THE 2019 EP ELECTION CAMPAIGN AS SEEN FROM TWITTER

The end of May 2019 saw European citizens go to the polls to choose a new European Parliament (EP). These elections marked a crucial moment for European democracy, as they took place in a context of significant challenges and upheavals, including the aftermath of the economic and financial crisis that had impacted the EU in the preceding decade, the refugee crisis that had tested the EU's capacity to manage migration and asylum, and the departure of the United Kingdom from the EU. These events had far-reaching implications for the political landscape of the EU, and the election campaign was shaped by the issues and concerns that they raised, as well as the parties and candidates who sought to address them. Against this backdrop, the 2019 EP election campaign was marked by a high degree of contentiousness and uncertainty, with many parties and candidates seeking to position themselves in response to the shifting political and economic landscape Moreover, the 2019 EP election campaign was characterized by the significant role of social media in political communication, a phenomenon that had not been as prominent in previous campaigns. This transformation was due to the widespread dissemination of digital communication technologies, which had granted European citizens unprecedented access to information and new opportunities for political participation. As a result, political parties and politicians have greatly increased the number of available communication channels in order to disseminate information online and mobilize party members and supporters, making the use of social media a critical aspect of election campaign activity. Virtually every political party in Europe had a presence on social media, highlighting the importance of these platforms in modern-day political communication. The literature has recognized the significance of social media in shaping the outcome of election campaigns and has examined its role in creating new opportunities for political participation, promoting transparency and responsiveness between citizens and their representatives. An online election campaign offers a “window of opportunity” where behaviors and interactions between parties, candidates and citizens can be observed in condensed form. In the case of European Parliament elections, the view can embrace 28 different countries simultaneously in the same campaign context and in the same digital arena. Despite its many promises however the online dimension of EP campaigns is still relatively understudied. In this dissertation I present three distinct studies, each addressing a different dimension of the multifaceted and complex phenomenon of the 2019 EP online campaign. In the first chapter I examine the role of candidates and parties in online election campaigns, given the increasing importance of social media platforms. Specifically, the study investigates the prevalence of Twitter usage among candidates for the 2019 European Parliament election, as well as the factors that influence adoption and use of the platform. The study takes a comparative approach, analyzing data from all 28 member states of the European Union, and it aims to contribute to the ongoing “equalization vs. normalization” debate on the impact of social media on election campaigns. In particular, I investigate whether social media can be viewed as a force that equalizes the resource gaps between candidates and parties, or its use is being normalized into the EP campaign. To accomplish this, the study considers both individual-level and party-level factors that may influence Twitter adoption and use during the EP 2019 campaign. Chapter 2 investigates the dynamic nature of online communication between European Parliament (EP) candidates and their audiences during the 2019 EP campaign in the five largest Western European countries: France, Germany, Italy, Spain, and the United Kingdom. The study uses online communication data from both candidates and regular users to investigate the question of "who leads whom" in the communication process. Drawing on the literature on competing principals, the study explores who the "principal" is in the online communication between candidates and their audiences. Rather than relying on self-reported policy positions or roll-call votes, the study proposes a text-based measure at the level of individual tweets to determine the distance between the positions of political elites and those of their audience. The measure is then used to assess the extent to which EP candidates and Members of the European Parliament (MEPs) are faced with competing pressures from their audience, their party, and their EP group. The findings of the study contribute to the literature at the intersection of EU studies and legislative studies that focuses on the role of MEPs as agents with multiple principals. Additionally, the study contributes to the debate on the potential of social media to promote responsiveness and transparency between citizens and their representatives. In Chapter 3, the focus shifts towards the audience of the Twitter campaign, analyzing how the public reacts to campaign content by examining the replies to candidates' Twitter messages. A significant concern in recent years has been the rise in animosity and hostility across partisan lines, which has developed in parallel with the changes in media environments. The investigation covers campaign communication originated by the candidates of four EU member states: France, Italy, Spain, and the United Kingdom. The study addresses how interactions between candidates and the public might fuel reinforcing spirals of polarization dynamics along partisan lines, highlighting patterns of affective polarization in the context of the EP 2019 election campaign. Research on affective polarization has typically relied on survey questionnaires or experimental settings to measure partisan animosity and positive partisanship. However, most existing studies lack a comparative perspective and tend to focus on the mass level, thereby neglecting the crucial aspect of the interaction between citizens and political elites. To address these gaps, the study leverages behavioral data from Twitter to test how citizens react to elite messages in the context of an election campaign. A text-as-data approach is developed to analyze Twitter communication from more than 350,000 Twitter users, collected as they interact with hundreds of political candidates in the four countries. This measure is then used to unpack the mechanisms behind affective polarization and to shed light on its main drivers.