Color, Gender, and Bias: Examining the Role of Stereotyped Colors in Visualization-Driven Pay Decisions

2025-09-29

AI Summary

This study investigates how gender–color stereotypes (e.g., “pink = female, blue = male”) interfere with salary decisions driven by data visualizations. Using a two-stage crowdsourced experiment, we orthogonally manipulated legend labeling (explicit vs. implicit) and color stereotypicality (stereotypic vs. non-stereotypic), and assessed their causal effects on salary adjustment judgments within a behavioral economics paradigm. Results reveal that explicit legends induce significant in-group bias—favoring same-gender candidates—whereas implicit legends eliminate this bias. Crucially, non-stereotypic colors consistently but modestly reduce salary allocation bias. This work provides the first causal evidence for color semantics as a design-level intervention in data visualization, identifying legend labeling as a critical moderating mechanism. Findings offer empirically grounded, actionable guidelines for developing fairer, less biased visualization practices in high-stakes decision contexts.

Technology Category

• Computer Vision: Bias, Fairness & Privacy
• Machine Learning: Ethics, Bias, and Fairness
• Game Theory and Economic Paradigms: Fair Division

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Web Mining and Content Analysis: Web data visualization

Abstract

We investigate the impact of stereotyped gender-color associations in a visualization-driven decision-making task. In the context of gender data visualization, the well-known "pink for girls and blue for boys" color assignment is associated with stereotypes that could bias readers and decision-makers. Understanding the effects of using stereotyped colors in visualizations for decision-making can help designers better choose colors in stereotype-prone contexts. We therefore explore the potential impact of stereotyped colors on compensation decision-making through two crowdsourced experiments. In these experiments, we evaluate how the association of color with gender (stereotyped vs non-stereotyped) affects the user's allocation decisions in the context of salary adjustments. Our results indicate that explicit expression of the color-gender associations, in the form of a legend on the data visualization, leads to in-group favoritism. However, in the absence of a legend, this in-group favoritism disappears, and a small effect of non-stereotyped colors is observed. A free copy of this paper with all supplemental materials is available at https://osf.io/d4q3v/?view_only=22b636d6f7bb4a7991d9576933b3aaad

Problem

• Examining stereotyped color-gender associations in visualization
• Investigating color bias in visualization-driven compensation decisions
• Analyzing legend effects on in-group favoritism in salary allocations

Innovation

• Used crowdsourced experiments to test color effects
• Examined stereotyped color-gender associations in visualizations
• Analyzed legend presence impact on decision bias