Evaluation of Large Language Models: STEM education and Gender Stereotypes

TL;DR

This study examines gender biases in large language models across multiple cultures and languages, revealing significant differences in STEM education suggestions based on gendered prompts, with implications for educational equity.

Contribution

It provides a cross-cultural, quantitative analysis of gender stereotypes in LLMs related to educational choices, highlighting persistent biases despite model updates.

Findings

Significant gender bias in STEM suggestions based on names.

Fewer STEM suggestions in Danish, Spanish, and Indian contexts.

Subtle profession biases identified and categorized.

Abstract

Large Language Models (LLMs) have an increasing impact on our lives with use cases such as chatbots, study support, coding support, ideation, writing assistance, and more. Previous studies have revealed linguistic biases in pronouns used to describe professions or adjectives used to describe men vs women. These issues have to some degree been addressed in updated LLM versions, at least to pass existing tests. However, biases may still be present in the models, and repeated use of gender stereotypical language may reinforce the underlying assumptions and are therefore important to examine further. This paper investigates gender biases in LLMs in relation to educational choices through an open-ended, true to user-case experimental design and a quantitative analysis. We investigate the biases in the context of four different cultures, languages, and educational systems (English/US/UK, Danish/DK, Catalan/ES, and Hindi/IN) for ages ranging from 10 to 16 years, corresponding to important educational transition points in the different countries. We find that there are significant and large differences in the ratio of STEM to non-STEM suggested education paths provided by chatGPT when using typical girl vs boy names to prompt lists of suggested things to become. There are generally fewer STEM suggestions in the Danish, Spanish, and Indian context compared to the English. We also find subtle differences in the suggested professions, which we categorise and report.