Search results

People rely on their experiences when making important decisions. In making these decisions, individuals may be significantly influenced by the timing of their experiences. Using administrative data, we study whether the order in which students are assigned courses affects the choice of college major. We use a natural experiment at the United States Military Academy in which students are randomly assigned to certain courses either during or after the semester in which they are required to select their college major. We find that when students are assigned to a course in the same semester as they select a major, they are over 100 percent more likely to choose a major that corresponds to that course. Despite low switching costs, approximately half of the effect persists through graduation. Our results demonstrate that the timing of when students are assigned courses has a large and persistent effect on college major choice. We explore several potential mechanisms for these results and find that students’ initial major choice best fits a framework we develop that incorporates salience and availability. Furthermore, our results suggest that once students select a major, they are less likely to switch majors than the standard model of economic choice predicts. Instead, students’ decision to remain in a major is more consistent with status quo bias.

The U.S. incarceration rate rose dramatically over the past 45 years, increasing the number of marriages and cohabiting unions disrupted by a jail or prison stay. But as some have pointed out, not all unions dissolve as a result of incarceration, and there seems to be racial–ethnic variation in this tendency, with Blacks displaying higher rates of dissolution than Whites and Hispanics. Yet it is unclear what explains racial–ethnic differences in union dissolution among the incarcerated. Drawing on the National Longitudinal Survey of Youth 1997 (NLSY97), we examine why racial–ethnic differences in union dissolution exist among a sample of individuals who had a marital or a cohabiting union interrupted by an incarceration spell. In doing so, we draw on social exchange theory and structural and cultural theories to suggest that racial–ethnic disparities in union dissolution are explained by differential exposure to protective relationship characteristics. The results of Cox hazard models reveal that Blacks have significantly higher hazards of union dissolution than do Whites and Hispanics. These results also indicate that being married, having a child together, having full‐time employment, a longer union duration, and a shorter incarceration spell may protect against dissolution and that these factors account, in part, for the greater risk of dissolution among Blacks relative to Whites and Hispanics.

This paper examines a training intervention aimed at boosting leadership and communication skills among employees of a large Latin American retailer. The identification exploits an experimental design in the context of a difference-in-difference strategy. Using longitudinal information obtained from the firm and two skills surveys, we document large positive effects of the training on store- and individual- level productivity. The intervention was more effective in boosting leadership than communication skills. Spillovers from trained managers to untrained sales representatives also contribute to the main effects. Our findings confirm the possibility of increasing productivity through training targeting critical soft-skills.

Amy L. Lewin, Kevin Roy, Family Science, individual and structural inequalities deriving from traumatic experiences among immigrant Latino youth

Predictive modeling methods from the field of machine learning have become a popular tool across various disciplines for exploring and analyzing diverse data. These methods often do not require specific prior knowledge about the functional form of the relationship under study and are able to adapt to complex non-linear and non-additive interrelations between the outcome and its predictors while focusing specifically on prediction performance. This modeling perspective is beginning to be adopted by survey researchers in order to adjust or improve various aspects of data collection and/or survey management. To facilitate this strand of research, this paper (1) provides an introduction to prominent tree-based machine learning methods, (2) reviews and discusses previous and (potential) prospective applications of tree-based supervised learning in survey research, and (3) exemplifies the usage of these techniques in the context of modeling and predicting nonresponse in panel surveys.

Kirsten Stoebenau and Sangeetha Madhavan examine impact of economic inequality through NICHD R03

College financial aid letters can provide a nudge toward borrowing, or not

When is Hope Enough? Hopefulness, Discrimination and Racial Disparities in Physiological Dysregulation

New National Institute on Minority Health and Health Disparities grant

Sonalde Desai and Reeve Vanneman study the "Determinants of Maternal and Child Health in India"