Theses Doctoral

School Closure in New York City

Silander, Megan Reilly

School districts and states have increasingly abandoned traditional school reform efforts in favor of simply closing low-performing schools. This movement reflects growing frustration among policymakers with the disappointing effects of previous school improvement policies, and the view that some schools may simply lack the capacity to undertake meaningful improvements. This paper focuses on arguably the most aggressive school closure policies in the nation--those in New York City. Over the past decade, New York City has closed over 100 schools. Using a longitudinal database of students and schools, I explore the implementation and effects of closure and reconstitution of middle schools in New York City, and assess the links between school closure and student academic development and behavior. My descriptive findings indicate that schools selected for closure have significantly lower school-average state test score exams and lower attendance rates compared to other middle schools for several years prior to closure, and that students who attend these schools are almost exclusively Hispanic and Black, more likely to come from low-income families, and more mobile than other middle school students in the district. I also find that students enter these middle schools already at a significant academic disadvantage. I examine characteristics of the reconstituted schools that replace the closed schools, and find that in terms of demographics, reconstituted schools enroll students similar to those served by the closed schools that they replaced. However, the reconstituted schools serve higher performing students with fewer absences and tardies in the year prior to enrolling in middle school. To assess the impact of school closure on student academic outcomes, I use propensity-score matching within a difference-in-differences framework. I find a small, positive effect of school closure on student test scores and rates of absences. As a robustness check, I conduct a second set of analyses using student fixed-effects models that produced similar results: students learn slightly less at chronically underperforming schools, compared to what would have happened had they attended an alternate school. School closure appears to be a somewhat effective in improving student academic outcomes. It is not clear, however, whether the policy is efficient given the small effects and the considerable disruption associated with the policy. Future research should examine the fiscal costs associated with closure, compared to costs of other policies with similar effects.

Geographic Areas


  • thumnail for Silander_columbia_0054D_11005.pdf Silander_columbia_0054D_11005.pdf application/pdf 1.29 MB Download File

More About This Work

Academic Units
Education Leadership
Thesis Advisors
Ready, Douglas David
Ph.D., Columbia University
Published Here
June 28, 2013