Loading...
Paper Type
Complete
Abstract
We develop a social network based model of social media engagement in the context of online health communities. Grounded in the social network theory, we hypothesize the differential impacts of an online health community member’s betweenness centrality and eigenvector centrality in the web-of-support on a) depth of engagement, and two features of locus of engagement namely b) other-centeredness and c) self-disclosure. Variables were operationalized using text mining and social network analyses. Using econometric modeling (dynamic panel data models and fixed-effects models), we tested our model using panel-data collected from an online health community for people with diabetes. We show that higher betweenness centrality results in deeper engagement from the member; however, the marginal effect is decreasing. Higher eigenvector centrality results in reduced engagement depth; however, the effect is increasing at higher levels. For locus of engagement, we find contrasting curvilinear effects of social network positions for other-centeredness and self-disclosure, respectively.
Recommended Citation
Parameswaran, Srikanth and Kishore, Rajiv, "Network Positions and Engagement in Social Media: An Empirical Examination in the Context of Online Health Communities" (2020). AMCIS 2020 Proceedings. 17.
https://aisel.aisnet.org/amcis2020/virtual_communities/virtual_communities/17
Network Positions and Engagement in Social Media: An Empirical Examination in the Context of Online Health Communities
We develop a social network based model of social media engagement in the context of online health communities. Grounded in the social network theory, we hypothesize the differential impacts of an online health community member’s betweenness centrality and eigenvector centrality in the web-of-support on a) depth of engagement, and two features of locus of engagement namely b) other-centeredness and c) self-disclosure. Variables were operationalized using text mining and social network analyses. Using econometric modeling (dynamic panel data models and fixed-effects models), we tested our model using panel-data collected from an online health community for people with diabetes. We show that higher betweenness centrality results in deeper engagement from the member; however, the marginal effect is decreasing. Higher eigenvector centrality results in reduced engagement depth; however, the effect is increasing at higher levels. For locus of engagement, we find contrasting curvilinear effects of social network positions for other-centeredness and self-disclosure, respectively.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.