Who Chooses in Hartford? A Statistical Analysis of School Integration Lottery Applicants and Nonapplicants

• In Event: 70.066 - Roundtable Session 28
  In Roundtable Session: 70.066-15 - Investigating Public School Choice in Connecticut at State, City, and School Levels

Abstract

Which families were more (or less) likely to apply to the state-run lottery for racially integrated magnet schools and interdistrict transfer programs in the metropolitan Hartford, Connecticut region, and how do they vary by student characteristics & achievement, school composition, and neighborhood demographics? This study seeks to answer these questions based on student-level data provided by the Connecticut State Department of Education (CSDE), the Regional School Choice Office (RSCO), and the Hartford Public Schools.

Over the past two decades, the range of public school choices for Hartford students has increased dramatically through three different policy changes. After the Connecticut Supreme Court's Sheff v O'Neill school desegregation ruling in 1996, and the court-approved Sheff I (2003) and Sheff II (2007) remedies, the state legislature funded the growth of voluntary integration through interdistrict magnet schools (with curricular themes designed to attract both city and suburban students) and the Open Choice program (where city students enroll in suburban district schools, and vice versa). When combined with the growth of city-wide district choice and interdistrict charter schools, the parent of a typical Hartford 6th grader is eligible to apply to over 40 different district and interdistrict public schools or programs in the metropolitan Hartford region.

The first installment of our "Who Chooses in Hartford?" offers a statistical analysis of RSCO applicants versus non-applicants among 6,673 Hartford-resident students enrolled in Hartford Public Schools (HPS) — both district schools and interdistrict magnet schools — from grades 3 through 7 in Spring 2012. Overall, we found that participation in the RSCO application process was not random, but linked to student socioeconomic characteristics that often showed higher participation by more privileged families. In this sample, there were statistically significant lower levels of RSCO participation among English Language Learners and those with special needs, and generally higher levels by students with high CMT scores, and those who live in census areas with higher median household incomes and higher percentages of owner-occupied housing. The report also evaluates statistically significant differences and the magnitude of numbers of expected versus actual applicants by race and ethnicity, school performance, location, and other characteristics.

Our findings raised several policy considerations. If both the State of Connecticut and the Sheff school desegregation plaintiffs desire more equitable choice participation, they could establish benchmarks by student characteristics (e.g. ELL, SPED, Income), in addition to the traditional total goal (e.g. 41% Hartford minority students). In addition, possible strategies to achieve equity participation benchmarks include: holding school choice fairs and door-to-door outreach in neighborhoods with lower lottery participation, designing magnet themes schools (e.g. Dual Language) to attract under-represented students, and creating a weighted lottery to increase odds for students from Hartford neighborhoods with low participation rates.

Authors

• Jack Dougherty, Trinity College

• Stephen Spirou, Trinity College

• Diane Zannoni

• Marissa Block