Dual identities in intergroup contact: Group status and size moderate the generalization of positive attitude change

To explore the eVects of various categorization strategies on intergroup bias within and beyond a contact situation, two experiments were conducted involving groups of diVerent size and/or status that worked together on a cooperative task. Three categorization strategies (decategorization, recategorization, and dual identity) were compared, and bias was measured through symbolic reward allocations to people who were and were not actually encountered. In Experiment 1 (ND 129), we varied group size (minority or majority) and found that it aVected bias within the contact situation—minority groups were more biased than majority groups. All of the categorization strategies limited bias and they did so equally well. Outside the contact situation, however, only the recategorization and dual identity strategies limited bias. In Experiment 2 (ND 156), we varied both group status (low or high) and group size. Both of these variables aVected bias within the contact situation—high status groups were more biased than low status groups, and minority groups were again more biased than majority groups. Once again, all three categorization strategies limited bias and they did so equally well. Outside the contact situation, however, an interaction among the independent variables was observed. For minority groups, only the dual identity strategy limited bias, but none of the categorization strategies limited bias for majority groups. © 2005 Elsevier Inc. All rights reserved.