Related Experiment Video
Updated: Jul 4, 2025

Integrating Computerized Linguistic and Social Network Analyses to Capture Addiction Recovery Capital in an Online Community
Published on: May 31, 2019
Language use on Twitter reflects social structure and social disparities
1University of Basel, Switzerland.
Social media language, particularly indicators of depression, reflects area deprivation and ethnic minority status. These linguistic markers can help prioritize public health efforts in disadvantaged communities.
Area of Science:
- Computational Social Science
- Public Health
- Psychology
Background:
- Mental health assessments increasingly use social media data.
- Minority group membership and social disparity impact mental health.
- The manifestation of these factors in social media language is not fully understood.
Purpose of the Study:
- To investigate the association between social media language, social disparity, and ethnic minority status.
- To determine if linguistic indicators of mental health on social media correlate with county-level social factors.
Main Methods:
- Spatial lag regressions were used to analyze county-level data.
- Examined associations between lexical indicators of depression (e.g., first-person singular pronouns, negative emotions) and social disparity markers (e.g., Area Deprivation Index-3, ethnic minority percentage).
- Utilized a dataset of approximately 30 million geo-coded tweets aggregated at the U.S. county level.
Main Results:
- County-level lexical indicators of depression were positively associated with area disparity and the percentage of ethnic minority groups.
- Predictive validity checks indicated that lexical indicators correlate with future health and mental health outcomes.
- Linguistic markers of depression and adjustment from tweets showed significant associations with socio-economic and demographic factors.
Conclusions:
- Social media language, specifically indicators of depression, reflects underlying social disparities and ethnic minority concentrations at the county level.
- These linguistic indicators show predictive validity for health outcomes.
- Lexical indicators from social media data can aid in prioritizing public health campaigns, especially in socially deprived areas.
More Related Videos
10:29A Complex Diving-For-Food Task to Investigate Social Organization and Interactions in Rats
Published on: May 8, 2021
09:40Measuring Neural and Behavioral Activity During Ongoing Computerized Social Interactions: An Examination of Event-Related Brain Potentials
Published on: November 15, 2014
Related Concept Videos
Social Scripts
Social Exchange Theory
Relationship Formation
Social Traps
The Stanford Prison Experiment
Stereotypes, Prejudice, and Discrimination