service navigation

EASP – European Association of Social Psychology

EASP Seedcorn Grant Report by Schumann, Ehrke and Dupont

31.03.2019, by Tina Keil in grant report

Sandy Schumann, University College London; Franziska Ehrke, University Koblenz-Landau; and Julia Dupont, University Koblenz-Landau

Sandy Schumann
Sandy Schumann

Selective Exposure and Populist Radical Right Attitudes: Assessing the Impact of Intergroup Contact

Support for populist radical right parties, such as the German ‘Alternative für Deutschland’ (AfD), has grown considerably in recent years. Previous research has highlighted that voter characteristics as well as social, economic and political conditions can explain this development. Less, however, is known about the role of individuals’ informational context. Our project addresses this gap in the literature and examined whether populist radical right attitudes become stronger over time due to selective exposure to belief congruent information, sources, and media. Specifically, we postulated that individuals who hold populist and exclusionist attitudes prefer information and sources that endorse popular sovereignty as well as anti-establishment, anti-minority, and anti-pluralist convictions. Populist attitudes further were thought to predict social media use and traditional media avoidance.

In addition, we assessed the extent to which selective exposure effects can be attenuated by frequent, positive contact with others who hold opposing political opinions as well as contact with ethnic minority groups members. It has been argued that threats to identity-central beliefs, for instance, perceived symbolic and realistic threats to one’s nation from external (e.g., refugees, immigrants) or internal out-groups (e.g., political opponents), enhance the selection of belief congruent information, sources, and media. In turn, efforts that reduce perceived threats—such as intergroup contact—should decrease preferences for selective exposure.

In order to investigate these hypotheses, we conducted a four-wave longitudinal survey study in Germany. We collected data between October 2018 and January 2019; the lags between waves were approximately three weeks. Doing so, we relied on an online opt-in access panel and oversampled respondents who indicated that they had sympathy for the German populist radical right party AfD as well as the populist left-wing party ‘die Linke’. Although we focused initially on populist radical right attitudes, we included the latter sub-sample to conduct robustness checks, generalise certain findings across the political spectrum (e.g., contact effects with political opponents), and explore populism as a thin ideology. The pre-registration of the main analyses is available here:

Thus far, results point to two important insights. First, we did not find support for belief congruent source selection but evidence of a reinforcing spiral of media selection and exposure effects. A latent curve analysis with structured residuals demonstrated that over nine weeks, populist attitudes facilitated social media use and social media use enhanced populist attitudes. Stronger endorsement of right-wing ideology was related with a stronger growth of populist attitudes; (social) media selection and exposure effects, however, were not contingent on political orientation. We are still in the process of analysing the open text answers that indicate which media content people consumed during the study period. Once completed, we will be able to examine selective exposure effects to belief congruent information. Findings further showed that intergroup contact quality but not quantity is associated with the growth of populist attitudes over time. Specifically, more positive contact with political opponents reduced the growth of populist attitudes (across the political spectrum). However, neither intergroup contact quantity nor quality moderated selective exposure (social) media effects. Intergroup contact effects on selective exposure will be examined again as soon as all media content data has been analysed.

Data is currently not yet publicly available, but we are keen to collaborate on joint projects with people who would like to work with the data. Please see an overview of all assessed variables here: and contact Sandy Schumann,, should you be interested in exploring a specific research question with the data.

Results of the study will be presented at the EASP meeting on Intergroup Communication in Bologna in June 2019. In addition, we submitted an abstract to this year’s meeting of the European Communication Research and Education Association. A first manuscript as well as a contribution for in-mind magazine are in preparation. Following the completion of the data analysis, we seek to develop a proposal to fund follow-up experimental studies in which belief congruent information exposure will be manipulated. We also aim to improve the survey design by introducing a more detailed method to assess information and source selection, for example, with a diary study.

The present study was initiated during a visiting fellowship that Sandy Schumann was awarded at the University of Koblenz-Landau, research focus ‘Communication, Media, Politics’ (KoMePol); KoMePol also partially funded the study. We further would like to thank Fabian Thomas, University Koblenz-Landau, for his support with the latent curve modelling.

Importantly, we are grateful to the EASP for their support of our work and thank Sibylle Classen in particular for helping us with some administrative hurdles.