This experiment built on research showing stereotype threat based on social class (Croizet & Claire, 1998). Similar to Croizet & Claire (1998), students who were low or high in SES participated in the study. Unlike Croizet & Claire, middle-class students were also included, and both White and minority students were included in the sample. Students completed a difficult math and verbal test with instructions suggesting that the test provided a "valid assessment of their abilities" and that "middle and upper income students consistently performed better than lower income students on standardized tests" (stereotype threat for low SES) or that the "purpose of the research is to understand the psychological factors involved in completing standardized tests" (control). After completing the tests, students completed scales designed to measure effort exerted, domain identification, test anxiety, and state self-esteem. Results indicated that lower SES students performed worse on both tests in the stereotype threat than in the control condition, whereas high SES students performed better on both tests in the stereotype threat compared with the control condition. Middle-class students did not perform differently on the tests in the two conditions. Results on anxiety mirrored these effects, but results regarding domain identification showed only that low SES students identified less with English and math under stereotype threat compared with the control condition. The only significant effect involving race showed that White lower income students identified less with English than did non-White lower income students. There were no significant effects of any variables on effort exerted and self-esteem. This study adds to the understanding of social class effects in stereotype threat.