Partisan Intensification in Campaigns: Proof of Concept

• In Event: Partisanship in Comparative Perspective

Abstract

Partisan Intensification in Campaigns: Proof of Concept

Erikson and Wlezien (2012) outline a model to account for diminution over the course of a campaign in the responsiveness of electorates to external forces. The key is the intensification of preferences. Within temporal units, variances across individuals in preferences (indicated for instance by thermometer-type readings) increase. This has the effect of damping the response the aggregate of individuals around some threshold. Similarly, it damps the impact of a shift in the threshold, as might be induced by some campaign event. The model is simple and plausible, but still rests on conjecture. This paper moves beyond conjecture to model the evolution of such variance with thermometer-type data on individuals from true rolling cross-section designs in three different contexts, the US, Canada, and Germany. For the US, the data are from the National Annenberg Election Survey, 2000-8; for Canada, Canadian Election Studies from 1988 to 2921; for Germany, the German Longitudinal Election Study, 2005-17. This range of sources also enables us to consider the implications of the number of parties and the electoral formula as moderators of the basic relationship.

Author

• Richard Johnston, University of British Columbia