Social class, lefthight political orientations, and subjective class voting in provincial and federal elections

Abstract

Nous proposons ici une étude comparative des niveaux de vote de classe subjectif (vcs) au Canada, aux échelles provinciale et fédérale, au moyen des données de l'Etude électorale nationale de 1984. Sont considérées les différences interprovinciales, et l'influence du statut socio‐éonomique (sse) et de certaines variables psycho‐sociales sur le niveau du vcs. Notre analyse indique que le vcs est généralement plus élevé lors déections provinciales que fédérales. Il tend aussi àêtre élevé pour les deux types délections en Colombie Britannique, en Saskatchewan et au Manitoba, et très bas au Québec. Lorsqu'on ajoute des variables subjectives (i.e., le répondant se dit‐il de gauche ou de droite) à celles, objectives, reliées au sse la pertinence du vcs est grandement renforcée. De tels résultats appellent leurs propres questions, quant aux moyens de maximiser ou minimiser le vote de classe, et au statut théorique de ce dernier au sein des théories du vote. Using data from the 1984 Canadian National Election Study, we tested predictions about the comparative levels of subjective class voting (scv) in provincial and federal elections, interprovincial differences in scv, and the contributions of ses and selected social psychological intervening variables to the explanation of scv. We found, in general, more evidence of scv in provincial elections than in federal elections. Also, provincial and federal scv was generally highest in British Columbia, Saskatchewan and Manitoba, and lowest in Quebec. Finally, the addition of two social psychological variables (respondents' social class self‐placements and self‐descriptions on a left/right political orientation rating scale) to the standard objective ses indicators (occupation, income and education) greatly increased the total variance explained in scv. We conclude with some observations about the circumstances under which class voting will be maximized vs. minimized, and about the desirability of viewing the specialized problem of class voting within the larger context of voting theory.

Class Position, Class Ideology and Class Voting: Mobilization of Support for the New Democratic Party in the Canadian Election of 1984*

“…However, Ogmundson (1975a;1975b;19761, taking account of respondents' views of the class positions of parties rather than relying solely on "expert" ratings of their positions, provided evidence that what had appeared to be "classless" voting could be explained by the disagreement between academics and individual voters on the political orientation of the Canadian political parties. In so doing, he shifted attention tothe importance ofvoters'perceptions in understanding class voting (see also Lambert et al, 1987;Myles and Forcese, 1981). Similarly, a study of self-rating on aleft-right scale by Lambert et al (1986) showed that self-ratings were correlated with party preference, except where New Democrats were concerned, and that the strength of the correlation varied directly with the strength of party identification.…”

Neoliberalism, Class and Culture: The 2008 Federal Elections in Canada and the United States

“…Yet, while the Republicans and Conservatives are bourgeois parties, simple electoral math dictates that they must appeal to the majority of voters in the working and 'new middle' Aronowitz (1998) From The Ashes of the Old: American Labor and America's Future for an Analysis of American Exceptionalism. For Canada, readers are direct to Brodie and Jenson (1991) Crisis, Challenge and Change: Party and Class in Canada;Lambert et. al.…”

Education, occupation and vote in Canada, 1965–1984

“…Ogmundson (1975a) also pioneered the investigation of 'subjective class voting,' in which votes are defined as 'for the working class' or 'for the middle class' depending upon how the respondents rate the parties for which they voted. Using this approach, a Conservative vote can count as support for the working class and an NDP vote as support for the middle class (see also Lambert et al, 1987;Lambert and Hunter, 1979;Ogmundson and Ng, 1982). Different choices are possible, but some simplifications are necessary in order to detect and describe weak statistical relations among categorical variables in the presence of many disturbing factors.…”