Self-Diagnosis on Social Media: Malingering or Identity Work?

• In Event: 20820 - Section on Medical Sociology Roundtables
  In Refereed Roundtable (90 minute): Table 6. Mental Health

Abstract

Self-diagnosis is often criticized as malingering, inaccurate, or the product of social media algorithms. But self-diagnosis may also represent an accurate interpretation of the self, as individuals deploy broadly available psychiatric language and biomedical thinking to better understand themselves. Further, online environments that allow anonymity and protect users from the social sanctions and stigma they may face when self-labeling offline provide new contexts to explore identity through self-diagnosis. Using a corpus of text from Reddit (N = 138,767) this research considers how social media users engage in and with self-diagnosis online. Through qualitative analysis of a small sample of posts (N = 762), I conceptualize self-diagnosis as two mutually occurring processes: a feeling or question of mental wrongness and the application of a particular diagnostic explanatory label. I then scale up qualitative identification of self-diagnosis using a large language model (LLM) to code the full corpus, enabling the application of computational methods of text analysis to explore how social media users engage in and with self-diagnosis differently from (or not) than professional diagnosis. This work has implications for better understanding why social media users engage in self-diagnosis and its meanings and implications for identity.

Author

• Amy L Johnson, Lehigh University