{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/understanding-and-measuring-psychological","title":"Understanding and Measuring Psychological Stress using Social Media","arxiv_id":"1811.07430","date":"2018-11-19","proceeding":null,"authors":["Sharath Chandra Guntuku","Anneke Buffone","Kokil Jaidka","Johannes Eichstaedt","Lyle Ungar"],"abstract":"A body of literature has demonstrated that users' mental health conditions,\nsuch as depression and anxiety, can be predicted from their social media\nlanguage. There is still a gap in the scientific understanding of how\npsychological stress is expressed on social media. Stress is one of the primary\nunderlying causes and correlates of chronic physical illnesses and mental\nhealth conditions. In this paper, we explore the language of psychological\nstress with a dataset of 601 social media users, who answered the Perceived\nStress Scale questionnaire and also consented to share their Facebook and\nTwitter data. Firstly, we find that stressed users post about exhaustion,\nlosing control, increased self-focus and physical pain as compared to posts\nabout breakfast, family-time, and travel by users who are not stressed.\nSecondly, we find that Facebook language is more predictive of stress than\nTwitter language. Thirdly, we demonstrate how the language based models thus\ndeveloped can be adapted and be scaled to measure county-level trends. Since\ncounty-level language is easily available on Twitter using the Streaming API,\nwe explore multiple domain adaptation algorithms to adapt user-level Facebook\nmodels to Twitter language. We find that domain-adapted and scaled social\nmedia-based measurements of stress outperform sociodemographic variables (age,\ngender, race, education, and income), against ground-truth survey-based stress\nmeasurements, both at the user- and the county-level in the U.S. Twitter\nlanguage that scores higher in stress is also predictive of poorer health, less\naccess to facilities and lower socioeconomic status in counties. We conclude\nwith a discussion of the implications of using social media as a new tool for\nmonitoring stress levels of both individuals and counties.","url_abs":"http://arxiv.org/abs/1811.07430v2","url_pdf":"http://arxiv.org/pdf/1811.07430v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"understanding-and-measuring-psychological","repo_url":"https://github.com/chandrasg/lexica","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1811.07430","atlas_url":"https://app.syntology.ai/?focus=1811.07430","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}