A Model of Network Dynamics: Tie Formation, Product Adoption, and Content Generation
58 Pages Posted: 26 Jun 2017 Last revised: 31 Jul 2018
Date Written: March 1, 2018
We develop a model for the co-evolution of individual users' friendship tie formation and their concurrent online activities (production adoptions and production of user-generated content) within an online social network. By explicitly modeling the endogenous formation of the network and accounting for the interdependence between decisions in friendship tie formation and in concurrent online activities, we are able to discover important drivers underlying individuals' friendship decisions while, at the same time, provide a clean identification of the resulting peer effects on individuals' actions. We estimate our model using a novel data set capturing the continuous development of a network and users' entire action history within the network. Our results reveal that, compared to a potential friend's product adoptions and content generation activities, the total number of friends and the number of common friends this potential friend has with the focal user are the most important drivers of friendship formation. Further, while having more friends does not make a user more active, having more active friends does increase a user's activity levels in terms of both product adoptions and content generation through peer effects. We conduct several counterfactual exercises to assess the effectiveness of various seeding and stimulation strategies in increasing website traffic while taking the endogenous network formation into account. We find that seeding to users with the most friends is not always the best strategy to increase users' activity levels on the website.
Keywords: Network Formation, Peer Influence, Product Adoption, UGC
JEL Classification: D83, L82, M31
Suggested Citation: Suggested Citation