Evaluation is necessary to produce stereotype threat performance effects
Researchers recruited undergraduate women (N = 52) to determine whether evaluation is necessary to produce stereotype threat. Participants completed 15 multiple-choice GRE math problems. Prior to testing, researchers told participants either that previous research demonstrated gender differences (stereotype threat) or previous research indicated no gender differences (control). In addition, they were told either that their test scores would be combined with all participants (anonymous) or segregated and evaluated based on gender (segregated). Performance was impaired only in the segregated/stereotype threat condition, even though the gender stereotype was salient in both scoring conditions. Overall, these results suggest that fear of evaluation is necessary for stereotype threat to impair performance.