September 14, 2021
Before John R. Lewis Elementary School opened its new permanent campus in fall 2019, officials for the Dekalb County School District in Atlanta needed to redraw attendance zones to assign students to the school. To do so, they used criteria previously set by the school board, including geography, school capacity, and public feedback, but under no circumstances could they consider race or ethnicity, income, or student performance.
District officials omitted race, ethnicity, and socioeconomic status from zoning criteria to prevent “bad actors” from using those data to create segregated schools, according to Dan Drake, the district’s former chief operating officer. But without those data, the district officials also excused themselves from having to create more integrated schools. And the final boundary for John R. Lewis Elementary School, named for the iconic civil rights leader, shows how ignoring race and ethnicity can reinforce segregation.
The John Lewis boundary borders five other elementary school attendance zones, four of which serve majority-Black or majority-Hispanic student bodies like John Lewis. But the fifth boundary, which narrowly snakes along a highway, separates John Lewis from the nearby Ashford Park Elementary School, which serves a majority-white student body.
That boundary dividing John Lewis from Ashford Park is one of the most racially segregating school boundaries nationwide, according to an Urban Institute analysis.
Unequal school boundaries are those that have stark demographic differences in the populations living on either side, resulting in similarly stark differences in enrollment demographics between neighboring schools. These invisible lines that segregate school systems at a hyperlocal level can prevent Black and Hispanic students, who have been historically underserved by the US education system, from accessing the same educational resources and opportunities as their white peers. With just small changes to these boundaries, district officials could create more integrated schools and potentially increase opportunities for Black and Hispanic students.
Before John Lewis opened, Woodward and Montclair Elementary Schools were overcrowded. Ashford Park was not.
To alleviate the overcrowding in the two predominantly Black and Hispanic districts, Dekalb County officials purchased land in the Ashford Park zone, which housed predominantly white residents.
The map shows the race and ethnicity of all residents 17 or younger.
Rezoning drew more than 200 students from Woodward and Montclair to John Lewis but drew only 22 from Ashford Park.
The new boundary between the two schools runs along the John Lewis campus, dividing the white families to the north from the Black and Hispanic families to the south.
For neighborhoods along the boundary itself, the starkness of this dividing line remains apparent. These households share parks, transit stops, and stores, with the public school boundary the main difference between them.
Black or Hispanic residents within 500 meters of the boundary
Seventy-nine percent of people living within 500 meters of the boundary on the John Lewis side are Black or Hispanic. On the Ashford Park side, that share falls to 13 percent.
Each bubble represents a census block.
By prioritizing existing boundaries and neighborhoods rather than race, ethnicity, or socioeconomics in zoning criteria, school districts can create attendance boundaries that reinforce segregation patterns stemming from decades of discriminatory policies. The resulting school segregation can have lasting effects. Research shows that education inequity can negatively affect test scores, graduation rates, and even adult income and life expectancy for students of color.
These data highlight attendance boundaries separating two schools in which there is at least a 25 percentage-point difference in the representation of Black or Hispanic students and residents and in the representation of white residents. Simply put, for boundaries to meet our definition, the share of Black and Hispanic students must vary substantially between the two schools, and the shares of Black, Hispanic, and white residents must vary substantially both near the boundary and within the entire school zone. Many of these boundaries, like the one between John Lewis and Ashford Park, are within one district, so these data can help district officials who care about integration understand where they can make small changes to produce large effects. Other boundaries are jurisdictional, crossing multiple school districts. In those cases, local and state officials could use these data to reconsider school systems at a macro level or better target underlying housing policies that have created segregated communities.
This tool highlights racially unequal school boundaries* nationwide. The map shows the race and ethnicity of residents 17 or younger in cities with unequal school boundaries, and the charts display school pairs in ranked order, starting with the most unequal boundary in each metro area, and indicate whether a boundary separates schools within one school district or across two districts.
Share of residents within 500 meters of boundary
Swipe and click to select a different boundary
Race and ethnicity of residents 17 or younger: Atlanta-Sandy Springs-Marietta, GA
Two or more races
*We define racially unequal school boundaries as those that separate schools with a 25 percentage-point difference in representation of Black or Hispanic residents and students and in the representation of white residents.
*We define racially unequal school boundaries as those that separate schools with a 25 percentage-point difference in representation of Black or Hispanic residents and in the representation of white residents.
For school officials and policymakers invested in integrating public schools, rezoning can offer an avenue to address the most unequal boundaries, but the inextricable links between school and neighborhood segregation can pose a seemingly insurmountable barrier. Hans Williams, the director of planning for the Dekalb County School District, believes school zoning is primarily a geographic process, and until that changes, “school boundaries will only be as diverse as the communities they’re serving.”
Although addressing housing segregation is necessary to create integrated neighborhoods and schools, policymakers and school officials have the tools to work around community segregation to address school segregation through rezoning. In places with these unequal boundaries, school district officials can make small changes without disrupting an entire school system and, as a result, can make a big difference in students’ lives.
Of course, sometimes it might be necessary to disrupt—or at least reevaluate—an entire school system. Considering demographic data in this process could remedy unequal boundaries and create more integrated school districts.
Over the past few years, the Minneapolis Public Schools (MPS) district has undertaken a larger systemwide redesign to integrate schools. To do so, officials commissioned a district analysis (PDF), receiving feedback from more than 10,000 families. Like all rezoning decisions, MPS considered facility capacity, enrollment trends, transportation barriers, and other factors, but it prioritized increasing equity.
“The first thing you have to do is acknowledge that our systems are racist,” said Eric Moore, MPS’s senior accountability, equity, and research officer. “The achievement gap is just a failure of our school systems being able to address the needs of students of color and low-income students, so the work has to be grounded in reality.”
As a result of its analysis, MPS redesigned its attendance boundaries and its school choice system, investing more into the community schools where students are zoned. Through the redesign, MPS expects to reduce its number of racially identifiable schools (defined as schools where more than 86 percent of students are students of color) from 21 schools to 10.
“We were backwards. We were investing in choice, and a lot of parents would leave schools that they didn’t feel met their needs,” said Moore. “You can’t just change boundaries. You also have to show folks that you’re taking those savings and investing in better schools.”
The Minneapolis redesign did not receive significant pushback from families, according to Moore, but rezoning plans often do face negative feedback, especially from those who have benefited from the system. Some parents say they’re worried that rezoning could lead to decreased home values, even though there is little evidence of real estate depreciation during localized school rezoning, and others point to fears of longer bus rides and broken-up neighborhoods.
These concerns—often rooted in racism and classism—are compounded by broader systemic inequities in education, housing, and social services. Families, both privileged and not, have learned to navigate a broken system to create the best situation for themselves. As a result, they have little faith that the system can fix itself, according to Moore.
By using rezoning to address unequal boundaries, school officials will likely invite negative feedback from families who have grown accustomed to the existing system. But rezoning, whether as part of a larger redesign or for a single school boundary pair, offers school districts an effective tool to address the most unequal boundaries.
Recently, Dekalb County has undertaken its own systemwide analysis to look at facilities across the district to see what capacity and rezoning needs may arise over the next 10 years. In terms of reducing segregation in Dekalb County schools, Williams said the comprehensive analysis is “a really good next step.”
But redrawing school boundaries will not be part of the analysis, and if any rezoning needs arise, boundary changes will have to adhere to the rezoning criteria. And that means, under current district rules, they can’t consider race.