User communities are increasingly becoming an essential element of companies' business processes. However, reaping the benefits of such social systems does not always prove effective, often because companies fail to stimulate members' collaboration continuously or neglect their social integration. Following communication accommodation theory, the authors posit that members' communication style alignment symbolically reflects their community identification and affects subsequent participation behavior. This research uses text mining to extract the linguistic style properties of 74,246 members' posts across 37 user communities. Two mixed multilevel Poisson regression models show that when members' linguistic style matches with the conventional community style, it signals their community identification and affects their participation quantity and quality. Drawing on an expanded view of organizational identification, the authors consider dynamics in members' social identification by examining trends and reversals in linguistic style match developments. Whereas a stronger trend of alignment leads to greater participation quantity and quality, frequent reversals suggest lower participation quantity. At a community level, greater synchronicity in the linguistic style across all community members fosters individual members' participation behavior.