Recent research calls for integration, but it won’t level the playing field unless we also change our approach to education to align with what scientists have discovered about how kids learn.
Getty
A recent study concludes that gaps in student test scores are driven by poverty, not race—but then says the solution must nevertheless be racial integration. More fundamentally, it overlooks current classroom practices that perpetuate income-based gaps even when schools are integrated.
Earlier this week, Stanford University sociologist Sean Reardon and some colleagues released a report using massive amounts of test-score data to investigate the effects of modern-day racial segregation. After Southern schools were desegregated in the wake of Brown v. Board of Education in 1954, test-score gaps between black and white students decreased. But with the decline of court-ordered integration, racial segregation in schools returned and has remained at high levels since the 1980s. The question the study set out to investigate is: does racial segregation still matter?
The answer, Reardon and his colleagues say, is yes. School systems that are more segregated have larger achievement gaps, and “their gaps grow faster during elementary and middle schools than in less segregated ones.” But it’s not because of race per se. The real problem, the researchers conclude, is poverty.
How did they figure that out? According to news reports, they zeroed in on three districts: New York City, Atlanta, and Detroit. (Detroit isn’t mentioned in the report itself.) In the first two, race and poverty are highly correlated—that is, poor students are overwhelmingly black and Hispanic, and wealthier students are almost all white. In Detroit, though, students of all races come from low-income families—and, unlike in New York and Atlanta, there was hardly any score gap between whites and students of color. That, perhaps, is the good news. The bad news is that scores in Detroit were low across the board.
The researchers also say it’s not the poverty of individual students that’s the problem—it’s the concentration of poor kids in particular schools. The researchers aren’t sure exactly why that’s a problem, because they don’t have good enough data. But, they say, “schools’ poverty rates, not differences in schools’ racial composition” are somehow linked to inequality.
You might expect this finding to lead to the conclusion that the way to address inequality is through socioeconomic rather than racial integration. But no. Instead the researchers argue that because race and poverty are so closely related, “the only way to close the gap is to racially integrate schools.” The reasoning here is hard to follow. It’s true that black and Hispanic families are disproportionately low-income in the United States, but why not aim for socioeconomic integration, if that’s the real problem? You’d probably end up with more or less the same result: schools that are both racially and socioeconomically integrated. But given that the basic problem is poverty—and that the Supreme Court has decreed that assigning students to schools on the basis of race is constitutionally prohibited—why focus on race?
In any event, the whole question is largely academic. Segregation and busing have been in the news a lot, but a return to mandated integration of any kind—which is what would be required to make it happen on a large scale—is highly unlikely.
That’s a pretty bleak picture, if you agree with Reardon that integration is our only hope of improving education for poor kids. “There’s a common argument these days that maybe we should stop worrying about segregation and just create high-quality schools everywhere,” he told one reporter. “This study shows that it doesn’t seem to be possible.”
But Reardon, like many other education researchers, is overlooking a fundamental problem: what is being taught and how, especially in elementary schools. What Reardon’s data doesn’t capture is the disconnect between the world of education and what scientists have discovered about how children learn. True, some kids do just fine in our education system—mainly those from higher-income families. But that may be despite rather than because of our prevailing curriculum and pedagogy.
High-poverty schools often face a host of challenges, including high numbers of kids who have experienced trauma that hasn’t been treated. Still, it’s possible to significantly raise the quality of the education they provide if we bring our teaching practices in line with cognitive science. That’s not to say, of course, that we shouldn’t try to reduce segregation. It just means we shouldn’t pin our hopes on something so difficult to achieve while overlooking other things that can work, right now—and in fact are beginning to produce results in the relative handful of high-poverty schools that are adopting a different approach.
The most egregious, and widespread, example of an educational practice that conflicts with science is the way American schools approach reading instruction. Many teachers haven’t received good training in teaching kids to sound out words, despite mountains of evidence on what works best. And nearly all have been trained to believe that the way to teach kids to understand what they read is to spend many hours every week on comprehension “skills and strategies”—things like “finding the main idea” of a text, or “making inferences”—and then have kids practice the “skills” on books on random topics that are easy enough for them to read on their own. But cognitive scientists have long known that the most important factor in reading comprehension isn’t generally applicable “skills,” it’s how much knowledge the reader has relating to the topic. The subjects that could build kids’ knowledge—history, science, the arts—are the very ones that have been cut to make room for comprehension instruction, especially in schools where test scores are low.
The result is a kind of intra-classroom tracking that begins in kindergarten. Children from less educated families, which also tend to be low-income families, are disproportionately likely to benefit from phonics instruction. They’re also more likely to lack the kind of knowledge and vocabulary needed to understand sophisticated texts. If they don’t get these things at school, they often remain permanently behind—regardless of whether they’re in a classroom with white or affluent peers.
What about the data showing benefits to black students from desegregation? Much of it comes from decades ago, before high-stakes tests and other factors made reading comprehension instruction the centerpiece of the elementary curriculum. Elementary schools may not have been providing a rigorous content-focused curriculum, especially in social studies, but at least students had a chance to acquire some academic knowledge and vocabulary. And white schools often had better textbooks and other amenities than black schools under segregation.
There’s some more recent evidence on the benefits of integration, but there’s also data pointing in the opposite direction. A 2017 study found “large differences” in the performance of wealthier and poorer students in the same schools, concluding that socioeconomic segregation is unlikely to reduce the test score gap. Similarly, a 2018 study of elementary schools in New York City found significant gaps in scores between students from different socioeconomic groups at the same schools. More anecdotally, there’s de facto tracking at many integrated high schools. Regular classes are populated by black and brown students, while honors classes are filled with white and Asian ones—a situation that has prompted some high schools to put everyone in honors classes in an attempt to further educational equity.
But if we really want to achieve that goal, we need to start questioning standard classroom practices that reinforce rather than reduce existing inequities, beginning in kindergarten. It’s fine to look at masses of test score data to document inequality, but if researchers want to understand how to address it, they’ll need to look elsewhere.
