Essays in education and health economics

Robert M Lantis, Purdue University

Abstract

This thesis is comprised of three independent chapters. The first chapter identifies the effect of peer ability on a student's effort choice and academic performance, while the second chapter identifies how the between and across effects of peer ability on academic performance varies. The third chapter identifies the effect that unemployment benefits have on individual's alcohol use and abuse following job loss. In the first chapter, I identify the effect that peer ability has not only on a student's academic performance but also on their effort choice as measured by hours of study; the latter of which is absent from the literature. The paper estimates these effects for exogenous peer group changes arising from school consolidations that create a set of "non-carryover" peers. We find the average effect of a change in the ability of non-carryover peers to be significantly positive on both a student's academic performance and hours studied per week. Unlike previous literature this paper emphasizes theoretically and empirically the how the relative ability position impacts the effect of peer ability. Consistent with the theoretical predictions the effect of peer ability on academic performance is found to be positive for students of any relative ability position, but greatest for those closest to their median peer, and decreasing as their relative ability moves away from this position. Also consistent with the theoretical predictions the effect of peer ability on hours studied is found to be negative for student's of the lowest relative ability, and greatest for those of the highest relative ability, which is a finding that is absent from the previous literature. In the second chapter, the effect of the ability of different racial groups on a student's academic performance, and how this varies with the student's own race is identified. Results show that a student's academic performance is affected more by same race peers as opposed to different race peers. Furthermore, this paper identifies how inter and intra race effects of peer ability are impacted by the racial composition of the school or peer group, which is not present in the current literature. Estimates show that that effect of peer ability from same race peers is greater when the student belongs to a racial group that is the minority of the student's school or peer group. Additionally, effects become even more pronounced for students belonging to a minority group in an environment with a strong majority. The results highlight the role race plays between peers, and suggests that instructors and administrators should be aware of the dynamics that exist between racial groups within the classroom. The final essay, written jointly with Brittany Teahan, investigates potential unintended consequences of unemployment insurance (UI) policy on alcohol use and abuse. Using NLSY data supplemented with Geocode data, we estimate the effect of benefit replacement rates on changes in individual alcohol consumption following job loss. Identification relies on variation in replacement rates across states and over time. We find evidence that income effects from increased benefits dominate potential stress reducing benefits of UI. Moreover, we find that increased benefits increase the likelihood an individual abuses alcohol following job loss. Individuals' responsiveness to changes in replacement rates varies based on drinking history. We find that individuals with no history of alcohol abuse are the most sensitive to changes in UI policy.

Degree

Ph.D.

Advisors

Mumford, Purdue University.

Subject Area

Educational sociology|Labor economics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS