Feeling Europe: political emotion, knowledge, and support for the European Union

Abstract

Theories of political emotion suggest that feelings towards an issue or candidate are often better predictors for support than attitudes or preferences. We investigate whether this conjecture also holds for more abstract political entities, such as the European Union (EU), and test whether EU citizens’ feelings toward the EU are significant predictors of their EU support. We first review existing research and provide theory-driven propositions of how positive and negative emotion may influence EU-related attitudes. Second, using multilevel regression models fitted to Eurobarometer data, we estimate how feelings toward the EU are associated with support for the EU. In line with our hypotheses, analyses show that positive emotions are positively associated with EU-support, while negative affect is negatively associated with it. Contrary to some theoretical predictions, however, these effects are not mediated by individuals’ use of EU-related information.

Keywords:

• affect
• emotion
• cognition
• attitudes
• European Union

Acknowledgements

We thank Sjoerd Stolwijk and Leonie Steckermeier for valuable comments on statistical analyses.

Supplemental data

Supplemental data for this article can be accessed http://dx.doi.org/10.1080/13511610.2017.1398074

Notes

1 We provide detailed information on variables in the supplementary materials.

2 We conducted a series of robustness checks using separate items instead of the additive index in additional regression models reported in Tables S1 and S2 in the supplementary materials.

3 See Scherer (2005) for a discussion of the use of forced choice options in the measurement of emotion.

4 Since all emotion items are dichotomous categorical variables, the EFA was conducted using a matrix of polychoric correlations.

5 As a robustness check, we also conducted select regression analyses with all seven discrete emotions as independent variables, supporting the rationale of our additive indices (see Models 2 and 9 in Tables S1 and S2).

6 The same pattern is evident when looking at separate indicators of support for the EU in Tables S1 and S2. Results of the regression analyses show that positive affect increases the probability of supporting the political union and evaluating EU-membership positively (in comparison to the “neither” category), but decreases the likelihood of evaluating EU-membership as unfavorable (again, with “neither” as the reference category). In contrast, negative affect decreases the probability to support the political union and to evaluate EU-membership positively and increases the likelihood to evaluate EU-membership negatively. The same pattern is evident when considering discrete emotions: enthusiasm, trust, and hope increase the likelihood of supporting the political union and evaluating EU-membership positively, whereas indifference, mistrust, rejection, and anxiety decrease it. Looking at the unfavorable evaluation of EU-membership, enthusiasm (not significant), trust, hope and indifference decrease the odds of this evaluation, whereas mistrust, rejection, and anxiety (not significant) increase the odds. The inconsistent impact of indifference confirms its position at the low activity end of positive emotions: it neither increases the likelihood of a positive nor of a negative evaluation of the EU.

7 This pattern remains stable when looking at separate indicators of EU-support in Models 2–6 in Tables S1 and S2.

8 This is in line with recent research indicating that even when considering the impact of European identification on the EU-support, utilitarian considerations remain the strongest predictor (van Klingeren, Boomgaarden, and de Vreese 2013).