Date of Award

Spring 5-15-2021

Author's School

Graduate School of Arts and Sciences

Author's Department

Business Administration

Degree Name

Doctor of Philosophy (PhD)

Degree Type



The broad research topic of my dissertation is individual’s decision making in the presence of peer interactions. I focus on empirical settings where individuals are connected with others in the peer network, and through peer interactions, their decisions are interdependent. Peer network structures in terms of who is connected with whom and the mechanisms of peer interactions convey valuable information on the decision making process. I employ quantitative empirical methods to uncover the information embedded in peer networks, provide insights on the mechanisms of peer interactions, and generate peer-network based managerial implications.

Chapter 1 studies how participants collaborate to compete in an online crowdsourcing competition platform. Participants have heterogeneous abilities and incomplete information on peer participants’ private abilities. We develop a structural matching model captures the decisions on forming collaboration, in which participants form beliefs on each other and face competitive pressure from peer participants. We provide implications on how platforms could better align individual incentives to improve collaboration efficiency.

Chapter 2 focuses on peer effect in workplace misconduct in the empirical setting of server theft in restaurants. We utilize the coworker network structure to employ an instrumental variables model that accounts for endogeneity in peer influence. We show that although servers are more likely to steal when working with high-theft peers, they steal less as peers steal more on a given day. We also show that this negative correlation in daily peer theft is higher under an IT system that increases managerial oversight by reporting likely theft to managers. Our results suggest that the costs of employing unethical workers is higher than the direct cost of those workers' misconduct because their behavior spills over into coworkers' actions and amplifies through reflection effects, yet this contagion can be mitigated by managerial oversight.

Chapter 3 addresses the endogeneity in social connections in empirical measurement of peer effect in customer churn. We first model the process of social network formation, which is based on observed consumer interactions during product use, and then model the consumers' choice to churn under peer influence. Modeling peer network formation allows us to recover the unobserved individual-specific parameters that might affect both peer network formation and decision to churn. We apply the model on data from the popular massively multiplayer online game World of Warcraft, where gamers form social groups to progress through the game content. Our results show that a significant amount of peer ties is explained by the latent gamer characteristics. Based on the estimated model, we run policy experiments to investigate how churn dynamics is affected by the structure of the generated peer network. We provide recommendations to firms on how they could induce formation of networks that suffer less from peer influenced churns.

Chapter 4 studies the consumption and subscription decisions of users in the gym under peer influence. We develop a structural model to empirically study how the attrition of a user in a gym influences not only her direct but also indirect peers, allowing the effects to spread within a social network. We utilize exogenous attritions of gym members in data to tackle the identification challenge. Estimation results help quantify how the social customer life-time value of a user varies depending on her position in the social network, after taking account of the interactive peer effects among users. We then use the estimated model to explore how the gym can prevent user attritions by offering individually-targeted membership discounts. Our results show that the gym can effectively mitigate the negative effects by utilizing the information of the network structure among members. Our results demonstrate the importance ofunderstanding the role of individual users in the social network for customer relationship management.


English (en)

Chair and Committee

Tat T. Chan

Committee Members

Raphael R. Thomadson, Yulia Y. Nevskaya, Lamar L. Pierce, Nitin N. Mehta,