5. Discussion
The primary objectives of the present study are to examine the dual effects of social identity on consumer online behaviors as well as to explore and investigate whether the three distinctive components of social identity suggested by the theory exert various influences on outcome behaviors. Research findings offer theoretical contributions in several ways. First, this study extends the customerebrand and customerecompany identification literature to the social media context. The results revealed that users develop social identification with virtual groups and even with platforms that foster these groups. Second, by integrating both use and purchase behavior in one model, this study can investigate the dual effects of social identity on two key consumer behaviors. When a member of an online social network develops identification toward social media platforms, he or she is more likely to continuously use the website and engage in purchase behavior. In other words, this finding confirms our proposition that social identity is able to serve as a common factor that drives both use and purchase behavior at the same time. Moreover, the effects of social identification on outcome behaviors are resulted from different identity components. Therefore, this study presents an integrative perspective on the role of social identity in the social media context. Third, this study examines the dimensional effect of social identity on two behaviors of interest. Interestingly, the analysis shows that the affective dimension of social identity has a greater influence on use behavior, accounting for 16% of variance of use behavior, whereas cognitive and evaluative dimensions of social identity have a more significant effect on purchase behavior, explaining 7% of the variance. To this end, this study confirms the theoretical distinction between three different components of social identity with empirical data. Moreover, different social identity aspects influence user behaviors in various ways; that is, only affective identity is able to influence users’ use behavior, whereas cognitive and evaluative identity can significantly influence purchase behavior. For purchase behavior, the evaluative component of social identity is twice as influential as the effect of cognitive dimension. This novel and interesting result provides additional support to prior research that the three dimensions are independent and non-interchangeable (Tajfel, 1982).