The effects of conflict type and conflict expression intensity on conflict management

Abstract

Purpose

This paper aims to add to the current knowledge about conflict management by examining the relationships between conflict type, conflict expression intensity and the use of the conflict management approach.

Design/methodology/approach

The authors test theory-based hypotheses using a field study of new product development teams in an interdisciplinary Masters program (Study 1) and an experimental vignette study (Study 2).

Findings

Results show that people are more likely to respond to task conflict and conflicts expressed with less intensity using collectivistic conflict management approaches (i.e. problem-solving, compromising and yielding), and to relationship conflicts and conflicts expressed with higher intensity through forcing, an individualistic conflict management approach. Information acquisition and negative emotions experienced by team members mediate these relationships.

Practical implications

Knowing how the characteristics of the conflict (type and expression intensity) affect conflict management, managers can counteract the tendency to use dysfunctional, forcing conflict management approaches in response to high intensity conflicts, as well as to relationship conflicts and support the tendency to use collectivistic conflict management approaches in response to low intensity conflict, as well as task conflicts.

Originality/value

The authors examine an alternative to the prevailing view that conflict management serves as a moderator of the relationship between conflict and team outcomes. The research shows that conflict type and intensity of conflict expression influence the conflict management approach as a result of the information and emotion they evoke. The authors open avenues for future research on the complex and intriguing relationships between conflict characteristics and the conflict management approach.

Citation

Todorova, G., Goh, K.T. and Weingart, L.R. (2022), "The effects of conflict type and conflict expression intensity on conflict management", International Journal of Conflict Management, Vol. 33 No. 2, pp. 245-272. https://doi.org/10.1108/IJCMA-03-2021-0042

Introduction

Conflict is ubiquitous in teams and teams represent the primary building block of most companies in the current dynamic environment (Rigby et al., 2016; Weiss and Hughes, 2005). However, the mounting evidence on both positive and negative effects of conflict on team performance puzzles managers and scholars alike (De Dreu and Weingart, 2003; de Wit et al., 2012; Toegel and Barsoux, 2016). To resolve the paradox of conflict and reconcile conflicting findings on the effects of conflict on team performance, researchers have examined the contingent effects of the conflict management approach (Blake and Mouton, 1964; De Dreu and Weingart, 2003; DeChurch and Marks, 2001; Jehn and Bendersky, 2003; Pruitt and Rubin, 1986; Rahim, 1983). For example, teams that use collectivistic conflict management approaches, such as problem-solving, experience more positive effects of task conflict on performance compared to teams that use more individualistic approaches, such as power tactics (e.g. forcing). Extant research assumes that the conflict management approach occurs independently of characteristics of the conflict and focuses on examining the best conflict management approach team members can use to successfully manage the conflict they experience. Extending prior theories, we propose that the conflict experienced by a team and the conflict management approach used by that team do not occur independently of one another, but rather, characteristics of the conflict being experienced influence the use of the conflict management approach. We examine what management approach team members actually use in response to two characteristics of conflict: its type and the intensity of its expression.

Research has primarily focused on two types of conflict – task and relationship and on how conflict management moderates their effects on team outcomes. Task conflicts involve disagreements regarding opinions about a group’s task (e.g. what the project’s objective should be) or debating over the best course of action (Jehn, 1995, 1997). Relationship conflicts involve opposition to the people involved, including interpersonal incompatibilities and personal attacks (Jehn, 1995, 1997). To our knowledge, only one empirical study has explicitly examined the effects of these conflict types on the conflict management approach (Maltarich et al., 2018). In their field study, Maltarich et al. (2018) found early-stage conflict type to be related to the use of conflict management approach – relationship conflict was positively related to the use of competitive conflict management and negatively related to the use of cooperative conflict management. Research examining the interactive effects of conflict type and conflict management approach on team outcomes often report correlations between conflict type and conflict management approach, providing initial evidence that the two may be theoretically and causally related. Task and relationship conflict is negatively correlated with collaborating and positively correlated with competing. Although DeChurch et al. (2013) report meta-analytic correlations of these relationships, like the research they summarized and examined, they do not discuss the possibility of a causal relationship between these conflict states and processes.

In the present research, we seek to add to the current knowledge base by developing theory and directly examining the relationship between conflict type, conflict expression intensity and conflict management approach. Using the framework by DeChurch et al. (2013) that differentiates between conflict states and conflict processes, we theorize and empirically examine whether and why team members’ perceptions of the type of conflict in their team (a conflict state) influence their conflict management approach (a conflict process). Our study differs from the (DeChurch et al., 2013) meta-analysis in two important ways. First, the two papers test different models. DeChurch et al. (2013) examine conflict type and conflict management as independent predictors of team outcomes, while we examine the effects of conflict type on conflict management. Second, we use a more fine-grained conceptualization of conflict states by distinguishing between type and intensity. DeChurch et al. (2013) combine the two when defining conflict states, stating:

We define (a) team conflict states as shared perceptions among members of the team about the intensity of disagreement over either tasks (i.e. goals, ideas, and performance strategies) or relationships (i.e. personality clashes, interpersonal styles) (DeChurch et al., 2013, p. 560).

In contrast, we distinguish between the amount of a given type of conflict (conflict type) and the strength with which opposition is communicated (conflict expression intensity) (Weingart et al., 2015). We draw from the novel framework advanced by Weingart et al. (2015) that differentiates conflict expression from conflict type and defines it along two dimensions – oppositional intensity and directness and discusses the effects of conflict expression on perception, emotion, information and conflict spirals.

Integrating cognitive dissonance theory (Festinger, 1962; Matz and Wood, 2005) with conflict theories (Cronin and Bezrukova, 2019; Jehn, 1995), we argue that conflict surfaces inconsistent views, perceptions, etc. in the team. These inconsistencies trigger cognitive dissonance that can evoke negative emotions, additional information search and acquisition and attempts at conflict management to reduce the resultant intra-individual tensions. As such, we propose that negative emotions and information acquisition represent two fundamental mechanisms that explain the effects of conflict type and conflict expression intensity on the conflict management approach (Figure 1).

We present two studies that examine team member’s use of conflict management approach in teams facing conflict. In Study 1, we investigate whether and how conflict type affects conflict management approach at the team level through the mechanisms of information acquisition and negative emotions in a naturalistic setting. In Study 2, we examine the individual level conflict management responses and mechanisms that underlie the team-level processes (Humphrey and Aime, 2014; Waller et al., 2016) using an experimental design and incorporating the distinction between conflict types and conflict expression intensity.

This multi-method approach has the potential to provide findings with high internal and external validity (Brewer and Hunter, 1989). The naturalistic setting in Study 1 allows us to examine actual behavior, but does not test for causality (higher external validity but lower internal validity). The experimental vignette study used in Study 2, allows us to test for causality but does not examine actual behavior (higher internal validity but lower external validity) (Aguinis and Bradley, 2014). The research designs of the two studies complement each other and the consistent findings across the studies increase the validity of our findings.

By differentiating between the effects of conflict type and conflict expression intensity, Study 2 incorporates the essential role of how conflict is communicated. While conflict type focuses on what the conflict is about (e.g. task or relationship), conflict expression intensity captures the strength with which opposition is expressed, reflecting the entrenchment in position and subversiveness of action of the conflict participants (Weingart et al., 2015). High intensity conflict expression is characterized by actions that defend positions (high entrenchment) and attack or undermine others’ positions (high subversiveness) and can be experienced as arguments (a more direct expression) or as undermining behavior (a less direct expression) (Weingart et al., 2015). Low intensity conflict expression expresses opposition using words and actions that are less entrenched and subversive and often demonstrate a willingness to consider others’ perspectives, for example, when expressed as a debate (Weingart et al., 2015). Higher conflict expression intensity has been shown to decrease information acquisition, resulting in more negative emotional responses and decrease job satisfaction (Todorova et al., 2014; Tsai and Bendersky, 2016). We argue below that irrespective of the type of conflict a team experiences, the intensity of the expression of opposition (conflict expression intensity) will influence the conflict management approaches taken by the team via its impact on the information and emotions that are experienced by the team members.

Our research contributes to and extends our understanding of conflict management in three ways. First, the study adds new insights into the complex relationships between conflict and conflict management approaches by examining how conflict type and expression intensity directly influence the conflict management approach used. This represents an alternative to the prevailing view that conflict management either moderate the relationship between conflict type and team outcomes or merely acts as an independent causal factor (DeChurch et al., 2013). Second, integrating literature on conflict (de Wit et al., 2012; Cronin and Bezrukova, 2019; Weingart et al., 2015) and cognitive dissonance (Matz and Wood, 2005), we develop and test two mechanisms that should underlie the link between conflict and use of conflict management approach, namely, acquired information and negative emotions. Third, our research provides evidence regarding the role of the intensity of conflict expression in team conflict. By examining both the effects of conflict expression intensity and conflict type in Study 2, we gain a better understanding of how conflict drives the use of conflict management approaches.

Approaches to managing conflict

Conflict management is a process where group members interact with the aim of working through the task and interpersonal disagreements (DeChurch et al., 2013). Typologies have identified five types of conflict management approaches or styles that differ in terms of their concern for self and concern for others, namely, problem-solving, forcing, yielding, avoiding and compromising (Blake and Mouton, 1964; Deutsch, 2002; Rahim, 1983; Pruitt and Rubin, 1986; van de Vliert and Euwema, 1994). Problem-solving occurs when team members attempt to resolve the conflict by integrating each other’s concerns. Yielding involves “giving in” and accepting others’ will, concerns or desires. Compromising is characterized by behaviors designed to find a middle ground. Forcing is evidenced by team members imposing their will and concerns on others. Avoiding involves behaviors that ignore the conflict, and thus, ignores the concerns of other parties.

In their meta-analysis, DeChurch et al. (2013) used the fundamental distinction between individualistic (concern for self) and collectivistic (concern for others/the team as a whole) conflict processes to categorize conflict management approaches within teams. Individualistic processes “preserve individuality and subjugate the entitativity of the team to safeguard the disparate views of its members” (DeChurch et al., 2013, p. 561) and include approaches, such as avoiding and forcing/competing. Collectivistic processes incorporate “differences in members’ viewpoints while still preserving the entitativity of the team” (DeChurch et al., 2013, p. 561) and include collaborative approaches that incorporate the needs of others, including problem-solving, yielding and compromising. This distinction captures differences between cooperative and competitive approaches to conflict management (Deutsch, 2002, for a discussion).

In our research, we examine the five conflict management approaches identified in the dual concern model and rely on DeChurch et al.’s (2013) distinction between individualistic approaches (in which we include forcing and avoiding) and collectivistic approaches (in which we include problem-solving, yielding and compromising) when developing our predictions. Doing so provides more parsimony in our predictions and allows us to focus on critical shared attributes of the different conflict management approaches that we study. At the same time, we measure each conflict management approach independently, allowing us to examine the efficacy of this theoretical distinction.

Conflict management approaches have been found to influence both performance and team attitudes. In their meta-analysis, DeChurch et al. (2013) found collectivistic approaches to be positively related to both team performance and attitudes, such as team satisfaction and trust (Behfar et al., 2008). In contrast, individualistic conflict management approaches of forcing and avoiding are likely to result in lower quality performance and attitudes because either the conflict is not resolved or one disputant feels overrun by the other (DeChurch et al., 2013; Pruitt and Rubin, 1986). Prior research on conflict management demonstrates the important role of the conflict management approach in understanding team outcomes – we shift the focus to its antecedents.

Conflict, information and emotion as drivers of conflict management approaches

Conflict interactions both provide information about inconsistencies among team members and generate emotions and together these two mechanisms influence a team’s ability to perform their task and resolve their conflicts (Cronin and Bezrukova, 2019; Weingart et al., 2015). In the sections that follow, we argue that these same two mechanisms influence the team’s choice of conflict management approach. Task conflicts and conflicts expressed with lower intensity are more likely to trigger the acquisition of information and as a result, will lead to more collectivistic conflict management approaches that incorporate the other party. In contrast, relationship conflict and higher expression intensity conflicts are more likely to trigger negative emotions, leading to a greater reliance on individualistic conflict management approaches.

We argue that the intra-individual and interpersonal tension generated by conflict will motivate team members to manage the conflict, and the type and intensity of conflict expression will trigger the specific approaches taken to do so. First, individuals will be motivated to reduce the negative intra-individual tension associated with the cognitive dissonance that is caused by disagreeing with others in a group (Matz and Wood, 2005). Cognitive dissonance occurs when people experience their behavior as discrepant from some standard of judgment (Matz and Wood, 2005; Stone and Cooper, 2001, 2003) and the social group can be the source of dissonance (Festinger, 1962; Matz and Wood, 2005). Research on group-induced cognitive dissonance showed that individuals who disagree with others in their group experience a state of dissonance (tension) due to their apparent inconsistencies in attitudes with others, and that tension results in “dissonance discomfort” they are then motivated to reduce using dissonance-reduction strategies (Matz and Wood, 2005). Similarly, we argue that individuals engaged in conflict will be motivated to use conflict management to reduce the cognitive dissonance created by the conflict. Reducing the intra-personal and interpersonal tensions enables team members to make progress with their tasks and/or perform group maintenance. Second, we argue that the type and expression intensity of a conflict will elicit cognitive dissonance in unique ways. Differences in cognitive dissonance will be reflected in the information that is elicited and the negative emotions that are triggered, which will, in turn, impact the use of different conflict management approaches. Prior research demonstrates that cognitive dissonance increases negative emotions (Elliot and Devine, 1994; Harmon-Jones, 2000) and information processing (Rydell et al., 2008). We expect that negative emotions and information acquisition, in turn, will affect whether team members will be more collectivistic or more individualistic in their approach to conflict management. Thus, cognitive dissonance theory provides a perspective that helps us examine the dual pathways of the effects of conflict type and conflict expression intensity on conflict management.

Information acquisition

We propose that information acquired as a result of task conflict and low conflict expression intensity will result in a better understanding of the opinions of other team members (Todorova et al., 2014) and this focus on others should both cue and support the use of collectivistic, team-oriented, strategies when trying to resolve the conflict, including problem-solving, compromising and yielding.

First, conflict focused on the task and expressed with lower intensity should result in more information acquisition about the others in the group because it surfaces inconsistencies around task-related opinions, creating a cognitive dissonance that is likely to elicit additional information search to gain a better understanding. Disagreements about the task are likely to facilitate further inquiries about the perspectives, ideas, and opinions of other team members (Amason et al., 1995; Jehn, 1995). Relationship conflicts do not provide such information and have been shown to reduce collaborative problem-solving and take time away from task-focused activities (De Dreu, 2006; Evan, 1965; de Wit et al., 2012). Similarly, conflict expression intensity should influence the acquisition of information, regardless of the conflict type (Bradley et al., 2015; Todorova et al., 2014; Tsai and Bendersky, 2016; Weingart et al., 2015). High-intensity conflict expression occurs when team members are entrenched in their positions, which interferes with information seeking and receptiveness. When conflict is expressed with less oppositional intensity such that positions are less entrenched, team members are more likely to request and integrate information from other team members (Weingart et al., 2015). In support of this supposition, workers in a long-term health care facility who reported that their task conflicts were more often expressed as arguments (high intensity) also reported obtaining less information from those conflicts compared to workers who reported that their task conflicts were more often expressed as debates (low intensity) (Todorova et al., 2014). Dissenting opinions are more likely to be accepted if team members engage in low intensity conflict expression such as debates (Tsai and Bendersky, 2016).

Second, acquiring information facilitates a more complete understanding of the issues, namely, team members gain information about different opinions, concerns and perspectives in the team (Jehn, 1995; Lovelace et al., 2001; Pelled et al., 1999), both sensitizing team members to others’ concerns and making it easier to find mutually beneficial solutions. This information is necessary for a collaborative approach to conflict management (Lovelace et al., 2001; Pelled et al., 1999; Pruitt and Rubin, 1986). Furthermore, with that additional information and the insight it provides, team members may be motivated to collaborate and use other collectivistic approaches, such as yielding or compromising (DeChurch et al., 2013), in their quest to reduce the discomfort and resolve the task conflict.

Thus, we predict:

H1a–c and H2a–c.

Conflict type and conflict expression intensity are likely to influence the use of collectivistic conflict management approaches. Team members who experience more task conflict (as opposed to relationship conflict) (H1) and who experience lower intensity conflict expression (H2) are likely to rely more on (a) problem-solving, (b) compromising, and (c) yielding to manage their conflict.

H3a–c.

The relationship between conflict type and collectivistic conflict management approaches is likely to be mediated by information acquisition. Team members who experience more task conflict (as opposed to relationship conflict) in their team are likely to acquire more information and thus will be more likely to engage in (a) problem solving, (b) compromising, and (c) yielding.

H4a–c.

The relationship between conflict expression intensity and collectivistic conflict management approaches is likely to be mediated by information acquisition. Team members who experience lower intensity conflict expression in their team are likely to acquire more information and thus will be more likely to engage in (a) problem solving, (b) compromising, and (c) yielding.

Negative emotions

We propose that the negative emotions resulting from the intra-individual tension generated by relationship conflict and high conflict expression intensity will trigger the use of more individualistic conflict management approaches, including forcing and avoiding.

Relationship conflict provides little information about the task at hand; instead, it focuses on the individual, arousing dissonance related to one’s self-concept resulting in discomfort – an aversive, negative emotional experience that people will want to mitigate by defending themselves and their self-identity. Relationship conflict is more subjective and personal than task conflict. Teams high in relationship conflict are beset by personality clashes and personal attacks, which are likely to put a person into a defensive posture and trigger negative emotions (Cronin and Bezrukova, 2019; de Wit et al., 2012; Jehn, 1995). While both task and relationship conflict evoke negative emotions (Cronin and Bezrukova, 2019; Syna Desivilya and Yagil, 2005; Greer and Jehn, 2007; Jehn, 1995; Pinkley, 1990; Jehn et al., 2008), relationship conflict do so more consistently. Similarly, higher intensity conflict expressions, such as arguments and undermining are expected to evoke more negative emotions than lower intensity conflict expressions. The subversiveness and entrenchment associated with arguments and undermining will make people feel angry and frustrated (Weingart et al., 2015).

Experiencing high expression intensity and relationship conflict should motivate one to reduce the discomfort and negative emotions arising from the dissonance created by the threat to one’s self-concept through individualistic conflict management approaches – forcing and avoiding – that place the individual above the team (DeChurch et al., 2013). In support of this supposition, Maltarich et al. (2018) found that relationship conflict early in the team’s life cycle was associated with more use of competitive conflict management approaches. Negative emotions toward other team members have been shown to reduce the willingness to collaborate and increase competitive behaviors (Knight and Eisenkraft, 2015).

Thus, we predict:

H5a and b and H6a and b.

Conflict type and conflict expression intensity is likely to influence the use of individualistic conflict management approaches. Team members who experience more relationship conflict (as opposed to task conflict) (H5) and who experience higher conflict expression intensity (H6) are likely to rely more on (a) forcing and (b) avoiding to manage their conflict.

H7a and b.

The relationship between conflict type and individualistic conflict management approaches is likely to be mediated by negative emotions. Team members who experience more relationship conflict (as opposed to task conflict) in their team are likely to experience negative emotions, and as a result will be more likely to engage in (a) forcing and (b) avoiding.

H8a and b.

The relationship between conflict expression intensity and individualistic conflict management approaches is likely to be mediated by negative emotions. Team members who experience higher intensity conflict expression are likely to experience more negative emotions, and as a result will be more likely to engage in (a) forcing and (b) avoiding.

Crossover and interaction effects of conflict type and conflict expression intensity

We proposed specific hypotheses regarding how conflict type and conflict expression intensity will influence conflict management strategies as a result of the information and emotion triggered by the conflict. We also explore two potential additions. First, in Study 1 we explore whether conflict type has crossover effects such that relationship conflict is associated with information acquisition and task conflict is associated with negative emotions [1]. Second, we explore possible interaction effects of conflict type and conflict expression intensity on the conflict management approach. It is possible that team members who engage in task conflict with high intensity do not use collectivistic conflict management approaches because they get entrenched in their positions. Therefore, it is interesting to explore whether task conflict might lead to more individualistic conflict management when the task conflict is also high intensity conflict.

Below we report the results of two studies testing our model. Study 1 examines the relationships between task and relationship conflict, information acquisition, negative emotion and conflict management approaches in naturalistic, intact student project teams. The second study differentiates conflict expression intensity from conflict type and uses a vignette experimental design to assess the causal effects of task and relationship conflict on information acquisition, negative emotions and conflict management approach.

Study 1

In Study 1, we test H1, H3, H5 and H7 in a naturalistic field setting examining how team conflict type, information acquisition and negative emotions relate to team conflict management.

Sample and procedure

Data was collected from members of interdisciplinary masters student project teams enrolled in a semester-long (15 weeks) course on developing interactive media (e.g. video games and interactive displays at theme parks). Students had work experience. Data was collected from six cohorts of students. In total, 13 teams were removed from the initial data set because their response rate was lower than 50%. Thus, the current sample used in this paper consists of 57 teams. Teams ranged in size from 3 to 12 members (Mode = 4). Demographic data were collected at the beginning of the semester. Conflict type, information acquisition, negative emotions and conflict management measures were administered in week 10 after mid-semester presentations.

Measures

See Appendix 1 for all scales used in Study 1. Task conflict was measured using a four-item scale (α = 0.87) (1 = not at all, 5 = very much) with which members reported the amount of task disagreement in their team using items that avoided the use of the word “conflict” (Behfar et al., 2017; Bendersky et al., 2014). Aggregation of these items to the team level was justified (mean r_wg = 0.84; ICC(1) = 0.15, F = 1.68, p < 0.01; ICC(2) = 0.41). Relationship conflict was measured using a four-item scale adapted from Jehn (1995) that allowed members to report on the amount of interpersonal and personality conflict while removing references to emotions (α = 0.74) (1 = Strongly disagree, 5 = Strongly agree). Aggregation of these items to the team level was justified (mean r_wg = 0.82; ICC(1) = 0.47, F = 3.83, p < 0.001, ICC(2) = 0.78).

Information acquisition and negative emotions were measured using seven-point Likert scales (1 = Strongly disagree, 7 = Strongly agree). Information acquisition was measured using a two-item scale adapted from Kraut and Streeter (1995) that we aggregated to the team level (mean r_wg = 0.70; ICC(1) = 0.74, F = 3.82, p < 0.001, ICC(2) = 0.77). Participants responded to statements regarding their personal experience of three fundamental negative emotions (afraid, hostile, distressed; Ekman, 1992) in response to relationship conflict in their team (α = 0.83). As we did not expect individual members to share a consensus about their experience of negative emotions, we used the additive composition model and treated the aggregated score as a summary index (Chan, 1998; Chen et al., 2004) of the negative emotions experienced in the team. Consistent with our expectations, the mean r_wg for negative emotions was low at 0.25. The expected difference in negative emotions for individuals within the team represents one of the motivations for testing the model on an individual level in Study 2. The ICC aggregation indices were satisfactory (ICC(1) = 0.64, F = 2.75, p < 0.001, ICC(2) = 0.72).

We measured conflict management approach using 20-item Dutch conflict management scale (1 = not at all, 5 = very much) (De Dreu et al., 2001) (Problem-solving: α = 0.85; Compromising: α = 0.77; Yielding: α = 0.66; Avoiding: α = 0.73; Forcing: α = 0.65). Aggregation of these items to the team level was also justified (Problem-solving: mean r_wg = 0.94; ICC(1) = 0.84, F = 6.23, p < 0.001, ICC(2) = 0.85; Compromising: mean r_wg = 0.93; ICC(1) = 0.72, F = 3.61, p < 0.001, ICC(2) = 0.77; Yielding: mean r_wg = 0.94; ICC(1) = 0.56, F = 2.25, p < 0.001, ICC(2) = 0.67; Avoiding: mean r_wg = 0.92; ICC(1) = 0.73, F = 3.67, p < 0.001, ICC(2) = 0.74; Forcing: mean r_wg = 0.92; ICC(1) = 0.56, F = 2.28, p < 0.001, ICC(2) = 0.65).

We controlled for gender composition as a percent of men on the team, as gender has been shown to be associated with conflict management approach (Bear et al., 2014; Rahim and Katz, 2019). We considered team size as a control variable but it did not affect the results in a meaningful way and so we did not include this variable in our analyzes of the hypothesized model.

Results

Table 1 shows descriptive statistics and correlations for all variables in Study 1.

We tested the direct and indirect effects of conflict type on conflict management approach using path analysis in MPlus. We chose to aggregate the data to the group level and use MPlus to test the model because it was the most effective way to test our mediation hypotheses. Aggregation was justified both theoretically and empirically for all the variables in the model. In addition to the hypothesized paths, we allowed for covariation between information acquisition and negative emotions. We also included control paths allowing for direct effects of conflict type on conflict management approach and crossover effects of relationship conflict on information acquisition and of task conflict on negative emotion. The overall goodness of fit of the model was satisfactory [χ² (12) = 26.806, p < 0.01; comparative fit index (CFI) = 0.92; root mean square error of approximation (RMSEA) = 0.15; standardized root mean square residual (SRMR) = 0.10]. The results from these analyzes are summarized in Table 2 and Figure 2.

In H1 and H5, we proposed that conflict type was associated with the type of conflict management approach. In H3 and H7, we proposed the mediating effects of information acquisition and negative emotions in the relationship between conflict type and conflict management approaches. To test these hypotheses, we used a bootstrapping approach (Preacher and Hayes, 2008). The 95% bootstrap confidence intervals were obtained using 5,000 samples. Results showed that there was a positive indirect effect of task conflict on problem-solving via information acquisition (indirect effect = 0.08, 95% bias-corrected bootstrap confidence interval (CI) [0.02, 0.14]) and a positive indirect effect of task conflict on yielding via information acquisition (indirect effect 0.05, 95% bias-corrected bootstrap CI [0.003, 0.10]). The indirect effect of task conflict on compromising via information acquisition was not statistically significant (indirect effect 0.01, 95% bias-corrected bootstrap CI [−0.04, 0.08]) but the direct effect of task conflict on compromising was statistically significant (β = 0.25, p < 0.05). The direct effects of task conflict on problem-solving (β = 0.31, p < 0.01) and yielding (β = 0.27, p < 0.05) were also statistically significant. Thus, H1a–H1c, H3a and H3c were supported, while H3b was not supported.

The results of the mediation analyze also showed that there were no significant indirect effects of relationship conflict on forcing via negative emotions (indirect effect = 0.07, 95% bias-corrected bootstrap CI [−0.06, 0.20]) and on avoiding via negative emotions (indirect effect = −0.10, 95% bias-corrected bootstrap CI [−0.22, 0.02]). Relationship conflict had a significant direct effect on forcing (β = 0.34, p < 0.01) and marginally significant effect on avoiding (β = 0.23, p < 0.10). Thus, H5a was supported while H5b, H7a and H7b were not supported. Finally, results also showed a significant negative path between relationship conflict and information acquisition (β = −0.52, p < 0.01). Groups that engaged in more relationship conflict reported acquiring less information from their team.

Supplementary analysis

We used Harmon’s single-factor test to assess whether our results were affected by common method bias because our data were collected using self-report questionnaires at a single point in time (Podsakoff et al., 2003). Results showed that the general common method factor explained only 23% of the variance, which is well below the threshold of 50%, suggesting that concerns about common method variance are low.

Study 1 discussion.

The results of study 1 provide support for the team-level relationships between conflict type and conflict management approach with regard to problem-solving, compromising and yielding. As hypothesized, teams with more task conflict used more problem-solving, compromising and yielding to manage their conflicts. Information acquisition mediated two of these relationships – teams with more task conflict acquired more information from their teammates and that information acquisition was related to the use of more problem-solving and yielding when managing their conflicts. As expected, teams with more relationship conflict experienced more negative emotions and engaged in more forcing, but negative emotions did not mediate the effect. Furthermore, while the direct effect of relationship conflict on avoiding was marginally significant, it was not mediated by negative emotions.

Due to the cross-sectional nature of our data collection in Study 1, we were unable to test for causality. Also, our Study 1 was limited to focusing on conflict type, without examining how differences in conflict expression intensity come into play. In Study 2, we use an experimental vignette method in which we manipulate the presence of task vs relationship conflict and levels of conflict expression intensity to better understand the relationships between conflict type, conflict expression intensity and conflict management approach, as well as their mechanisms.

Study 2

Study 2 uses an experimental design to assess the causal relationships between conflict type, conflict expression intensity and conflict management approach and examines the mediating mechanisms of information acquisition and negative emotions at the individual level. We also disentangle what the conflict is about from how intensely it is expressed.

We used an experimental vignette methodology (EVM), which allowed us to exercise control over independent variables and examine causal relationships (Cavanagh and Fritzsche, 1985), complementing our field research design in Study 1, which did not allow us to do so. EVM was also warranted given the questionable ethicality of experimentally manipulating relationship conflict in a laboratory setting (Aguinis and Bradley, 2014).

Participants and procedure

Participants were recruited via Amazon’s Mechanical Turk (Buhrmester et al., 2018), which is a reliable source of high-quality data when appropriate quality checks are used (Bendersky and Shah, 2013; Buhrmester et al., 2018; Campagna et al., 2016). To help ensure data quality, we restricted participation to US-based workers who had completed more than 5,000 prior tasks with an approval rate of at least 99%. Participants were compensated $2 for approximately 20 min of their time.

A total of 262 individuals responded to the survey and usable data was provided by 251 respondents. Specifically, 11 respondents were removed for responding carelessly (missing attention checks or completing the survey at an unrealistically fast pace) or having duplicate entries from the same IP location. Participants’ average age was 34.6 years (SD = 10.19 years). The sample included 59% male and 41% female participants. In total, 45.4% of respondents reported completing a 4-year undergraduate degree program, 27.1% reported to have attended some college and 13.9% reported completing high school or equivalent.

We randomly assigned the participants to one of four conditions in a 2 (conflict type: task vs relationship) × 2 (conflict expression intensity: lower vs higher) between-subjects design. We used a type of EVM known as a “paper people” study in which participants were first presented with vignettes in written form and then asked to report their explicit responses to the hypothetical scenario. This type of EVM has been used extensively, especially in fields, such as business ethics, leadership, executive behaviors and organizational citizenship behavior (Aguinis and Bradley, 2014).

Manipulations and measures

To manipulate conflict type and conflict expression intensity, we presented participants with scenarios that asked them to imagine they were a member of a three-person project team (Appendix 2). We based the scenarios on incidents we had observed in Study 1. In the relationship conflict conditions, the other team member made negative comments that were directed at the participant personally, followed by their interpretation of the comments. In the low intensity expression condition, these comments were described as being delivered lightheartedly (“really? I know you can come up with something better than that”). While in the high intensity expression condition, the comments were more derogatory (“really? That is completely unrealistic, just like all your ideas”) and framed as personal attacks.

Participants in the task conflict condition read about another group member’s opposition to one of their ideas, followed by their defense of the novelty and usefulness of their idea. In the low intensity task conflict expression condition, this interaction was framed as another group member expressing a differing view followed by an explanation by the participant to counter the points raised. By contrast, in the high intensity task conflict expression condition, this interaction was framed as an argument in which the participant had to fight to defend their point [2].

After reading the assigned scenario, participants evaluated their reactions, feelings, behaviors and responses toward the other team member who participated in the conflict episode. The scales are listed in Appendix 1. Information acquisition was assessed using a four-item scale (Todorova et al., 2014) measuring their expected acquisition of information resulting from the interaction (α = 0.89) (1 = never, 5 = all of the time). We chose this scale for Study 2, rather than the scale we used in Study 1 because it focuses more specifically on individual-level information acquisition from engaging in conflict. Participants responded to four statements regarding the negative emotions they would feel in response to the interaction (frustrated, angry, annoyed, tense (α = 0.90) (1= strongly disagree, 5 = strongly agree) (Todorova et al., 2014). These negative emotions are different from those assessed in Study 1, allowing us to assess our hypotheses using a broader range of negative emotions. Conflict management approach was assessed using the same measure as in Study 1 (De Dreu et al., 2001), this time asking them how they, as an individual, would have resolved the conflict with the other person. Participants were asked how much they disagreed or agreed (1 = Strongly disagree, 7 = Strongly agree) with 20 statements (Problem-solving: α = 0.88; Compromising: α = 0.88; Avoiding: α = 0.81; Forcing: α = 0.83; Yielding: α = 0.80).

We controlled for the effects of gender and work experience. We controlled for gender to be consistent with Study 1. We controlled for years of work experience because research shows that individuals learn from experience (Tjosvold, 2000) and the participants in the experiment had a wide range of work experience. Prior work experience might influence the conflict management approach.

Manipulation checks

To assess the validity of our manipulation of conflict type, at the end of the survey we asked participants how strongly they agreed (1 = Strongly disagree, 7 = Strongly agree) with statements that the scenario was an example of task conflict (e.g. “this conflict is about the work itself, that is, it is an example of disagreement among people about the tasks being performed or how to perform them”) or relationship conflict (e.g. “this conflict is interpersonal, that is, it involves personal attacks or people challenging each other such that the conflict gets personal”). We examined the efficacy of the conflict type manipulation within each conflict expression intensity condition. Independent samples t-tests demonstrated that all comparisons were significant and in the expected direction (Table 3).

We also tested the validity of our manipulation of the intensity of conflict expression in our scenarios by asking participants to rate the intensity of conflict expression. The intensity was defined as the forcefulness with which opposition is conveyed by both parties in a conflict (five-point scale, 1 = Very Low, 5 = Very High). We examined the efficacy of the conflict expression intensity manipulation within each conflict type condition. Independent samples t-tests demonstrated that all comparisons were significant and in the expected direction (Table 4).

Results

Table 5 shows descriptive statistics and correlations for all variables.

We conducted multiple analysis of variance to test the direct effects of conflict type and conflict expression intensity on conflict management approaches (H1, H2, H5 and H6). This analytical approach was appropriate because we manipulated the presence of task vs relationship conflict and low vs high conflict expression intensity. The analysis revealed a significant main effect for conflict type, F(5, 241) = 4.77, p < 0.001 and for conflict expression intensity, F(5, 241) = 2.83, p < 0.001. Individuals in the task conflict condition reported they would use more problem-solving (H1a: F(1, 250) = 13.35, p < 0.001), compromising (H1b: F(1, 250) = 5.24, p = 0.01) and yielding (H1c: F(1, 250) = 7.69, p = 0.01) and less forcing (H5a: F(1, 250) = 4.024, p = 0.03) than individuals in the relationship conflict condition. Thus, we find support for H1a–H1c and H5a. There were no significant effects of conflict type on avoiding (H5b). As predicted, individuals in the low intensity condition reported that they would use more yielding (H2c: F(1, 250) = 8.23, p = 0.003) and less forcing (H6a: F(1, 250) = 5.36, p = 0.02) than individuals in the high intensity condition. Counter to predictions, there was a significant negative effect of conflict expression intensity on avoiding (H6b: F(1, 250) = 5.25, p = 0.02) and no main effect of conflict expression intensity on problem-solving (H2a) and compromising (H2b). Thus, we find support for H2c and H6a while H2a, H2b and H6b are not supported.

To test the mediating role of information acquisition and negative emotions (H3, H4, H7 and H8), Hwe examined the indirect effects of conflict type and conflict expression intensity on the conflict management approach (Rucker et al., 2011). We conducted path analysis with bootstrapping using MPlus (Muthén and Muthén, 2005). We controlled for the effects of gender and experience. Consistent with Study 1, we allowed information acquisition and negative emotions to covary. We also included direct effects of conflict type and conflict expression intensity on the conflict management approach. The overall goodness of fit of the model was satisfactory (χ² (7) = 23.25, p < 0.01; CFI = 0.98; RMSEA = 0.10, RMSEA 90% CI [0.06, 0.14]; SRMR = 0.03). The results from the path analyses are summarized in Table 6 and Figure 3.

Mediation results from bootstrapping analysis with 5,000 samples showed that individuals in the task conflict condition (coded as 0) would use each of the collectivistic conflict management approaches more than would individuals in the relationship conflict condition (coded as 1) as a result of acquiring more information from the type of conflict (problem-solving: indirect effect = −0.15, 95% bias-corrected bootstrap CI [−0.23, −0.07]; compromising: (indirect effect = −0.11, 95% bias-corrected bootstrap CI [−0.18, −0.05]; yielding: indirect effect = −0.10, [95% bias-corrected bootstrap CI = −0.15, −0.06]). Thus, H3a, H3b and H3c were supported.

As for the effect of relationship conflict on forcing and avoidance via negative emotions, we found individuals engaged in relationship conflict were more likely to engage in forcing because of more negative emotions (indirect effect = 0.05, 95% bias-corrected bootstrap CI [0.02, 0.09]). However, individuals in the relationship conflict condition would not use more avoiding than individuals in the task conflict condition because of more negative emotions (indirect effect = 0.03, 95% bias-corrected bootstrap CI [−0.01, 0.06]). Thus, H7a was supported, but H7b was not supported.

The intensity of conflict expression was found to significantly affect collectivistic (problem-solving, compromising and yielding) conflict management via information acquisition during the conflict. Individuals in the high-intensity conflict condition (coded as 1) were less likely to engage in these approaches than individuals in the low intensity condition (coded as 0) because of less information acquisition (problem-solving: indirect effect = −0.06, 95% bias-corrected bootstrap CI [−0.10, −0.01]; compromising: indirect effect = −0.04, 95% bias-corrected bootstrap CI [−0.08, −0.01]; yielding: indirect effect = −0.04, 95% bias-corrected bootstrap CI [−0.07, −0.01]). Therefore, H4a–H4c were supported.

Finally, we predicted that in comparison to individuals in the low-intensity conflict condition, individuals in the high-intensity condition would be more likely to engage in forcing (H8a) and avoiding (H8b) because of more negative emotion. The analysis revealed that participants in the high-intensity conflict condition would use more forcing than participants in the low-intensity conflict condition because of more negative emotion (indirect effect = 0.05, 95% bias-corrected bootstrap CI [0.01, 0.09]). Conflict expression intensity did not affect the likelihood of avoiding via its effect on negative emotion (indirect effect = 0.03, 95% bias-corrected bootstrap CI [−0.01, 0.06]). H8a was supported. We did not find evidence in support of H8b.

Exploratory analysis

Although the hypothesized model proposed independent and direct effects for conflict type and conflict expression intensity, we performed an exploratory analysis to probe potential interactive effects of conflict type and conflict expression intensity on the conflict management approach. The interaction terms were not statistically significant.

Study 2 discussion

The results of Study 2 provide support for our hypotheses that conflict type and conflict expression intensity influence the use of the conflict management approach and these effects are driven by differences in information acquisition and negative emotions that arise. Higher intensity conflict expression and conflicts that were focused on the relationship rather than the task were less likely to provide information and more likely to evoke negative emotions, decreasing the use of collectivistic conflict management approaches and increasing the likelihood of using force, an individualistic approach. Exploratory analysis revealed no interactive effects of conflict type and conflict expression intensity on the conflict management approach.

Counter to predictions, higher intensity conflict decreased the use of avoiding as a preferred conflict management approach. While high intensity conflict expression increased negative emotions, those emotions did not result in more avoidance. Instead, participants would avoid a conflict expressed with low intensity to a greater extent than one of high intensity. This finding suggests that there could be moderating variables and also supports separating avoiding from forcing when examining the use of individualistic conflict management strategies. While high intensity conflict expression might lead to more forcing, the decision of whether to avoid addressing a high intensity conflict is more complex.

General discussion

Prior research focuses on the conflict management approach as a moderator of the conflict type-team outcome relationship (Alper et al., 2000; Rahim, 1983; Tekleab et al., 2009) and often assumes that conflict type and conflict management approaches are not interrelated. In contrast, we provide evidence on both the individual and the team level that the conflict management approach is influenced by the type of conflict experienced.

People responded to task conflicts and conflict expressed with less intensity by acquiring information, which is associated with a greater reliance on problem-solving, yielding and in one of our studies, compromising, to manage the conflict at hand. We predicted and found support for the prediction that task conflict and conflict expressed with less intensity increased information acquisition and collectivistic conflict management where people consider problems from multiple perspectives and are more willing to compromise and accommodate others. In our naturalistic field setting, we saw evidence of this when a participant from one of the recounted how disagreements and debates about the features to include in their project were resolved by going back-and-forth between the client’s priorities and assessing the work that each team member had to perform.

In contrast, the negative emotions that people experience with relationship conflict and with conflict expressed with higher intensity appear to increase the reliance on forcing to manage the conflict. These results suggest that relationship conflict triggers negative emotions, such as anxiety, frustration and anger that focus people on forcing their positions and perspectives on others. We observed this dynamic in our field setting when a software engineer was labeled by his team as uncooperative, difficult to work with and lazy. Team members were angry and frustrated that the engineer frequently produced work that was different from project requirements resulting in many unexpected revisions of the individual’s work and delays in project deliverables. After several unsuccessful attempts at addressing these issues themselves, the team escalated the conflict by asking their advisor to evaluate each individual member’s contributions to the project rather than evaluating the group product as a whole (a forcing approach to resolving the conflict).

Theoretical implications

Our findings demonstrate a new way to conceptualize the relationship between conflict type and conflict management, suggesting the need for more theory and research about the role of conflict management as a mechanism that helps to explain the effects of conflict on team outcomes. Knowing whether and why conflict type triggers specific conflict management approaches might allow disputants to interrupt or even reverse that link and prevent the selection of dysfunctional conflict management approaches. Below we explore the implications of our findings for theory and practice.

The effects of conflict type on conflict management approach.

New theoretical developments about the antecedents of the conflict management approach could further advance team research. While we know that the conflict management approach impacts team outcomes over and above conflict type, there is presently scarce research on why team members engage in specific conflict management approaches (DeChurch et al., 2013). Our research addressed this question by examining the effect of conflict type on conflict management. In our first study, we asked team members what conflict management approach they used. In our second study, we manipulated conflict type, and thus, were able to directly examine the causal effects of conflict type on the conflict management approach.

A natural extension of our study is to examine the effects of other conflict types, for example, process conflict and status conflict (Behfar et al., 2011; Bendersky and Hays, 2012; Greer and Dannals, 2017; Pai and Bendersky, 2020) on the use of conflict management approach. These additional types of conflict have been more recently identified and there is scarce research on the effects of process and status conflict on information acquisition and emotions.

Mechanisms of the link between conflict type and conflict management approach.

Our research provides insight into the mechanisms that drive individuals and teams to gravitate toward specific conflict management approaches in response to a task or relationship conflict. Our findings regarding the mechanisms contribute to research on conflict, which increasingly looks at both its positive and negative effects (De Wit et al., 2012). Consistent with the recent advances, our study contributes to the understanding of how tensions experienced by team members engaged in conflict can lead to functional and dysfunctional behavior. We found that while the tension arising from relationship conflict triggers negative affect and dysfunctional conflict management approaches, tension arising from task conflict produces beneficial outcomes. Discomfort resulting from task disagreements motivates team members to acquire new information and engage in collectivistic conflict management approaches.

Knowing the mechanisms that drive the use of the conflict management approach provides clues for moving the team from relying on one approach to another. For example, increasing information acquisition during task conflict might increase the group’s natural tendency to rely on collectivistic conflict management approaches, while interventions to reduce the negative emotions experienced from relationship conflict might result in people engaging in more productive conflict management approaches.

Future research should also examine boundary conditions of the effects of information acquisition and negative emotions on the conflict management approach. For example, the quality or type of information that team members acquire might impact their conflict management approach. If the information is less useful or not relevant, team members might switch from a problem-solving approach to more individualistic approaches, such as forcing.

Our research provides insights into the individual-level cognitions, emotions and behaviors that underlie the team’s tendency to use a specific conflict management approach. Our approach allows us to go beyond aggregate effects at the team level, to account for how team conflict directly affects individuals and their subsequent interactions with others in the team. By testing the individual level reactions and behaviors directly in Study 2, we provide evidence about how and why specific conflict management approaches are adopted at the collective level. Building on recent theorizing on multilevel processes and micro-dynamics in teams (Humphrey and Aime, 2014; Waller et al., 2016), we contribute to the understanding of how individuals respond to events in the team, such as team conflict.

The importance of conflict expression intensity.

Our research highlights the importance of examining conflict expression intensity when we study conflict management, team processes and team outcomes. In Study 1, we used existing measures of conflict type. These measures ignore and potentially conflate the intensity of conflict expression with the type of conflict, not allowing us to examine their independent effects (Bendersky et al., 2014). By manipulating the intensity of conflict expression independently of conflict type in Study 2, we extended and tested theories on conflict expression on conflict management, information acquisition and negative emotions. For example, our findings demonstrate that the intensity of conflict expression might counteract the benefits of task conflict because it limits information acquisition and triggers negative emotions regardless of the type of conflict.

The findings of conflict expression intensity challenge the way we think about conflict management. For example, although applying a problem-solving approach to task conflict might lead to more beneficial outcomes, our findings demonstrate that team members are less likely to collaborate under intense conflict. Future research should examine what predicts the intensity of conflict expression.

Managerial implications

Our findings suggest an alternative approach to how we think about managing conflict in teams. Current recommendations regarding conflict management approaches are based on people’s orientation to one another – whether they are concerned about themselves, the other party or both, ignoring how the characteristics of the conflict might predispose a given approach. Knowing that characteristics of the conflict (type and intensity of expression) drive people’s conflict management approaches, team leaders and members have more avenues to set up a team for success. This could be done by supporting the tendency to use collectivistic conflict management approaches in response to low intensity conflict expression, as well as to task conflicts and by counteracting the tendency to use dysfunctional, forcing conflict management approaches in response to high intensity conflict expressions, as well as to relationship conflicts.

Managers and employees should better understand mechanisms through which conflict type and conflict expression intensity drive the use of the conflict management approach. For example, our study shows that teams and team members who experience relationship conflict use forcing to manage conflict because of more negative emotions. We also know from prior research that forcing reduces trust and team satisfaction (DeChurch et al., 2013). Therefore, managers and employees should learn skills to effectively manage their negative emotions, such as developing emotional intelligence (Farh et al., 2012; Goleman, 1995), which can reduce the use of forcing approaches to manage conflict in teams.

Finally, managers should proactively aim to prevent the escalation of low-intensity conflict into high-intensity conflict (Cronin and Weingart, 2019; Weingart et al., 2015). As we show, high-intensity conflict expression reduces the use of problem-solving, compromising and yielding and stimulates the use of forcing, independent of conflict type. One approach for preventing escalation is helping teams develop cognitive integration and affective integration (Cronin and Weingart, 2019).

Limitations and directions for future research

Our studies tested the theoretical model in a field and an experimental setting. Our field study sample consisted of students. The students were involved in new product development projects and worked together over a 15-week semester. This approach allowed us to control for the variation in the external context, team inputs and the team selection process. There were also several proxies for organizational realities: students worked under time pressure, they were accountable to different stakeholders and worked on other projects simultaneously. Interestingly, task and relationship conflict were not significantly correlated in the project teams, which is an uncommon finding in teams research. Corroborating this finding, results also showed that task conflict did not increase negative emotions nor was it associated with forcing and avoiding. These findings speak to the positive culture of collaboration within their master’s program. As task and relationship conflict tend to be positively correlated in teams (see the meta-analysis by de Wit et al. (2012) in which correlations ranged from −0.69 to +0.93 with a mean of 0.52), this might affect the generalizability of our findings. The co-occurrence of conflict types is important to consider because it has been shown to be a significant contextual moderating factor when assessing the impact of task conflict on team performance (De Dreu and Weingart, 2003; de Wit et al., 2012). We encourage future research involving organizational teams to examine the generalizability of the findings.

The specific content of the scenarios in Study 2 also influences the generalizability of our Study 2 findings. Specifically, the relationship conflict scenarios included personal attacks – a form of relationship conflict that is likely to threaten an individual’s self-concept. The personal attack in the low-intensity relationship conflict scenario was lighthearted (teasing) and was rude (derogatory) in the high-intensity relationship conflict scenario. Relationship conflict also encompasses interpersonal incompatibilities related to non-task relevant issues (such as politics). We encourage future research that examines the impact of different forms of relationship conflict on conflict management approaches.

More experimental research is needed to test the causal mechanisms underlying the effects of conflict on the conflict management approach. Our study included more interpersonal interaction in our task conflict scenario than in our relationship conflict scenario. Vignettes could be constructed that vary the descriptions of the conflict interactions that occur within task and relationship conflict beyond the ones we used in this study to test for generalizability and limits to the results.

Future research could also examine the possibility of reverse causality between the conflict management approach and conflict type. In our research, we hypothesized that conflict type would influence the use of the conflict management approach. In Study 2, we tested this relationship using an experimental approach that provides evidence for this direction of causality. It is also possible that the conflict management approach used by a team (e.g. forcing) could lead to more conflict (e.g. relationship) (Tjosvold et al., 2006). We suggest that future longitudinal research should address the reciprocal relationships between conflict type and conflict management.

It is also possible that the mechanisms that link conflict type to the conflict management approaches do not function independently. For example, emotions may influence information processing (Tiedens and Linton, 2001). We controlled for this possibility by allowing information acquisition and negative emotions to co-vary in our models. In addition, we added control paths from task conflict to negative emotions and from relationship conflict to information acquisition in Study 1. Results showed that relationship conflict negatively impacted information acquisition while task conflict did not affect negative emotions on the team level. We speculate that when team members engage in a relationship conflict, they are less likely to acquire information because they are less willing to do so. During relationship conflict, team members’ discomfort is related to a threat to their self-concept and acquiring any additional information from other team members might seem threatening. In addition, more relationship conflict was associated with less collectivistic conflict management approaches of problem-solving and yielding (Table 1). Together, the results thus, suggest that this negative relationship is due to relationship conflict’s interference with information acquisition. We believe that future theorizing and research on the relationships between conflict types, information acquisition and negative emotions will provide new insights.

We acknowledge there might be other mechanisms that help to explain how conflict type and conflict expression intensity influence conflict management approaches. For example, it is possible that people experience ego threat in response to conflict (de Wit et al., 2012). Disputants who experience ego threat during a conflict event might adopt individualistic conflict management approaches, such as forcing and avoiding as a way to defend their self-image. We encourage future research on the mechanisms linking conflict characteristics with the use of a conflict management approach.

Conclusions

By integrating the recent reconceptualization of conflict management as a team process and conflict type as team state with theories on cognitive dissonance, we developed and tested a model of what conflict management approach team members actually use when they engage in conflict. Our results suggest that the conflict management approach is influenced by the type of conflict and the conflict expression intensity and provide evidence regarding mechanisms that explain these relationships.

Figures

Figure 1.

Theoretical model of the effects of conflict type and conflict expression intensity on conflict management

Figure 2.

Summary of results: team level (Study 1)

Figure 3.

Summary of results: individual level (Study 2)

Table 1.

Summary of means, standard deviations and correlations (Study 1)

Variable	M	SD	1	2	3	4	5	6	7	8	9	10
1. Gender composition	0.25	0.25	1
2. Task conflict	3.73	0.44	−0.14	1
3. Relationship conflict	2.13	0.54	0.08	−0.08	1
4. Information acquisition	5.27	0.65	−0.14	0.24	−0.54**	1
5. Negative emotions	3.31	0.64	0.14	−0.02	0.45**	−0.38**	1
6. Problem-solving	3.69	0.45	0.06	0.33*	−0.47**	0.52**	−0.29*	1
7. Compromising	3.28	0.38	0.05	0.15	−0.17	0.15	−0.30*	0.67**	1
8. Yielding	3.42	0.44	−0.08	0.28*	−0.27*	0.39**	−0.33*	0.73**	0.69**	1
9. Forcing	2.46	0.40	−0.14	0.16	0.38**	−0.42**	0.30*	−0.20	−0.11	−0.08	1
10. Avoiding	3.01	0.37	−0.16	−0.09	−0.01	0.08	−0.29*	0.39**	0.51**	0.48**	0.17	1

Notes:

N = 57. *Correlation is significant at the 0.05 level (2-tailed). **Correlation is significant at the 0.01 level (two-tailed)

Table 2.

Indirect effects from conflict types to conflict management approaches (Study 1)

IV	Path	Standardized path estimate	95% CI
Task conflict	Information acquisition → problem-solving Information acquisition → compromising Information acquisition → yielding	0.08 0.01 0.05	[0.02, 0.14] [−0.04, 0.08] [0.003, 0.10]
Relationship conflict	Negative emotions → forcing Negative emotions → avoiding	0.07 −0.10	[−0.06, 0.20] [−0.22, 0.02]

Notes:

CI = Bias-corrected bootstrap confidence interval, p < 0.05

Table 3.

Manipulation checks for conflict type

	Task conflict		Relationship conflict
Variable	M	SD	M	SD	t-test
Low intensity conflict
Task content	5.57	1.36	4.25	1.56	5.00**
Relational content	2.50	1.48	5.05	1.51	−9.46**
High intensity conflict
Task content	5.29	1.48	3.59	1.71	6.05**
Relational content	3.59	1.68	6.21	0.92	−10.93**

Notes:

**p < 0.001. M = Mean, SD = Standard deviation. The scales range from 1 (Strongly disagree) to 7 (Strongly agree)

Table 4.

Manipulation checks for conflict expression intensity

	Low intensity conflict		High intensity conflict
Variable	M	SD	M	SD	t-test
Conflict type
Task conflict	2.87	0.74	3.51	0.69	−5.04**
Relationship conflict	2.59	0.76	3.97	0.84	−9.55**

Notes:

**p < 0.001. M = Mean. SD = Standard deviation. All scales range from 1 (Strongly disagree) to 5 (Strongly agree)

Table 5.

Summary of means, standard deviations and correlations (Study 2)

Variable	M	SD	1	2	3	4	5	6	7	8	9	10	11
1. Gender	0.59	0.49	1
2. Experience	13.05	9.41	−0.09	1
3. Conflict type	0.49	0.50	0.03	0.10	1
4. Conflict expression intensity	0.51	0.50	−0.09	0.09	−0.01	1
5. Information acquisition	3.30	0.92	−0.10	0.06	−0.43**	−0.07	1
6. Negative emotions	3.19	0.89	−0.08	−0.02	0.13*	0.23**	−0.29**	1
7. Problem-solving	5.23	1.18	−0.20**	0.10	−0.23**	−0.10	0.54**	−0.22**	1
8. Compromising	4.92	1.19	−0.15*	0.10	−0.15*	−0.09	0.40**	−0.07	0.79**	1
9. Yield	3.67	1.19	−0.06	−0.21**	−0.17**	−0.19**	0.33**	−0.08	0.40**	0.43**	1
10. Force	4.19	1.21	0.24**	−0.10	0.01	0.05	−0.03	0.09	−0.20**	−0.12	−0.27**	1
11. Avoid	4.05	1.54	−0.21**	−0.01	0.07	−0.13*	0.01	0.18**	0.17**	0.26**	0.45**	−0.33**	1

Notes:

n = 251. *p < 0.05, **p < 0.01. Conflict type: relationship conflict was coded as 1, task conflict was coded as 0. Conflict expression intensity: High intensity was coded as 1, low intensity was coded as 0. Gender was coded as 1 for male and 0 for female

Table 6.

Indirect effects from conflict types and conflict expression intensity to conflict management approaches (Study 2)

IV	Path	Standardized path estimate	95% CI
Task conflict (0) vs relationship conflict (1)	Information acquisition → problem-solving Information acquisition → compromising Information acquisition → yielding Negative emotions → forcing Negative emotions → avoiding	−0.15 −0.11 −0.10 0.05 0.03	[−0.23, −0.07] [−0.18, −0.05] [−0.15, −0.06] [0.02, 0.09] [−0.01, 0.06]
Low intensity conflict expression (0) vs high intensity conflict expression (1)	Information acquisition → problem-solving Information acquisition → compromising Information acquisition → yielding Negative emotions → forcing Negative emotions → avoiding	−0.06 −0.04 −0.04 0.05 0.03	[−0.10, 0.01] [−0.08, −0.01] [−0.07, −0.01] [0.01, 0.09] [−0.01, 0.06]

Notes:

n = 251; CI = Bias-corrected bootstrap confidence interval, p < 0.05

Notes

1. The design of Study 2 compares types of conflict and levels of conflict expression intensity, which precludes the ability to test these crossover paths.

2. To ensure that directness of conflict expression was comparable across conditions, we included both direct and indirect conflict expressions in all scenarios (Weingart et al., 2015).

Appendix 1

Scales study 1

Task conflict

1. We often debate different ideas when solving a problem.

2. We argue the pros and cons of different opinions.

3. We frequently discuss evidence for alternative viewpoints.

4. We often engage in debates about different opinions or ideas.

Relationship conflict

1. We often have arguments that get personal.

2. In our team, it is common to offend one another.

3. Personality conflicts are evident in my team.

4. There are often personality clashes within my team.

Information acquisition

1. I always receive the information I need from other team members on time.

2. It is very easy to get information from other team members when I need it.

Negative emotions

“When the team engages in relationship conflict, […].”

1. […] I feel afraid.

2. […] I feel hostile.

3. […] I feel distressed.

Conflict management approach

“When our team has a conflict, we do the following:”

[Problem-solving]

1. […] examine issues until we found a solution that really satisfies all of us.

2. […] stand for all our goals and concerns.

3. […] examine ideas from different sides to find a mutually optimal solution.

4. […] work out a solution that serves all interests as much as possible.

[Compromising]

1. […] try to realize a middle-of-the-road solution.

2. […] emphasize that we have to find a compromise solution.

3. […] insist we all give in a little.

4. […] strive whenever possible toward a 50-50 compromise.

[Yielding]

1. […] give in to other team members’ wishes.

2. […] concur with the other team members.

3. […] try to accommodate each other.

4. […] adapt to other team members’ goals and interests.

[Forcing]

1. […] push our own points of view.

2. […] search for (personal) gains.

3. […] fight for a good outcome for ourselves.

4. […] do everything to win.

[Avoiding]

1. […] avoid a confrontation about our differences.

2. […] avoid differences of opinion as much as possible.

3. […] try to make differences loom less severe.

4. […] try to avoid confrontations with one another.

Scales study 2

Information acquisition

“If I had been in this meeting, after this interaction, I would…”

1. […] have received information about the concerns of the other person.

2. […] be able to consider the other person’s opinions.

3. […] have gained a better understanding of the issues.

4. […] have better understood the position of the other person.

Negative emotions

“If I had been in this meeting, after this interaction, I would feel…”

1. […] frustrated.

2. […] angry.

3. […] annoyed.

4. […] tense.

Conflict management approach.

Same items as study 1. The prompt for Study 2 was:

“Keeping in mind the interaction you experienced in the scenario you have just read, how would you have resolved the conflict with the other person?”

Appendix 2

Study 2 scenarios

Imagine that you are a member of a three-person project team tasked with identifying ways to solve a problem faced by a client. The norm in these teams is for every member to contribute their ideas without criticizing and for the team to collectively agree on the best ideas to be presented to the client.

Relationship conflict scenarios

Low intensity expression.

In response to one of your ideas during a brainstorming session, one group member jokingly remarked, “really? I know you can come up with something better than that.” Just as you try to respond to this teasing comment, that group member lightheartedly adds, “my five year old has more novel ideas!” You realize your coworker was joking around, even though the comments were focused on you personally.

High intensity expression.

In response to one of your ideas during a brainstorming session, one group member remarked, “really? That is completely unrealistic, just like all your other ideas.” Just as you try to respond, the same group member cuts you off and adds, “you are talking without thinking. This is just another example of your unprofessional approach to our work.” You viewed the comments as an attack on you personally.

Task conflict scenarios

Low intensity expression.

In response to one of your ideas during a brainstorming session, another group member expresses his differing view that he is reluctant to consider your idea because he thought that the client had already rejected similar versions of your idea as a potential solution. You then present a counter argument and you explain in detail why your idea is novel and useful for the client. You and the group members further debate and deliberate about this idea.

High intensity expression.

In response to one of your ideas during a brainstorming session, one group member is unwilling to consider this idea and argues that the client had already rejected similar versions of your idea as a potential solution. You ignore his arguments and fight for the point that the client had rejected a very different idea for problem solution and that your idea is novel and useful. You and the group member get into an argument about this idea.