Association for Public Policy Analysis & Management

Across the country, voluntary desegregation busing programs aim to ameliorate racial and socioeconomic imbalances and disparities. A longstanding Massachusetts program, METCO, serves as a national model for other desegregation programs throughout the country. Founded in 1966 in response to the contentious attempt to desegregate the Boston Public Schools. METCO buses non‐white K‐12 students from Boston and Springfield to 38 suburban districts that voluntarily enroll urban students. METCO grapples with issues that affect school systems across the country: how to improve access to high quality schools and reduce racial isolation and prejudice.

To improve understanding of the impact of integration and access to higher quality schools, this paper identifies the causal effect of attending a suburban, high performing school on the academic outcomes of minority students. Students and families who choose to apply for METCO program may be different than the average student. Therefore, to properly estimate the program's effect, we need to compare students who applied to METCO and randomly received an offer to enroll to those that applied, but did not receive offers.

Springfield runs a lottery for METCO admissions, so I compare students with random lottery offers to those without offers using two-stage least squares. In Boston, parents apply to METCO anytime after their child's birth and students are admitted on a first-come, first-served basis. Since children who applied slightly earlier than others are more likely to be admitted, but otherwise similar, I use enrollment offers as an instrument in a two-stage least squares analysis. Offers are made within grade, gender, and ethnicity categories, so I use separate offer instruments for each group.

I will estimate the effect on aspects of the students' experience: measures of school quality, peer characteristics, access to advanced coursework, teacher characteristics, likelihood of special education classification, time spent on the bus, and school climate. Intermediate and long-term outcomes including test scores, attendance, grade progression, suspension, perception of school climate and experience of bullying, advanced course taking, SAT and AP scores, high school graduation, college enrollment, persistence, and graduation. To understand mechanisms behind the effects, I will investigate how effects vary by demographic groups, district attended, time in program, and METCO entry grade and analyze which district, school and classroom characteristics are correlated with METCO students' success.

I merge over 28 years of historical METCO application and waitlist records with state administrative data. I have set up the data for analysis and matched the applicants to the state data. I will have a full set of findings by July, leaving plenty of time to refine before the APPAM conference.

This paper fits nicely into the conference theme since it analyzes the impact of increasing diversity in schools. It also employs quantitative and qualitative methods to understand the mechanisms behind the effects. The research contributes to the literature on the effects of court‐ordered and voluntary desegregation and to the literature on how school versus neighborhood quality impact student outcomes by isolating the impact of enrolling in a high‐performing suburban school while holding studentsʹ neighborhood fixed.