Franziska Barbara Keller

I am Assistant Professor at Hong Kong University of Science and Technology's Division of Social Science, and a Faculty Associate at its Institute for Emerging Market Studies, specializing in Comparative Politics and Social Network Analysis (SNA). Applying SNA to informal institutions in authoritarian regimes, I explain why some elites become part of the inner circle while others get purged, how leaders can rule without holding official positions, and why being well-connected is sometimes less important than having well-connected friends. More recently, I have examined political AstroTurfing (hidden propaganda campaigns) on social media. My research has appeared in Political Communication, the Journal of East Asian Studies, in the Proceedings of ICWSM, in Political Studies and the Journal of Intercultural Studies. I've also co-authored a textbook on SNA and contributed to the Palgrave Handbook of Political Elites.

In my research (Journal of East Asian Studies) I argue for conceptualizing elite politics as happening in networks instead of between factions. I show that it is possible to construct political elite networks in closed regimes using publicly available data, and evaluate different approaches to do so. In research awarded the John Sprague Award of APSA's Political Networks Section, I test a series of such more complex hypotheses on a network constructed from the career paths and promotion patterns of the top Chinese political elites. I show that while direct connections to alleged faction leaders raise the chance of entering the inner circle (the Politburo), so do ties to other elites. Central network positions predict appointments up to a decade ahead – without having to rely on insiders identifying powerful figures first. But not all friends in high places are helpful. In a bureaucracy riven with struggles such as China's, connections to anyone but the most powerful figure in a province may actually reduce promotion chances (Political Studies).

My second research focus is on hidden disinformation campaigns on social media. I examine one of the earliest known Twitter AstroTurfing campaign - implemented by the South Korean secret service during the 2012 Presidential election - and show that networks of coordinated messaging help us identify such campaigns and are the result of principal-agent problems theorized in social science. Such network patterns are present even in the most recent such campaigns, as we discuss in our contribution to the Washington Post.

I have received my PhD from New York University's Department of Politics in 2015, was a postdoctoral research fellow at Columbia University's Harriman Institute in 2015/2016 and a visiting scholar at UCSD's School of Global Policy & Strategy.