Author ORCID Identifier

https://orcid.org/0000-0002-7959-503X

Year of Publication

2021

Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation

College

Business and Economics

Department/School/Program

Economics

First Advisor

Dr. Kenneth Troske

Abstract

The social multiplier effect of peers has the potential to influence all members of a group through changing the outcomes of a single member and thus poses a strong, low-cost channel for increasing student performance. As such, peer effects are much studied in education and inform important policy decisions such as classroom composition, ability tracking, and magnet schools. These studies are often limited in scope to the contemporaneous period—how one’s peers affect them while they are around one another, but we should also be concerned with whether peer group continues to influence student outcomes after their inevitable separation. I use the National Longitudinal Study of Adolescent to Adult Health (Add Health) and its panel structure to further extend the study of peer effects to long run outcomes, focusing on employment status and earnings. Additionally, I contribute to this nascent branch of the literature by tightening my definition of peer to the individuals a student declared are friends. Using linear-in-means estimation, I find that peer participation in some risky behaviors has a persistent effect on outcomes whereas peer academic performance does not. Lastly, the effect of opposite gender peers has been a topic of interest for many years. It informs single-sex schooling decisions, classroom composition, and other education interventions. A common finding is that a greater share of female classmates increases academic achievement, but it is important to ask how female classmates are differentially affecting their peers. In other words, which characteristics and behaviors of peers affect achievement and how does the effect differ based on the peer’s gender? I use higher-order spatial autoregressive estimation (SAR) to analyze how the influence of friends’ characteristics and behaviors differs based on gender. I find that female friends’ and male friends’ peer effects do differ and the extent of which depends on the outcome analyzed.

Digital Object Identifier (DOI)

https://doi.org/10.13023/etd.2021.339

Share

COinS