To estimate peer effects in college achievement we exploit a unique dataset in which individuals have been exogenously assigned to peer groups of about 30 students with whom they are required to spend the majority of their time interacting. This feature enables us to estimate peer effects that...

Understanding whether racial attitudes are malleable is critical for addressing the underlying causes of racial discrimination. We examine whether white males' stated attitudes and behavior toward African Americans change based on the number and type of black peers to whom they are exposed. To...

Why aren't there more women in science? Female college students are currently 37 percent less likely than males to obtain a bachelor's degree in science, technology, engineering, and math (STEM), and comprise only 25 percent of the STEM workforce. This paper begins to shed light on this issue by...

It is difficult to measure teaching quality at the postsecondary level because students typically "self-select" their coursework and their professors. Despite this, student evaluations of professors are widely used in faculty promotion and tenure decisions. We exploit the random assignment of...

This paper examines the effect of alcohol consumption on student achievement. To do so, we exploit the discontinuity in drinking at age 21 at a college in which the minimum legal drinking age is strictly enforced. We find that drinking causes significant reductions in academic performance,...

We take cohorts of entering freshmen at the United States Air Force Academy and assign half to peer groups with the goal of maximizing the academic performance of the lowest ability students. Our assignment algorithm uses peer effects estimates from the observational data. We find a negative and...

The increase in obesity over the past thirty years has led researchers to investigate the role of social networks as a contributing factor. However, several challenges make it difficult to demonstrate a causal link between friends' physical fitness and own fitness using observational data. To...

We use data from a natural experiment in which retirement benefits were reduced by congressional legislation and later restored to estimate the effect of future retirement benefit eligibility upon the decision of whether to remain in the U.S. military. We find that the generosity of retirement...

It is estimated that between ten and twenty percent of children in the United States are exposed to domestic violence annually. While much is known about the impact of domestic violence and other family problems on children within the home, little is known regarding the extent to which these...

We present evidence from a recent field experiment in college coaching/ mentoring. We find surprisingly large impacts on college attendance and persistence. We test several theories as to why a short lived intervention has large impacts on lifetime human capital investments. We do not find...