Working Papers

Are Measured School Effects Just Sorting?

Identifying Causality in the National Education Longitudinal Survey

Abstract

Youth sharing a school and neighborhoods often have similar academic achievement. This correlation between neighborhood quality and youth achievement holds even when controlling for many observable features of a family. Nevertheless, the correlation is not entirely causal because families and youth are sorted into relatively homogeneous groups. Thus, the quality of the school or neighborhood in part acts as a proxy for hard-to-measure attributes of the family.

To enable separate estimation of sorting and school effects, we use the characteristics of the high school students will attend as an additional indicator of family background. When we compare youth who are at the same junior high school, the above measure is an appropriate instrument to identify family background separately from neighborhood and junior high school effects. Even after this correction, the point estimate of school effects on student achievement remains large and is statistically significant.