In search of the key to closing achievement gaps

This post originally appeared on U.S. News and World Report’s Knowledge Bank blog. 

In early January 2016, Dr. John B. King Jr. stepped into the role of acting secretary of Education, following Arne Duncan’s resignation last month. As a vocal proponent of reducing inequalities in American schools and a person of color, Dr. King’s leadership may open the door for the Department of Education to take a stronger position on one of the most persistent problems in American schools: segregation. He has been a proponent of integrated schools in the past and as State Education Commissioner in New York, he promoted integration as part of the state’s school turnaround efforts.

Dr. King’s new role, in light of today’s holiday celebrating another Dr. King, warrants the question: Would greater school integration actually be successful in closing long-standing achievement gaps based on race and income?

Policy research and debates about this question have been going on for years, and a full discussion deserves more space than I can devote to it here. Yet, I’d like to come at this question from a slightly different angle than what has been commonly considered, based on my research work focused on equalizing student access to quality teaching. 

By way of background, American schools have had an unremarkable track record in achieving greater levels of school integration, in spite of long-time policy interests to do so. Though schools did show an increasing trend towards integration in the decades immediately following the 1954 Brown v. Board of Education decision, that trend has reversed over the last 20 years, and schools in many states are now only slightly less segregated than they were before the decision.

Given the intransigence of the segregation problem over time, policymakers in recent years appear to have become content with monitoring equality in educational resources across segregated schools rather than promoting student integration. And many acknowledge that schools serving large numbers of economically disadvantaged and minority students tend to be under-resourced in a myriad of ways; hence, the resurgence of the “separate and unequal” claim in recent years.

But even more important than dollars and textbooks, teachers themselves are a key educational resource; indeed, they are the primary school-based input into student learning. Proponents of school integration argue that equalizing access to school resources (with a big emphasis on teachers) is one of the key mechanisms through which integration will narrow achievement gaps. After all, we know from prior research evidence that teachers in disadvantaged schools tend to have less experience and fewer credentials than those in more affluent schools. 

Recent findings from a large project I am involved with, however, throw a bit of a wet blanket on that argument. I have been part of the research team for the Study of the Distribution of Effective Teaching, a large five-year project funded by the U.S. Department of Education’s Institute of Education Sciences that investigates whether disadvantaged students have equal access to quality teaching.

Our study took a different approach from prior research on the distribution of teachers. Instead of quantifying teacher quality based on experience and credentials (which are decidedly not the best indicators of student learning), we estimated the size of inequalities in access based on student learning gains on test scores. Using data from 29 large participating school districts, we did find gaps in access to effective teaching in English Language Arts in all districts, though not all were statistically significant. We also found gaps in access in math in most, though not all, districts.

To the surprise of those of us on the research team, though, the estimated gains from equalizing access to effective teaching across all students was not nearly as large as we had expected: We estimated student achievement gaps in test scores could be reduced by 2 percentile points in both subjects. This amounts to less than one-tenth of the current achievement gaps in either subject. 

What does this mean for integration? Well, inasmuch as the bulk of learning gains to integration are predicated on equalizing access to teachers for all students, our findings suggest that integration would help in a modest way, but very large gaps in achievement would remain for disadvantaged students even in integrated schools. Given how difficult it is to translate short-term learning gains into persistent gains across students, even if students went through all of grades K-12 in integrated schools, achievement gaps would be markedly narrowed but still very large. 

Given the large achievement gaps that accompany students when they enroll in kindergarten, it is not too surprising that simply equalizing access to quality teachers (or any other school resource) will not remediate those gaps. Under integration, however, at least U.S. schools would no longer be reinforcing these gaps students enter with.

Interestingly, we also found a surprising amount of variability across the 29 study districts, in which disadvantaged students had significantly lower access to effective teaching in a few of the districts. In these districts, equalizing access to teachers would make a more consequential and immediate dent in achievement gaps and put disadvantaged students on a very different learning trajectory. Yet, current gaps in these districts are presumably already larger than they otherwise would be due to the current state of inequalities.

I do not mean to suggest that reducing the level of segregation in schools is not worth trying. Equalizing student access to teachers should improve disadvantaged students’ learning now, along with other longer-term educational outcomes. Other mechanisms beyond effective teaching – like peer effects – could contribute to academic gains in integrated schools, not to mention other socio-emotional benefits that may accrue to students as well. However, I argue that the academic gains from this strategy may not be as transformational to the distribution of student achievement as some media accounts have suggested.

To conclude, I offer my congratulations to Acting Secretary King and look forward to this new chapter of education policy in America. I offer my genuine support in his efforts to raise awareness of and reduce inequalities in American schools, and hope that our schools can live up to the aspirations of both Drs. King.