Top management teams make strategic decisions that are critical for organizational performance. The intragroup interaction of management team members is directly related to decision quality and financial performance (Eisenhardt & Zbaracki, 1992; Peterson, Owens, Tetlock, Fan, & Martorana, 1998). Conflict is an inevitable relationship in groups (Amason, 1996). Jehn (1995) distinguished two types of intragroup conflict: task conflict and relationship conflict. There is empirical evidence that in nonroutine tasks like decision making, task conflict is functional and relationship conflict will lead to poorer implementation of the decision (Jehn, 1995). Relationship conflicts arise from task conflicts in team discussions (Jehn, 1995). Task conflicts will be positively associated with team performance when relationship conflict levels are controlled (Higashide & Birley, 2002). However, meta-analysis shows that it is still unclear how the two kinds of conflicts correlate, and how they influence decision performance (De Dreu & Weingart, 2003; Simons & Peterson, 2000).

In this study we looked at the role of conflicts in the process of strategic decision making in teams. We were especially interested in different decision-making patterns during the process. It has been suggested that the success of a decision depends on the process used to reach it (Dean & Sharfman, 1996). It has been shown in some studies that the regular conventional interacting consensus approach was less effective than other approaches (Van De Ven & Delbecq, 1974). The mechanism of conflicts in different decision-making processes and their effect on decision quality has not been examined in previous research.

Currently, a two-stage decision-making process is widely used in management teams in China. In the first stage, which we have called the consensus discussion stage, the team leader allows team members to share and discuss their opinions without commenting on them (as in the consensus approach). In the second stage which we have called the leader decision stage, the leader joins in the discussion and makes the final decision (as in the hierarchical distributed expertise approach). Because conflicts changed in this process we proposed that relationship conflict would be induced by, and positively correlated to, task conflict in the consensus discussion stage. In the leader decision stage relationship conflict would be reduced, while task conflict would facilitate decision performance.

Literature Review

The Relationship Between Task and Relationship Conflicts

Jehn (1995) divided conflict into task conflict and relationship conflict. It is widely agreed that task conflict can make members consider the task thoroughly and think over the task-related information more carefully, a process that facilitates decision making. But relationship conflict, which has arisen from task conflict, has been found to decrease team members’ satisfaction and commitment and lead to poorer implementation of the decision (Jehn, 1994, 1995; Jehn, Chatwick, & Thatcher, 1997; Jehn & Mannix, 2001). Simons and Peterson (2000) found an average correlation of 0.47 between task and relationship conflicts (range -0.17 to 0.88). De Dreu and Weingart (2003) gained results ranging from 0.18 to 0.84 (average = 0.54). We reviewed five empirical studies in Chinese journals, and gained an average correlation of 0.43 (ranged 0.13 to 0.59). Although the average correlation was around 0.5, there was considerable variation across studies.

Effects of Conflict on Decision Performance

Amason (1996) found implementation was critical and that members’ satisfaction and commitment affected decision implementation and performance. On the basis of these findings, we used members’ satisfaction, commitment, and implementation as the criteria of decision quality. Some researchers have found that relationship conflict will decrease satisfaction (Simons & Peterson, 2000). Task conflicts, on the contrary, are thought to benefit performance by leading to deep thinking and thorough consideration of information (Jehn et al., 1997). Although empirical evidence supports a strong positive correlation between task conflict and team performance (Nijdam, 1998), some researchers have found a negative correlation (Jehn, Northcraft, & Neale, 1999; Lovelace, Shapiro, & Weingart, 2001), or no correlation (Pelled, Eisenhardt, & Xin, 1999). There is no strong evidence that task conflict must be functional and relationship conflict must be dysfunctional (Amason & Schweiger, 1997).

Intragroup Goal Conflict

One possible explanation for the inconsistency of previous results, is that there are different goals set (Locke, Smith, Erez, Chah, & Schaffer, 1994) and competing interests (Eisenhardt & Zbaracki, 1992) among team members. Goal incompatibility could be a component of task conflict (Higashide & Birley, 2002) and this, in turn, would increase potential conflicts (Schmidt & Kochan, 1972). Goal conflict has been found to be negatively related to performance (Locke et al., 1994). Previous experimental studies on conflict were based on a fundamental assumption that there was only one team goal and, therefore, goal conflict was ignored (Eisenhardt & Zbaracki). We included goal conflict in the experimental design in order to explore how it affected conflicts and decision performance.

Two-Stage Decision Process

Authors of both experimental and field studies have suggested that decision technique could influence group decision making (Kiesler & Sproull, 1992) and that decision-making processes are related to decision quality and organizational success (Dean & Sharfman, 1996). The common conventional interacting consensus approach has been found to be less effective than other approaches (Van De Ven & Delbecq, 1974) and it has also been found that teams in which expertise was hierarchically distributed performed differently (Hollenbeck et al., 1995).

In the study by Hollenbeck and colleagues (1995) most management teams worked on a system of hierarchical distribution of expertise. In China, organizations have developed a more specific technique of management practice, which has become very popular. Members are first allowed to share and discuss their opinions without the team leader making comments, and then the leader joins in the discussion and makes the final decision. This two-stage process can be seen as a combination of consensus and hierarchical distribution of expertise. We have named the two stages of the process consensus discussion and leader decision.

This process was developed to prevent the negative influence of power distance and facilitate the decision performance in the Chinese culture where team members in a hierarchical team would be reluctant to contribute their opinions for fear of conflict with the team leader and other members of the group. A major purpose of this team decision process was to encourage team members’ debate on the decision alternatives, which, according to the conflict theory, can improve the decision quality.

Self-report questionnaires could merely reflect respondents’ perceptions of conflicts in previous tasks but might not focus on specific tasks and also might ignore the effect of task types (De Dreu & Weingart, 2003). In this study our aim was to explore conflict relationships and their effects on decision performance, based on the decision process for one specific task. We hypothesized that decision-making processes would moderate the relationship between task and relationship conflicts and their effect on decision performance. Specifically, we hypothesized that the consensus discussion stage might induce both task and relationship conflicts of significant gravity while a hierarchical discussion would reduce relationship conflict; decision processes would moderate both the positive relationship between task conflict and decision performance and the negative relationship between relationship conflict and decision performance.

Method

Participants

We recruited 120 graduate, doctoral, and Master of Business Administration degree course students (66 female) with majors in psychology and management from two universities in China. The ages of participants ranged from 22 to 32 years (M = 24.5, SD = 1.42) and over 70% had work experience.

Design and Task

We used two types of discussion process: consensus discussion vs. total process × 2 (goal conflict: with or without) between-subjects design and randomly divided participants into 40 teams with 3 members in each. For the consensus discussion, members had equal status and took part in an unstructured discussion to reach a consensus decision. For the total process, at the first stage the team was allowed to have a consensus discussion with no input from the leader, and in the second stage the team leader joined in the discussion and made the final decision. Dependent variables, including task and relationship conflict, and decision performance were assessed from the answers to the post discussion.

The task scenario was developed based on interviews with enterprises. Participants were assigned to act as a top manager of a clothing original equipment manufacture (OEM). The company was facing a financial crisis and had two alternative ways to transform itself. One was to move to a remote area for lower costs and continue doing the OEM, the other was to upgrade to using their own brand and develop domestic markets. We provided each member with three advantageous and three disadvantageous pieces of information for each alternative. For goal conflict, we provided information about influences on members’ family and the career of the individual. For one member transferring meant promotion and increased income and upgrading meant poor competency to fit the new job. For another member transferring led to poor education for his/her children and meant he/she would have to live away from family, and upgrading meant good use of client resources.

Procedure and Measures

Participants in each group first read the material for 10 minutes. Before discussion, in goal conflict groups two members were randomly assigned the goal conflict materials and had to complete a questionnaire about their attitudes to the alternatives with responses rated on a 7-point scale ranging from 1 = totally support the transfer plan to 7 = totally support the upgrading plan. For the consensus discussion groups, members discussed and reached a decision in 15 minutes. In the total process groups, two members first held a discussion for 10 minutes then the leader joined in the discussion for five minutes and made the final decision. After the decision was made, participants had to complete a questionnaire on conflicts and performance.

A conflict scale was developed, based on Jehn’s (1995) intragroup conflict scale. Jehn’s scale was designed to measure integrated impression of past team experiences rather than specific perception of a recently completed discussion, so several terms were not suitable for use in our study. Therefore, we used only six items and modified the scale to focus on the task just completed by our participants. For the decision performance we used satisfaction, commitment, and implementation items from a model developed by Zhou (2005). Responses to both these questionnaires were rated on a 7-point scale.

Results

Manipulation Checks and Factor Analysis

In groups where there was a situation of goal conflict, the mean attitude for upgrading treatment was 6 (SD = 0.79) and 2.86 (SD = 1.31) for transferring; the difference was significant (t(38) = 9.2, p < .01). Thus, we believe that our manipulation of goal conflict was valid. Factor analysis showed KMO of conflict scale was .772 and three items were aggregated to form task conflict (α = .887) and the three formed relationship conflict (α = .891). Three items formed the performance scale (α = .941) and KMO was 0.768. Therefore, we could use the mean of these items as variables for the next analysis.

Interaction Effect of Discussion Types and Goal Conflict On Performance

There was a significant interaction effect (F(1, 116) = 15.68, p < .01) using a 2×2 analysis of variance (ANOVA) (see Figure 1). When we examined the simple effect we found that consensus discussion resulted in significantly higher performance than did the total process when there were no goal conflicts (F(1, 116) = 11.57, p < .01). The result was reversed where there was goal conflict in the teams (F(1, 116) = 4.84, p < .05). The performance of teams without goal conflict was significantly higher in the consensus discussion (F(1, 116) = 9.41, p < .01), but the result was reversed in the total process group (F(1, 116) = 6.41, p < .05).

Effect of Discussion Types and Goal Conflicts on Relationship Conflicts

The main effect of goal conflict on task conflict was significant (F(1, 118) = 73.17, p < .01) and task conflict in the goal conflict group was much greater. But neither the main effect of discussion type (F(1, 118) = 1.35, p > .05) nor the interaction (F(1, 116) = 3.50, p > .05) was significant.

For relationship conflict, the interaction was significant (F(1, 116) = 14.57, p < .01) and relationship conflict in consensus discussion was significantly higher both in groups with (F(1, 116) = 62.12, p < .01), and without (F(1, 116) = 6.17, p < .05), goal conflict. While the relationship conflict of the goal conflict groups in consensus discussion was significantly higher (F(1, 116) = 20.10, p < .01), no significant difference was found in the total process group (F(1, 116) = 0.84, p > .1) (see Figure 2).

Effect of Conflicts on Decision Performance

Based on the literature review and discussion above, we considered that teams with goal conflicts could better represent the true behavior. Therefore, for the analysis of the influence of conflict on true performance of the team we used only data from goal conflict groups. Results show that there was a significant positive correlation between task and relationship conflict in the consensus discussion type (r = .63, p < .01) but neither discussion process showed significant correlations with decision performance. A significant positive correlation was found between task conflict and decision performance in the total process group (r = .62, p < .01) and no other significant results were recorded.

Discussion

The main purpose in this study was to explore how conflict influences team decision performance. Previous researchers had found that the correlation between task and relationship conflicts ranged from negative to significantly positive (De Dreu & Weingart, 2003; Simons & Peterson, 2000) and many researchers believe that there must be some moderating variables between them (Carnevale & Probst, 1998; Jehn & Mannix, 2001; Simons & Peterson, 2000). We hypothesized different decision-making processes as an alternative explanation. Our findings demonstrated that the task conflicts produced in the consensus discussion remained in place until the end of the total process, which means that the leader decision stage did not remove task conflicts. However, the relationship conflicts decreased significantly during the leader decision stage. Therefore, we can infer that the leader decision stage can increase the decision performance by reducing the dysfunctional relationship conflict.

For the consensus discussion, task and relationship conflicts operated similarly to the process observed in previous studies. There was a significant positive correlation but neither task nor relationship conflict showed a significant correlation with performance. This may be because the positive effect of task conflict on performance was neutralized by the negative effect of relationship conflict. For the teams using the total discussion process, task and relationship conflict showed no significant correlations. Relationship conflict reduced significantly in the leader decision stage while task conflict remained the same as in the previous stage of group discussion. The consistency of task conflicts can be explained by the fact that the consensus discussion had raised all the possible task conflicts. Relationship conflict, which was correlated with task conflict, reached a significant level in consensus discussion and this could account for the low performance at this stage. When the leader joined in the discussion a harmonious atmosphere pervaded the team (Briley, Morris, & Simonson, 2000; Briley & Wyer, 2001; Fong & Wyer, 2003; Markus & Kitayama, 1991) and members felt fewer relationship conflicts and greater satisfaction with the team leader’s decision.

Based on our findings, we believe that the two-stage team decision making process works more efficiently and produces better team performance than does the consensus discussion process in the Chinese cultural setting. We cannot expand this finding to other cultures because the specific process has developed during years of management practices in China and fits the culture very well. It can satisfy both the team members and team leaders and can also result in high quality decisions and productive implementation. This kind of decision-making process makes effective use of intragroup functional task conflicts and allows control to be effectively exerted over dysfunctional relationship conflicts.

Amason, A. C. (1996). Distinguishing the effects of functional and dysfunctional conflict on strategic decision making: Resolving a paradox for top management teams. Academy of Management Journal, 39(1), 123-148.

Amason, A. C., & Schweiger, D. (1997). The effect of conflict on strategic decision-making effectiveness and organizational performance. In C. K. W. De Dreu & E. Van de Vliert (Eds.), Using conflict in organizations (pp. 101-115). London: Sage.

Briley, D. A., & Wyer, R. S. (2001). Transitory determinants of values and decisions: The utility (or nonutility) of individualism and collectivism in understanding cultural differences. Social Cognition, 19(3), 197-227.

Briley, D. A., Morris, M., & Simonson, I. (2000). Reasons as carriers of culture: Dynamics versus dispositional models of cultural influence on decision making. Journal of Consumer Research, 27(2), 157-178.

Carnevale, P. J., & Probst, T. M. (1998). Social values and social conflict in creative problem solving and categorization. Journal of Personality and Social Psychology, 74(5), 1300-1309.

De Dreu, C. K., & 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.

Dean, J. W., Jr., & Sharfman, M. P. (1996). Does decision process matter? A study of strategic decision-making effectiveness. Academy of Management Journal, 39(2), 368-396.

Eisenhardt, K. M., & Zbaracki, M. J. (1992). Strategic decision making. Strategic Management Journal, 13(1), 17-37.

Fong, C., & Wyer, R., Jr. (2003). Cultural, social, and emotional determinants of decisions under uncertainty. Organizational Behavior and Human Decision Processes, 90(2), 304-322.

Higashide, H., & Birley, S. (2002). The consequences of conflict between the venture capitalist and the entrepreneurial team in the United Kingdom from the perspective of the venture capitalist. Journal of Business Venturing, 17(1), 59-81.

Hollenbeck, J. R., Ilgen, D. R., Sego, D. J., Hedlund, J., Major, D. A., & Phillips, J. (1995). Multilevel theory of team decision making: Decision performance in team incorporating distributed expertise. Journal of Applied Psychology, 80(2), 292-316.

Jehn, K. (1994). Enhancing effectiveness: An investigation of advantages and disadvantages of value- based intragroup conflict. International Journal of Conflict Management, 5(3), 223-238.

Jehn, K. (1995). A multimethod examination of the benefits and detriments of intragroup conflict. Administrative Science Quarterly, 40, 256-282.

Jehn, K., Chatwick, C., & Thatcher, S. (1997). To agree or not to agree: The effects of value congruence, individual demographic dissimilarity, and conflict on workgroup outcomes. International Journal of Conflict Management, 8(4), 287-305

Jehn, K., Northcraft, G. B., & Neale, M. A. (1999). Why differences make a difference: A field study of diversity, conflict, and performance in workgroups. Administrative Science Quarterly, 44, 741-763.

Jehn, K., & Mannix, E. (2001). The dynamic nature of conflict: A longitudinal study of intragroup conflict and group performance. Academy of Management Journal, 44(2), 238-251.

Kiesler, S., & Sproull, L. (1992). Group decision making and communication technology. Organizational Behavior and Human Decision Processes, 52(1), 96-123.

Locke, E. A., Smith, K. G., Erez, M., Chah, D., & Schaffer, A. (1994). The effects of intra-individual goal conflict on performance. Journal of Management, 20(1), 67-91.

Lovelace, K., Shapiro, D. L., & Weingart, L. R. (2001). Maximizing cross-functional new product teams’ innovativeness and constraint adherence: A conflict communications perspective. Academy of Management Journal, 44(4), 779-793.

Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98(2), 224-253.

Nijdam, N. E. (1998). The functioning of work teams. Unpublished master’s thesis. University of Amsterdam: The Netherlands.

Pelled, L. H., Eisenhardt, K. M., & Xin, K. R. (1999). Exploring the black box: An analysis of work group diversity, conflict, and performance. Administrative Science Quarterly, 44, 1-28.

Peterson, R. S., Owens, P. D., Tetlock, P. E., Fan, E. T., & Martorana, P. V. (1998). Group dynamics in top management teams: Groupthink, vigilance, and alternative models of organizational failure and success. Organizational Behavior and Human Decision Processes, 73(2/3), 272-305.

Schmidt, S. M., & Kochan, T. A. (1972). Conflict: toward conceptual clarity. Administrative Science Quarterly, 17(3), 359-370.

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(1), 102-111.

Van De Ven, A. H., & Delbecq, A. L. (1974). The effectiveness of nominal, delphi, and interacting group decision-making processes. Academy of Management Journal, 17(4), 605-621.

Zhou, J. B. (2005). Multilevel decision model and its mechanism of decision performance in entre- preneurial teams. Unpublished doctoral dissertation. Zhejiang University: People’s Republic of China.

This research was supported by the Chinese National Scientific Foundation (70732001). Appreciation is due to reviewers including

Ana Jimenez-Zarco

Open University of Catalonia

Barcelona

Spain

Email

[email protected]

Long Cheng, 16#203, Xixi Campus, Zhejiang University, Hangzhou 310028, People’s Republic of China. Phone: +86-1358-8745608; Email: [email protected]