| David Figlio
University of Florida
Education policy and school choice
FINAL REPORT:
Study 1: Are Private Schools Really Better? (Co-Author J.A. Stone)
Are private schools really better than public schools, or is it simply that better students attend private schools? To address this question, this study employs uniquely detailed local instruments and jointly models selection into religious and nonreligious private high schools, to measure the treatment effects of both types of private schools on test scores, as well as the probability of high schooi completion, completion of two years of college, college selectivity, and college major. Failing to correct adequately for selection leads to a systematic upward bias in the estimated treatment effect for religious schools (a bias that is exacerbated by using religion to instrument for sector selection), but a small downward bias for nonreligious private schools. With adequate correction, neither religious nor nonreligious private schools have a significantly positive treatment effect on either test scores or high school completion (the two outcomes explored in the literature before). However, private schools, and particularly nonreligious schools, increase the probability of staying on in college, and of attending a selective college. Our test score results, however, mask considerable heterogeneity in the treatment effect: students in large central cities, particularly African-Americans, fare much better in private schools than in public schools. Other factors that may make both religious and nonreligious private schools attractive include increased security and discipline, and greater opportunities for a variety of specialized school-day and extra-curricular activities.
Study 2: Can Public Policy Affect Private-School Cream Skimming? (Co-Author J.A. Stone)
We investigate how key school and community characteristics interact with the characteristics of individual students and families in determining the enrollment pattems in public and private schools. Using unique, nationally-representative, individual-level data, we find evidence that a number of factors plausibly influenced by public policy (e.g., public school student-teacher ratios, school district concentration, teacher unionization and local violent crime rates) have substantial effects on the composition of public and private schools, especially in communities with proportionately large minority populations.
Back to Funded Research Grants Page |