Attribution Biases in the Evaluation of New Product Development Team Members*

Managers are often concerned with the potential negative reputation impact of being assigned to a new product development project. Social psychology theories, and in particular the group attribution error theory, suggest that their worries might be justified, with individual team members being evaluated on the basis of the overall project performance, without regard for the processes by which the team outcome was reached. The objective of this paper is to empirically test for the existence of such biases in the evaluation of new product development team members. For this purpose, three independent experiments based on scenarios test the extent to which the group attribution error is at play in the evaluation of new product development team members and the extent to which it can be removed. Overall, this paper indicates that this bias does indeed affect the evaluation of new product development team members as well as decisions based on these evaluations. In the studies presented in this paper, analysis of variance showed that subjects inferred that team members' attitudes were consistent with the decision made and failed to adjust adequately for the decision rule used. Subjects then used these summary judgments as the basis for deciding on reward allocations and making competence attributions about the team members. In Study 1 , the decision rule used was either a vote or a team leader decision, and therefore the bias might have been explained by the lack of information available. Study 2 , however, provided unambiguous information about team members' positions, yet subjects did not adequately take this information into account. Study 3 replicated these results with experienced new product team managers, suggesting that theses biases are likely to be at play in the workplace. Moreover, subjects in Studies 2 and 3 felt quite confident that their judgments were being fair, even in the cases where these judgments truly were not, which suggests a lack of awareness of the bias on their part. The robustness of this bias should be cause for concerns for managers working in new product development teams or involved in the evaluation of the performance of such teams. The studies conducted in this paper suggest that team members can get unfairly rewarded or punished for decisions over which they have little or no control and that their reputation can also get affected by these decisions. Moreover, the fact that the group attribution error affected evaluations even in the case where experienced participants had specific information about team members' positions suggests that this bias will not be easy to remove.