Socially Nudged: A Quasi-Experimental Study of Friends' Social Influence in Online Product Ratings

Social-networking functions are increasingly embedded in online rating systems. These functions alter the rating context in which consumer ratings are generated. In this paper, we empirically investigate online friends’ social influence in online book ratings. Our quasi-experiment research design exploits the temporal sequence of social-networking events and ratings and offers a new method for identifying social influence while accounting for the homophily effect. We find rating similarity between friends is significantly higher after the formation of the friend relationships, indicating that with social-networking functions, online rating contributors are socially nudged when giving their ratings. Additional exploration of contingent factors suggests that social influence is stronger for older books and users who have smaller networks, and relatively more recent and extremely negative ratings cast more salient influence. Our study suggests that friends’ social influence is an important consideration when introducing social-networking functions to online rating systems.