When populists become popular: comparing Facebook use by the right-wing movement Pegida and German political parties

ABSTRACT

Previous research has acknowledged the use of social media in political communication by right-wing populist parties and politicians. Less is known, however, about its pivotal role for right-wing social movements which rely on personalized messages to mobilize supporters and challenge the mainstream party system. This paper analyzes online political communication by the right-wing populist movement Pegida and German political parties. We investigate to which extent parties attract supporters of Pegida, to which extent they address topics similar to Pegida and whether their topic use has become more similar over a period of almost two years. The empirical analysis is based on Facebook posts by main accounts and individual representatives of these political groups. We first show that there are considerable overlaps in the audiences of Pegida and the new challenger in the party system, AfD. Then we use topic models to characterize topic use by party and surveyed crowdworkers to which extent they perceive the identified topics as populist communication. The results show that while Pegida and AfD talk about rather unique topics and smaller parties engage to varying degrees with the topics populists emphasize, the two governing parties CDU and SPD clearly deemphasize those. Overall, the findings indicate that the considerable attention devoted to populist actors and shifts in public opinion due to the refugee crisis have left only moderate marks in political communication within the mainstream party system.

KEYWORDS:

• Political communication
• populism
• protest movements
• party politics
• topic modeling
• Facebook

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Sebastian Stier is a postdoctoral researcher in the Department Computational Social Science at GESIS – Leibniz Institute for the Social Sciences in Cologne. He studies the role of the Internet in political communication relying on theoretical approaches and methods from Comparative Politics, Communication Sciences and Computational Social Science [email: [email protected]].

Lisa Posch is a doctoral student in the Department Computational Social Science at GESIS. Her research is focused on human intelligence tasks and topic models, and their application within the Computational Social Science domain [email: [email protected]].

Arnim Bleier is a postdoctoral researcher in the Department Computational Social Science at GESIS. His research interests are in the field of Computational Social Science. In collaboration with social scientists, he develops Bayesian models for the content, structure and dynamics of social phenomena [email: [email protected]].

Markus Strohmaier is a Full Professor of Web-Science at the Faculty of Computer Science at the University of Koblenz-Landau and scientific director of the Department Computational Social Science at GESIS. His main research interests include Web-Science, Computational Social Science and Data Science [email: [email protected]].

ORCID

Sebastian Stier http://orcid.org/0000-0002-1217-5778

Notes

1. In further research, we will extend our approach to positional competition.

2. There are indications that the newly created Pegida branches are more radical than the Dresden chapter (Vorländer et al., 2016, p. 69). However, since the accounts are listed and liked by the main Pegida site, they are clearly regarded as part of the movement by its leaders.

3. We relied on several data sources. We thank Martin Fuchs and his website Pluragraph for providing us with lists of the social media accounts of sitting parliamentarians in the federal parliament (Bundestag) and leading politicians of the non parliamentarian parties AfD and FDP. The list of AfD politicians also contains the candidates for the German federal election 2013 (Kaczmirek & Mayr, 2015) except the ones who have left the AfD and joined ALFA, the new party of AfD founder Lucke. To increase the share of messages coming from official party accounts, which we assume communicate more strategically than individual politicians, we included the accounts of the parties in the federal states.

4. The Facebook Graph API does not provide information on who likes Facebook pages themselves.

5. Pegida's account was deleted by Facebook on 22 July because of ‘instances of hate speech’. Therefore we could only conduct the behavioral analysis based on data retrieved in a previous data crawl. Figure 1 is consequently based on all unique users engaging with a post by one of the eight main accounts before February 20, 2016.

6. The results are robust when using the Jensen-Shannon divergence as a distance metric.

7. As a robustness test, we compared the cosine similarities between all groups in the two models with 50 and 100 topics resulting in a Spearman rank correlation of . The party specific topic distributions are therefore very similar independent of the number of topics.

8. https://www.crowdflower.com