Characterizing Social Structural and Linguistic Behaviors of Subgroup Interactions: A Case of Online Health Communities for Postpartum Depression on Facebook

Online health communities (OHCs) have become a major source of sharing knowledge and social support for people with health concerns. The present paper aimed to extend the previous understanding of community dynamics of two types of members, contributors and lurkers, in OHCs for postpartum depression (PPD). Multi-level analyses were conducted to identify subgroup formation and different roles of members and their interaction patterns within subgroups. Specifically, social structural behaviours in OHCs on Facebook were analysed at both network and node levels in addition to members' sentiment and linguistic behaviours which were analysed in light of members' roles and structural behaviours. Results suggest that structural as well as sentiment and linguistic behaviours of members in OHCs for PPD varied across different groups and roles. While contributors tended to be highly influential as information/support givers, lurkers still formed subgroups to seek for support and information. These findings not only articulated the underlying mechanism of community networks and subgroup formation in OHCs for PPD, but also shed light on ways of facilitating prosperity and sustainability of OHCs.