Talk more about yourself: a data-driven extended theory of reasoned action for online health communities

Online health communities (OHCs) are able to facilitate social support exchange among people with similar health concerns, and the relationship between self-disclosure and social support exchange has been widely discussed. Build upon the previous studies, we extend the seminal theory of reasoned action (TRA) in this research, through the adoption of an salient but understudied construct, user role, which captures user inclination toward providing or seeking social support, and further explore how it impacts OHC users’ self-disclosure intention. We adopt a mixed methods approach including manual coding, machine learning algorithms, and econometric analyses, and investigate over 4 million posts over 17 years from a well-developed breast cancer OHC to uncover both direct and indirect effects of user role on OHC users’ intention of future self-disclosure intention. The results reveal both the main and moderating effects of user roles regarding users’ self-disclosure intention. First, user with a more clear intention of social support seeking or providing are more likely to disclose personal information. Second, the relative tendency of support provision magnifies the positive effects of attitudinal and normative factors toward future disclosure. We also discuss the findings and implications of the extended TRA framework to generate some actionable insights of OHC design.