Dynamic Populism Rhetoric Measurement and its impact on vote choice
No dia 8 de maio, o Grupo de Investigação SPARC organiza um seminário sobre o tema Dynamic Populism Rhetoric Measurement and its impact on vote choice. Catarina Pereira (University of Zurich) será a oradora convidada. A partir das 14h30, na Sala 2 do ICS-ULisboa e online (ID: 975 6636 2593, pass 124396).
Anti-elite feelings are one of the most important concepts used to understand the behaviour of parties and politicians and their interactions with multiple audiences. Despite a rich theoretical literature discussing anti-elite feelings, such as populism, we do not have a comparable measure across parties, between countries, and over time. I present a new strategy to measure populism from legislative speeches with word-embedding techniques. I introduce an original dataset on the rhetoric of 7 million speeches over 10 years, and across two countries—the U.K and Germany. I perform several validation tests and present an application of this multi-lingual method to investigate which component of populism matters for party choice. I find that the Manichean component is what mobilises voters the most. I created a measurement of moral polarisation and test whether partisanship and economic performance matters for rhetoric moral polarisation. Finally, I discuss how moral polarization in rhetoric can contribute to classical and new questions of party competition.
I am a Political Science PhD candidate at the University of Zurich (UZH), Switzerland. My research focuses on political competition, specifically on the use of rhetoric as a tool for parties to attract voters. I am interested in questions such as how populism can be identified in political speech. Under what conditions does populism successfully attract voters, and why? Methodologically, I primarily use machine learning and causal inference to answer these questions.





