Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Shiferaw Gurmu

Second Advisor

Mesfin Bezuneh

Third Advisor

Rachana Bhatt

Fourth Advisor

Barry Hirsch


The essays in this dissertation explore the challenges of primary school attendance and the timing of enrollment in primary school in a typical developing country where child labor is widely practiced and poor households have limited access to school. The first essay assesses if when a child was born relative to his/her siblings affect whether the child attends school or participates in child labor. I investigate this question by estimating the effect of birth order on the probabilities of school attendance and child labor participation. Endogeneity of family size may bias the coefficient estimate of birth order since high birth order children are observed only in larger families, and parents who choose to have more kids may be inherently different and children in these families would have worse outcome regardless of family size and birth order. To address the endogeneity of family size, I use instrumental variable approach where the proportion of boys in the family is used to instrument family size. Using a longitudinal household survey data from Ethiopia, I estimate unobserved effect bivariate probit instrumental variable model of school attendance and child labor choices. The results suggest that the probability of child labor participation decreases with birth order, but I find no evidence that suggests birth order affects the probability of school attendance. However, among children who are going to school, hours spent studying increases with birth order.

The second essay offers empirical evidence on whether access to primary school induces children to enroll in primary school at the legal enrollment age using household survey data from Ethiopia. I exploit the variation in the intensity of the impact of the education reform across districts in Ethiopia to identify the effect of access to school on the timing of enrollment. Using prereform enrollment rate in primary school to measure the variation in the intensity of the impact of the reform, I estimate difference-in-differences models. The results suggest that the reform has substantially increased the probability the child enrolls in grade 1 by age 7. It is also found out that the reform has decreased age at enrollment in grade 1 by about 4 months. These estimates highlight an important role that access to school plays in inducing parents to enroll their kids in primary school at the legal enrollment age.