This dissertation's essays exploit longitudinal data sets to provide evidence on education economics topics of school choice, social-emotional learning curriculum, and teacher hiring.

Chapter 1 estimates the causal effect of full-time virtual school attendance on student outcomes. I use a longitudinal data set composed of individual-level information on all public-school students and teachers throughout Georgia from 2007 to 2016 and implement individual fixed effect and semi-parametric cell analysis to investigate how attending virtual schools influences student outcomes. I find that attending a virtual school leads to a reduction in English Language Arts, Mathematics, Science, and Social Studies achievement test scores for students in elementary and middle school. I also find that ever attending a virtual school is associated with a 10-percentage point reduction in the probability of ever graduating from high school.

Chapter 2 examines the impact of Social-Emotional Learning (SEL) curriculum on student achievement over a three-year period in an urban district. I use a longitudinal data set composed of individual-level information of students and teacher. I implement a staggered difference-in-difference approach to estimate the causal effect of implementing SEL program on student outcomes. I find that the program does not impact attendance, discipline, nor test scores across the elementary and middle school grades. For high school students, the program leads to a reduction of the number of incidents, an increase in attendance, and no statistical impact on end-of-course exams nor on graduation.

Chapter 3 evaluates the predictive power of the non-cognitive traits measured in TeacherInsightTM (TI) testing tool in comparison to other measures of prospective teachers' abilities, like educational credentials, and certifications. I implement regression analysis to evaluate the relationship between teachers' non-cognitive skills (TI score), value-added test score, and observational score. I find that the Teacher Insight Score does not do a good job at predicting which teachers will be effective as measured by the teacher's value-added score. In contrast, the Teacher Insight Score and the observational score have a positive relationship. More specifically, a one-point increase in Teacher Insight score is associated with a .04 increase in teacher observation score.