Conceptualizing Political Orientation in Canadian Political Candidates: A Tale of Two (Correlated) Dimensions

Article excerpt

Political orientation is often operationalized as a unidimensional left-right continuum. However, some research suggests that this conceptualization might be overly simplistic. The present study examined the structure of political orientation in a sample of 190 politicians who were candidates in the 2006 Canadian federal election. Participants completed measures of attitudes toward specific political issues (social conservatism issues, economic competition issues), ideological beliefs (right-wing authoritarianism, social dominance orientation), and abstract values (conservation, self-enhancement) as indicators of political orientation. Confirmatory factor analyses demonstrated that the structure of political orientation was explained best by 2 moderately correlated dimensions: social left-right and economic left-right. Differences in the political orientation indicators between political parties are also discussed.

Keywords: political ideology, values, political attitudes, right-wing authoritarianism, social dominance orientation

The political orientations of individuals and of parties are frequently described in terms of the well-known left-versus-right continuum (e.g., Jost, 2006; Jost, Glaser, Kruglanski, & Sulloway, 2003a, 2003b; see also Knight, 1999). This left-right spectrum has been described as "the single most useful and parsimonious way to classify political attitudes for more than 200 years" (Jost, 2006, p. 654). In the United States, for example, self-placement on the left-right, or liberal- conservative, axis has consistently been strongly related to voting behaviour in national elections (Jost, 2006). This same dimension has also been found to be applicable in many other countries including Canada (Jost et al., 2003b). Moreover, this axis has shown a consistent pattern of correlations with a diverse array of psychological variables, including perceptions of mortality salience, perceptions of instability or threat to the social system, uncertainty avoidance, and openness to experience (see Jost et al., 2003b). In short, the left-right spectrum is particularly useful in measuring what Jost and his colleagues refer to as "core" political ideology, or the aspects of political ideology that are replicated across time and place. Despite the consistent ability of the left-right continuum to predict various criteria, however, there are several reasons to believe that the structure of political orientation is more complex than this single dimension, at least within certain samples. In the present study, we investigate the structure of political orientation in a novel sample - a group of political candidates running in the 2006 Canadian federal election. Our expectation is that political orientation amongst these individuals is best represented by a more complex structure than the ubiquitous left-right continuum.

Two Dimensions Associated With the Left-Right Spectrum

Notwithstanding the familiarity and the utility of the single left-right continuum, there is much evidence that this vector is itself a compound of two constructs. For example, Jost and his colleagues (Jost et al., 2003a; 2003b; Jost et al., 2007) have contended that political conservatism comprises two core components: resistance to change and acceptance of inequality. The view of the political spectrum as representing a blend of two underlying factors has a long history in political psychology research. To investigate the structure of political orientation in the present sample, we administered measures of the two dimensions that have been observed in three domains: attitudes toward specific issues, ideological beliefs, and social values. We describe each of these domains in this article.

Ferguson (1939) assessed attitudes toward specific political issues in the United States and reported two factors, which he called religiosity and humanitarianism. Later, Eysenck (1954) similarly found two orthogonal factors of conservatism (vs. …