This article lays out the elementary logic of age structures in party preference data and proposes a simple estimation model with demographic and historical elements. As voters age their preferences intensify. But they do not intensify much and generational differences in the direction of party preferences are correspondingly weak. The Canadian electorate does not seem all that strongly anchored by the accumulated experience of the individuals that make it up. The major source of long-term electoral change, therefore, is conversion in the existing electorate. Consideration is given to how distinctive the Canadian pattern is.
* Department of Political Science, The University of British Columbia. This is a greatly revised version of a paper presented at the Conference on Political Generations and Political Change, Université Laval, Quebec City, Quebec, 19–20 June 1986, and published as ‘Générations politiques et changement électoral au Canada’ in Jean Crête and Pierre Favre, eds, Générations et Politique (Paris: Economica; and Quebec: Les Presses de L'Université Laval, 1989). The data were furnished by the Interuniversity Consortium for Political and Social Research and by the University of British Columbia Data Library. Computer assistance was provided by the University of British Columbia and the California Institute of Technology computing centres. The author wishes to thank André Blais, Henry Brady, Jean Crête and Daniel Latouche for comments and advice. None of these individuals or institutions is responsible for any errors of analysis or interpretation in this article.