Polarized appraisals of out-group members

Developed and tested a model that assumes that people have a more complex schema regarding in-groups than out-groups and consequently, that appraisals of out-group members will be more extreme or polarized than appraisals of in-group members. Four experiments with 415 White male and female undergraduates tested this model, as well as predictions derived from attribution principles. In Exp I, Ss read and evaluated a law school application containing incidental information about the applicant's race and gender. A Black applicant with strong credentials was judged more favorably than an identical White applicant, supporting a prediction derived from the augmentation principle. In Exp II, an applicant with weak credentials was included in the design. Results support the prediction that out-group members would be evaluated more extremely: When the application credentials were positive, the out-group member (a Black or opposite-sex applicant) was evaluated more favorably than the in-group member (a White or same-sex applicant). When the application credentials were weak, the out-group member was evaluated more negatively. Exp III and IV provided support for the 2 assumptions underlying the complexity-extremity hypothesis: First, White Ss demonstrated greater complexity regarding Whites than Blacks. Second, greater complexity resulted in evaluative moderation. (20 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved). © 1980 American Psychological Association.

Duke Scholars

Author Patricia W. Linville Fuqua School of Business