Research
Gender Peer Effects in Post-Secondary Vocational Education (Job Market Paper)
This paper presents evidence that women benefit from having a higher percentage of female peers in post-secondary vocational STEM programs. I use idiosyncratic variation in gender composition across cohorts within majors within branches (campuses) for identification. A 10 percentage point increase in the proportion of women in a STEM major cohort has a statistically significant positive effect on female students. It decreases female dropout rates by 9.6% and increases GPA by 0.05 standard deviations. The evidence suggests peer effects are mediated by the gender of the instructors: as female students have fewer female instructors, the effect of having more female peers intensifies.
Gender segregation on college major choice remains high, with adverse effects for women and possible inefficiencies in allocating students to majors where they have comparative advantages. A partial explanation of the college major gender gap is self-confidence biases in some areas of study. In this paper, I evaluate an intervention that informed outstanding students in STEM about their relative performance. The results show that women and men were affected differently by the intervention, and those good-performing women that are not at the top of the performance distribution increase their probability of applying to any major. The increase in application probability comes from an increase in health applications. Men in this group decrease their chances of applying to any major.
Personalized Information Provision: Does Gender Play a Role?
(With Christopher Neilson)
This paper studies the effect of providing information on major choice and college enrollment, and it analyzes if there is a difference in the effect between women and men. According to our results, information provision increases STEM enrollment, incentivizing women's enrollment in STEM. On the other hand, information provision discourages men from enrolling in humanities \footnote{For the sake of brevity, I refer to the social sciences and the Humanities (Arts, Architecture, Humanities, Law) as Humanities.} majors. The fact that the effect is different for STEM and the humanities , and that the STEM effect is significant only for women and the humanities effect is significant only for men suggest that information provision is acting through different channels depending on students' gender.
(With Andres Barrios-Fernandez and Josefina Eluchans-Errazuriz)
Using a randomized controlled trial (RCT), this project aims to assess whether the provision of targeted information and mentoring to students attending vocational high schools affect the quantity and type of post-secondary studies they pursue and their higher education aspirations. The RCT randomly assigned 80 high schools to the control group, 80 high schools to an information-only treatment group, and 80 high schools to a combination of information and mentoring treatment group. The mentoring treatment is randomly assigned to only a few students per class, a feature that will allow us to study the spillovers of the mentoring program on the social network of treated individuals. With this last part of the experiment, we aim to understand to which extent social spillovers could be used to design more efficient and effective interventions to help students to make informed decisions about their post-secondary education trajectories.
Textbook Gender Bias and Educational Gender Gap
This paper tests the hypothesis that a policy aimed at lowering gender bias in textbooks will reduce the achievement gap between boys and girls in math. Using a dataset that contains information for all students in 10 cohorts of the Chilean school system, I exploit the variation in implementation timing of a policy aimed to reduce textbook gender bias to identify its effects in the classroom gender gap, using reading and math standardized test scores. Using a difference-in-difference approach to estimate the effect, my results suggest that a policy aimed to lower textbook gender bias is associated with a decrease in the reading gender gap.
Do Cohort Peer Diversity Matters and For Whom? Evidence from Chile on Students’ Academic Outcomes
(With Francisco Lagos)
The effect of peers on short term human capital formation has been largely studied in the education literature. However, the evidence in this context is still mixed, depending on the definition and operationalization of peer effects. Using national longitudinal data from Chile, in this study, we aim to contribute to this literature by focusing on the effects of peer socioeconomic diversity (rather than peer average socioeconomic status) as a function of the student's own socioeconomic status. Preliminary results show that the effect of peers' diversity varies depending on the student's own socioeconomic status. Results suggest that more disadvantaged students benefit from having a socioeconomically diverse group of peers.