Using Linear Regression to Examine Bias in Home Valuation

Presenter: Dirk Tolson III

Co-Presenter(s):
Serina Cabrera, Anabel Camarena, Sydney Hernandez, Joseph Immel, Salvador Ochoa Zavalza, Madelyn Elena Williams

Presenter Status: Undergraduate student

Academic Year: 22-23

Semester: Spring

Faculty Mentor: Omayra Ortega

Department: Mathematics

President's Strategic Plan Goal: Connectivity and Community Engagement

Screenshot URL: https://drive.google.com/uc?id=16sQIX2FkEA5Qels41jQVO5L_MWrIaANA

Abstract:
In a capitalist society, the main tool to determine someone’s wealth is to look at the price on their home. But the price of a home, however, seems to not be weighted the same for all individuals. Home valuation has been determined historically by a percentage of the value of the home in the past transactions. But due to redlining and possible bias from home appraisers, given the lack of diversity among them, enduring bias in home valuation has negatively impacted communities of color. We explore bias through a series of ordinary linear regression models and use the recently released data set shared in the article, Appraised: The Persistent Evaluation of White Neighborhoods as More Valuable Than Communities Color, written by Howell & Korver-Glenn (2022).