match students and instructors by race

9/6/2011: Here’s a new carefully done study on the effects of matching the races of instructors and students. By three NBER economists with data from a CA community college. As paraphrased by Scott Jaschik at Insidehighered.com:

Among all nonwhite groups, the study found a gain of 2.9 percentage points in the proportion of students completing courses taught by instructors of the same race as students — cutting in half the gaps in minority vs. white course completion rates. (Among all students in all non-recreational courses, 24 percent of white students drop out, compared to 26 percent of Asian students, 28 percent of Latino students, 30 percent of black students and 28 percent of other, nonwhite students.)

Among students who don’t drop out, there are also gaps between the performance of white and Asian students and that of other groups, especially black students. For instance, of those students who don’t drop out, 89 percent of white and Asian students pass, compared to 82 percent of black students; and 68 percent of white and Asian students who complete courses earn at least a B, while only 53 percent of black students do. For black students taught by a black instructor, there was a gain of 13 percentage points — among those who completed the course — in the proportion earning a B or higher.

At the same time, the authors note that there were declines in various performance measures for white students taught by non-white instructors.

The logical policy conclusion is for more matching of students and instructors by race, i.e. more segregation. Or we could try to figure out what drives this effect, and help students and teachers offset it. Presumably both groups are already making some efforts to do this – but not entirely successfully. Or we could cry racism, bury our heads in the sand, and continue our current policies.