Laurence Vardaxoglou is a PhD candidate at Université Paris 1 Panthéon-Sorbonne (Centre d'economie de la Sorbonne) and the Paris School of Economics. He has a fully funded position (CIFRE) with the research firm Verian (formerly Kantar Public), where he also works as a Research director. He previously worked for various WPP communication agencies in London and Paris. He has advised clients including the European Commission, the European Central Bank, and the French government on issues related to communication and public opinion.

Laurence is studying the impact of misinformation on public opinion and behaviours during elections, using survey data and experimental methods. His thesis comprises two single-authored chapters that rely on nationally-representative online surveys carried out during the 2022 French presidential election, and a third chapter based on a laboratory experiment using an abstract election setting.

Chapter 1: Free, fair, and fraudulent elections?

Public opinion about electoral fraud is well covered in the context of US politics. But less so in Europe, where misinformation that challenges the integrity of elections is becoming more commonplace. I carried out two nationally representative online surveys to explore these issues in the context of the 2022 French presidential election. In Study 1, I exploit a post-election survey to show that 19% of the public believed that the election was rigged. And that this is linked to belief in misinformation about electoral fraud. In Study 2, I conduct a randomised experiment to show that misinformation about voting machines significantly increases public concern about the integrity of the election, and damages trust in political institutions. The results highlight the fragility of trust in democracy in France.

Chapter 2: Voting under the influence of far right misinformation

Despite widespread concern, there is mixed evidence as to the effect of far right misinformation on voting behaviour. I conducted an online experimental survey (N = 3000) during the 2022 French presidential election, exposing some participants to misinformation in a randomly assigned treatment. I find that there is no effect of misinformation on voting intention for far right candidates, but rather a significant increase in support for centrist candidates. I provide evidence that these effects are mediated by positive changes in opinion with respect to the incumbent president Macron. The results suggest that misinformation can be understood as an example of negative communication, which has been shown to backfire for politicians.

Chapter 3: How to win votes and influence strategic voters (co-written with Lily SAVEY)

Voters often strategically desert their preferred candidate if they deem them too extreme to stand a viable chance of winning. To counter this extreme or populist candidates have in recent years shared misinformation that play up their chances of winning the election, especially under unfavourable electoral systems. We conduct a laboratory experiment using an abstract election setting to test the effect of misinformation about an extreme candidate being more viable. We find that misinformation not only increases the vote share of the extreme candidate, but it also significantly reduces the likelihood of strategic voting. The results shed new light on optimal strategies for extreme candidates in plurality elections.