A new framework for studying dynamic multi-party competition скачать в хорошем качестве

A new framework for studying dynamic multi-party competition

2021-12-08 | Input Talk | Denis Cohen Abstract Large literatures on party competition and voting behavior focus on voter reactions to parties' policy strategies, agency, or legislative performance. While many inquiries make explicit assumptions about the direction and magnitude of voter flows between parties, comparative empirical analyses of vote switching remain rare. In this talk, I present a new approach that overcomes three challenges that have previously impeded the comparative study of dynamic party competition based on voter flows: A newly compiled data set that marries comparative vote switching data with information on party behavior and party systems in over 200 electoral contexts across 36 OECD countries, a novel conceptual framework for studying how party behavior affects voter retention, defection, and attraction in multi-party systems, and a statistical model that renders this framework operable. An applied walkthrough showcases the data set and a newly developed R package for the estimation of the newly developed statistical model, along with functions for the calculation and visualization of substantively meaningful quantities of interest. Presenter(s) Denis Cohen is a postdoctoral fellow in the Data and Methods Unit at the Mannheim Centre for European Social Research (MZES), University of Mannheim. His research focus lies at the intersection of political preference formation, electoral behavior and political competition. His methodological interests include quantitative approaches to the analysis of clustered data, measurement models, data visualization, strategies for causal identification, and Bayesian statistics.