ReducingStereotypeThreat.org

Major, Spencer, Schmader, Wolfe, & Crocker, 1998

Two experiments tested the notion that stereotype threat can produce disengagement from domains implicated by stereotype threat. In Experiment 1, Black and White undergraduates were told they were to take a test that either "may be biased against people belonging to certain ethnic and/or racial groups" (stereotype threat for Blacks) or "is culturally unbiased...the scores are a true test of intelligence, regardless of ethnic background" (control). The students then completed either an easy or a difficult task involving word associations. Based on whether they had been given the easy or difficult task, students were told that they had done well above average or well below average, respectively. The students' performance self-esteem measured after they had received feedback was the primary dependent measure. Results showed that Whites reactions to the feedback were more extreme; compared with Black students, they felt better after positive feedback and thought they actually performed better but felt worse and thought they had performed worse after negative feedback. Although these results were consistent with the hypothesis that Black's self-esteem can become decoupled from academic performance, it was surprising that the stereotype threat manipulation did not moderate this effect. Experiment 2 addressed the possibility that the mere mentioning of race in both conditions of Experiment 1 might have primed racial stereotypes, inducing threat in both conditions. To do so, Black and White undergraduates who varied in their pre-existing levels of identification with academics completed the difficult test from Experiment 1 either to "determine whether [it] is racially biased against certain minority groups" (stereotype threat for Blacks) or to "gather more information on this important test" (control). Although all students were told they had done poorly, the affect of that feedback on self-esteem varied by race and stereotype threat condition. Under stereotype threat, Black's self-esteem was higher than for Whites after failure feedback. In the control condition, however, failure feedback caused Blacks to exhibit lower self-esteem than Whites. The consequences of feedback were also somewhat moderated by chronic disengagement from academics among Black, but not White, students. The findings confirm that African-American self-esteem tends to be disengaged from academic performance, particularly in conditions that highlight the potential role of racial bias in accounting for poor performance.