More than 60 years ago, the courts deemed school segregation unconstitutional, yet across the country many students still attend school primarily with students who look like them.
Research shows that the racial composition of the public school student population has changed substantially over the past 25 years, but student racial sorting among schools has remained relatively stable. A growing body of research shows that school segregation matters for the educational and socioeconomic outcomes of students of color. To fix it, however, we have to understand why racial segregation has persisted.
School segregation is a complex problem, rooted in history, structural racism, school assignment policies, and parental behavior. In this piece, we focus on one lever that captures many of those factors: residential segregation.
The chart above compares the residential segregation of school-age kids in a city with the segregation at public and most private schools in that same area in 2015. For the purposes of this piece, we consider the segregation of black or Hispanic students from their peers from other racial and ethnic groups.
Our segregation measure ranges from zero to one, with zero representing a perfectly integrated city in which each neighborhood or school is representative of the racial composition of the city or school system and 1 corresponding to a perfectly segregated world in which there is no interaction across racial groups at all. This index measures how far away the system is from being perfectly segregated, in proportional terms. A segregation level equal to 0.3, for instance, means the city is one-third as segregated as the maximum it could be given its racial composition. If residential integration were the only determinant of school integration, we’d expect cities to fall right on the 45 degree line. Where does your city land?
The role of residential segregation
School choice programs have become more common recently, but by and large kids attend school in the neighborhood in which they live. This means that school integration depends significantly on neighborhood integration.
You can see this positive relationship in the chart, for every percentage point increase in neighborhood segregation, school segregation increases 1.04 points on average. But the fact that most cities fall above or below the 45 degree line says that neighborhood integration is not the only factor. In fact, we estimate that neighborhood segregation—rooted in a long history of racism and discrimination—explains about 76 percent of the variation in school segregation across cities.
What explains the rest? For each city, it’s a different combination of district enrollment policy, parental preferences, and other factors we can’t measure. Though we don’t have to the data to say definitively which one of these peripheral factors most affects school segregation, we can look at examples to begin to understand how policies, choice, and residential integration interact.
Similarly integrated schools, very different stories
Schools in San Francisco and Charlotte had similar levels of integration in 2015, but neighborhoods did not. San Francisco, in fact, was more integrated residentially—so why weren’t its schools more integrated, as well?
In the 1970s, San Francisco Unified School District made an effort to address school segregation by introducing desegregation busing. This program slowly chipped away at segregation, until the state got rid of all affirmative action programs in public education in 1996. In 2001, the district introduced a school choice system allowing parents to apply to any school. Students were assigned to schools by a lottery process that included a “diversity index” based on socioeconomic status and achievement. This was intended to ensure diversity in schools without considering race. But in 2010, the diversity index was dropped, giving priority to parental choice and school attendance boundaries, and school segregation has been on the rise since. (Our data measures segregation for the entire San Francisco metropolitan statistical area, so SF Unified School District’s policy only partly explains our results.)
Charlotte, meanwhile, has a different story. Like San Francisco, Charlotte has a history of prioritizing desegregation. In 1971, Charlotte-Mecklenburg school district was a defendant in a key Supreme Court case regarding school desegregation. In that case, the court held that busing was an appropriate remedy to school segregation, and Charlotte introduced a highly influential desegregation busing plan that was imitated by school districts around the country.
But a 1999 lawsuit brought an end to busing in 2002, and rather than move toward school choice, as in San Francisco, Charlotte, at the order of the court, redrew school attendance boundaries based exclusively on neighborhood. Thus, Charlotte schools are almost exactly as integrated as its neighborhoods.
Highly segregated neighborhoods and attempts at school integration
In both Springfield, Illinois, and Newark, New Jersey, neighborhoods are severely segregated. Though Springfield has used district policy to successfully integrate schools, relatively speaking, Newark has in fact made segregation worse.
Springfield resembles the archetypal American city: minorities live in the city center, surrounded by a predominantly white suburban area. In 1976, a lawsuit brought to federal court resulted in a mandated desegregation plan. Springfield agreed to redraw school boundaries every four to five years to maintain racial balance in schools, even as neighborhoods changed. That plan is still in place, and the results are clear when you compare neighborhood integration with school integration in Springfield.
Newark, which has similarly segregated neighborhoods, though is a much larger city than Springfield, has not had similar success. Newark is one of the most segregated cities in the country, a result of decades of redlining and other policies that discriminated housing access based on race. It is also one of the most historically poor performing school districts.
Newark’s school enrollment policy combines school choice and neighborhood attendance zones. If students live nearby or have a sibling in a school they are guaranteed a seat in that school, otherwise they rank schools by preference and enter a lottery. The neighborhood-based zones replicate the high levels of neighborhood segregation, and families’ choices do not appear to mitigate this segregation.
Our estimates show that school segregation has been on the decline in Newark since the early 2000s, however. Though we can’t say exactly why, we know that charter schools have been steadily growing as a share of enrollment since 2002. Additionally, in 2010, a large donation prompted the district to close poor performing schools, shift enrollment patterns, and introduce universal school choice for both district and charter schools. Which of these changes spurred the decline in school segregation should be examined in future research.
How can we address school segregation?
Without remedying residential segregation—likely a very difficult and expensive task—school segregation is unlikely to ever go away.
But the examples and data above offer some hope. We see that though residential integration is an important driver of school integration, it is not the only one.
District enrollment policy can promote school integration, as seen in Springfield. Thoughtful school zoning can also ensure that schools are at least no more segregated than neighborhoods. School choice programs, meanwhile, have coincided with increased segregation in some cities, such as San Francisco, and increased integration in others, like Newark. We need more research to understand exactly to what extent residential choice and school choice affect school integration. We also need to understand why the programs that work are successful and what it means for students’ outcomes.
School segregation is not a problem we can solve in isolation, but better understanding the mechanisms that drive it will help to chart the path forward.
This feature was funded by the Overdeck Family Foundation, with additional support from the Walton Family Foundation for Urban’s research on school segregation. We are grateful to them and to all our funders, who make it possible for Urban to advance its mission. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Funders do not determine our research findings or the insights and recommendations of our experts. More information on our funding principles is available here. Read our terms of service here.
Chart notes updated to clarify the meaning of the city dot sizes (updated 11/28/2018).
RESEARCH: Tomas Monarrez
DESIGN: Christina Baird
DEVELOPMENT: Dan Wood
ILLUSTRATION: Allison Feldman
To read the methodology, click here.
Copyright © November 2018. Urban Institute.