Separating Homophily and Peer Influence with Latent Space

by Joseph P. Davin, Sunil Gupta & Mikołaj Jan Piskorski

Overview — People are often more willing to try new things when they see others doing so. This phenomenon, which academics call "social influence", has a profound impact on many aspects of customer decision-making and marketing. For example, social influence affects consumers' willingness to take up new technologies, adopt and use social networks, and ask their physicians for particular prescription medicines. Marketers are thus eager to understand how and to what extent social influence affects people's consumption decisions. To date, however, it has been difficult to pinpoint the effects of social influence, as researchers have struggled to separate it from a simple fact that like-minded people tend to enjoy the same things, per the adage "Birds of a feather flock together." The authors use the field of mobile app adoption in Japan to examine this problem. Japan is an ideal testing ground because approximately 80-85 percent of all page views occur through mobile. In addition, mobile apps are often social in nature, especially those that are linked to a social network platform. The authors devise a new method to assess social influence by controlling for other factors that usually complicate the picture. Overall, the findings show that peer usage accounts for more than a quarter of all mobile app adoptions. The paper also highlights a risk that firms could overestimate social influence by 40 percent on average, even up to 100 percent in certain cases. The authors' method helps overcome this risk. Key concepts include:

Social influence is clearly important in customer decision making and marketing. It has a profound impact on the adoption of new technologies, social network usage and adoption, eCommerce, new prescriptions of pharmaceutical drugs, ad effectiveness, group decision making, and customer retention.
Using a new approach, this paper documents the impact of peer behavior on the adoption of mobile apps in a social network.
Results show that peers drive more than a quarter of all mobile app adoptions.
Results also show that firms risk overestimating the impact of social influence by 40% on average, even up to 100% in certain cases. The authors' method helps overcome this risk.

Author Abstract

We study the impact of peer behavior on the adoption of mobile apps in a social network. To identify social influence properly, we introduce latent space as an approach to control for latent homophily, the idea that "birds of a feather flock together." In a series of simulations, we show that latent space coordinates significantly reduce bias in the estimate of social influence. The intuition is that latent coordinates act as proxy variables for hidden traits that give rise to latent homophily. The approach outperforms existing methods such as including observed covariates, random effects, or fixed effects. We then apply the latent space approach to identify social influence on installation of mobile apps in a social network. We find that peer influence accounts for 27% of mobile app adoptions and that latent homophily inflates this estimate by 40% (to 38%). In some samples, ignoring latent homophily can result in overestimation of social effects by over 100%.