Better Use of Information Could Help Agencies Identify Disparities and Address Racial Discrimination

In April, the Government Accountability Office (GAO) released a report examining recent trends in the racial and socioeconomic composition of America’s public schools. Between the 2000–01 and 2013–14 school years, the study finds, the fraction of U.S. schools that were both high-poverty (75 percent or more eligible for free or reduced-price lunch, or FRPL) and high-minority (75 percent or more African American or Hispanic students) rose from 9 to 16 percent.

While the GAO analysts caution that their analyses “should not be used to make conclusions about the presence or absence of unlawful discrimination,” to headline writers at the Washington Post, USA Today, and the Los Angeles Times, the findings suggest “resegregation” in American schools. The Post editorial board declared a “resurgence of resegregation.” But is this a fair interpretation?

There are at least two problems with drawing such a conclusion. The first is that the GAO analysis doesn’t take into account overall demographic trends. During this time period, student demographics were changing in America. As a share of the national student population, Hispanic students increased from 16 percent to 25 percent from 2000 to 2014 (though African American pupils remained virtually unchanged as a fraction of the population). Due to the increase in the Hispanic population, we shouldn’t be surprised that the percentage of high-minority schools increased over time as well.

A more refined gauge of whether schools are becoming increasingly segregated over time—or, in more neutral language, “increasingly racially isolated”—would examine how the demographic makeup of America’s schools compares to the national composition of students. The more “dissimilar” a school’s makeup is relative to the overall population, the more racially isolated it is—and vice versa. Applying this type of analysis in a recent issue of Education Next, economist Steve Rivkin found no evidence that school segregation increased from 2000 to 2012, at least for African American students.

The second problem is the GAO’s use of FRPL eligibility as a marker for poverty. That’s important because the fraction of pupils eligible for FRPL also noticeably increased during the study period—from 38 percent to 52 percent. The increase in FRPL eligibility could be partly (or even largely) attributable to the 2010 enactment of the Community Eligibility Provisions, a program that allows qualifying districts to deem non-low-income kids as FRPL-eligible. (Rising childhood poverty, or near-poverty, because of the recession and its aftermath could also be a contributor.) Given these trends, one might naturally expect to see more U.S. schools meeting the 75 percent thresholds that GAO uses to identify an economically isolated school.

In sum, Matt DiCarlo of the Shanker Institute asks about the GAO findings, “Does it really represent ‘resegregation?’ Not necessarily.” Almost everyone can agree that school segregation is an important but complicated matter. Let’s not reduce it to a headline based on a questionable interpretation of the data.

SOURCE: “Better Use of Information Could Help Agencies Identify Disparities and Address Racial Discrimination,” United States Government Accountability Office (April 2016).

Aaron Churchill
Aaron Churchill is the Ohio Research Director of the Thomas B. Fordham Institute.