Evidence that Subjects' Expectancies Confound Intergroup Bias in Tajfel's Minimal Group Paradigm

Reports that persons discriminate against out-group members even at cost to the in-group refer primarily to subjects' distribution of money in minimal group situations. Expectancies that might confound that measure of discrimination are examined here. First divided into minimal groups, subjects divided money between in-group and out-group members so as to (a) confirm whatever they believed to be the study hypothesis (positive demand), (b) disconfirm that hypothesis (negative demand), and (c) express their own unrestricted preferences (intergroup bias). The demand measures preceded, followed, or were placed between blocks of unrestricted trials. There were no effects of demand placement, but significant negative correlations and differences in means were found between amounts awarded under positive and negative demand instructions. Correlations between positive demand and bias scores, questionnaire reports of expectancies favoring discrimination, and relationships between reported expectations and bias scores are consistent with "demand" explanations of early minimal intergroup studies.