Town twinning and political support

ABSTRACT

Does town twinning foster citizens’ political support of the European Union? This study investigates the relation, which is maintained in politics and academics but still calls for empirical evidence. We distinguish between two levels of analysis, namely individual and context level, and differentiate in accordance with David Easton’s modes of political support specific from diffuse support. The corresponding hypotheses are tested with a unique dataset, which embraces survey findings of more than 12,000 respondents from 28 randomly selected municipalities in Germany. The main conclusions that emerge from the analysis are: First, town twinning does promote political support of the European Union on the individual level. Second, there are positive effects on specific and diffuse support. Thirdly, there is no relationship between local activity in twinning affairs and political support.

KEYWORDS:

• Town Twinning
• European Union
• Political Support
• Multi-Level Analysis
• Governance

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The correlations between the additive indexes and the factor scores (regression-based approach) are very high (r > .99). We apply additive indexes as these appear handier than factor scores.

2. We also check the construction of the two independent variables with a principal component analysis with all six items. In such a calculation, in a first step only one component with an eigenvalue greater than, 1’ is extracted. With an eigenvalue of .96, the second component is only slight below the threshold of ‚1’. Moreover, by taking account of this second factor, the explained variance increases from 60 to 76 percent. In addition, the scree plot of the eigenvalues of each component does also suggest to choose two components as it is only afterwards that a flattening of the curve does appear. For these reasons, a second principal component analysis had been conducted which extracted two components (eigenvalue > 0.9). Regardless of the kind of rotation (orthogonal or oblique), the results reveal a clear and plausible structure. Items for specific support do load on factor 1, items for diffuse support load on factor 2.

Additional information

Notes on contributors

Markus Tausendpfund

Markus Tausendpfund is a postdoctoral researcher in the Faculty of Humanities and Social Sciences at the University of Hagen, Germany. His main research interests are political attitudes, political sociology and comparative politics. More specifically, he focuses on political support and political knowledge. Previously, he has published in American Behavioral Scientist and the Journal of Experimental Political Science.

Lisa Schäfer

Lisa Schäfer is research assistant and Ph.D. candidate in comparative politics at the University of Hagen, Germany. Her current research focuses on comparative politics, Turkey-EU relations and European identity.