Is Race a Determinant of Student Performance in Economics?

Abstract

While several authors have examined gender as a determinant of student performance in introductory economics, few have considered race as a contributing factor. In this study, data collected on over 5,000 students enrolled at the University of Texas at Austin indicate that on average minority students earn significantly lower grades than non-minority students. A probit model is developed to control for a set of variables that measure or proxy student characteristics, academic maturity, previous coursework, and relative high school quality. Race remains statistically significant. A decomposition framework, commonly used in labor economics to study wage gaps, is adapted to predict success in economics classes based on mean variable characteristics. Details on the probabilities of earning specific grades are obtained through development of an ordered logit model and decomposition techniques. Results of these analyses indicate significant and unexplained differences in average grades earned by minority students, relative to non-minority students.