A Descriptive Bias in Trait Attributions following Decisions Favoring One Role Candidate Over Another

An experiment tested the hypothesis that deciding in favor of one role candidate over another would subsequently cause people to view the chosen individual in more role-stereotypic terms, and the rejected individual in less role-stereotypic terms, even if the initial decision itself were based on a factor (e.g., likability) unrelated to considerations of personality–role congruence. Subjects read materials on two college students; some of them were later asked to indicate which would be more deserving of admission to (and likely to succeed in) programs in certain career fields. Each student's “likability” and “personality” were independently manipulated so that the unlikable student's personality fit the type of career in question better than did the likable student's. Nevertheless, subjects making the career-related judgments tended to favor the likable student in their decisions (as intended). These subjects' subsequent trait attributions to the likable student were biased toward the career stereotype, and their attributions to the unlikable student were biased away from that stereotype, relative to the attributions of subjects who had not made the career-related judgments. The findings are discussed in relation to previous studies of initial-judgment effects, and their possible significance for cognitive, motivational, and cultural views of trait-attribution processes is considered.