Quantifying Social Norms and Anxiety in Social Media Text

ACL ARR 2025 May Submission8058 Authors

Abstract: Social norms can induce anxiety within members of a society when they feel pressured to conform. While researchers have previously examined the psychological impact of specific norms or norms collectively, little is known about how different types of norms vary in association to anxiety. In this work, we propose a framework to extract and categorize social norms and their sources (norm drivers) from social media using large language model (LLM). We conduct a human evaluation to assess the reliability of LLM annotations on the obtained categories and systematically examine the relationship between different norm types, norm drivers, and the users' levels of anxiety. Our findings reveal that romantic partners and norms concerning physical appearance are most strongly linked to anxiety. We share the norm types, norm drivers, their rankings in association to anxiety, and the social norms extraction tool to help advance the study of social norms found through language.

Paper Type: Short

Research Area: Computational Social Science and Cultural Analytics

Research Area Keywords: human behavior analysis; stance detection;NLP tools for social analysis

Languages Studied: English

Submission Number: 8058