A zoomed detail of the full retention curves of the two groups. For ease of comparison, the treatment curve is shifted on the right of 3 days. Error bars are 95% confidence intervals. The inset shows the full curve with no shifting. The red dashed line is the treatment group.
A zoomed detail of the full retention curves of the two groups. For ease of comparison, the treatment curve is shifted on the right of 3 days. Error bars are 95% confidence intervals. The inset shows the full curve with no shifting. The red dashed line is the treatment group.
MoodBar: Increasing New User Retention in Wikipedia Through Lightweight Socialization
Abstract
Socialization in online communities allows existing members to welcome and recruit newcomers, introduce them to community norms and practices, and sustain their early participation. However, socializing newcomers does not come for free: in large communities, socialization can result in a significant workload for mentors and is hard to scale. In this study we present results from an experiment that measured the effect of a lightweight socialization tool on the activity and retention of newly registered users attempting to edit for the first time Wikipedia. Wikipedia is struggling with the retention of newcomers and our results indicate that a mechanism to elicit lightweight feedback and to provide early mentoring to newcomers improves their chances of becoming long-term contributors.
This research was performed while at the Wikimedia Foundation. For more information of the initial study, please see our blog post on the official Wikipedia blog titled “What MoodBar tells us about registered editors”.
The paper presented at CSCW’15 expands upon those findings and concludes that lightweight socialization does have a positive effect on long-term editor retention on Wikipedia.
