Social Media Polarization in the US Election

Presentation of a research project as part of the "Jeudis de la Maison Suger", a residents' research seminar.

Michael Bailey (in residence until April 2023) will speak on the "Social Media Polarization in the US Election".

The research project

American politics has become polarized in recent years to a disturbing degree. This project uses text-as-data tools to analyze the language used on social media by all 2020 congressional candidates to assess ideology and polarization. We show that the algorithm is able to convert the language into reasonable measures of ideology, identifying the most polarizing terms. We also show that the most extreme candidates did worse in the 2020 general election. Why then do candidates use such polarizing language? We see signs that they are raising more money and gaining more followers. This talk should appeal to those interested in American politics, political polarization, or automated text analysis.

Michael A. Bailey is the Colonel William J. Walsh Professor of American Government in the Department of Government and McCourt School of Public Policy at Georgetown University. Bailey teaches and conducts research on public opinion, American politics and political economy. He directs the Data Science for Public Policy program at the McCourt School. He is working on three projects:

• a book that calls for a new paradigm for public opinion polling;
• analysis polarization in elections using social media;
• directing the Environmental Impact Data Collaborative, a platform for data sharing and code collaboration.

Bailey received M.A. in Economics and Ph.D. in Political Science from Stanford University. He is also conversational in Japanese and interested in Japanese politics.

→ See his researcher page