Stereotype threat, inquiring about test takers' ethnicity and gender, and standardized test performance

This paper reported on two field studies examining the consequence of making inquiries about demographic information prior to taking the Advanced Placement Calculus examination and the Computerized Placement Test. In Experiment 1, individuals who were taking the AP Calculus AB Examination either completed the standard answer sheet that inquires about the test taker's race and gender (stereotype threat, n = 897) or responded to an answer sheet with those sections masked (experimental, n = 755). Overall, girls performed worse than boys, and Blacks performed worse than Whites. There were no effects of soliciting race and gender information on test performance for any group. In Experiment 2, individuals who were taking the Computerized Placement Test at community colleges during a two-week period did not respond to standard questions including items about race and ethnicity before taking the exam (experimental, n = 632). Their performance was compared with the performance of students who took the test in the next two weeks under standard conditions (control and stereotype threat, n = 709). Women performed more poorly than men in algebra, and Blacks performed generally worse than Whites. Again, there were no appreciable effects associated with soliciting social identity information on any groups. The authors suggest that the high motivation present in high-stakes testing situations might be responsible for negating stereotype threat effects. However, a more recent re-analysis of these data (Danaher & Crandall, 2008) suggest that there is evidence of reduced performance based on the timing of identity solicitation, using somewhat more liberal statistical decision rules, and that these stereotype threat effects might have substantial consequences.

Stricker, L. J., & Ward, W. C. (2004). Stereotype threat, inquiring about test takers' ethnicity and gender, and standardized test performance. Journal of Applied Social Psychology, 34, 665-693.