Browsing by Author "Chen, Haohan"

Streaming social media network data have been used in recent studies on political behavior and institutions. Modeling time dynamics in such data helps political scientists produce robust results and efficiently manage their data collection process. However, existing political science methods are yet to provide researchers with the tools to analyze and monitor streaming social media network data. In this thesis, I introduce Bayesian dynamic network modeling for political science research. An extension of the recent development of dynamic modeling techniques, the method enables political scientists to track trends and detect anomalies in streaming social media network data. I illustrate the method with an application to an original dataset of political discourse from a Chinese social networking site. The model detects citizens' behavioral responses to political and non-political events. It also suggests the Chinese government censors and fabricates online discourse during politically sensitive periods.

How do citizens in authoritarian China talk about politics with one another in the social media era? Political talk is a social activity in authoritarian regimes as much as it is in democracies. However, the scholarship has so far dominantly focused on state-citizen interactions in political communication in authoritarian regimes, overlooking the dynamics of citizen-citizen interaction. In this dissertation, I present a new theory, original data, and two novel methods to understand social media political talk in authoritarian China. I argue that citizens in the social media era are engaged in a new form of preference falsification: expressing truthful political opinions to strangers outside their network while lying to “friends” in their network. I theorize that the behavior is attributable to a combination of psychological rewards for being truthful and social punishment for being a dissident. A consequence of the behavior, I posit, is discouragement of collective action, which stabilizes authoritarian rules. I test the theory with an original dataset I collected from Chinese social media. In addition, I develop two novel methods to analyze big social media data of political communication in authoritarian China and beyond. I develop ATIOS, a system based on distributed semantics that generates valid and replicable text-as-data measurement. I introduce Bayesian Dynamic Network Modeling, a method that efficiently models time series of social media networks. With this dissertation, I contribute new theories and methods for the study of contemporary Chinese politics and comparative political behavior.