ABSTRACT
In this paper, we analyze a natural experiment regarding anonymity options and participation in a large, successful online community of practice (CoP) for U.S. soldiers. We study the impacts of changes of anonymity options on comment quality for productive discussion and professionalism. Four levels of personal attribution or anonymity of comments are significantly correlated with comment quality under some, but not all, circumstances. Eliminating anonymity options produced significantly fewer antisocial comments and fewer comments overall, although it did not affect overall peripheral participation as measured by logins and page views. Online identity or reputation appears to be more of a factor than external culpability in shaping user behaviors. Attitudes of participants and the evolution of norms over time are presented, and implications for the design of online learning communities are discussed.
