Skip to main content
. Author manuscript; available in PMC: 2018 Jan 19.
Published in final edited form as: Proc Int AAAI Conf Weblogs Soc Media. 2017 May;2017:377–386.

Figure 2.

Figure 2

A: The monthly retention rate for communities differs drastically according to their position in our identity-based typology, with dynamicity being the strongest signal of higher user retention (x-axes bin community-months by percentiles; in all subsequent plots, error bars indicate 95% bootstrapped confidence intervals). B: Dynamicity also correlates with long-term user retention, measured as the number of months the average user spends in the community; however, distinctiveness does not correlate with this longer-term variant of user retention.