It is widely accepted that survey data tend to underestimate the levels of support for political parties or other groups that are perceived to advocate ‘extreme’ views. In this Note I propose the use of latent class analysis as a means by which this difficulty might be overcome and I illustrate it by an application to the case of support within Northern Ireland for Sinn Féin from survey data and apply latent class analysis to a recent dataset from Northern Ireland. The note concludes with some general remarks on the question of why surveys and elections disagree and the role of latent class analysis in explaining this phenomenon.
No takes yet. Share an insight, caveat, or question.
Richard Breen (2000) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: