Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detection

TL;DR

This study explores how annotator identities and beliefs influence toxicity ratings in language, revealing biases that impact dataset quality and suggesting the need for context-aware annotation practices.

Contribution

It provides empirical evidence linking annotator social identities and beliefs to toxicity judgments, highlighting biases in toxic language datasets and detection systems.

Findings

Conservative annotators less likely to label anti-Black language as toxic.

Annotators with racist beliefs more likely to label AAE as toxic.

Toxicity detection systems reflect specific social perspectives.

Abstract

The perceived toxicity of language can vary based on someone's identity and beliefs, but this variation is often ignored when collecting toxic language datasets, resulting in dataset and model biases. We seek to understand the who, why, and what behind biases in toxicity annotations. In two online studies with demographically and politically diverse participants, we investigate the effect of annotator identities (who) and beliefs (why), drawing from social psychology research about hate speech, free speech, racist beliefs, political leaning, and more. We disentangle what is annotated as toxic by considering posts with three characteristics: anti-Black language, African American English (AAE) dialect, and vulgarity. Our results show strong associations between annotator identity and beliefs and their ratings of toxicity. Notably, more conservative annotators and those who scored highly on our scale for racist beliefs were less likely to rate anti-Black language as toxic, but more likely to rate AAE as toxic. We additionally present a case study illustrating how a popular toxicity detection system's ratings inherently reflect only specific beliefs and perspectives. Our findings call for contextualizing toxicity labels in social variables, which raises immense implications for toxic language annotation and detection.