top of page
Writer's pictureJoanne Jacobs

Students don't learn more from a same-race teacher

Elementary students don't learn more from a teacher who matches them in race or ethnicity, concludes a new study. There is little difference in behavior, executive functioning or placement in gifted or special education classes, write Paul L. Morgan, a Penn State professor of education and demography and Eric Hengyu Yu, a graduate student.


Nationwide, 79 percent of teachers are white, 7 percent black, 9 percent Hispanic and 2 percent Asian American. By contrast, 45.2 percent of K-12 public school students are White, 14.9 percent are Black, 28.4 percent Hispanic and 2 percent Asian.


However, "calls to diversify the teacher workforce are unlikely to meaningfully address large racial and ethnic educational inequities in U.S. elementary schools," they write in USA Today.


In theory "a teacher of the same race or ethnicity might help lessen biases and cultural misunderstandings, increase access to role models and mentors, and foster student engagement in classroom activities," they write. There's evidence that black students who've had a black teacher go farther in school.


However, national studies have found only small benefits for students of color," write Morgan and Yu. Effects are "more often observed on subjective measures like classroom behavior than on objective measures of academic achievement."


Student-teacher racial matching seems to have more impact for students attending schools in the U.S. South, they note.


That could be explained by research by Lavar Edmonds, a Stanford graduate student. He found black elementary students didn't do any better with a black teacher, but earned higher math scores if their teacher, of any race, had attended a historically black college or university (HBCU), reports Hechinger's Jill Barshay. There was no effect for white or Hispanic students.


Why? Edmonds doesn't know.


His study found "Black boys were more likely to be suspended with white teachers than with Black teachers," writes Barshay. But "Black boys were less likely to be suspended by an HBCU-trained white teacher than a white teacher who trained elsewhere."

5 Comments


Guest
May 21, 2023

I can understand that the race of a specific teacher might not have any effect on a Black child's performance, but do believe that having Black teachers in the school (or a Black principal) would make that child feel more positive towards school, and make him or her feel more like a part of the school community. I certainly felt that way about female professors in college. I never had one, but I knew they were there (though not in great numbers).

Like

Steve Sherman
Steve Sherman
May 19, 2023

The obsession with race is a no win zero sum nightmare where the kids always are the losers. And it will stay that way because there's a lot of interested parties who want it that way

Like

Guest
May 19, 2023

One of the districts my kids attended, for most of their k-12 years, had a large percentage of Asian kids; Chinese, Korean and Vietnamese. There was ONE Asian (Filipino) teacher; in the middle school and none in ES or HS. HS grad classes were about 400, increasing to 500ish after we moved. The school was very high-performing and was especially strong in math/science. I was told that the increase in students included even more Asians; because of the strong academics. Somehow, they managed to do particularly well; even without role models who looked like them.

Like

Guest
May 19, 2023

If you run enough tests of hypotheses, sooner or later you'll get a statistically significant result. These strange, highly specific findings suggest that the researchers ran a LOT of tests, so it's not surprising that they ended up with weird and inexplicable conclusions.

Like
Guest
May 20, 2023
Replying to

It used to be called a multiple comparisons problem. These days, one would have to control the experiment statistica rate to 95% or calculate a p-value considered all of the comparisons. There used to be tests that accounted for multiple comparisons but still calcuated the 95% CI.

Like
bottom of page