Abstract:

In Essay 1, we investigate the determinants of teacher attrition in five large metropolitan areas in upstate New York State. We focus on a teacher's decision to leave a school district or to leave teaching using the Prentice-Gloeckler-Meyer technique for proportional hazards with unobserved heterogeneity. We find that teachers in districts with higher salaries relative to non-teaching salaries in the same county are less likely to leave teaching, and that a teacher is less likely to change districts when he or she teaches in a district near the top of the teacher salary distribution in that county. We also find, however, that very large salary increases would be required to offset the impact of concentrated student disadvantage on the attrition of female teachers.

Starting in 1999, New York State implemented class size reduction policies targeted at early elementary grades. Due to funding limitations, most schools reduced class size in some grades and not others. In Essay 2, I use within school variation in class size induced by the policies to construct instrumental variable estimates of the effect of class size on teacher attrition. I also examine whether class size reduction affected the quality of teacher hires. Teachers with smaller classes were less likely to leave a school in some districts. However, class size reductions increase the share of teachers hired with no previous teaching experience.

In Essay 3, I investigate how physical and mental health effect student performance and whether accounting for health disparities changes the trajectory of the (regression adjusted) black-white test score gap as children age. I use data on a cohort of children in kindergarten, first grade, third grade, and fifth grade from the Early Childhood Longitudinal Study-Kindergarten Cohort (ECLS-K). I find that differences in health status partially explain the divergence of black and white test scores as students age. After including socioeconomic background variables, accounting for physical and mental health measures further reduces the black-white test score gap by eight to 31 percent. For third and fifth grade, the largest reduction in the test score gap occurs when student reported mental health measures are included.