Calculus as a Gatekeeper in Selective College Admissions?: Insights from a Lab-in-the-Field Experiment with Admissions Officers

• In Event: Test-Optional Admissions Policies and Other Trends in Selective College Admissions

Abstract

Access to calculus in high school is highly unequal across socioeconomic and racial lines. Nationwide, only 50% of public high schools offer calculus—and the percentage is even lower among schools with high Black and Latine student enrollment (U.S. Department of Education, 2018). Despite this disparity, calculus is widely perceived to carry significant weight in selective college admissions, particularly for STEM applicants. This raises pressing equity concerns: How do admissions officers evaluate otherwise high-achieving students who could not take calculus because it was not offered at their school? To address this, our study examines whether applicants who did not take calculus in high school (because it was not offered) are evaluated more negatively in terms of admissions recommendations (H1) and admissions scores (H2) compared to applicants who took calculus.

We also investigate whether Asian American applicants face greater penalties than comparable White applicants if they did not take calculus due to lack of course offerings. This hypothesis is grounded in the Shifting Standards Model (Biernat et al., 1994; Foschi, 2000), which posits that evaluators unconsciously adjust their standard of evaluation based on stereotypes associated with different groups, leading to different evaluations of similar qualifications. Prior research indicates that Asian Americans are stereotyped as being good at math (Shah, 2019). Consequently, Asian American applicants may be held to a higher standard for math coursework and penalized more severely for not meeting these expectations—even when such coursework was not available to them. We thus examine whether Asian American applicants who did not take calculus because it was not offered are evaluated more negatively than comparable White applicants in terms of both admissions recommendations (H3) and admissions scores (H4).

To test these hypotheses, we conduct a lab-in-the-field experiment with admissions officers at selective four-year institutions around the country. Each participant evaluates four simulated applications that vary on three dimensions: (1) whether the applicant submitted SAT scores, (2) whether the applicant took calculus or could not due to lack of availability, and (3) whether the applicant is White or Asian American. Applications are tier-adjusted to reflect each institution’s selectivity level. To confirm the comparability of our mock applications, we conducted a small pilot study with admissions officers, which suggested that the applications were balanced. This indicates that differences observed in the main study are likely due to our manipulated variables (SAT, calculus, race) rather than imbalances in applicant profiles. Because SAT submission is not the focus of this paper, its effects will be analyzed separately. For all four hypotheses, we will estimate mixed models with random intercepts for both admissions officers and institutions. This approach allows us to model unobserved heterogeneity at both levels as well as account for clustering in the data. As of April 10, we have collected data from 297 admissions officers; We anticipate reaching our target of 450 by July 2025, and will present complete findings at the time of the conference.

Authors

• BK Byun, University of Michigan Ann Arbor
  Presenting Author

• Jeremy Wright-Kim, University of Michigan Ann Arbor
  Non-Presenting Co-Author