The Effect of Social Categorization on Coalition Formation.

Abstract

The purpose of the present experiment was to determine the effect of group membership on coalition formation. Subjects rank-ordered ingroup and outgroup members of laboratory groups as preferred coalition partners. A coalition game then followed in which bogus offers were made by these members to the subjects. These offers in all cases were unfair. Hypothesis One stated that participants would prefer to form coalitions with members of the ingroup, and that this ingroup preference would be stronger for those who have few resources than for those who have many resources. Coalition preferences formed three clusters: (i) high and medium resource ingroup members, (ii) high and medium resource outgroup members, and (iii) low resource members of both groups. In addition, low resource members were found to prefer forming coalitions with members of the ingroup more than did the high resource members. However, ingroup preference of high resource members increased when strength of categorization increased, while ingroup preference of low resource members decreased when strength of categorization increased. Thus, the results confirm the hypothesis, and suggest that the location of the minimum-sized-winning coalition will also influence coalition preference. According to Hypothesis Two, participants will accept coalition offers from ingroup members over outgroup members, provided ingroup members make fair offers. However, violations of these expectations by ingroup members will be seen as grounds for rejecting an offer, while comparable offers by outgroup members will not be seen as unusual. It was found that high resource members accepted more unfair offers from ingroup members than from outgroup members, while low resource members accepted fewer such offers from ingroup members than from outgroup members. These results confirm the hypothesis, and suggest that participants will reject coalition offers from the ingroup that are unfair primarily if they have few resources, or when the best coalition, that which is minimum-sized and winning, can be formed with an ingroup member. Hypothesis Three stated that participants will perceive ingroup members as more heterogeneous in personal attributes than outgroup members, but that this perception would be diminished when no utilitarian advantages are associated with ingroup membership. In support of the third hypothesis, it was shown that when the minimum-sized winning coalition was in the ingroup, members saw more ingroup heterogeneity than did those when it was equally available in either group or only available in the outgroup. (Abstract shortened with permission of author.)