Saved in:
Bibliographic Details
Main Authors: Kwegyir-Aggrey, Kweku, Durvasula, Naveen, Wang, Jennifer, Venkatasubramanian, Suresh
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.01984
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929484441059328
author Kwegyir-Aggrey, Kweku
Durvasula, Naveen
Wang, Jennifer
Venkatasubramanian, Suresh
author_facet Kwegyir-Aggrey, Kweku
Durvasula, Naveen
Wang, Jennifer
Venkatasubramanian, Suresh
contents In many domains, it is difficult to obtain the race data that is required to estimate racial disparity. To address this problem, practitioners have adopted the use of proxy methods which predict race using non-protected covariates. However, these proxies often yield biased estimates, especially for minority groups, limiting their real-world utility. In this paper, we introduce two new contextual proxy models that advance existing methods by incorporating contextual features in order to improve race estimates. We show that these algorithms demonstrate significant performance improvements in estimating disparities on real-world home loan and voter data. We establish that achieving unbiased disparity estimates with contextual proxies relies on mean-consistency, a calibration-like condition.
format Preprint
id arxiv_https___arxiv_org_abs_2409_01984
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Observing Context Improves Disparity Estimation when Race is Unobserved
Kwegyir-Aggrey, Kweku
Durvasula, Naveen
Wang, Jennifer
Venkatasubramanian, Suresh
Computers and Society
In many domains, it is difficult to obtain the race data that is required to estimate racial disparity. To address this problem, practitioners have adopted the use of proxy methods which predict race using non-protected covariates. However, these proxies often yield biased estimates, especially for minority groups, limiting their real-world utility. In this paper, we introduce two new contextual proxy models that advance existing methods by incorporating contextual features in order to improve race estimates. We show that these algorithms demonstrate significant performance improvements in estimating disparities on real-world home loan and voter data. We establish that achieving unbiased disparity estimates with contextual proxies relies on mean-consistency, a calibration-like condition.
title Observing Context Improves Disparity Estimation when Race is Unobserved
topic Computers and Society
url https://arxiv.org/abs/2409.01984