Monday, November 4, 2019

We Need a More Productive Conversation About Charter Schools and Segregation

Editor's note: This blog series examines common narratives about charter schools, separates facts from fiction, and highlights legitimate issues that must be addressed. The full series is available here.


In 2009 the UCLA Civil Rights Project published a report accusing charter schools of resegregating public education. Although that charge was quickly and thoroughly refuted, critics of charter schools continue to say they deprive black children of opportunities because they are more segregated than traditional public schools. 

In this blog I look closely at these accusations, and at much sounder and more nuanced evidence that has emerged recently.

The UCLA report compared the percentage of black children in traditional public schools with the percentage attending charter schools. They found a big difference, as black children in charters attended schools that were more heavily segregated than did black children in traditional public schools. From this the authors claimed that charter schools are causing segregation. 

The facts are as the Civil Rights Project presented, but their statement of cause and effect is misleading. The authors used the oldest statistical fallacy in the book: comparing a group of charter schools— which, by design, work in particular neighborhoods—with all the schools in much larger and more diverse places (districts, metropolitan areas, states). Thus, in a city whose population is 30 percent black, charter schools in neighborhoods whose populations are 94 percent black are much more segregated than the average of all schools. That is true of all schools in such neighborhoods; charters are not the cause of segregation. 

Others have done a thorough job debunking these claims, but charter school critics continue citing the Civil Rights Project’s publications—and other analyses that suffer similar flaws—as if they were fresh and unsullied. Ironically, the charter schools most committed to creating quality schools in underserved communities are unfairly stigmatized.  Students who transfer from a district-run school to a charter school move from one school where the students are almost all black to another school with essentially the same racial composition. Those neighborhoods are not diverse enough for schools to be desegregated, absent much broader efforts to improve integration.

Nationally, racial isolation in schools is driven by racial isolation in neighborhoods, the cost of housing, and a transportation system that makes cross-town movement difficult. A new Washington Post report shows that segregation is decreasing nationwide but is still severe in the big cities.

The same year the initial Civil Rights Project report was released, a RAND study followed students transferring from traditional public schools to charter schools. These results differ by city: in one city, black students are moving to less segregated charter schools; in another, they move to schools with a 15 percent greater concentration of black students. However, across the cities studied, “the average increase in the African American concentration experienced by an African American transfer student was 3.8 percent.” 

In the first nationally representative study of charter schools’ contribution to segregation, Matt Chingos and colleagues at the Urban Institute come to similar conclusions in their study about school districts:

"[O]n average, an increase in the share of enrollment attending charter schools increases the segregation of black, Hispanic, and white students. Nonetheless, the magnitude of this average effect is small. Our preferred estimates suggest that a 1 percentage point increase in the segregation of black or Hispanic students. . . For the average district in our sample, this implies that eliminating charter schools would lead to a 5 percent decrease in the segregation of black or Hispanic students."

Clearly, charter schools are not solving the problem of school segregation, and in some places they are making it a bit worse. 

The fact that charter schools haven’t reduced segregation is disappointing to those who hoped charters, as public schools of choice, would attract more diverse groups of students than traditional neighborhood schools. But these results are driven by state incentives for charter schools to serve the most disadvantaged, and by current charter operators’ commitment to children whose neighborhood schools aren’t working for them. 

Putting aside questions of blame for current and past segregation, it’s fair to ask: Can charter schools play a more positive role in the future? This is a pertinent question for three reasons. First, because ongoing population changes in cities like New York City, Los Angeles, Washington, D.C., Philadelphia, Baltimore, Seattle, San Diego, New Orleans and Chicago mean there will be more white and Asian students living in neighborhoods that formerly consisted of all black or Hispanic residents. Second, increasing numbers of charter schools are being designed from scratch to attract and keep racially and economically diverse student populations. Third, as charter schools pioneer personalization and individualized pathways for students, they could attract and effectively serve a more diverse group of students than those attracted to the basic skills-focused charter schools that now predominate in underserved neighborhoods.

Good results are possible, but not inevitable. The Century Foundation’s recent report about diverse-by-design charter schools shows that only a minority of these schools have been able to attract the diverse population they sought. Of 181 charter schools rated highly committed to diversity, 89 still were rated “low diversity” in enrollment. Intentions matter, however, as “diverse by design” charter schools are nearly 10 times as likely as all charter schools to be desegregated. As Patrick Denise and Michael DeArmond found in their analysis of Kansas City, two such schools proved to be more segregated than the surrounding district because they were too successful in attracting white students, who are normally a small minority in the city’s schools. 

Creating integrated schools is difficult, even in localities where the population is varied enough to make desegregation numerically possible. And little if anything is understood about how an intentionally desegregated school comes about—whether by attracting just the “right” diversity of students from the start, or by building parents’ confidence that a diverse setting can work, before striving to meet numerical targets. 

Families—black, white, and Hispanic—can (and should) be skittish. In a gentrification scenario, unless the school options are highly attractive, advantaged families might go to private schools or flee to the suburbs when their children reach school age. Alternatively, populations of students of color might dwindle as schools and other neighborhood conditions change beyond recognition and no longer feel welcoming. 

In an earlier blog post I discussed gentrification as a possible opportunity for school integration. I posed, but could not answer, the big questions: Can district or charter schools attract and keep diverse students regardless of race, ethnicity, or income? How much can better parent information, fairer admissions, and more accommodating transportation systems increase school integration?

Much depends on whether schools are designed to support students who enroll with diverse levels of academic preparation, build organizational cultures that embrace students with diverse backgrounds, and provide all students with challenging programs that prepare them for college or a rewarding career pathway. 

And this depends on investment in school designs that support these goals, talent strategies to make the ranks of excellent teachers more closely match the backgrounds of their students, professional development that guides teachers and school leaders through the mid-course corrections that will inevitably be necessary, and replication of successful practices in other schools. 

This will require school systems to support innovation and adult learning—and new investments that neither government nor philanthropies are now making.

Wed, 11/20/2019

Our network participants are already showing the potential benefits of the shift toward local problem solving.

Mon, 11/18/2019

Settling for cherry-picked or incomplete evidence isn't necessary.