120405_7892_shugars011 I am doctoral candidate studying American political behavior, political networks, and methodology in Northeastern’s Network Science program. My research focuses on developing natural language processing and network analysis techniques to better understand political dialogue and deliberation; modeling the way an individual reasons as a network of interconnected ideas and studying deliberation as process in which groups exchange ideas and collectively create new solutions.

View my full CV.

Microblog Conversation Recommendation via Joint Modeling of Topics and Discourse
Xingshan Zeng, Jing Li, Lu Wang, Nick Beauchamp, Sarah Shugars
Proceedings of the North American Chapter of the Association for Computational Linguistics (NAACL), June 2018

Games for Civic Renewal
Joshua A. Miller, Sarah Shugars, and Daniel Levine
The Good Society, June 2018

Winning on the Merits: The Joint Effects of Content and Style on Debate Outcomes
Lu Wang, Nick Beauchamp, Sarah Shugars, Kechen Qin
Transactions of the Association for Computational Linguistics (TACL), July 2017.

Why Keep Arguing? Predicting Participation in Political Conversations Online
Sarah Shugars and Nick Beauchamp
SAGE Open: Social Media and Political Participation Global Issue,

Dissertation Topic
Reasoning Together:
Network Methods for Political Talk And Normative Reasoning
Proposed May, 2018. Expected completion, Spring 2020.


  • Chair: Nick Beauchamp, Assistant Professor of Political Science, Northeastern University
  • Co-Chair: David Lazer, Distinguished Professor of Political Science and Computer and Information Science, Northeastern University
  • Lu Wang, Assistant Professor of College of Computer and Information Science, Northeastern University
  • Peter Levine, Associate Dean for Research, Tisch College of Civic Life, Tufts University

The slides from my proposal defense can be downloaded in Keynote or as a pdf. For those of you who have asked about my use of animation and transitions, note those can only be viewed in the original keynote file.