Microeconometric Essays in Education

Abstract

Chapter 1 addresses a long-standing problem in econometrics, that of identifying social interactions or peer effects. The estimation strategy is based on an estimator developed by Klein and Vella (2010) that uses information from the conditional distribution of error terms. This strategy exploits the excess variance in outcomes induced by social interactions and has the potential to identify the endogenous and exogenous channels. I demonstrate the proposed estimator using both simulated data and actual data from Project STAR, the classroom-size reduction experiment.

Chapter 2 evaluates the impacts of a conditional transfer program designed to increase primary school participation in Cambodia. The absolute impact of the program is estimated, as well as the relative performance of food and cash transfers. Results indicate that both transfer modalities decrease the probability of primary school dropout. No adverse impact from the receipt of cash transfers was observed.

Chapter 3 applies a structural econometric approach to the Cambodian conditional transfer program, developing and estimating a dynamic discrete choice model. With further development, this model will enable the simulation of a wide range of counterfactual policy alternatives. This work contributes to the underutilized set of tools of structural estimation, by providing further evidence of its complementarity with experimental program evaluation.