News, misinformation and support for the EU: exploring the effect of social media as polarizing force or neutral mediators

As social media platforms have become a staple news source for many EU citizens, we build on extant research of news media effects on the perceived legitimacy of the EU to explore the polarizing effects a social media news ‘diet’ – and particularly the ‘fake news’ aspect of this - may have on public opinion about European integration. Through multilevel modeling of repurposed Eurobarometer data, we first investigate whether those relying on social media for news about the EU express less support for the EU overall and its migration policies compared to those relying on other news sources. As migration policies are a crucial case for a study of social media effects , due to their media salience and potential for polarization, we are interested in measuring the effects in periods where the chance of external factors like excessive migration inflows acting as omitted variables is as small as possible. We thus analyse surveys fielded after the Brexit vote (2016) and the waning of the migration crisis (2018). We test whether individual attitudes commonly found to predict Euroscepticism, such as low knowledge about the EU, become a more potent driver of Eurosceptic beliefs when they co-exist with a reliance on social media as a news source. This causal mechanism could be assumed due to how online debates frequently act to reinforce existing beliefs. We control for a range of variables common in the literature on public support for the EU (people's experience of feeling European, their knowledge of EU politics, trust in national institutions and already existing image of the EU), which are nonetheless rarely used when investigating the effects of social media use on trust. In a second step, we probe the link between social media news consumption, fake news and polarization by expanding the cross-sectional analysis with country-level analysis of the countries where the largest (Hungary) and smallest (Finland) percentage report encountering news they believe to be fake on at least a weekly basis. Our paper thus contributes to a still sparse literature on social media effects on support for European integration. Preliminary results of our study show no negative correlation between social media use and support for the EU. However, when we interact social media use with common predictors of Euroscepticism, we find positive – albeit not statistically significant - correlations between the interacted variables and supporting the EU. By controlling for an extensive number of variables that confound the correlation between social media use and support for the EU, we show that the polarizing effect often ascribed to social media may be less relevant in the case of the EU. This calls for further research into whether the effect of social media use on political polarization is overstated in general, or if features of the EU as a political system makes it a less likely target for polarization.