Relating Word Embedding Gender Biases to Gender Gaps: A Cross-Cultural Analysis

2026-01-23

Proceedings of the First Workshop on Gender Bias in Natural Language Processing

AI Summary

This study quantifies gender bias in word embeddings and investigates its association with real-world gender disparities across societies. Leveraging Twitter data from 51 U.S. regions and 99 countries in 2018, the authors construct a metric of gender bias in word embeddings and systematically correlate it with 18 international and 5 U.S.-specific gender gap indicators spanning education, politics, economics, and health. The analysis reveals significant cross-cultural correlations and strong predictive power, demonstrating for the first time that gender bias embedded in language models serves as a valid proxy for societal gender inequality. These findings establish a novel paradigm for monitoring social biases through computational linguistic methods at a global scale.

Technology Category

• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy
• Machine Learning: Ethics, Bias, and Fairness
• Computer Vision: Bias, Fairness & Privacy

Application Category

• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Modern models for common NLP tasks often employ machine learning techniques and train on journalistic, social media, or other culturally-derived text. These have recently been scrutinized for racial and gender biases, rooting from inherent bias in their training text. These biases are often sub-optimal and recent work poses methods to rectify them; however, these biases may shed light on actual racial or gender gaps in the culture(s) that produced the training text, thereby helping us understand cultural context through big data. This paper presents an approach for quantifying gender bias in word embeddings, and then using them to characterize statistical gender gaps in education, politics, economics, and health. We validate these metrics on 2018 Twitter data spanning 51 U.S. regions and 99 countries. We correlate state and country word embedding biases with 18 international and 5 U.S.-based statistical gender gaps, characterizing regularities and predictive strength.

Problem

• gender bias
• word embeddings
• gender gaps
• cross-cultural analysis
• social inequality

Innovation

• word embeddings
• gender bias
• cross-cultural analysis
• gender gap prediction
• computational social science