White Representation in Neighborhood Schools: School Funding, Nonprofit Investment, and Academic Outcomes

Abstract

My dissertation examines the enrollment patterns of White children in traditional U.S. public schools in 2010. I link schools to their attendance boundaries to compare the percentage of White children living in a catchment area to the percentage of White children who attend the local neighborhood school. I find that just under a third of schools are roughly representative of their catchment area (29%), the plurality are underrepresented White (40%), and the remaining 31% are overrepresented White. Descriptive analyses determine that White underrepresentation is more common in urban schools. White underrepresented schools tend to be in poorer neighborhoods and have a higher-than-average share of students in poverty and students with limited English proficiency. I investigate whether there is a connection between White representation and school quality outcomes. I focus on four facets of school quality that I hypothesize might be responsive to White representation: 1) school funding metrics, 2) school-supporting nonprofit presence, 3) standardized test scores, and 4) Gifted and Talented programming. Overall, the findings here offer mixed support for the theory of “opportunity hoarding,” in which White underrepresented schools receive fewer resources. Taken together, descriptive analyses find that White underrepresentation is largely associated with negative outcomes. White underrepresented schools have less public and charitable funding than their peers. White underrepresented schools are lower performing academically than White overrepresented schools, although they are not clearly academically different from representative White schools. White underrepresented schools are not necessarily less likely to have a GAT program, but when they do have a GAT program, it disproportionately targets White students. Furthermore, multivariate analyses reveal that the bivariate relationships between White representation and school outcomes are not entirely explained by the percentage of White students in a school, nor other covariates. This suggests that there is a meaningful distinction between White representation and the percentage of White students in a school. In other words, White representation tells us something about a school, net of the presence of White students. However, this was not the case for every multivariate model in the study. I find a significant negative association between White representation and school funding. White underrepresented schools have significantly lower mean teacher salaries and per-pupil salary expenditures, net of the percentage of White students within the school. This could be evidence that disproportionately low White enrollment leads to diminished school resources or less experienced teachers. Alternately, it could be that White families are more adept than non-White families at avoiding under-resourced schools. I find no evidence of a connection between White representation and whether a school has a school-supporting nonprofit. Instead, the economic composition of the school appears to be a more important driver of school nonprofit presence and nonprofit revenue. I also find no connection between White representation and test scores. However, White representation appears to influence the racial composition of GAT programs. Schools that are less White than their neighborhoods tend to have GAT programs that are significantly Whiter than the schools.