Networked Legislatures: Identifying Legislative Coalitions

• In Event: Factions and Coalitions in Legislative Politics

Abstract

When studying such diverse topics as democratization (Ahmed 2013; Ziblatt 2017) and polarization (McCarty et al. 2006), political scientists have examined how legislators cooperate to pass policy and how patterns of alliance between legislators influence the agenda. We propose network analysis as a useful way to examine these coalitions such that the structure of alliances and voting decisions which underpin them are incorporated into the object of analysis. Using a unique dataset of roll call votes from the German Imperial Reichstag during the Wilhelmine Period, we lay out the process of creating networks from roll call data and finding coalitions from these networks. Then, using network measures like betweenness centrality, we demonstrate some advantages of retaining the structure within the object of analysis. When testing the quality of the method, network simulations show the coalitions to be statistically distinguishable from noise. And comparisons with scaling find that these networked coalitions are a useful alternative when the main research goal is to detect coalitions and to analyze their structures and compositions. The network approach discussed and tested in this paper offers a method uniquely suited to address a wide variety of critical research questions.

Authors

• Stephanie Chan, Princeton University

• Amel F. Ahmed, University of Massachusetts, Amherst