While the population of minority students in the US continues to grow, the number of minority teachers has not kept pace, in spite of the fact that the percentage of minority teachers has actually more than doubled since 1988 (a little known fact surfaced by Richard Ingersoll.) A new study from a group of researchers from Harvard, University of Arkansas and University of Colorado offers more evidence that the lack of minority teachers is hurting student achievement. It's a study that plays well and has gotten a lot of attention—but one whose findings should still be put in perspective. 

The researchers took advantage of Florida's large dataset, finding some limited evidence that matching teacher race with student race can improve outcomes. In reading, for African American and white students, a .004 to .005 standard deviation bump in scores was achieved. In math, a slightly stronger effect size was picked up for not only African American and white students, but also Asian/Pacific Island students, who experienced a .007 to .041 standard deviation bump. (No effect was found for Hispanic students; but since the study did not control for assignment to ELL classes, we think the lack of findings should be attributed to the methodology employed here, and not because Hispanic students don't actually benefit from having Hispanic teachers.) 

As these results suggest, certainly having race-congruent teachers appears to nudge the needle on student achievement, but what gets overlooked is that other interventions can move it more. Here we compare the effect sizes of teachers of the same race as their students with the effect sizes of a few other interventions, mostly achieved when schools have altered the curriculum.

It's hard not to notice that choosing a better math curriculum yields effects seven times greater (using the most conservative calculation) than matching teacher and student race.

Getting districts to embrace the importance of strong curriculum in an era of curricular indifference may be a fool's errand, but count us in.