BackHome
Refresh: Off
Session Submission Summary
Share...

Direct link:

Conflict, Coercion, and Radicalization through the Lens of Text as Data

Fri, September 6, 12:00 to 1:30pm, Pennsylvania Convention Center (PCC), 111A

Session Submission Type: Full Paper Panel

Session Description

Text as Data methods are gaining increasing popularity among political scientists due to their potential in uncovering patterns and trends in large corpora with relatively high accuracy and efficiency. The panel aims to explore and evaluate how Text as Data methods can be employed to answer questions in the interrelated studies of conflict, coercion, and radicalization, the subfields of political science and international relations that still use Text-as-Data methods relatively rarely in comparison to other domains. The panel seeks to assess both the strengths and weaknesses of using Text-as-Data methodology in these subfields.

Paper 1: Aid, Alliances, Framing: Coevolution of International Ties and Terror Rhetoric

Existing scholarship accepts that states frame terrorism in ways which are compatible with their national interest. However, the extent to which this is conditioned by international networks is underexamined. To address this gap, the article uses à la carte word embedding techniques to derive 2,181 state-year measurements of terrorism rhetoric from 3,537 speeches delivered between 1994 and 2012 by representatives of 185 states to the United Nations General Assembly.

Paper 2: ‘I Say, You Say’: Transitional Justice and Normative Polarisation in Parliamentary Interactions
Scholars have analysed how different actors in post-conflict societies support or resist reckoning with war crimes, but have paid little attention to how these actors engage with each other. To address this gap, this paper uses an original dataset of speeches in the Serbian Parliament from 2003-2009 and applies computational text analysis, including topic models and sentiment measures, to study patterns of MPs’ discourse about transitional justice and their drivers.

Paper 3: Nationalism in Transnational Arenas: PRC Supporters’ Online Responses to Uyghur Activists-in-Exile

This paper examines how supporters of nondemocratic governments respond to prominent exiled dissidents on Twitter. The analysis draws on pro-PRC (People’s Republic of China) tweets constructed with digital traces of online responses to exiled Uyghur dissidents that are identified based on incidents reported in the China's Transnational Repression of Uyghurs (CTRU) dataset. The paper constructs the text corpus employed for analyses using a pro-PRC stance classifier based on GPT-4 architecture.

Paper 4: Co-optation and Coercion of Online Influencers: Evidence from Saudi Wikipedia

How do authoritarian regimes use co-optation and coercion of influential internet users to control online information? Drawing on a recent ban of Saudi Wikipedia users for coordinated inauthentic activity, the paper uses a two-way fixed effects design and quantitative text analysis of Wikipedia edits to evaluate how banned users’ behavior compares to the activity of non-banned users before and after their reported co-optation.

Paper 5: Cross-Platform Spread of Online Polarization

How news spreads and is consumed on social media and what individual characteristics or behaviors drive migration to radical echo chambers is a topic of much debate in political science. Using text classification of transcripts to label videos as political or non-political as well as by political leaning, the paper explores how and when individuals consume news online, whether partisan news behaviors are consistent across platforms, and whether certain individuals are consuming increasingly extreme content over time.

Sub Unit

Individual Presentations

Chair

Discussants