Federico Vegetti (University of Turin) (webinar)

Abstract: Scholarly interest in the several interpretations of the populist phenomenon has long focused on the communication dimension. In this respect, studies agree that social media are extremely conducive to the spread of populism, as they provide an environment where political actors may directly address and mobilize their supporters. This paper aims at clarifying the impact of populist discourse on social media by looking at the engagement from citizens on Facebook, a platform playing a central role in the public discourse in many countries. Specifically, the paper investigates which features of Facebook posts – namely the topic and the language used – are more likely to boost the number of reactions, shares and comments among social media users. We collected data including all the Facebook posts published by the official pages of all major political leaders and parties in France, Germany, Greece, Hungary, Italy, Poland, Turkey and UK from August 2019 to October 2020. We use text analysis techniques (structural topic modeling and dictionary approach) to assess for each post: (1) the topic (e.g. whether it refers to immigration, economy, Covid-19, and others) and (2) the prevalence of populist language (e.g. the usage of anti-elitist terms, captured using a validated multi-language dictionary). These predictors, extracted from the content analysis, are included in a set of multilevel models where the dependent variables are the number of likes, shares, and comments by the users. Our analysis sheds light on the relative importance of the content and the language in producing engagement among social media users.