I study role-model effects at scale by analyzing how brief presentations by STEM professionals influence high-school students’ STEM participation in college. I link data on over 1,500 female and male speakers at 183 STEM events to administrative records on the educational trajectories of all Swiss high-school graduates. Using an event-study design, I find that event exposure increases college STEM enrollment and leads to higher STEM graduation rates. I then exploit event-level variation in speaker composition to show that events with a higher share of female speakers have a stronger impact on both female and male students. This effect is largely explained by gender differences in presentation style rather than by a role-model mechanism: female speakers are more likely to give interactive presentations, which both female and male students respond positively to. Controlling for presentation style reduces the female-speaker effect by 51%. Thus, while female speakers do not narrow the STEM gender gap, the results suggest that selecting engaging speakers — who in this setting are more likely to be female — can boost students' overall STEM participation.

I analyze a policy reform that shifts public lunchtime childcare in kindergartens and elementary schools in the city of Zurich, Switzerland, from an opt-in to an opt-out system. In the new system, children are automatically signed up for low-cost lunchtime childcare unless parents actively deregister their child. Leveraging the staggered rollout in an event-study design, I document that the policy increases public childcare enrollment by on average 2.3 additional hours per week, compared to a baseline of 6.0 hours. The policy increases the employment rate of mothers with children in kindergarten by 5% and total earnings by 7.7%, and has a small positive effect on the extensive margin of fathers. The effects are particularly pronounced among low-income families and parents born abroad. Investigating the underlying mechanism, I document that parents react to the default component of the policy, suggesting that behavioral barriers play an important role in public childcare uptake.

I investigate how mentoring programs shape female students' educational choices. Existing evidence across studies suggests that female mentors are most likely to impact female students’ outcomes in contexts in which students have few other role models. Collaborating with the Swiss Academy of Engineering Sciences, this ongoing work investigates the impact of a seven-month mentoring program that pairs female students with female STEM professionals, focusing on two groups: students with and without STEM role models prior to the intervention. Exploiting oversubscription, I randomize access to the program and assess its impact on students’ college educational choices, tracked through administrative data, as well as on their knowledge and self-concept in STEM. In June 2024, a first cohort of 203 students completed the baseline survey and was randomized into the program. 

I study how the availability of high-school specialization tracks affects students’ participation in STEM in high school. Using Swiss administrative data and a fixed-effects strategy, I compare STEM enrollment of students who have different high-school tracks available at the high school that is closest to their place of residence, while controlling for a large set of fixed effects. My results show that while male students are significantly more likely to enroll in a STEM track when the track is available, STEM track availability is not correlated with female students' choices. Conversely, both male and female students are more likely to enroll in STEM when the language specialization, the most popular non-STEM track, is not offered. The findings suggest that the structure of students' choice sets can be an important tool for increasing female students' STEM participation.