Inside the Black Box: Detecting and Mitigating Algorithmic Bias Across Racialized Groups in College Student-Success Prediction
Inside the Black Box: Detecting and Mitigating Algorithmic Bias Across Racialized Groups in College Student-Success Prediction
 
Print

Published Online in:
AERA Open
July 11, 2024

Denisa Gándara, University of Texas at Austin
Hadis Anahideh, University of Illinois Chicago
Matthew Ison, Northern Illinois University
Lorenzo Picchiarini, University of Illinois Chicago

Colleges and universities are increasingly turning to algorithms that predict college-student success to inform various decisions, including those related to admissions, budgeting, and student-success interventions. Because predictive algorithms rely on historical data, they capture societal injustices, including racism. In this study, we examine how the accuracy of college student success predictions differs between racialized groups, signaling algorithmic bias. We also evaluate the utility of leading bias-mitigating techniques in addressing this bias. Using nationally representative data from the Education Longitudinal Study of 2002 and various machine learning modeling approaches, we demonstrate how models incorporating commonly used features to predict college-student success are less accurate when predicting success for racially minoritized students. Common approaches to mitigating algorithmic bias are generally ineffective at eliminating disparities in prediction outcomes and accuracy between racialized groups.

Read the full open-access article

Read the press release: "Study: Algorithms Used by Universities to Predict Student Success May Be Racially Biased"

Video: Co-authors Denisa Gándara and Hadis Anahideh discuss findings and implications of the study

Study citation: Gándara, D., Anahideh, H., Ison, M., & Picchiarini, L. (2024). Inside the black box: Detecting and mitigating algorithmic bias across racialized groups in college student-success prediction. AERA Open, 10(1), 1–15. https://doi.org/10.1177/23328584241258741
 
 
News Mentions
Print