Patrick Park (Carnegie Mellon University): The Centre Cannot Hold: Online Polarization and the Paradox of Network Diversity

Abstract | How does online political polarization emerge? I develop a network-based explanation focusing on the relational constraint that well-connected social media users experience from context collapse, the situation in which a user’s interactions with one group of friends is easily observed by other groups of friends. In such a setting, misunderstandings and conflicts are bound to occur, thereby constrain the user’s social relationships. To avoid such relational constraint, the users with diverse networks and the political moderates with more balanced connections to conservatives and liberals might self-censor more frequently. Political polarization, then, can appear aggravated as an unintended consequence of the self-censoring choices made by individual political moderates, which, in the aggregate, biases the observed opinion distribution toward bimodality, even when the underlying opinion distribution is stable. I test this explanation using the tweet deletions as a measure of self-censorship for 26M U.S. Twitter users and find that network diversity is associated with more frequent tweet deletions. These findings offer an alternative network-based mechanism in opinion dynamics and raises the need to account for survivorship bias in the empirical data used in the research of opinion polarization.

Bio | Patrick Park is a computational social scientist with research interests in the structure and evolution of large-scale social networks. His research focuses on how people form and maintain social ties and how the broader social, economic and natural environments affect this process. He conducts social network research at the intersections of social contagion, economic sociology, social psychology, the diffusion of innovation, and social movements using empirical data that capture population-scale online social interactions. His works were published in Science, Social Networks, PLoS One, Lecture Notes in Computer Science, and Big Data and Society. Currently, he is a tenure-track assistant professor in the Software and Societal Systems Department (S3D) in Carnegie Mellon University School of Computer Science. Before joining CMU, Patrick was a postdoctoral fellow at the University of Michigan and Northwestern University, and received his doctoral degree in sociology at Cornell University.