School segregation has remained a hot-button political issue since Brown vs. Board of Education, a landmark 1954 decision of the U.S. Supreme Court in which the justices ruled that state laws establishing racial segregation in public schools are unconstitutional. New research from a Penn State College of Education professor sheds light on how school district leaders’ use of diversity-focused student assignment policies could disrupt the link between residential and school segregation.
“Yes, it can be harder in places that have high residential segregation (to desegregate schools) but there can be different strategies you can use to try to design policy to decouple the link between residential and school desegregation,” said Erica Frankenberg, PRI associate, professor of education, and director of the Center for Education and Civil Rights, which seeks to promote research-based actions that address the complicated nature of racial and ethnic inequality in the 21st century.
In a new article, “Student Assignment Policies and Racial and Income Segregation of Schools, School Attendance Zones and Neighborhoods,” Frankenberg and lead author Kendra Taylor, an independent researcher who received a doctorate from the College of Education’s Department of Education Policy Studies, examine the relationship between educational and residential segregation in three school districts that each has its own unique approach to student assignment. Using a relatively recently released federal data set, the School Attendance Boundary Survey, along with Census and Common Core of Data, their study examines racial and income segregation at the neighborhood, school zone and school levels in the districts to explore the relationship between districts’ diversity policies and school, attendance zone and residential segregation.
“If we saw more integrated schools than the neighborhoods would suggest, we wanted to understand why,” said Frankenberg. “We can try to look at the ‘why’ in ways that I think will hopefully inform for other districts what they might do if they wanted to create more integrated schools.”
According to Frankenberg and Taylor, school attendance zones are utilized by virtually every district that has more than one school for each grade level; and districts are mostly responsible for designing school boundaries. Historically, the most common paradigm has been the neighborhood school model, with students attending the school zoned for their neighborhood.
“School attendance zones therefore gain powerful social meaning and can themselves spur residential stratification that furthers both residential and school segregation,” Taylor and Frankenberg wrote.
To conduct their study, the researchers selected three large countywide school districts in the southern United States that are currently or have a history of pursuing racial or socioeconomic diversity: Jefferson County, Kentucky, race-conscious diversity policy; Wake County, North Carolina, socioeconomic diversity policy; and Charlotte-Mecklenburg, North Carolina, no current diversity policy (but formerly pursued racial diversity). All three counties have a predominant racial composition of Black/white students, Frankenberg said, with a growing share of students of color who are not Black in each.
According to the researchers, Jefferson County Public Schools (JCPS) has integrated socioeconomic factors into its race-conscious student assignment policy. Charlotte-Mecklenburg in 2009-10 largely assigned students to school by proximity, while at the time of data collection, Wake County drew attendance boundaries to minimize concentration of low-income students and had previously implemented racial desegregation policies.
In 2009-10, according to Frankenberg and Taylor, JCPS implemented a new student assignment policy to address concentration of JCPS’s large low-income population. The policy was a new iteration of its longtime integration-focused controlled choice policy, which allowed students to submit school preferences, and the district granted students’ choices according to their policy. The method that JCPS employed, Frankenberg said, was drawing six clusters designed to be racially and economically diverse in the entire county. The clusters consisted of subsets of elementary schools and a family got preference if they chose a school within the cluster. If the district had more people apply for a particular cluster than they had room for, the decision on whether to admit the student would be based on racial and socioeconomic characteristics of the student’s small neighborhood in order to achieve the district’s diversity goals without using the child’s racial status as a determining factor — which, as part of the district’s previous plan, had been ruled unconstitutional.
Like JCPS, Wake County also operated a race-based student assignment policy at one time, Frankenberg and Taylor wrote in their paper. At the time in which the researchers collected their data, Wake County Public School System, a countywide district including Raleigh, “operated one of the largest socioeconomic diversity policies in the country.” Due to growing concern about the legality of race-conscious assignment policies and because of research indicating the benefits of economically diverse schools, the district had switched to an assignment policy using boundary lines to de-concentrate low-income students in 2000.
“(The district) drew boundaries around schools including some that were noncontiguous to create attendance zones so that they didn’t have too many children from low-income families in the same school because of challenges that can mean for schools in terms of having higher achievement,” said Frankenberg. “As Wake County’s population boomed, this policy required a fair amount of redrawing boundaries to redistribute kids relatively equally in the district.”
Unlike the other two districts in the study, Charlotte-Mecklenburg had no diversity policy in place at the time of data collection, having ended its desegregation plan due to a court ruling in 2002. In 2009-10, the policy was essentially neighborhood school assignment through the use of attendance zones.
After analyzing the data, Frankenberg and Taylor found that in the context of similar residential segregation, educational segregation varies. In JCPS and Wake County Public School System, the two districts with diversity policies, educational segregation was lower than Charlotte-Mecklenburg Schools, which had no diversity policy in place. Even though Charlotte-Mecklenburg is less segregated residentially than the other two districts, Frankenberg said, it still had more segregated schools. Additionally, the link between attendance zone and school segregation was weakest in JCPS, the district operating a race-conscious controlled choice policy.
“This highlights that there can be a complex relationship between the different types of segregation that exist within a school district, including neighborhood, catchment and school, and that comparing segregation at these multiple scales can help us understand the dynamics of segregation in school districts and best target potential solutions,” the researchers wrote.
According to Frankenberg, the results of the study suggest that districts have different ways of forming equitable school attendance policies. One option is focusing on catchment — the area from which schools draw students. School boards typically determine catchments, she added, and Wake County, which drew catchments designed to be less segregated, had lower segregation in its school enrollment.
On the other hand, Frankenberg said, the school choice policy that JCPS implemented also demonstrated success. The district had high levels of segregation in its catchments but lower segregation in school enrollment due to the way students selected into schools.
One surprising result from the study, said Frankenberg, was the success that Wake County had in lowering racial segregation even though the district switched in 2000 from considering race to income in its desegregation plan.
“We were surprised that the way in which they drew catchments in regard to income also were effective with race,” she said.
One of the key takeaways of the study, Frankenberg said, is the importance of accurate data on school boundaries and attendance zones, especially since the federal government stopped collecting attendance zone data around 2015. The paper she wrote with Taylor “fueled a current National Science Foundation (NSF) grant to collect attendance zone boundaries so that we can have more information on these boundaries and can look at them over time to see how changing catchment zones may affect some of these relationships.”
In support of that effort, CECR has launched the Longitudinal School Attendance Boundaries Study (LSABS), a novel data collection effort that aims to collect longitudinal data on school attendance zone boundaries from as many school districts as possible across the country. Co-led by principal investigator Christopher Fowler, an associate professor of geography at Penn State, and Frankenberg, the project will lead to a database of attendance zone boundaries from 1990 to 2020, which will be made available to the public for research. The project has received funding from the NSF and Penn State.
“We need to support the development of tools like (LSABS) to support districts around the country that are trying to grapple with what they can do to desegregate schools,” said Frankenberg. “Federal and state governments and foundations like NSF can help support the collection of data on school attendance zone boundaries so districts can try to learn from each other.”