This paper reports two field studies in which students were led to self-affirm to determine the consequences of stereotype threat on academic performance. In these studies (separated by a year and comprised of a separate set of students), seventh grade students at a racially-diverse middle school in the Northeast U.S. were randomly assigned to self-affirm or not to self-affirm as part of a brief classroom exercise. Students who did not self-affirm indicated their least important values and wrote an essay regarding why those values might be important to others. Students who did self-affirmed indicated values that were important to them and wrote an essay indicating why those values were important. The effects on academic performance during the semester were dramatic. Black who had been led to self-affirm performed 0.3 grade points better during the semester than those who had not, and those benefits occurred both in the class where the intervention took place and other classes. Moreover, benefits occurred regardless of pre-intervention levels of demonstrated ability. The self-affirmation intervention appears to have attenuated the drop in performance that typically occurs for Black students over time.