Quantifying the Effect of Informational Support on Membership Retention in Online Communities through Large-Scale Data Analytics

Wanli Xing, Sean Goggins, Josh Introne

Research output: Contribution to journalArticle

12 Scopus citations

Abstract

Participating in online health communities for informational support can benefit patients in various ways. For the online communities to be sustainable and effective for their participants, membership retention and commitment are important. This study explores how informational support requesting and providing by users holding different social roles (core user and periphery user) are related with participants’ retention in the community. We first crawled six years of data in the WebMD fibromyalgia forum with around 200,000 posts and 10,000 users. Then a supervised machine learning model is trained and validated to automatically identify the requesting and providing informational support posts exchanged between the members in the community. Lastly, survival analysis was employed to quantify how the informational support requesting and providing by different social roles predicts the member's continued participation in the online community. The results reveal the different influencing mechanism of requesting and providing support from different social roles on the patients’ decision to stay in the community. The findings can aid in the design of better support mechanisms to enhance member commitment in online health communities.

Original languageEnglish (US)
Pages (from-to)227-234
Number of pages8
JournalComputers in Human Behavior
Volume86
DOIs
StatePublished - Sep 2018
Externally publishedYes

Keywords

  • Informational support
  • Membership retention
  • Online communities
  • Social role
  • Survival analysis
  • Text mining

ASJC Scopus subject areas

  • Arts and Humanities (miscellaneous)
  • Human-Computer Interaction
  • Psychology(all)

Fingerprint Dive into the research topics of 'Quantifying the Effect of Informational Support on Membership Retention in Online Communities through Large-Scale Data Analytics'. Together they form a unique fingerprint.

  • Cite this