Unraveling the “Social” in Social Norms: The Conditioning Effect of User Connectivity

Operating on the basic principle of “telling people about what lots of other people do,” social norms interventions have demonstrated efficacy in inducing behavior change in diverse settings. However, very few studies examine how the effect of social norms is differentially manifest across individuals, especially in the contemporary socially connected digital world. We provide new empirical evidence from a randomized field experiment that included more than 7,000 individuals on an online physical activity community observed for a two-month period. We find that connectivity plays an important role in moderating the effectiveness of social norms: individuals with higher levels of social connectivity are more susceptible to a social norms message. Additional analysis reveals that individuals who engage in information sharing (high followers and low followees) are the most susceptible to the social norms message. This study has important implications for the effective and safe use of social norms in nudging people’s behavior. Our finding that social norms do not affect all users equally should help optimize interventions by focusing on users with high susceptibility based on easily observable measures. Abundant empirical evidence supports the overall efficacy of social norms as a strategy to induce behavior change. However, very few studies examine how the effect of social norms is differentially manifest across individuals, especially in the contemporary socially connected digital world. We conjecture that the effects of social norms are conditional on an individual’s digital social ties and provide new empirical evidence from a randomized field experiment that included more than 7,000 individuals on an online physical activity community observed for a two-month period. In our investigation of the effect of social norms on users’ goal-setting and goal attainment behaviors, we find a significant moderating role for social connectivity: individuals with higher levels of social connectivity are more susceptible to a social norms message containing information indicating the number of users in this community who set a goal in the pretreatment month. Additional analysis reveals that individuals who have many followers (i.e., high in-degree) but do not follow many others (low out-degree) are the most susceptible to the social norms treatment. Strikingly, we find that social norms also lead to a substantially lower rate of goal attainment compared with the control m