Integrating Large-Scale Online Surveys and Aggregate Data at the Constituency Level: The Estimation of Voter Transitions in the 2015 British General Elections

Abstract

What have been the underlying voter shifts that led to the victory of the Con- servative Party in the 2015 British general election – against all predictions by pollsters? Analyses of voter transitions based on (online) surveys and recall ques- tions are plagued by sampling and response biases, whereas aggregate data analyses are suspect of the well-known ecological fallacy. We propose a systematic statistical combination of individual and aggregate data at the constituency level to identify regional electoral shifts between the 2010 to 2015 British general elections, with a particular focus on England. Large-scale individual data collected by the British Election Study Internet Panel (BESIP) allow us to locate more than 28,000 respon- dents in their constituencies. We estimate transitions based on a recently developed Bayesian Hierarchical Hybrid Multinomial Dirichlet (HHMD) model. We discover a clear deviance from pure RxC ecological inference and from pure online panel- based estimations of transition matrices. Convergence diagnostics corroborate the superiority of the hybrid models.

Keywords

• voter transitions
• ecological inference
• British election study internet panel
• hybrid models