Why do women underperform under stereotype threat? Psychological Science, 16, 572-578

In this experiment, women undergraduate students were given a difficult math task after being told either that "there are clear differences in the scores obtained by men and woman" (stereotype threat) or that "there are no differences between men and women" (control) in logical-mathematical tasks. Interspersed throughout the testing booklet were blank sheets on which students were encouraged to "write anything that comes to your mind." These thought listings were classified by coders who were unaware of the students experimental condition as reflecting i) negative math related thoughts, ii) hatred of math, iii) general distress, iv) level of self-confidence, or v) neutral references to the test. Consistent with previous studies, women performed more poorly under stereotype threat than in the control condition, and performance differences increased from the beginning to the end of the test. Significant differences were found between conditions regarding the frequency of negative math-related thoughts, but there were no significant differences in the frequency of the other categories. Moreover, the number of negative thoughts regarding incompetence and low ability in math appeared to account for the decreased performance in the stereotype threat condition. Whereas other work has suggested that stereotype threat might increase cognitive burdens, this experiment demonstrated that burden might reflect increased concerns with one's ability in the domain in question.

Cadinu, M., Maass, A., Rosabianca, A., & Kiesner, J. (2005). Why do women underperform under stereotype threat? Psychological Science, 16, 572-578.