Gendered Prompting and LLM Code Review: How Gender Cues in the Prompt Shape Code Quality and Evaluation

TL;DR

This study investigates how gendered language in prompts influences code generation and review by LLMs, revealing subtle prompt differences and biases in evaluation that could impact fairness in automated programming tools.

Contribution

It provides a comprehensive analysis of gendered linguistic effects on LLM code tasks through real-world data, controlled experiments, and simulation, highlighting biases in code review.

Findings

Female prompts are more indirect and involved.

No consistent difference in code correctness or quality.

Models tend to approve female-authored code more.

Abstract

LLMs are increasingly embedded in programming workflows, from code generation to automated code review. Yet, how gendered communication styles interact with LLM-assisted programming and code review remains underexplored. We present a mixed-methods pilot study examining whether gender-related linguistic differences in prompts influence code generation outcomes and code review decisions. Across three complementary studies, we analyze (i) collected real-world coding prompts, (ii) a controlled user study, in which developers solve identical programming tasks with LLM assistance, and (iii) an LLM-based simulated evaluation framework that systematically varies gender-coded prompt styles and reviewer personas. We find that gender-related differences in prompting style are subtle but measurable, with female-authored prompts exhibiting more indirect and involved language, which does not translate into consistent gaps in functional correctness or static code quality. For LLM code review, in contrast, we observe systematic biases: on average, models approve female-authored code more, despite comparable quality. Controlled experiments show that gender-coded prompt style affect code length and maintainability, while reviewer behavior varies across models. Our findings suggest that fairness risks in LLM-assisted programming arise less from generation accuracy than from LLM evaluation, as LLMs are increasingly deployed as automated code reviewers.