Subgroup Reporting and School Segregation
An Unhappy Pairing in the No Child Left Behind Equation
One of the more prominent features of the federal No Child Left Behind Act is the requirement that schools and districts track the performance of subgroups of students. While the law identifies several subgroups, including low-income and English-language learners, the low performance of black and Hispanic students should be of particular concern. Ironically, because of the way the law is written, the schools and districts that could end up being most heavily penalized are those that are the most heavily integrated.
The impact of these new subgroup-reporting provisions will be felt most keenly where there is new information and new sanctions. While the provisions of the No Child Left Behind law will require some states to create accountability systems from scratch, prior to the federal legislation many states already publicly reported school performance, and many also imposed some type of sanctions on low-performing schools. So then, where will the new subgroup reporting yield new information? Only in integrated schools, which, of course, are the only ones to have racial subgroups. And where will there be new sanctions? Only in integrated schools that are performing adequately but have sizable subgroups that are not, or in integrated schools that are only occasionally performing below par, but have subgroups that are persistently performing inadequately. Unfortunately, there will be no new information based on subgroup reporting in segregated schools. New accountability for a “race gap” in performance is limited to a subset of integrated schools and districts. The law brings little additional pressure to bear on the significant gap in performance between schools that are racially segregated, at least in the early years before strict sanctions take place.
But this is far from being just a theoretical reporting problem. Our research in New York state reveals that the problem is large indeed. More than 40 percent of New York state schools with 4th grades served only white students in 2001-02, and over 20 percent served only nonwhite students, meaning that less than 35 percent of the state’s schools served both white and nonwhite students. Schools with 8th grades were not much different. Over a third of the state’s elementary and middle schools were essentially all-white in 4th and 8th grades and would be exempt from having any racial subgroup to report.
Making matters worse, we found that the biggest gaps in test scores are between schools that only serve white students and those that do not serve any white students. And well over half the 4th grades fell into these groups. In segregated white 4th grades, 74 percent of students passed the reading test; in segregated nonwhite schools, 40 percent of the students passed; in integrated 4th grades, 75 percent of white and 55 percent of nonwhite students passed. Strikingly, white performance did not vary much across types of schools, but nonwhite performance was much worse in all-nonwhite grades.
Will reporting at the district level remedy the problem? Not much. At the district level, nearly a quarter of schools with 4th grades and more than one-third of schools with 8th grades served segregated populations.
Moreover, new accountability for the performance of minority students will be unevenly distributed across districts, falling disproportionately on urban schools, and substantially ignoring rural and suburban schools. And whatever thresholds individual states use to determine whether there are “enough” students in the subgroup to provide statistically reliable subgroup-performance scores, schools—and perhaps districts—will face an incentive to stay below the threshold to avoid the increased pressure of No Child Left Behind subgroup accountability. The end result? Schools and school districts could actually benefit from segregation.
So what can be done? In our opinion, the glare of public reporting is a good tool to help solve the problem of racial test-score disparities. Since school and district subgroup reporting will not be entirely effective because of segregated schools and districts, individual states are going to need to step in. Perhaps high-performing schools could each adopt an all-nonwhite school for reporting and remedy. Or maybe states, or even the federal government itself, could be held accountable for racial test-score gaps.
Of course, there might be a far more basic solution: Work harder to integrate our schools.
Vol. 24, Issue 28, Page 31Published in Print: March 23, 2005, as Subgroup Reporting and School Segregation