Students’ academic achievement is crucial to life opportunities and chances, yet minorities continue to underperform compared to whites in reading and math. One mechanism that makes a difference in student success is teachers, and especially teacher-student relationships. Since research emphasizes that feelings of social belonging positively influence student achievement, some suggest that good teacher-student relationships, potentially cultivated by sharing race (race congruence), may improve students’ achievement. This study uses data on a census of elementary and middle school students (white, Black, and Hispanic) in a large urban school district in Texas from 2009-10 through 2010-11 to investigate the potential benefits of race congruence between teachers and students on a standardized math and reading assessment. This study contributes by focusing on a heavily segregated school district. Using regression modeling to isolate the effects of congruence independent of other student and teacher characteristics, I examine race differences in any effects of congruence. The results indicate little evidence for race congruence benefiting students except for one situation: black students in reading, which may be driven by school racial composition.