Electronic copy available at: http://ssrn.com/abstract=1298620
The effects of conflict asymmetry in mediation
Karen A. Jehn and Joyce Rupert
University of Amsterdam and Randstad HR Solutions
Seth van den Bossche
TNO Work and Employment
Electronic copy available at: http://ssrn.com/abstract=1298620
The effects of conflict asymmetry in mediation
Our main research question is how the asymmetry of conflict between two parties
involved in mediation will affect the outcomes of the mediation. Conflict asymmetry is the
difference in perceptions of conflict among the parties; that is, one person experiences high
levels of conflict while the other person perceives that there is little or no conflict. In this
multimethod study of 54 individuals involved in matched-pair mediations in an
organizational setting, we examine the effects of conflict asymmetry on satisfaction with the
process and results of the mediation, as well as their recommendation of mediation to
others. We find that when the two people involved in mediation have asymmetrical conflict
perceptions there is less satisfaction with the result and the process and this is partly due to
their view of the mediator being biased. In addition, we find that the person who
experienced more conflict is more likely to recommend mediation as a successful process to
Electronic copy available at: http://ssrn.com/abstract=1298620
The effects of conflict asymmetry in mediation
In this study we examine conflict asymmetry in the context of organizational
mediations. Conflict asymmetry is the difference in perceptions of conflict among the
parties involved in the conflict (Jehn, Rupert, & Nauta, 2006; Kluwer & Mikula, 2002).
This is important to consider in the context of mediation because mediation presupposes
equal commitment of the parties involved, which may be seriously threatened in cases
where conflict is asymmetric (Jehn et al. 2006). For example, in the educational setting in
which our study takes place a teacher who has his teaching plan rejected by another
teaching team member may be much more emotionally involved (and perceives a higher
level of conflict) than the member who rejected him. We propose that asymmetric conflicts
are the rule rather than the exception; therefore, it is very important to systematically
investigate the impact of the degree of conflict asymmetry upon the mediation process and
its success. Our main research question is how the asymmetry of conflict perceptions
between the two parties involved in mediation influences the outcomes of the mediation.
Our study has three main contributions. First, we contribute to the research on
mediation by empirically examining an assumed, but often ignored, aspect of conflict
mediations; that is, that people perceive different levels of conflict. The second contribution
of our article is that we advance the research on organizational conflict that often assumes
that people perceive the same level of conflict (c.f., Jehn & Rispen, 2007; e.g., Amason,
1996; De Dreu & Weingart, 2003; Pelled, 1996) by examining asymmetrical perceptions of
conflict and how they affect outcomes in organizations. Third, we also investigate the
direction of asymmetry - what are the different individual effects on the person who
perceives more conflict versus the person who perceives less conflict? Asymmetric conflict
may lead to asymmetric outcomes, with the person perceiving more conflict being more
satisfied with the outcomes than the person who perceives less conflict. Because of these
expected differences between the two parties involved in the conflict, it is interesting to
study asymmetric conflict both at the dyadic level and at the individual level. Therefore, we
use a multi-level approach in this study, that is, we examine both the individual effects of
conflict perceptions and asymmetry, as well as the dyadic level of asymmetry with matched
pairs (which the asymmetry research has yet to do; c.f., Jehn, Rupert, & Nauta, 2006). In
sum, we examine real-life conflicts and how they are differentially perceived by the two
parties involved, and what effects this may have on the mediation process and success.
The Mediation Context and Asymmetry
Very few studies investigate perceptual differences between parties involved in
mediation (e.g., Lind, Erickson, Friedland, & Dickenberger, 1978; Peirce et al., 1993;
Shestowsky, 2004; Tyler, Huo & Lind, 1999). These studies typically look at preferences
for conflict resolution procedures, and generally have found that so-called ‘plaintiffs’ and
‘defendants’ differ in preferences for conflict resolution procedures. For example, Lind et
al. (1978) and others (Shestowsky, 2004) find that plaintiffs are more likely to perceive
high levels of conflict (thus initiating the proceedings), and therefore prefer to maintain the
struggle in a proactive way that allows them to win (Peirce et al., 1993). Defendants are
less likely to take active rolls and prefer slow-moving formal procedures that allow them
control over a situation in which they prefer inaction (Lind et al., 1978; Peirce et al., 1993;
Tyler, 1986; stalling, pleas, or even avoidance and withdrawal). In this research, we build
on this literature regarding different preferences of mediating parties by examining
asymmetric perceptions of the conflict and how this influences individual and dyadic
We specifically look at three components of mediation success: satisfaction with the
process and the results of the mediation, and whether the person involved in mediation
would recommend mediation to another. These three factors make up critical aspects to
assess the success of an organizational mediation – whether the individuals involved are
satisfied with the process, as well as with the outcome, and whether or not they are willing
to involve others in this process since whether mediation is perceived as a success can also
be drawn from whether the disputants would recommend mediation to other people, if
necessary. In essence, mediation success is the subjective experience of parties that justice
has been done, that is, that the process and outcomes of the mediation are perceived as fair
and ultimately a success for the parties involved (cf. Greenberg & Colquitt, 2005).
Asymmetrical Conflict and Conflict Types: Relationship and Task Conflict
An important distinction revealed in past research on organizational conflict is the
difference between relationship and task conflict (e.g., Amason, 1996; c.f. De Dreu &
Weingart, 2003; Jehn, Northcraft & Neale, 1999). Relationship conflicts are disagreements
and incompatibilities between two or more parties about personal issues, such as social
events, gossip, and world news. Task conflicts are disagreements about the task being
performed, for example about the strategy of an organization or about the type of data that
should be included in a report. A recent review on task- and relationship conflict (Jehn &
Bendersky, 2003) suggests that task conflict can produce positive outcomes under certain
conditions (e.g., more creative ideas), while relationship conflict often has negative effects
such as dissatisfaction and decreased performance (Amason, 1996, Jehn, 1995; Pelled,
However, this past research on conflict has largely ignored the fact that different
parties are involved in a conflict, and that these parties may have different perceptions of
what the conflict is about and how serious the issue is. Specifically, they may perceive
different levels and types of conflict which, according to Pruitt (1995) makes a conflict
asymmetrical. We propose that research should pay more attention to the asymmetric
aspect of conflict (c.f., Jehn & Chatman, 2000; Sanchez-Burks et al., 2007), especially in
the mediation arena. The differences in the way the two disputants perceive the conflict
initiating the mediation process can have different effects on the individuals and the dyads
mediating regarding the level of commitment, cohesiveness, satisfaction, and individual and
dyadic outcomes such as absenteeism and prolonged interactions (Jehn & Chatman, 2000;
Jehn, Rupert, & Naukje, 2006).
Given the nature of conflict, cognition, and affective processes, we predict that there
will be more asymmetry in perceptions of relationship conflict than in perceptions of task
conflict. Conflicts are seldom either purely task or relationship-oriented. Indeed, research
demonstrates moderately-to-high correlations between task and relationship conflict (De
Dreu & Weingart, 2006; Simons & Peterson, 2000). But, we propose, the degree to which
each party perceives the conflict as task-oriented and relationship-oriented may differ. In a
rational sense, as soon as conflict arises openly, both disputants are likely to perceive that
their interests diverge regarding the task (task conflict symmetry). However, the emotional
attachment to these interests are likely to be different for both parties. The example that we
described above demonstrates why: both the team member and the teacher may well
perceive that their rational interests diverge in the same way and both will perceive task
conflict (thus, task conflict symmetry): the teacher wants a different curriculum and the
other teacher wants to maintain the status quo. In this example, then, the emotional
consequences of the conflict may be more serious for the teacher wanting the change than
for the other teacher. The first experiences that his new personally developed and initiated
plans are blocked, whereas the other experiences no specific personal consequences of this
conflict (relationship conflict asymmetry). Both parties may equally perceive that their task-
related interests diverge, but the emotional attachment to these interests are likely to differ
more strongly between parties thus causing relationship conflict asymmetry. We propose
that there will be more relationship conflict asymmetry in work relationships than task
conflict asymmetry for several reasons.
First, when a conflict occurs, emotion can get in the way of accurate perception
(Daly, 1991; Jehn, 1995; Staw & Barsade, 1993; Thomas and Pondy, 1977) so if the
conflict is about more personal, or relationship conflict issues, then we believe that
perceptions will be more likely to vary. Research on emotion and cognition shows that
perceptions are more disperse across individuals in the same situation when it is a personal
or emotional issue that is involved, compared to a more rational and task-related issue
(Yang & Mossholder, 2004; Campbell, Simpson, Boldry, Kashy, 2005). By definition,
relationship conflict is personal, so people will have personal views and so their
perceptions of the level of relationship conflict, even in the same conflict mediation
situation, will be more varied.
Secondly, in organizational settings, where the work is task-focused, people
working together have common task goals (McCann & Ferry, 1979; Thompson, 1967;
Victor & Blackburn, 1987; Wageman, 1995). This, just by the nature of the task
environment, will create less task-focused issue disparity. People working (e.g., teaching)
together must have a common view of task issues to complete their work. In our setting,
the teachers work together to design the curriculum and are thus interdependent on the tasks
they must complete. There is therefore more motivation for employees to get the reality
correct regarding task issues for both parties to succeed in their jobs. This is not
necessarily the case with personal issues that are not related to the task goal of the
employees. So while we believe there will be task disagreements, we propose that there
will be more consensus around the task issues and conflicts than there will be regarding
personal, non-task issues (i.e., relationship conflict). In sum, based on the above, we
propose that there will be more asymmetry about relationship conflicts than task conflicts.
Hypothesis 1 (H1): There will be more asymmetry regarding relationship conflict
than task conflict in mediating dyads.
Asymmetry and Mediation Outcomes
As can be drawn from the procedural justice literature, equity is an important criteria
for people to feel justly treated (Lind & Tyler, 1988; Tyler, 1986). We argue that when one
party believes there is conflict and the other does not, or to a lesser extent, more discomfort
and inequity will exist between the parties. This will cause the disputants to be less satisfied
than when a symmetrical conflict situation exists. Furthermore, if one party believes that
the other party does not validate her view of the situation, she may question her own view.
Drawing from self-verification theory, this decreases satisfaction, motivation, effort, and
performance (Swann, 1999). Individuals search for coherence in their interactions and
social environment. Inconsistencies in these social interactions can negatively affect the
processes and outcomes of the parties involved (Swann, Rentfrow, & Guinn, 2002).
Therefore, in the mediation context of our study, we predict that asymmetry of conflict
perceptions leads to decreased satisfaction with the mediation process and result and to a
smaller probability that parties will recommend mediation to others.
Similar processes exist in situations of asymmetrical conflict structures, as
suggested by past research. Kluwer and Mikula (2002) define an asymmetrical conflict
structure as a situation in which one party wants to change the status quo, whereas the other
party wants to keep it. Typical asymmetrical conflicts exist within the domain of close
relationships of different-sex couples, where men and women often fight over the division
of family work. Research in this setting suggests that cognitions of injustice cause more
problems than actual injustice (Major, 1987; Thompson, 1991). This distributive justice
framework states that feelings of injustice are caused by a comparison process, in which
one party – usually the woman – is especially upset because of the indifference that the
other party shows with regard to issues of household work and childcare. The unfairness
that individuals feel can cause decreased motivation, depression, and dissatisfaction with
the relationship (see Kluwer & Mikula, 2002 for a review of the research in the marital
The literature on dyadic negotiations also suggests that parties involved often
assume that they have opposing views of the conflict situation and the potential outcome
(De Dreu, Koole, & Steinel, 2000; Schelling, 1960). This fixed-pie bias (“If you win, I
lose”; Bazerman & Neale, 1983; Thompson & Hastie, 1990) leads to suboptimal outcomes
in negotiation settings. Problems arise due to inferior cooperation strategies (Bazerman,
Curhan, Moore, & Valley, 2000; c.f. De Dreu, 2007; Thompson, 1991; Thompson &
Hastie, 1990) and biased judgments about the accuracy of information (Carnevale & Pruitt,
1992; Thompson, 1990). The differing views of conflict can impede the information
sharing and processing that ultimately leads to superior, integrative agreements (Neale &
Bazerman, 1991) in which all parties benefit without competing in a win-lose fashion
(Thompson & Hastie, 1990). This is the goal in a common-goal dyad, such as in our
setting, where mediation is utilized to get the parties working together to settle their conflict
in a manner that facilitates their future working relationship
The idea of cognitive consistency also suggests that similar views about the conflict
will facilitate the reconciliation of the conflict. Cognitive consistency theory states that
individuals strive for consistency in their social interactions (Freedman, Carlsmith, & Sears,
1970). When environmental signals are contradictory or ambiguous, individuals have more
difficulty processing and are more likely to have negative affective responses (Phillips,
2003). So, in the case of conflict asymmetry, there are mixed signals about whether or not
there is a problem. This inconsistency in the work environment will decrease the cognitive
processing necessary for successful mediation and conflict resolution. The research by
Roy and Sawyers (1990) shows that individuals can more quickly and clearly attribute
meaning to consistent messages. Inconsistent messages coming from one’s immediate
environment are processed with much less consensus and clarity. Therefore, we propose
that symmetry of perceptions within the group regarding conflict will facilitate a successful
Drawing from the different literatures described above, we develop hypotheses
about disputants at work, whose asymmetric perceptions of conflict may influence the
outcomes of the mediation process in which they are involved. A mediator helps disputants
come to agreement without imposing a settlement on them by controlling how the parties
interact (Bazerman & Neale, 1992; Elangovan, 1999; Gewurz, 2001). The mediator is an
impartial third party (Bush, 1996) who facilitates communication and understanding to help
the conflict parties find a solution that they can both agree upon (Gewurz, 2001). The
success of mediation is usually measured by the perception of this success by the parties
involved and their satisfaction with the settlement of the dispute, including how likely they
are to recommend mediation to others (Lewicki, Weiss, & Lewin, 1992).
We propose that mediation success is influenced by the level of asymmetry of the
conflict. When party A perceives more conflict than party B, there is a discrepancy over the
level of conflict the two parties experience, thus there are often inconsistencies in the social
interactions, and perceived inequality and injustice. We propose that it is less likely that it
will result in successful outcomes as perceived by the conflict parties.
Hypothesis 2: Asymmetrical conflict in mediating dyads is negatively associated
with mediation success (i.e., satisfaction with the process and result of mediation,
and the likelihood that parties will recommend mediation to others).
The Mediating Effect of Mediator Bias
In order to successfully reach a solution which is mutually acceptable for parties
involved in mediation, conflict parties must perceive the mediator to be unbiased
(Bercovitch, 1996). We suggest that the literature on dyadic negotations (De Dreu, Koole,
& Steinel, 2000; Schelling, 1960; Thompson & Hastie, 1990) and cognitive processes
(Freedman, Carlsmith & Sears, 1970; Phillips, 2003) also suggest that, in our mediation
context, that the results of the effects of asymmetry of perceptions is mediated through the
perceptions that the parties involved have of the mediator and their objectivity or bias in the
mediation. By role definition (Bercovitch, 1996), mediators should not favor one party
over the other. Past research shows that perceptions of the mediator and their personal
interests can affect the attitudes, motivation, and behaviors of the parties in the mediation
(Conlon & Ross, 1993; Pruitt & Carnavale 1993). Wittmer, Carnevale and Walker (1991)
found that conflicting dyads did not trust suggestions being made by a mediator that they
felt was biased. Other research has found that conflict parties who perceive their mediator
to favor the interests of one party over the other are less likely to follow the mediator’s
recommendations for settlement (Arnold, 2000; Conlon & Ross, 1993; Welton & Pruitt,
1987). Therefore, we suggest that asymmetric perceptions will increase mediator bias
which will in turn decrease the likelihood of satisfaction with the mediation process and
outcome, and that the likelihood that mediation will be recommended to others in the
organization will be unlikely.
Hypothesis 3: Mediator bias mediates the effect of asymmetrical conflict in
mediating dyads on mediation success; that is, when there is asymmetry of
perceptions regarding the conflict it is more likely that the parties will perceive a
bias on the part of the mediator which will lead to decreased mediation success.
Directional Asymmetry: High and Low Perceivers of the Conflict Situation
The scarce research on asymmetrical conflicts (Bijleveld, 2005; De Dreu, Kluwer &
Nauta, 2007; Jehn & Chatman, 1994; Jehn, Rupert, & Nauta, 2005; Rispens, Jehn, &
Thatcher, 2005) has focused mainly on the magnitude of the asymmetry, but has not
examined the direction of asymmetry. That is, it looks at the degree to which disputants
perceive differences in the amount of conflict, but it has not yet investigated the differential
outcomes for those parties perceiving more conflict compared to parties perceiving less
conflict. We propose that this so-called directional asymmetry leads to differential
perceptions of mediation success between the two parties.
We propose that the party who experiences more conflict than the other will be
more satisfied with the process and results of the mediation, and will also be more likely to
recommend mediation the others. We propose, based on the concept of positive illusions
(Taylor & Brown, 1988; 1994) and research on the differences between parties involved in
the mediation (e.g., Lind, Erickson, Friedland, & Dickenberger, 1978; Peirce et al., 1993;
Shestowsky, 2004; Tyler, Huo & Lind, 1999) that the conflict party who experiences more
conflict than the other will be more satisfied with the mediation outcome than the other
party. Positive illusions are beliefs that are more optimistic than reality. Colleagues may
sometimes have such positive illusions, as when they interpret their mutual collaboration as
better than it actually is (Taylor & Brown, 1988). In the case of asymmetrical conflict, it
could be said that the party perceiving less conflict has a more optimistic, illusive view of
the level of conflict compared to the party perceiving more conflict. People who hold
positive illusions, or a more positive view of the situation than those around them,
according to a review by Taylor and Brown (1988), are more content, more considerate of
others, and less likely to want change to occur. Therefore we propose that those we
experience positive illusions (i.e., low levels of conflict, all is well) will be less impressed
with mediation and a positive mediation outcome; in fact, they may be disturbed by the
wasted time, as in their opinion, there was no conflict to begin with. Based on this, we
expect that the high conflict perceiver has more to gain and will be more motivated and thus
more satisfied with the mediation outcome and procedures compared to the low perceivers.
In addition, research on plaintiffs and defendants shows that the different
perceptions and expectations around the conflict lead to different desired outcomes. For
example, Lind et al. (1978) examined differences in preferences for conflict resolution
procedures between ‘plaintiffs’ and ‘defendants’, where a plaintiff (high conflict perceiver)
is the person who accuses a defendant of having harmed him and is aiming to win the
contest. Peirce et al. (1993) also studied differential preferences for conflict resolution
procedures between ‘complainants’ and ‘respondents,’ and showed that complainants (high
conflict perceivers) liked struggle and mediation more, and inaction less, than defendants
did. In other words, those who perceive more conflict prefer active contending and
resolution strategies such as mediation.
Also, in this context of mediation, the complexity of a mediator’s job stems
especially from the party perceiving the most conflict. The mediator has to spend more
effort to give this party opportunity to vent his or her anger and frustration (thus the
perceptions of mediator bias we mention earlier). Thus this person will be more satisfied
with the process and effort put in by the mediator to resolve the conflict. On the other hand,
the party perceiving the least conflict will be somewhat less committed to, and dependent on
the mediation. It takes less effort from the mediator to satisfy the needs of this party, and
this decreased effort on the mediators behalf can influence the perceptions of sastisfaction
with process and outcome of the mediation, as well as members willingness to refer
mediation to others in the organization. If they do not feel the mediator was helpful to
them, there is no reason to recommend to others this cumbersome, time-consuming
Hypothesis 4 (H4): An employee involved in mediation who perceives more
conflict than the other party will report higher levels of mediation success compared to
the party who perceives less conflict.
Our sample consisted of fifty-four individuals (27 matched pairs of conflict parties)
all working in the education sector within the Netherlands. The conflict parties equally
represented the sub-sectors of primary education, secondary education, senior secondary
vocational education, and university education and all had experienced a conflict at work for
which they became involved in mediation. The participants were recruited from the database
of the Dutch Institute for Conflict in Public Services and Employment Relations (NICOA)
and the Association of Universities in the Netherlands (VSNU) These two institutes started
a mediation pilot to introduce mediation in the education sector. All participants gave
written permission to be approached for research purposes. The mean age of the
participants was 53 years and 46% were female.
The data collection consisted of evaluation surveys that conflict parties in the
educational sector were asked to fill in regarding the mediation. This evaluation survey was
part of an applied TNO research project (TNO Work and Employment is a Dutch research
institute that conducts applied research in organizations and the public sector) in which the
goal was to evaluate how mediation was introduced into the educational sectors. The large-
scale mediation program was implemented by the Dutch Ministry of Education so that
employees were given the opportunity to consult a mediator if they experienced a severe
conflict at work that they were not able to solve themselves. The implementation of the
project consisted of 3 phases. In phase 1, employees were informed about mediation and
the program initiatives. Mediators were assigned to various schools, where human resource
personnel were trained to refer conflict parties to mediation. In phase 2, the actual
mediations took place. This study examines data collected during phase 3 of the TNO
project, in which the mediation success was evaluated. Conflict parties were asked to fill in
a mediation evaluation survey, consisting of open-ended, as well as Likert scale questions,
in which they were asked about the sort of conflict they experienced and how satisfied they
were with the mediation procedure and outcome.
Asymmetry of conflict. In the survey, there was an open-ended question in which
parties had to describe what the subject of the conflict was when the mediation started. We
had two independent coders, blind to the hypotheses, code the descriptions from each
individual conflict party into task and relationship conflict (Jehn, 1995; De Dreu &
Weingart, 2003), using a 1-5 Likert scale. The raters were trained in multiple rounds to
ensure their coherence with the theoretical constructs in the model (Jehn & Shah, 1997).
The inter-rater agreement was 98%. Since we had the ratings of each individual conflict
party within a dyad, we were able to calculate difference scores (with values ranging from 0
to 4) between the conflict parties in a dyad, which indicated the level of asymmetry
represented in the dyad. We calculated asymmetry scores for both task conflicts and
Mediation success. Mediation success was measured on three dimensions;
satisfaction with the process of the mediation, satisfiaction with the mediation outcome, and
whether or not the parties would recommend mediation to others in their organization.
Participants were asked to indicate on a 1- 4 Likert scale (ranging from not satisfied at all to
very satisfied) to what extent they were satisfied with the process of the mediation and
whether they were satisfied with the outcome of the mediation. We measured whether the
parties involved would recommend mediation to other people in their organization by
asking if the participant would recommend that other people solve a conflict by participating
in mediation. The participants were asked to answer this question with ‘yes/no’.
Mediator bias. Participant were asked to rate to what extent they found the mediator
impartial on a scale from 1- 4 (1 = not biased at all to 4 = very biased).
Controls. In our analyses we controlled for the general levels of task conflict and
relationship conflict to show that asymmetry explains mediation success above and beyond
the basic level of conflict in the dyad. This was assessed based on the coding described
Correlations and Associations Among Variables
In table 1, the means, standard deviations, and correlations between the variables in
our study are displayed. Given that one of our dependent variables is dichotomous
(mediation recommendation), we are required to perform different tests of association for
the various combinations of continuous and categorical variables with this dependent
variable. To examine the associations between our continuous variables (age, task and
relationship conflict, task and relationship conflict asymmetry, satisfaction with the
mediation process and satisfaction with the mediation outcome) and the dichotomous
dependent variable, mediation recommendation, we performed analyses of variance (Field,
2005). We found a positive marginal association between satisfaction with the mediation
process and mediation recommendation (F (3,49) = 3.22, p <.05), showing that the more
conflict parties were satisfied with the mediation process, the more likely it was they would
recommend mediation to others. We did not find any significant associations between the
other continuous variables and mediation recommendation. The association between gender
and mediation recommendation, which we tested using a Chi-squared cross-tabulation test,
was also not significant. Men and women did not differ in their intention to recommend
mediation to others.
--------------------- Insert Tables 1 and 2 about here ---------------------------
Our first hypothesis proposes that there will be more asymmetry regarding
relationship conflict than task conflict. Our data showed that 10 out of 27 pairs had
asymmetrical perceptions of task conflict, while twice as many (20 out of 27 pairs) had
asymmetrical perceptions of relationship conflict. We conducted a t-test to examine whether
conflict parties perceived more asymmetry in relationship conflict than in task conflict and
the results show that there is significantly more asymmetry in relationship conflict than in
task conflict (t = 2.32, p <.05) in these mediations, providing support for hypothesis 1.
Regarding hypothesis 2, we conducted regression analyses to test our second
hypothesis in which we proposed that asymmetrical conflict in mediating dyads is
negatively associated with mediation success (i.e., satisfaction with the mediation process,
outcome, and recommendation to others; see Table 2). We found a significant negative
relationship between asymmetry of relationship conflict and party’s satisfaction with the
mediation process (β = -.31, p < .05) and the result of mediation (β = -.28, p <.05). We also
found a significant negative effect of asymmetry of task conflict on satisfaction with the
mediation result (β = -.36, p < .05), and a marginal negative effect on the mediation process
(β = -.29, p = .07). We found these effects after controlling for the level of task and
relationship conflict. These findings give partial support for our hypothesis, showing that
asymmetry of task and relationship conflict have a negative impact on satisfaction with
mediation, above and beyond the main effects of task and relationship conflict.
In hypothesis 3 we proposed that the degree to which the parties both perceive the
mediator to be biased will mediate the effect of conflict asymmetry on mediation outcomes.
We tested this mediation effect according to the procedure described by Baron and Kenny
(1986). Following this procedure, four conditions must be met. The first two conditions are
that there must be a correlation between the independent variable and the dependent variable
and between the independent variable and the mediator. Third, there should be an
association between the mediator and the dependent variable. Finally, when including the
mediator in the regression model, the significant relationship between the independent and
the dependent variable should decrease substantially or disappear (in case of full
mediation). Table 1 shows marginal correlations between asymmetry of relationship
conflict and satisfaction with the mediation process and result. So the first conditions are
met for both dependent variables. The second condition is also met, because of a significant
correlation between asymmetry of relationship conflict and mediator bias (r = .35; p < .01).
We also find a significant correlation between mediator bias and satisfaction with the
mediation process and result. To test the fourth condition, we conducted two regression
analyses in which we regressed asymmetry of relationship conflict and mediator bias on
satisfaction with the mediation process and result. The regression analysis for satisfaction
with the result of mediation was also significant (F (2, 52) = 5.26, p <.05), R² = .17),
which was due to a significant effect of mediator bias (β = -.35, p <.05) on satisfaction
with the mediation result. Asymmetry of relationship conflict became non-significant (β = -.
14, p = ns). The Sobel-test for this mediation was marginal (z = -1.83, p = .07). The
regression model for satisfaction with the mediation process was significant (F (2, 51) =
11.58, p <.001), R² = .32), which was due to a significant effect of mediator bias (β = -.54,
p <.001) on satisfaction with the mediation process. Asymmetry of relationship conflict
became non-significant (β = -.07, p = ns). The Sobel-test for mediation (Sobel, 1982) was
significant (z = -2.26, p< .05). Thus, we found support for hypothesis 3 that mediator bias
mediated the relationship between asymmetry of relationship conflict and satisfaction with
the mediation process and result.
In hypothesis 4, we proposed that an employee involved in mediation who
perceives more conflict than the other party will report higher levels of mediation success
than a member who perceives less conflict compared to the other party. We performed t-
tests comparing high and low perceivers on satisfaction with process, outcomes, and
mediation recommendation. We found a significant negative relationship between parties
perceiving high levels versus low levels of task conflict on the intention to recommend
mediation to others (t = -2.45, p <.05). Crosstabs showed that all high perceivers
recommended mediation to others (less than 40 percent of low perceivers did). Thus,
parties in mediation who perceived more task conflict than the other party where more
likely to recommend mediation to others than parties who perceived less conflict than the
In this study of fifty-four individuals involved in matched-pair mediations in an
organizational setting, we utilized survey and qualitative data to examine the effects of
conflict asymmetry on mediation success (satisfaction with the process of the mediation,
satisfaction with the results of the mediation, parties’ recommendation of mediation to
others). In general, we found that conflict asymmetry (one party perceiving more conflict
than the other) was more the norm, especially regarding relationship conflict issues, than
was symmetrical views of conflict. We also found that this asymmetry of conflict
perceptions was negatively associated with mediation success; that is, when mediating pairs
experienced different levels of conflict (e.g., one party perceiving a high level of conflict
and the other perceiving no or little conflict), they were less satisfied with the mediation
result, as well as being less satisfied with the mediation process.
This negative effect of conflict asymmetry on mediation success was mediated by
the amount to which the parties felt the mediator in the process was biased. In mediations
where the parties believed that the mediator favored one party over another, the overall
evaluation of the mediation indicated a low level of satisfaction with the process and the
outcome of the mediation. The unfairness the parties felt during the process decreased
their satisfaction and evaluation of the success of the mediation. This finding has important
implications for the training of mediators. If the mediation and the mediator involved are
not perceived as being fair, the mediation process may do more harm than good. Mediators
should be made aware that there is a high probability of unequal levels of conflict as
perceived by both parties, and that this asymmetry increases the likelihood of mediator bias
and, hence, less positive outcomes.
A first contribution of this line of research, and consideration for re-evaluating past
research on conflict in dyads in groups, is that it challenges the assumption of much past
research that people involved in the same conflict perceive it in the same way (e.g., Jehn,
1995; c.f., Jehn & Rispens, 2008). In our research, we found that if the parties involved in
the conflict experience different levels of conflict, the mediation experience is less likely to
have a successful process and a successful outcome.
We also contribute to the research on conflict, mediation, and asymmetric
perceptions by investigating an aspect of conflict asymmetry that past research has often
ignored (e.g., Jehn & Chatman, 2000; Jehn, Rupert, & Nauta, 2006), the direction of the
asymmetry; that is, whether there are different effects on those individuals who perceive
high levels of conflict versus those who perceive low levels of conflict. This is especially
relevant in the mediation context, where the goal is to have the conflict parties reach a
common understanding of the situation and continue working together in the future in an
effective manner (e.g. Pruitt, Rubin, & Kim, 2003). We find that the individuals who
perceive the most conflict are the most likely to recommend mediation to others within their
organization. These individuals are the ones who find the most use out of the mediation by
having the expectations for conflict resolution through this procedure. However, more
research is needed to replicate this finding. It may also be likely that the party that perceives
the most conflict will be less likely to profit from the mediation because it is more difficult
for this party to gain from the mediation due to his or her greater frustration. Thus, future
research should examine the role that emotions and perceptions play in determining the
likelihood of whether an employee will recommend mediation to others.
Limitations and Future Research
There are a number of limitations of the present research. First, this study had a
small sample size (N = 54 individuals). However, it has an advantage over past research on
asymmetric perceptions in mediation (Jehn et al., 2006) by having both parties involved in
the mediation respond to the survey and provide qualitative data. This allowed us to
directly assess asymmetries of perceptions of conflict reported by the two parties involved
in the conflict. The use of multiple methods also prevents common method bias common to
much survey research.
A second limitation of this data is that it is cross-sectional. While the actual
mediation program did occur over time (3 distinct phases), we were only allowed to collect
data during the evaluation of the mediation, which occurred after completion of the
mediation. We suggest that future research conduct longitudinal studies to assess the causal
effects of asymmetry on the processes involved in mediation, the mediator’s involvement
(and potential bias), and the outcomes of mediation. In addition, some of the survey items
were designed by the organization that implemented the mediation practices. In this early
implementation phase, the researchers were not yet involved and therefore, were not able to
design and implement valid measures of variables relevant for the mediation. For example,
the concept of recommendation of mediation to others was measured with one item only
that had only two categories (yes/no) instead of a five-point Likert scale. Future research
should use established scales as well as a broader range of mediation outcomes (e.g.,
absenteeism, future conflicts).
An important area for future research on asymmetry and mediation includes the
investigation of the status of the participating parties, as well as the role of the mediator in
assessing the status and influence of the parties involved. For example, Shestowsky
(2004) examined the effects of social status (equal versus different status of both conflict
parties) and role (defendant versus plaintiff). She found that lower status disputants found
the process option in which they would present information about their own position to a
neutral third party somewhat less attractive compared to equal-status participants, showing
that they have less confidence that they will be able to defend themselves successfully,
compared to equal-status disputants. The results on role and status differences were
marginally significant however, making further research necessary.
Our research findings suggest that a mediator has a more complex job if involved in
an asymmetrical conflict compared to a symmetrical one, because he or she first has to
invest in the asymmetry, restoring balance and making the conflict more symmetrical. For
example, Shestowsky (2004) wrote that in cases of power imbalance, disputants need a
highly skilled and perceptive mediator to deal with the power imbalance. He or she can, for
example, invite the lower status party to speak first and to guide communication in a certain
way; however, as our results show, the perception that the mediator is biased in such a way
may influence the outcomes of the mediation in a negative way. Mediation is simply a more
complex job to perform in cases of asymmetrical rather than symmetrical conflict, especially
where status differences and other structural differences (e.g., roles, reward structures) may
be involved. Future research warrants a closer examination of the role of the mediator in
asymmetric conflict situations to provide more specific guidance to mediators and mediator
trainers in how to successfully manage these situations.
Practical Implications for Mediators and Organizations
The findings in this study suggest that mediators in organizations should be more
aware that the parties involved in the conflict have different perceptions about the level of
conflict. Unfortunately, past research has paid little attention the asymmetrical aspects of
conflicts that can hinder mediation success. We recommend that mediators, the parties
involved in mediation, as well as those training mediators acknowledge that asymmetries of
perception exist and can directly influence the process and outcome of mediation. For
example, it may be necessary for mediators to allow the person perceiving higher levels of
conflict to vent their anger to allow the content issues to arise, but also to allow the other
party the opportunity to see the different point of view (Galinsky, Ku, & Wang, 2005).
However, our findings show that mediator bias is a critical source of mediation failure and
therefore the impartiality of the mediator is important for the mediation to succeed. Thus,
mediators need to balance their asymmetrical attention by also allowing the view that there
really is no conflict, as the appearance of bias toward one party can defeat the main purpose
of the mediator – a successful common understanding and resolution of the situation.
Organizations considering offering mediation (and mediation training) must assess the
potential benefits of mediation, as well as these potential harms and revise their training and
mediation strategies accordingly.
Amason, A. (1996). Distinguishing the effects of functional and dysfunctional conflict on
strategic decision making: Resolving a paradox for the top management teams.
Academy of Management Journal, 39, 123-148.
Arnold, J. A. (2000). Mediator insight: Disputants’ perceptions of third parties’ knowledge
and its effect on mediation negotiation. The International Journal of Conflict
Management, 11, 318-336.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, 1173-1182.
Bazerman, M. H., Curhan J. R., Moore D. A., & Valley, K. L. (2000). Negotiation. Annual
Review of Psychology, 51, 279-314.
Bazerman, M. H., & Neale, M. (1983). Heuristics in negotiation: Limitations to dispute
resolution effectiveness. In M. H. Bazerman & R. Lewicki (Eds.), Negotiations in
organizations, 51-67. Beverly Hills, CA: Sage
Bazerman, M. H. & Neale, M. (1992). Negotiating rationally. New York: The Free Press.
Bercovitch, J. (1996). Understanding mediation’s role in preventive diplomacy. Negotiation
Journal, 8, 241-258.
Bijleveld, E., Jehn, K. A., & Barreto, M. (2005). Asymmetries of perception in
organizational conflict. Working paper, Leiden University, The Netherlands.
Bush, R. A. B. (1996). What do we need a mediator for? Mediation’s value added for
negotiators. Ohio State Journal on Dispute Resolution, 12(1), 1-55.
Campbell, L., Simpson, J. A., Boldry, J., & Kashy, D. A. (2005). Perceptions of conflict
and support in romantic relationships: The role of attachment anxiety. Journal of
Personality and Social Psychology, 88, 510–531.
Carnevale, P. J., & Pruijtt, D. G. (1992). Negotiation and mediation. Annual Review of
Psychology, 43, 531-582
Chatman, J. A., & Jehn, K. A. (1994). Assessing the relationship between industry
characteristics and organizational culture – How Different can you be? Academy of
Management Journal, 37(3), 522-553.
Conlon, D. E. & Ross, W. R. (1993). The effects of partisan third parties on negotiation
behavior and outcome perceptions. Journal of Applied Psychology, 78, 1-11.
Daly, J. P. (1991). The effects of anger on negotiations over mergers and acquisitions.
Negotiation Journal, 7, 31-39.
De Dreu, C. K. W. (2007). The virtue and vice of workplace conflict: food for (pessimistic)
thought, Journal of Organizational Behavior, in press.
De Dreu, C. K. W., Kluwer, E.S., & Nauta, A. (2007). The structure and management of
conflict: fighting or defending the status quo. Group Processes and Intergroup
Relations, in press.
De Dreu, C. K. W., Koole, S. L., & Steinel, W. (2000). Unfixing the fixed pie: A
motivated information-processing approach to integrative negotiation. Journal of
Personality and Social Psychology, 79(6), 975-987.
De Dreu, C. K. W. & Weingart, L.R. (2003). Task versus relationship conflict, team
performance, and team member satisfaction: A meta-analysis. Journal of Applied
Psychology, 88(4), 741-749.
Elangovan, A. R. (1999). The manager as the third party: Deciding how to intervene in
employee disputes. In R.J. Lewicki, D.M. Saunders, et al. (Eds.), Negotiation:
Readings, exercises, and cases (3rd ed., pp. 458-469). Boston, MA: Irwin/
Field, A. (2005). Discovering statistics using SPSS (2nd edition). London: Sage
Freedman, J. L., M. J. Carlsmith, & Sears, D.O. (1970). Social psychology. Oxford,
Galinsky, A. D., Ku, G. & Wang, C. S. (2005). Perspective-taking and self-other overlap:
Fostering social bonds and facilitating social coordination. Group Processes and
Intergroup Relations, 8(2), 109-124.
Gewurz, I. G. (2001). (Re)Designing mediation to address the nuances of power
imbalance. Conflict Resolution Quarterly, 19(2), 135-161.
Greenberg, J., & Colquitt, J.A. (2005). Handbook of organizational justice. Mahwah, NJ:
Jehn, K. A. (1995). A multimethod examination of the benefits and detriments of
intragroup conflict. Administrative Science Quarterly, 40, 256-282.
Jehn, K. A. & Bendersky, C. (2003). Interpersonal conflict, Research in Organizational
Behavior, 25, 187-242.
Jehn, K. A. & Chatman, J.A. (2000). The influence of proportional and perceptual conflict
composition on team performance. The International Journal of Conflict
Management, 11(1), 56-73.
Jehn, K. A., Northcraft, G., & Neale, M. (1999). Why differences make a difference: A
field study of diversity, conflict, and performance in workgroups, Administrative
Science Quarterly, 44, 741-763.
Jehn, K. A., & Rispens, S. (2008). Negotiation and conflict. In C. I. Cooper & J. Barlings
(Eds.), Handbook of Organizational Behavior, forthcoming.
Jehn, K. A., Rupert J., & Nauta, A. (2006). The effects of conflict asymmetry on mediation
outcomes - Satisfaction, work motivation and absenteeism. International Journal
of Conflict Management, 17(2), 96-109.
Jehn, K. A.& Shah, P. (1997). “Interpersonal relationship and task performance: An
examination of mediating processes in friendship and acquaintance groups.”
Journal of Personality and Social Psychology, 72(4), 775-790.
Kluwer, E. & Mikula, G. (2002). Gender-related inequalities in the division of family work
in close relationships: A social psychological perspective. European Review of
Social Psychology, 13, 185-216.
Lewicki, R. J., Weiss, S. E., & Lewin, D. (1992). Models of conflict, negotiation and third
party intervention: A review and synthesis. Journal of Organizational Behavior,
Lind, E. A., Erickson, B. E., Friedland, N., & Dickenberger, M. (1978). Reactions
to procedural models for adjudicative conflict resolution: A cross-national study.
Journal of Conflict Resolution, 22, 318-341.
Lind, E. & Tyler, T. (1988). The social psychology of procedural justice. pp. 318-341.
Major, B. (1987). Gender, justice, and the psychology of entitlement. In: P. Shaver & C.
Hendrick (Eds.), Review of personality and social psychology (pp. 124-148).
Beverly Hills, CA: Sage Publications Inc.
McCann, J. E., & Ferry, D. L. (1979). An approach for assessing and managing inter-unit
interdependence. Academy of Management Review, 4, 113-119.
Neale, M. A., & Bazerman, M. H. (1991). Cognition and rationality in negotiation. New
York: Free Press.
Pelled. L. H. (1996). Demographic diversity, conflict, work group outcomes: An
intervening process theory. Organization Science, 7, 615-631.
Peirce, R. S., Pruitt, D. G., & Czaja, S. J. (1993). Complainant-respondent differences in
procedural choice. The International Journal of Conflict Management, 4, 199-222.
Phillips, K. W. (2003). The effects of categorically based expectations on minority
influence: The importance of congruence. Personality and Social Psychology
Bulletin, 29(1), 3-13.
Pruitt, D.G. (1995). Process and outcome in community mediation. Negotiation Journal on
the Process of Dispute Settlement, 11 (4), 365-377.
Pruitt, D.G. & Carnevale, P. J. (1993). Negotiation in social conflict. Pacific Grove, CA:
Brooks / Cole.
Pruitt, D.G., Rubin, J.Z., & Kim, S.H. (2003). Social conflict: Escalation, stalemate, and
settlement (3rd edition). New York: McGraw-Hill.
Rispens, S., Jehn, K. A., Thatcher, S. (2005), “Asymmetric conflict in workgroups: effects
on group processes and performance”, paper presented at the Work and
Organizational Psychology Conference, Rotterdam.
Roy, L. & Sawyers, J. K. (1990). Interpreting subtle inconsistency and consistency: A
developmental-clinical perspective. The Journal of Genetic Psychology, 151, 515–
Sanchez-Burks, J., Lee, F., Nisbett, R & Ybarra, O. (2007), Cultural training based on a
theory of relational ideology. Basic and Applied Social Psychology, 29(3), 257-268.
Schelling, T.C. (1960). The strategy of conflict. Boston: Harvard University Press.
Shestowsky, D. (2004), “Procedural preferences in alternative dispute resolution: A Closer,
Modern Look at an Old Idea. Psychology, Public Policy, & Law. 10(3), 211-249.
Simons, T. L. & Peterson, R. S. (2000). Task conflict and relationship conflict in top
management teams: The pivotal role of intragroup trust. Journal of Applied
Psychology, 85, 102-111.
Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equations
models. In S. Leinhart (Ed.), Sociological methodology 1982 (pp.290-312). San
Staw, B. M. & Barsade, S. G. (1993). Affect and managerial performance: A test of the
sadder-but-wiser vs. happier-and-smarter hypotheses. Administrative Science
Quarterly, 38, 304-331.
Swann, W. B. (1999). Resilient identities: Self, relationships, and the construction of social
reality. New York: Basic Books.
Swann, W. B., Jr., Rentfrow, P. J., & Guinn, J. (2002). Self-verification: The search for
coherence. In M. Leary & J. Tagney (Eds.), Handbook of self and identity. New
Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological
perspective on mental health Psychological Bulletin, 103(2), 193-210.
Taylor, S. E., & Brown, J. D. (1994). Positive illusions and well-being revisited: A
separating fact from fiction. Psychological Bulletin, 116(1), 21-27.
Thomas, K. W. & Pondy, L. R. (1977). Toward an “intent” model of conflict management
among principal parties. Human Relations, 30, 1089-1102.
Thompson, J. D. (1967). Organizations in Action. New York: McGraw-Hill.
Thompson, L. (1990). Negotiation behavior and outcomes: Empirical evidence and
theoretical issues. Psychological Bulletin, 108(3), 515-532.
Thompson, L. (1991). Family work: Women’s sense of fairness. Journal of Family Issues,
Thompson L, & Hastie R. (1990) Social perception in negotiation, Organizational
Behavior and Human Decision Processes, 47(1), 98-123.
Tyler, T. R. (1986). Procedural justice in organizations. In R. J. Lewicki, B. H. Sheppard &
M. H. Bazerman (Eds.). Research on negotiation in organizations (Vol. 1, pp
7-23). Greenwich, CN: JAI Press.
Tyler,T. R., Huo Y.J., & Lind, E. A. (1999) The two psychologies of conflict resolution:
Differing antecedents of pre-experience choices and post-experience evaluations.
Group Processes Intergroup Relations, 2, 99-115.
Victor, B., & Blackburn, R. S. (1987). Interdependence: An alternative conceptualization.
Academy of Management Review, 12, 486-498.
Wageman, R. (1995). Interdependence and group effectiveness. Administrative Science
Quarterly, 40, 145-180.
Welton, G. L. & Pruitt, D. G. (1987). The effects of mediator bias and disputant power
over the mediator on the mediation process. Personality and Social Psychology
Bulletin, 13, 123-133.
Wittmer, J. M., Carnevale, P. J. & Walker, M. E. (1991). General alignment and over
support in biased mediation. Journal of Conflict Resolution, 35, 594-610.
Yang, J., & Mossholder, K. W. (2004). Decoupling task and relationship conflict: the role
of intragroup emotional processing. Journal of Organizational Behavior, 25,
Means, Standard Deviations, and Correlations among the Variables
MSD 1234567891. Gender1.46.50 -2. Age53.417.5 .19- 3. Task conflict1.531.09-.20-.26†-4.
Relationship conflict3.451.60 .20.18 -.52***- 5. Asymmetry of task conflict.721.15 -.34**-.26† .
52***-.23†-6. Asymmetry of relationship conflict 1.351.31-.24-.03-.03-.18.25†- 7. Mediator bias
1.66.68-.10-.06-.06-.04.19 .35**- 8. Satisfaction mediation process3.04.51 .11.04.06-.17-.19-.26† -.
56***-9. Satisfaction mediation result2.94.51 .16.15.02-.02-.26†-.26† -.40***-.76***- Note. N = 54
*** p <.001, ** p <.01, * p <.05, † p <.10
Regression Analyses of Asymmetry of Task and Relationship Conflict on Satisfaction with Mediation
Satisfaction process Satisfaction result
conflict -.26 -.17 -.09 -.01
Task conflict -.08 .12† -.04 .20
conflict -.31* -.28*
task conflict -.29† -.36*
F2.18† 1.66 1.35 1.74
R².12* .09† .08* .10*
∆R².09* .04† .07* .09*
Note. N = 54
* p <.05, † p <.10