Abstracts

Essays on human capital formation policies

by Flavio Luiz Russo Riva

This thesis consists of three papers on human capital formation policies and their impact. The first chapter investigates how educational technologies that use different combinations of artificial and human intelligence are incorporated into classroom instruction, and how they affect learning. We conducted a field experiment to study two technologies that allow teachers to outsource grading and feedback tasks on writing practices. The first technology provides instantaneous grades and feedback using only artificial intelligence. The second uses human graders as an additional resource to enhance grading and feedback quality in aspects in which artificial intelligence arguably falls short. Both technologies significantly improved students’ essay scores, and the addition of human graders did not improve effectiveness in spite of increasing perceived feedback quality. Both technologies also similarly helped teachers engage more on personal discussions on essay quality with their students. The second chapter investigates the educational impacts of a policy-driven change in health services available to public elementary school students in Brazil. We study a nationwide program (Programa Saúde na Escola) designed to induce activities of primary health care professionals at schools and to refer children to other professionals of the public health care network when needed. Exploring variation in the timing of participation induced by rules that prioritized some municipalities first, we show that the program reduced retention and early withdrawal rates. The third chapter presents identification results for the probability of causation when there is sample selection. We show that the probability of causation is partially identified for individuals who are always observed regardless of treatment status and derive sharp bounds under three increasingly restrictive sets of assumptions. The first set imposes an exogenous treatment and a monotone sample selection mechanism. To tighten these bounds, the second set also imposes the monotone treatment response assumption, while the third set additionally imposes a stochastic dominance assumption. Finally, we use experimental data from the Colombian job training program Jóvenes en Acción to empirically illustrate our approach’s usefulness. We find that, among individuals who are always employed regardless of treatment, at least 12% and at most 19% transition to the formal labor market because of this training program.