Understanding Interest Groups’ Information Networks in the EU Using Twitter Data

• In Event: New Methods for Studying European Politics

Abstract

The study examines the information networks established in the EU system of interest representation by identifying ‘who follows whom’ on Twitter within the community of EU private actors and lobbying organizations. Empirically, the analysis builds on the INTEREURO interest groups population dataset to which it adds original data gathered by web-scrapping interest groups’ Twitter accounts and information about their direct meetings with high-level European Commission officials available online. This allows identifying the presence of a directed information tie between organizations included in this comprehensive dataset and to examine the Twitter followers of each organization in the dataset, as well as the importance of their network ties with executive decision-makers in determining their centrality in the Twitter information network. Theoretically, the study tests a key argument made in the classic network theory whose empirical test in the EU policymaking setting is long due: that preference similarity is a strong predictor of a private actor’s decision about whom to follow/monitor in information networks, while accounting for the strength of ties an organization has with European executive decision-makers. Despite its prominence and intuitive appeal, this argument has not yet been tested to explore the formation of network ties within the whole EU lobbying community. I use two key proxy measures for preference similarity: the interest type represented by an organization and the country of its organizational headquarters. I examine the extent to which these two sources of preference similarity predict an interest group’s decision to follow another on Twitter, and control for the effects of network structural features. I test my theoretical argument with the help of exponential random graph models across 10 policy sectors, providing thus one of the very first comparative, cross-policy domain analyses of EU lobbying information networks. The findings show that preference similarity is indeed a strong predictor of an organization’s Twitter-following behaviour. Sharing the same country of organizational headquarters constitutes also a particularly strong predictor of forming a Twitter information tie, indicative of the importance of shared national politics, identities and interests when lobbying at supranational level. Interesting empirical variations are observed across policy sectors regarding (1) the Twitter following behaviour of interest organizations and (2) the pattern and frequency of their ties with European Commission officials, which are extensively discussed and explained in light of theories of policy networks in the context of multilevel governance.

Author

• Adriana Bunea, University of Bergen