Reassessing Arrest Disparity Using Skin Tone as a Continuous Variable

Racial disparities in arrest remain a persistent feature of the U.S. criminal justice system. Prior studies often rely on broad racial categories that obscure within-group differences, whereas emerging evidence suggests that skin tone, independent of race, influences perceptions and outcomes. This s...

Full description

Saved in:  
Bibliographic Details
Main Author: Brockdorf, Soren (Author)
Format: Electronic Article
Language:English
Published: 2025
In: Race and social problems
Year: 2025, Volume: 17, Issue: 4, Pages: 625-640
Online Access: Volltext (kostenfrei)
Check availability: HBZ Gateway
Keywords:
Arrest disparities
Colorism
Discretion
Policing
Racial inequality
Skin tone
Description
Summary:Racial disparities in arrest remain a persistent feature of the U.S. criminal justice system. Prior studies often rely on broad racial categories that obscure within-group differences, whereas emerging evidence suggests that skin tone, independent of race, influences perceptions and outcomes. This study investigates whether arrest disparities can be more accurately assessed using a continuous measure of skin tone. Drawing on 1086 arrests in Alachua County, Florida, during 4 months surrounding the implementation of a 2020 county mandate, this study operationalized skin tone through RGB coding of mugshot images and applied spline regression models to test for threshold effects. Results indicate that disparities persist across the skin tone spectrum, suggesting that discretionary enforcement remains influenced by phenotypic cues or possibly the probability of these cues. Findings support looking away from historical accounts of stereotype threat, self-categorization theory, and social dominance theory, while also highlighting limitations of demographic groupings. Incorporating skin tone into disparity research offers a more precise approach to understanding colorism in policing and informs theoretical, methodological, and policy debates on equity and reform. Although the model in the study explained 69.17% of the variance in DI (R2 = .69), this study implies a redirection of future studies, focusing on suspicion and the rarity of skin tones in the population.
ISSN:1867-1756
DOI:10.1007/s12552-025-09464-y