Edmondson (1999).
The questionnaire measured the frequency of these behaviors occurred in team
The questionnaire consists 7 items and adopts Likert 7 point scale. Cronbach
α index of the questionnaire is 0.82.
Team satisfaction questionnaire developed by Brayfield
and Rothe (1951) was used to access level of team satisfaction. The questionnaire
consists 7 items, uses Likert 7 point scale. Cronbach α of the questionnaire
Participants: In this study, 292 individuals have been recruited from
54 self-management teams as participants.
All individual participants were invited to lab to finish questionnaires. Six
senior students helped participants to finish questionnaire and took questionnaires
back after completion.
In 292 individual participants, all of them are Chinese. There into, 104 are
male, 188 are female, means of age is 19.75, SD is 1.14. Mean of working time
in team is 1.02 year, SD is 0.60.
In 53 self-management teams (team participants), every team have several members
to finish valid questionnaires, average numbers of questionnaire per team is
Data has been imported into computer and analyzed by SPSS 20.0.
Transition from individual-level data to team level data: In this research,
the teams other than the individuals are the subjects for analysis. Thus individual-level
data should be converted into team-level data.
Rwg is abbreviation for reliability of within-group judgers and it is used
to measure the statistical agreement to which individual members perceptions
converge in assessing aspect their team. The range of Rwg is from 0 to 1. If
Rwg was more than 0.7, the variable will be thought to be a team-level variable
and means of team members individual data are used to represent the teams
condition (James et al., 1984).
Data of a team (team No. 45) didnt
meet this criterion; these data were excluded out of further statistical analysis.
Common method biases: Common method biases is a kind of systematic error
due to same source of data. All data of this research came from same participants
by questionaire (Zhou and Long, 2004). So, common method
biases must be tested and be proved not too strong to affect accuracy of results.
Common method biases in this research were examined using statistical remedies
(Harmans single factor analysis
technique) and it were acceptable. (χ2/df = 5.97, CFI = 0.75,
NFI = 0.72, RMSEA = 0.22).
Level of task conflict is significantly higher than level of the other two
types of team conflict: Team conflict questionnaire contain three dimensions:
relationship conflict, task conflict and process conflict. Calculate three dimension
means, SD and correlation matrix (Table 1).
All correlation coefficients are significant at 0.01 level, and the correlation
between task conflict and process conflict is high (r = 0.73), the correlation
between relationship conflict and task conflict is also high (r = 0.65). Maybe
there are highly internal connections among three types of conflict.
Explore the differences among the three types of conflict by paired-sample
t-test. The scores of task conflict is significantly higher than scores of task
conflict and process conflict (t = 9.58, p<0.01). There is no significantly
difference between scores of relationship conflict and process conflict (t =
Three types of conflict are all negative correlated with team learning and
team satisfaction: Relationship conflict is medium negative correlated significantly
with team learning and team satisfaction, so are task conflict and process conflict
Three types of conflict are medium negative correlated with team learning and
team satisfaction significantly. Hypothesis 1 is verified.
Full mediating effect of team trust exists between team conflict and team
learning: Baron and Kenny (1986) developed a method
to test the mediating effect. In this study, Barons method was used to
test mediating effect of team trust between team conflict and team learning.
Calculate regression coefficient of team trust on team learning (β = 0.638,
p<0.01), team conflict on team learning (β = -0.359, p<0.01), team
conflict on team trust (β = -0.489, p<0.01), all of them are significant.
|| Three dimensions of team conflict and its correlation matrix
|*Mean p<0.05, **Mean p<0.01
||The regression coefficient between team conflict and team
Put three variables into the regression equation, coefficient of team trust
(β = 0.607, p<0.01) still significant, but coefficient of team conflict
(β = 0.063, p>0.05) becomes quiet. So, team trust is a full mediator
between team conflict and team learning.
Partial Mediating effect of team trust exists between team conflict and
team satisfaction: Calculate regression coefficient of team trust on team
satisfaction (β = 0.533, p<0.01), team conflict on team satisfaction
(β = -0.507, p<0.01), team conflict on team trust (β = -0.489,
p<0.01), all of them are significant.
Put three variables into the regression equation, coefficient of team trust
(β = 0.319, p<0.01) still significant; coefficient of team conflict
(β = -2.527, 0.01<p<0.05) is significant too but level of significant
decreases. So, team trust is a partial mediator between team conflict and team
satisfaction. Hypothesis 2 is also verified.
Pervious research supported the idea that task conflict may be beneficial for
team effectiveness, whereas relationship conflict and process conflict are harmful.
This conclusion has little practical implications. Three types of team conflict
work as a whole. Avoid relationship conflict and enhance task conflict
is far more easily said than done (Rispens, 2012).
Discussion about component ratio of team conflict types are more instructive
than focusing on team conflict categories and its effects.
Component ratio of team conflict categories may due to team character or team
context (Wu et al., 2013). The 53 self-management
teams acted as participant in this research were formed by team members themselves.
Before these teams established, members are familiarized with one anther and
initiate interpersonal relationship are set up among them. When organization
allowed them found free-combinational team, they gathered together quickly.
So, team members have little disagreement regarding values, cultures, task assignments
and power distributions. It is reasonable that level of relationship and process
conflict are significantly lower than level of task conflict.
In this research, team conflict is negative correlated with team effectiveness,
no matter what type of conflict. It isnt in conformity with some authority
literatures from western countries. The authors speculated that culture differences
may play an important role in this disaccord. All individual participants are
Chinese. In Chinese culture, saving face is very important in interpersonal
skill (De Dreu and Weingart, 2003). If some one makes
other people losing face, their friendly interpersonal relationship would be
injured or ruined (Wu, 2004). Criticism toward others
opinion, even showing disagreement with others in public are typical behavior
of making others lose face. So, Chinese tend to hide the different standpoints
with others, express their own thought cautiously, in order to avoid all kinds
of conflict, including task conflict. In China, all conflict would be regarded
as personal attack and destroy interpersonal cooperation in teams, then team
effectiveness would be reduced.
Team conflict contains a lot of activities of information sharing and perspective
clarifying. Sometimes, conflict is anther form of discussion if personal attack
and negative affection arousing are excluded. In a sense, team conflict dont
reduce team effectiveness directly, it declines team effectiveness by diminishing
team members willingness to
work together on the task. Team trust acts as bond among team member. Effect
of violent team conflict would weaken if high team trust has been established
in the team. So, Team trust is a mediator between team conflict and team effective.
In self-management teams, level of task conflict is significantly higher than
level of relationship conflict and process conflict.
Team conflict is negative correlated to team effectiveness, no matter what
type of conflict.
Team trust is mediator between team conflict and team effectiveness.
This research is supported by National Social Science Fund (No. 12 CGL050)
from National Planning Office of Philosophy and Social Science.
This research also supported by Jiangsu University Philosophy and Social Science
Research Fund of Education Department, Jiangsu Province (No. 2012 SJB 190012).
Thank for the academic support from Prof. Tian Xiaoming (Soochow University),
Prof. Duan Jinyun (Soochow University) and Prof Yu Jiayuan (Nanjing Normal University).