Katie Worth

Schools in the United States, despite decades of nominally trying to diversify their workforce, still mostly employ white teachers. In a 2011 report titled “Teacher Diversity Matters” by the Center for American Progress, author Ulrich Boser noted that at some point in the next several years, the number of non-Hispanic white children in America’s public schools will be outnumbered by the number of children of color – and in fact, that’s already the case in some states, like California, where 72 percent of students are of color. This diversity is not reflected among the teachers who provide these children with an education: Only 17 percent of the country’s teaching force is non-Hispanic white. In California, just 29 percent of teachers are non-Hispanic whites. In fact, more than 20 states have a disparity of more than 25 percentage points between the diversity of students and teachers.

There is evidence that this yawning gap between the ethnic racial heritage of the student body and the people who teach them has an impact on the quality of education. A follow-up to the 2011 study, “Teacher Diversity Revisited,” published in 2014, notes that teachers of color can serve as role models for students of color, and help them feel more at home in schools.  Further, non-white students have better educational outcomes if they are taught by teachers of color. The benefits aren’t just for students of color, either: White students with a diverse teachers profit educationally from interacting with people and authority figures who look differently than they do.

But this phenomenon hasn’t been explored on a data level before, so I decided to gather the data from New York State and see whether we could find traces of the phenomenon there. Does the city’s teaching force reflect the rainbow of its student body? Are there some schools that are doing better than others? Are there some schools that are nearly all students of color but are taught by an all-white faculty? Are there any instances of the reverse? Which schools are doing the best at attracting teachers of color? Is there evidence that the diversity differential affects student outcomes? In which ways?

What I found is that if you are a student, your race is highly determinant of who you will be taught by – this is especially true if you are white. Schools with a large percentage of white students are taught by a large percent of white teachers. There is not a single school among the 4,500 that I analyzed that has a student body more than 50 percent white taught by less than 50 percent white teachers.

The same is not true for students of color, who are far more likely to be taught by people who don’t look like them. There is a relatively linear relationship between the percent of black students and percent of black teachers, but even schools that were nearly 100 percent black were taught by less than 60 percent black teachers.

The discrepancy was even greater among Latinos and Asians: Even the most densely Latino schools were taught by an average of about 30 percent Latino teachers. Asian students are likely to have less than 10 percent of their classes taught by Asian teachers, even if they are in a majority Asian school.

The next step was to look at student outcomes. I borrowed the Center for American Progress’s method of calculating a “diversity differential,” which subtracts the percentage of teachers of color at each school from the percentage of teachers of color. The result is a virtually always-positive figure that describes the difference in the racial makeup of students and teachers at the same school.

I plotted this against attendance rates and suspension rates, to see if there was a relationship.

And in fact, there was – the higher the diversity differential, the more likely were the students to have absences, or suspensions. Nonetheless, the relationship was only slight, with R-squared figures of 0.08 and 0.04. This is far too little to reach any conclusions, even about correlation.

That doesn’t mean the diversity differential does not affect student outcomes, or that our data couldn’t show it – simply that the regressions I ran did not do so.

A lot more work could be done on this – more outcomes examined, more thoughtful regression analysis run, etc. As it stands, our data tells a story of segregation in schools and racial disparity, a world where white students hardly ever contend with non-white authority figures, while students of color are overwhelmingly taught by white authority figures. This in itself is a story worth telling.