Previous research demonstrates that students taught by teachers of the same race and ethnicity receive more positive behavioral evaluations than students taught by teachers of a different race/ethnicity. Many researchers view these findings as evidence that teachers, mainly white teachers, are racially biased due to preferences stemming from racial stereotypes that depict some groups as more academically oriented than others. Most of this research has been based on comparisons of only black and white students and teachers and does not directly test if other nonwhite students fare better when taught by nonwhite teachers. Analyses of Asian, black, Hispanic, and white 10th graders in the 2002 Education Longitudinal Study confirm that the effects of mismatch often depend on the racial/ethnic statuses of both the teacher and the student, controlling for a variety of school and student characteristics. Among students with white teachers, Asian students are usually viewed more positively than white students, while black students are perceived more negatively. White teachers' perceptions of Hispanic students do not typically differ from those of white students. Postestimation comparisons of slopes indicate that Asian students benefit (perceptionwise) from having white teachers, but they reveal surprisingly few instances when black students would benefit (again, perceptionwise) from having more nonwhite teachers. (Contains 4 tables.)