Degendering Resumes for Fair Algorithmic Resume Screening

TL;DR

This study explores removing gendered information from resumes to reduce bias in algorithmic screening, finding that while lexicon-based obfuscation reduces gender cues, it can impair screening performance, and NLP debiasing methods are ineffective.

Contribution

It introduces a systematic approach to quantify and reduce gendered information in resumes, evaluating the impact on screening accuracy and debiasing techniques.

Findings

Lexicon-based gender obfuscation significantly reduces gender cues.

Obfuscation can impair resume screening performance.

Standard NLP debiasing methods are ineffective for resumes.

Abstract

We investigate whether it is feasible to remove gendered information from resumes to mitigate potential bias in algorithmic resume screening. Using a corpus of 709k resumes from IT firms, we first train a series of models to classify the self-reported gender of the applicant, thereby measuring the extent and nature of gendered information encoded in resumes. We then conduct a series of gender obfuscation experiments, where we iteratively remove gendered information from resumes. Finally, we train a resume screening algorithm and investigate the trade-off between gender obfuscation and screening algorithm performance. Results show: (1) There is a significant amount of gendered information in resumes. (2) Lexicon-based gender obfuscation method (i.e. removing tokens that are predictive of gender) can reduce the amount of gendered information to a large extent. However, after a certain point, the performance of the resume screening algorithm starts suffering. (3) General-purpose gender debiasing methods for NLP models such as removing gender subspace from embeddings are not effective in obfuscating gender.