Learning from conflicts? The relations between task and relationship conflicts, team learning and team performance

Abstract Team-based working has become a popular model of work organization. One of the expected yields of working in teams is collective learning. As team members work closely together they are likely to exchange information, which can enhance learning. One of the processes that may facilitate team learning is team conflict. In this study, 471 respondents, working in 84 teams in 38 organizations, indicated their perceptions of task conflict, relationship conflict, team learning, and their rating of the performance of the team. Task conflict was not related to team learning but was negatively related to the team members' rating of team performance. As was hypothesized, relationship conflict negatively affected team learning. Importantly, team learning was a strong predictor of the perceived team performance and partially mediated the relation between relationship conflict and performance. Keywords: Team learningTask conflictRelationship conflictTeam performance Acknowledgment We would like to thank our master students Sonja Van Ettinger, Marloes Theuns, Elske Van Herk, Astrid Beekhuis, Jonne Van Der Loo, Nicole Kivits, Manon Sulkers, Hanneke Van Woezik, and Melissa Van Der Krogt who helped us with the data collection. Special thanks to Josje Dikkers for her help with the CFA, three anonymous reviewers and the editor for their helpful comment, our colleagues from the department Human Resource Studies for their feedback and Marjolein Van Offenbeek for letting us use her scales for team learning activities. Notes 1The large teams of 30–40 members were all part of healthcare organizations in which team members would usually work in smaller subsets of the team depending on the shift they would be working in. 2We also explored whether a curvilinear relation exists between task conflict and team learning (cf. De Dreu, Citation2006). We computed the squared term for task conflict and entered this additional term in the regression analyses predicting team learning (e.g., De Dreu, Citation2006, Jehn, Citation1995). Adding the squared term to Step 2 in Table 2 did not yield an improvement of fit (R 2 change = .002, ns), thus no curvilinear relationship exists. 3We also explored whether the relationships between task conflict and perceived team performance is curvilinear as suggested by research from De Dreu (2006). We followed De Dreu's procedure of adding the squared term for task conflict in a next step into the analysis (see also Footnote 2). No evidence was found for a curvilinear relationship. The model with the additional squared term showed no improvement of fit to Step 2a in Table 4 (R 2 change = .004, ns).