Social media, legislative discourse, and lobbying: Twitter/X networks of policy actors in the context of German law-making

Research highlights the potential role of social media platforms, and Twitter/X in particular, as forums for policy and professional discourse, allowing actors such as politicians, interest groups and experts, to connect directly with each other. This paper shifts the focus from more event-centered accounts of these networks to the study of day-to-day political processes (Stier et al. 2018). Using a random sample of federal government bills submitted to the German Bundestag in 2019, I compare the related Twitter discussions across a set of macro policy subjects during the 2017-2021 legislative period. When political elites, organizations and professionals regularly interact in the negotiation of policy, such exchanges come close to the notion of policy networks (Adam & Kriesi 2007). My study analyses the extent to which these online structures are shaped by factors identified in research on policy networks from the offline world, or by logics specific to social media (Kane et al. 2014). Drawing on interest group and policy studies as well as the literature on social media networks, I translate propositions about biases and alignments of actors into network configurations. Do networked discussions reflect the importance of business interests in accessing policymakers, or the traditional alliance of these interests with conservatives and liberals, while cause group actors are more likely to align themselves with left-wing politicians (e.g. Eising et al. 2019)? I expect that certain types of actors, such as sectional groups or government actors, will be central to the discussions, mediated by the policy area, and that networks will show homophilic tie formation within and, according to alignments, across certain actor types. Methodologically, I rely on Exponential Random Graph Models to test the antecedents of policy-related mention and retweet networks. The analysis includes comparisons both within and across policy areas, to examine policy as a contextual factor (Beyers et al. 2016). Macro topic areas summarize all the major topics from the Comparative Agendas Project (Baumgartner et al. 2019), and I compare networks (Faust & Skvoretz 2002) related to the subjects of economy, welfare, environment, as well as law and order (Breunig 2014), presumably three cases each. I used the Twitter API v.2 (Academic Research Track) to retrieve tweets and user data. The study also aims to contribute to research on political social media networks by focusing on multiple actors, both organizations and individual professionals, such as interest group representatives. Actor types are identified based on external ID lists of politicians’ Twitter/X accounts, a self-collected ID list of accounts covering the German interest group population, and hand-coding of profile descriptions. In addition, the paper adds to the study of interest groups by exploring social media activity as an in-between form of inside and outside lobbying (Figenschou & Fredheim 2019).