Stereotype threat: The effect of expectancy on performance
One reason that stereotype threat might interfere with performance is because it lowers expectations. Two studies examined the consequences of stereotype threat on expectations of performance quality. In Experiment 1, female undergraduates completed a math task after being told either that on tasks of logicalmathematical abilities "women obtain lower scores than males" (stereotype threat), "women obtain higher scores than males" (no stereotype threat), or that "no differences between men and women emerge" (control). For women low in math identity, expectations of performance and actual performance were unaffected by the manipulation. In contrast, for women high in math identity, expectations of performance were affected by the manipulation (higher expectations under no stereotype threat, followed by control, followed by stereotype threat), and performance mirrored these expectations. Expectations partially mediated the effects of stereotype threat on performance, but only for women identified with math. Experiment 2 attempted to assess whether expectations and performance might vary depending on the particular identity made salient through specific intergroup comparisons. Black Americans living in Italy experienced a manipulation to highlight either their race or their nationality, and they were told that this identity produced either superior or inferior performance on a test of verbal ability. When individuals were told that their a highlighted identity was associated with superior performance, individuals expected to do better on the verbal task and indeed did perform better than those whose identities were associated with poor performance. These effects emerged regardless of whether race or nationality was highlighted. However, the strength of racial (but not national) identity did interact with the other variables, leading to stronger decrements in performance when highly-identified Blacks were told that their racial group did not perform well. Expectancies again partially accounted for the decreased performance under stereotype threat but only among Black participants. These studies show that lowered expectations might account for performance decrements under stereotype threat, particularly among stigmatized and low-status groups.