Discounting the difficult: How high math identified women respond to stereotype threat
In this experiment, men and women who varied in their level of math identity completed a difficult math test either under stereotype threat (where participants were told that males and females perform differently on the test) or no stereotype threat (where participants were told there are no gender differences in test performance). After controlling for self-reported math SAT scores, a typical stereotype threat effect emerged; women performed more poorly under stereotype threat than when not under stereotype threat, but men showed no difference in performance. In addition, highly identified women in the stereotype threat condition discounted the test more than other participants. Specifically, they were more likely to agree with the statements, "this test is not an accurate measurement of my math ability," "I feel that I am better at math outside of this test," and to disagree with the statements, "tests like this one are not biased against certain groups of people," and "my score on this test will likely be an accurate measure of my math ability." Moreover, these differences in discounting emerged even when controlling for actual test performance, suggesting that task disparagement arises from stereotype threat and does not simply reflect reactions to poor performance.