| This dissertation is comprised of three empirical essays on the economics of education. The first two chapters analyze the effect of schools on student outcomes. In Chapter 1, I analyze a program that was implemented in schools serving underprivileged populations in Texas that pays both students and teachers for passing grades on Advanced Placement (AP) examinations. Using a Difference in Differences (DID) identification strategy, I find that adoption of the program is associated with large increases in AP participation for minority students, improvements in SAT performance and increases in college matriculation. The per-student costs of the program are small relative to the implied lifetime benefits that accrue to affected students, and the program may narrow educational achievement gaps between whites and ethnic minorities at affected schools.; The streaming of students by ability across secondary schools in Trinidad and Tobago generates large differences in school quality. In Chapter 2, I estimate the effects of school quality, as measured by historical examination pass rates, on student examination performance. I use the student assignment rules to implement a "fuzzy" regression discontinuity (RD) design and to construct simulated instruments to remove the bias that occurs when students of higher unobservable ability or motivation systematically attend "better" schools. While there is evidence that "elite" secondary schools improve student exam performance, attending a low-performing secondary school is strongly related to worse academic outcomes.; In Chapter 3, I analyze how students affect the schooling environment. In 2001 Charlotte-Mecklenburg school district terminated its long standing race-based school busing policy. I use the sudden changes in schools' student demographics due to the policy change, that were uncorrelated with changes in neighborhood and school characteristics to make inferences about teachers' preferences for student characteristics. Using a simulated instrumental variables approach, I find that schools that experienced an inflow of black students gained black teachers and lost experienced white teachers. This in turn, led to increased teacher turnover, and a lower teacher qualifications. The results suggest that teacher quality is endogenous to student demographics. |
