This event is organized by the Schwartz Reisman Institute for Technology and Society.
Note: Event details may change. Please refer to the Schwartz Reisman Institute for Technology and Society’s events page for the most current information.
Our weekly SRI Seminar Series welcomes Semra Sevi, assistant professor of political science at the University of Toronto, cross-appointed with the Munk School of Global Affairs and Public Policy, and a Schwartz Reisman Institute faculty fellow. A leading scholar on electoral behaviour and political representation, Sevi uses experimental and survey methods to examine how citizens engage with democratic institutions.
In this talk, Sevi will present findings from a new working paper evaluating a chatbot-driven Voting Aid Application (VAA) designed to deliver balanced, personalized political information. Through a series of experiments with young adult participants, the study shows that these tools significantly increase users’ knowledge of party platforms—but have limited effects on how people evaluate or choose between parties. Sevi will discuss the implications of these results for civic education, AI-mediated political communication, and the design of democratic technologies.
Moderator: Avery Slater, Department of English & Drama
Location: Online
Talk title:
“Chatbot voting advice applications inform but seldom sway young unaligned voters”
Abstract:
Voting Advice Applications (VAAs) are interactive tools that communicate information about elections, yet their effectiveness in enhancing political knowledge and participation remains understudied. Moreover, traditional VAAs may disproportionately attract politically engaged users with already well-formed ideological views, limiting their potential to inform a broader and less engaged electorate. This paper introduces a novel “VAA Bot” that employs large language models (LLMs) and retrieval-augmented generation (RAG) to deliver balanced, personalized information drawn from official party platforms and public documents. We evaluate the VAA Bot's impact across three experimental studies aimed at young politically unaffiliated adults. The findings provide evidence that the VAA Bot improves knowledge of party stances on issues of great importance to each user. However, when we focus on respondents whose primary issue position aligns closely with one of the parties, we observe weaker effects on downstream outcomes such as vote preferences and party evaluations. These findings contribute to ongoing debates about the role of political information in shaping behavior and underscore both the promise and the limitations of LLM-based tools for civic learning.
Suggested reading:
Yamil Velez, Don Green, Semra Sevi, “Chatbot-Driven Voting Aid Applications Increase Knowledge about Party Positions but Do Not Change Party Evaluations,” APSA Preprints, working paper, 13 June 2025.
About Semra Sevi
Semra Sevi is an assistant professor in the Department of Political Science at the University of Toronto, cross-appointed with the Munk School of Global Affairs, and a faculty fellow at the Schwartz Reisman Institute for Technology and Society. Before coming to the University of Toronto, she was a Banting postdoctoral researcher in the Department of Political Science at Columbia University. She serves as associate editor at Research & Politics, and has published in journals including Canadian Journal of Political Science, Electoral Studies, Legislative Studies Quarterly, Political Science Research and Methods, and Politics & Gender. She earned her PhD in political science from l’Université de Montréal and holds an Honours BA and MA from the University of Toronto.
Sevi's research focuses on electoral behaviour, political representation, public opinion, electoral institutions, women and politics, and Canadian politics. She employs a variety of quantitative methods, including survey and experimental approaches. Sevi’s dissertation, What Voters Want: Identifying Voter Preferences for Candidates, examined how voters decide whom to support, with a focus on candidate characteristics such as gender, age, occupation, and incumbency. To pursue this research, she built the largest candidate-level database of Canadian federal elections since 1867, downloaded nearly 4,000 times on Dataverse. The dataset standardizes candidate names across time and includes variables such as riding identifiers, demographic characteristics, occupations, party affiliations, incumbency status, vote shares, and representation of Indigenous and LGBTQ2+ candidates. She received the 2022 SSHRC Impact Award in the Talent Category in part for this work.
About the SRI Seminar Series
The SRI Seminar Series brings together the Schwartz Reisman community and beyond for a robust exchange of ideas that advance scholarship at the intersection of technology and society. Seminars are led by a leading or emerging scholar and feature extensive discussion.
Each week, a featured speaker will present for 45 minutes, followed by an open discussion. Registered attendees will be emailed a Zoom link before the event begins. The event will be recorded and posted online.