Abstract:  Scholars have started investigating the link between social network site (SNS) use and depression among adolescents, but findings remain inconsistent. In order to find out the reasons, researchers conduct lots of studies, trying to give an interpretation for the relationship and its complex mechanisms from a different perspective. Among many factors influencing adolescents’ depression, the role of upward social comparison on SNS has increasingly received attention of both practitioners and researchers over the past few years. There is substantial literature documenting that upward social comparison on SNS has an important influence on adolescents’ depression, but little is known about the mediating and moderating mechanisms underlying this relation. Therefore, based on the social comparison theory of envy and the response styles theory of rumination, the present study constructed a moderated mediation model to examine the effect of upward social comparison on SNS and the underling mechanisms. Specifically, the present study examined whether upward social comparison on SNS would be indirectly related to adolescents’ depression through the experience of envy during SNS using, and whether this indirect association would be moderated by the adolescents’ rumination. This integrated model can address questions about both mediation and moderation in one model. Its results would be helpful to answer the issues such as “what works for whom”, and provide valuable information for early identification and prevention that cannot be obtained by separately testing the two questions. A total of 836 high school students (mean age = 16.34 years, SD = 1.06) participated in this study. Adolescents’ upward social comparison on SNS was measured by the revised upward social comparison questionnaire. Envy was assessed with the SNS envy scale. The short form of ruminative responses scale was used to assess rumination traits. Adolescents’ depression was assessed with the Center for Epidemiologic Studies Depression Scale. All the measures have good reliability and validity. The results showed that: (1) after controlling for age, grade, number of friends on SNS and the frequency and time of SNS use, the upward social comparison on SNS has a positive direct effect on adolescents’ depression. (2) The positive association between upward social comparison and adolescents’ depression was mediated by envy, and there is gender difference in this indirect effect. The indirect effect was stronger for girls than boys. (3) For boys, the mediating effect of envy was moderated by rumination. The indirect effect was stronger for boys with high rumination than for those with low rumination. However, the moderating effect of rumination cannot be found in the indirect effect of envy for girls. These findings contribute to our understanding of how and when upward social comparison on SNS impacts adolescents’ depression. On the one hand, the experience of envy during SNS using relative to the upward social comparison on SNS, may serve as a more important role on the emergence and development of adolescents’ depression. On the other hand, more attention should be paid to the boys with high level of rumination (especially reducing their frequency of upward social comparison on SNS) and boys with high frequency of upward social comparison on SNS (especially reducing their rumination). Last but not the least, the prevention and interventions for adolescents’ depression should pay attention not only to the effect of behavior factors, experience factors and personality trait factors, but also to the combined influence of those factors.

Key words:  depression, social network site (SNS), upward social comparison, envy, rumination, adolescents