Commitment of Newcomers and Old-timers to Online Health Support Communities
summary
Analysis of 1 million messages from the American Cancer Society Cancer Survivor Network (users registered 2002–2015, ~48,000 users) uses machine-learned labels for informational and emotional support and three commitment measures (self-reported attachment, continued participation, replying to others). Finding: more replies per month reliably increase retention (hazard ratio ~0.8), while higher average informational or emotional content per reply predicts higher churn (about a 15% increase in leaving); quantity helps old-timers more, content helps newcomers.
AI-written summary · about
chapters (11)
- 0:00Introduction to research
- 1:09Group socialization model
- 2:15Social support in health groups
- 4:22Research setup and data
- 6:48Newcomers' informational needs
- 8:11Communication and commitment
- 9:57Survival analysis findings
- 12:00Differences by tenure
- 13:46Summary of findings
- 14:47Implications and future work
- 15:07Q&A session
related talks
52mEmpowering Engagement: Introducing a Dynamic Dashboard for Proactive Retention Strategies
35mUsing Survival Analysis to understand customer retention
20mIdentification of Imminent Suicide Risk Among Young Adults using Text Messages
37mData Driven Diagnosis of Alzheimer's Disease
37mData Driven Diagnosis of Alzheimer's Disease
19mWhat Healthcare Can Learn from Netflix
32mAnalyzing social media for suicide risk using natural language processing
34mCommunity Data: What to Measure and Why?