Does High School Size Matter for Risky Behaviors and Mental Health of Low-Income Teenagers? Evidence from New York City

Abstract

We evaluate the impacts of small high schools on youth risky behaviors and mental health in New York City, using rich student-level administrative and survey datasets and Medicaid claim data. We use distance between student residence and school to instrument for endogenous school enrollment, using a two-sample-instrumental-variable approach that combines enrollment information from the educational administrative data with information on outcomes from the Medicaid claim data. We find that girls enrolled in old small schools (opened before 2003) are less likely to get pregnant before age 18 or to be diagnosed with substance use disorders between age 15-18. Boys in old small schools are less likely to be diagnosed with mental and substance use disorders. By contrast, we find that enrollment in more recently opened small schools increases the likelihood of violence, for both girls and boys, and the likelihood of mental disorder diagnoses for girls. Utilization of contraception treatment and access to school-based health facilities do not explain the findings. A close look at school characteristics suggests that the higher per student expenditures and lower student-teacher ratios in older small schools may contribute to the finding that they improve health outcomes. Conversely, lower teacher quality and higher probability of co-location with large schools in newly-opened small schools may explain their relatively poor performance. These results indicate that improving a package of both schooling organization and inputs is likely more effective than reducing school size alone in addressing youth risky behaviors.