Effectiveness of Quota Policies Across STEM, Biological, and Humanities Programs

TL;DR

This study analyzes more than ten years of Unesp quota policy data to assess how admission tracks (SU, EP, PPI) relate to academic outcomes in Physics, Biology, and Pedagogy. The authors distinguish Humanities from STEM and rely on Calculus I as a key gateway course to obtain sufficient statistics, applying Poisson-based uncertainties with a 3% systematic error and analyzing grade distributions and repeat attempts. They find that track effects are field-dependent: negligible in Physics, visible in Biology where SU/EP outperform PPI, and minimal in Pedagogy due to high overall pass rates; combining Calculus I data across STEM programs strengthens the SU>EP>PPI hierarchy. The results also show that higher entrance scores predict higher course success, and that the probability of passing declines with the number of attempts, signaling persistent early gaps that education alone does not readily close.

Abstract

We examine more than a decade of quota policy at Unesp, analyzing Physics, Biology, and Pedagogy as representative programs of distinct assessment styles. Quotas show little impact in Physics, where the admission barrier is low, and in Pedagogy, where high pass rates make it difficult to differentiate students, but they reveal systematic differences in Biology. Focusing the analysis on Calculus I - an introductory course in Physics and other Science, Technology, Engineering, and Mathematics (STEM) programs - for which much larger statistics are available, a clear hierarchy emerges: students admitted through open competition perform best, those from public schools achieve intermediate results, and students from racial quotas perform worst. When students are divided directly by the admittance exam grade, the performance difference is even clearer. Statistical analysis also shows that, contrary to expectation, the probability of passing decreases as the number of attempts increases, indicating that initial educational gaps are difficult to overcome within higher education.

Figures (11)

• Figure 1: Average cohort evolution for the three admission tracks (defined in the main text) in the Physics programs of Unesp. The average was taken over the cohorts of year 2013 through 2023, and the curves show the average percentage of students successfully graduating after a number of years in the course. The shaded regions show the uncertainty of each curve.
• Figure 2: Average cohort evolution for the three admission tracks (defined in the main text) in the Biology programs of Unesp. The average was taken over the cohorts of year 2013 through 2023, and the curves show the average percentage of students successfully graduating after a number of years in the course. The shaded regions show the uncertainty of each curve.
• Figure 3: Average cohort evolution for the three admission tracks (defined in the main text) in the Pedagogy courses of Unesp. The average was taken over the cohorts of year 2013 through 2023, and the curves show the average percentage of students successfully graduating after a number of years in the course. The shaded regions show the statistical uncertainty of each curve.
• Figure 4: Ratio $P \over E$ for Calculus I in Physics courses, where $P$ is the number of students passed and $E$ is the number of students enrolled (each year and in each category). The categories represent different admission tracks, as defined in the main text.
• Figure 5: Ratio $P \over E$ for Celular Biology in Biology courses, where $P$ is the number of students passed and $E$ is the number of students enrolled (each year and in each category). The categories represent different admission tracks, as defined in the main text.