1 Schwartz, Mildred, Politics and Territory (Montreal: McGill-Queen's University Press), 1974CrossRefGoogle Scholar; Elkins, David J., “The Perceived Structure of the Canadian Party Systems,” this JOURNAL 7 (1974), 397–437Google Scholar; Wilson, John, “The Canadian Political Cultures: Toward a Redefinition of the Nature of the Canadian Political System,” this Journal 7 (1974), 438–83Google Scholar.

2 On the need for studies of the attitudes and beliefs of Canadian political elites see Shiry, John, “Mass Values and System Outputs: A Critique of an Assumption of Socialization Theory,” in Pammett, Jon H. and Whittington, Michael S. (eds.), Foundations of Political Culture: Political Socialization in Canada (Toronto: Macmillan, 1976)Google Scholar, chap. 2.

3 See, for example, Porter's, John classic, The Vertical Mosaic (Toronto: University of Toronto Press, 1965)CrossRefGoogle Scholar. See also the more recent literature on “consociational democracy” in Canada: Noel, S. J. R., “Consociational Democracy and Canadian Federalism,” this Journal 4 (1971), 15–18Google Scholar; Presthus, Robert, Elite Accommodation in Canadian Politics (Toronto: Macmillan, 1973)Google Scholar; Presthus, Robert, Elites in the Policy Process (London: Cambridge University Press, 1974)Google Scholar; McRae, Kenneth, Consociational Democracy (Toronto: McClelland and Stewart, 1974)CrossRefGoogle Scholar.

4 A few relevant empirical studies do exist. See Chi, N. H., “Class Voting in Canadian Politics,” in Khruhlak, O. et al. (eds.), The Canadian Political Process (2nd ed., Toronto: Holt, Rinehart and Winston, 1973), 226–47Google Scholar; Stein, Michael B., The Dynamics of Right-Wing Protest (Toronto: University of Toronto Press), 1973Google Scholar; Presthus, Elite Accommodation, chaps. 11, 12; Kornberg, Allan, Mishler, William and Smith, Joel, “Political Elite and Mass Perceptions of Party Location in Issue Space: Some Tests of Two Positions,” British Journal of Political Science 5 (1975), 153–77CrossRefGoogle Scholar.

5 Basic studies of the CCF include Lipset, Seymour Martin, Agrarian Socialism (Garden City: Doubleday, 1968)Google Scholar; Young, Walter, The Anatomy of a Party (Toronto: University of Toronto Press, 1969)Google Scholar. On the Social Credit party see MacPherson, C. B., Democracy in Alberta: Social Credit and the Party System (Toronto: University of Toronto Press, 1953)Google Scholar; Irving, John A., The Social Credit Movement in Alberta (Toronto: University of Toronto Press, 1959)Google Scholar; Pinard, Maurice, The Rise of a Third Party (Englewood Cliffs: Prentice-Hall, 1971)Google Scholar; Stein, The Dynamics of Right-Wing Protest. On the decline of ideology in the CCF see Zakuta, Leo, A Protest Movement Becalmed (Toronto: University of Toronto Press, 1964)Google Scholar.

6 Alford, Robert R., Party and Society (Chicago: Rand McNally, 1963), 13Google Scholar; Porter, The Vertical Mosaic, 373; Horowitz, Gad, “Conservatism, Liberalism and Socialism in Canada: An Interpretation,” Canadian Journal of Economics and Political Science 32 (1966), 144–71CrossRefGoogle Scholar; Mallory, J. R., “The Structure of Canadian Politics,” in Thorburn, Hugh(ed.), Party Politics in Canada (2nd ed.. Scarborough: Prentice-Hall, 1967), 26Google Scholar; Engelmann, Frederick C. and Schwartz, Mildred A., Political Parties and the Canadian Social Structure (Scarborough: Prentice-Hall, 1967)Google Scholar, Kornberg, Allan, Canadian Legislative Behavior (New York: Holt, Rinehart and Winston, 1967), 33Google Scholar; Christian, William and Campbell, Colin, Political Parties and Ideologies in Canada (Toronto: McGraw-Hill Ryerson, 1974), ixGoogle Scholar.

7 An exception is Wilson's “Canadian Political Cultures,” See also Engelmann, F. C. and Schwartz, M. A., Canadian Political Parties: Origin, Character, Impact (Scarborough: Prentice-Hall, 1975)Google Scholar, chap. 15.

8 In addition to the works cited in note 1 above, see Pammett, Jon H., “Public Orientations to Regions and Provinces,” in Bellamy, D. J. et al. (eds.), The Provincial Political Systems: Comparative Essays (Toronto: Methuen, 1976), 86–99Google Scholar.

9 Edinger, Lewis J. and Searing, Donald, “Social Background in Elite Analysis: A Methodological Inquiry,” American Political Science Review 61 (1967), 428–45CrossRefGoogle Scholar.

10 Some relevant data are contained in Presthus, Elite Accommodation, chaps 11 and 12. Considerable attention to the linkages between federal legislators' role orientations (as opposed to basic belief system variables) is paid by Komberg in his Canadian Legislative Behavior, passim. Komberg's limited exploration of MPs' positions on a “welfare state” index (121–26) does not contain an analysis in terms of sociodemographic characteristics. It might also be noted that similar to most “elite theorists” Porter, in The Vertical Mosaic, assumes that social background variables are important determinants of elite beliefs and behaviour.

11 For an example of a study taking this approach to studying ideology at the elite level in Canada, see N. H. Chi's “Class Voting in Canadian Politics,” 25–36. See also Komberg et al., “Political Elite and Mass Perceptions.”

12 For previous studies utilizing similar scales at the mass and/or elite levels, see Laponce, Jean A., “Note on the Use of the Left-Right Dimension,” Comparative Political Studies 2 (1970), 481–502CrossRefGoogle Scholar; Barnes, Samuel, “Left, Right, and the Italian Voter,” Comparative Political Studies 4 (1971), 157–76CrossRefGoogle Scholar; Barnes, Samuel and Pierce, Roy, “Public Opinion and Political Preferences in France and Italy,” Midwest Journal of Political Science 15 (1971), 643–60CrossRefGoogle Scholar; Klingemann, Hans D., “Testing the Left-Right Continuum on a Sample of German Voters,” Comparative Political Studies 5 (1972),. 93–106CrossRefGoogle Scholar; Laponce, Jean, “In Search of Stable Elements of the Left-Right Landscape,” Comparative Politics 4 (July 1972), 455–76CrossRefGoogle Scholar; Putnam, Robert, The Beliefs of Politicians (New Haven: Yale University Press, 1973)Google Scholar; and Finlay, David J. etal.,“The Concept of Left and Right in Cross-National Research,” Comparative Political Studies 7 (1974), 209–21CrossRefGoogle Scholar. For Canadian data see Meisel, John, Working Papers on Canadian Politics (Montreal: McGill-Queen's University Press, 1972)CrossRefGoogle Scholar, chap. 2; Elkins, “Perceived Structure,” passim.

13 Laponce, “Note on the Use of the Left-Right Dimension,” 195.

14 Kornberg, Mishler and Smith, “Political Elite and Mass Perceptions,” 169.

15 The term “centre-left” rather than “left-centre” is used to indicate which category on the left-right scale has the largest percentage of Liberal MLAs. Similarly, the PCs are labelled “right-centre” rather than “centre-right.”

16 The weakness of the relationship between ideological self-perceptions and region is evidenced by summary measures of association, for example, Cramer's V = .14; Goodman and Kruskal's Tau = .01.

17 Concerning policy and more basic ideological differences between parties with the same label in different provinces, see the various essays in Robin, Martin (ed.), Canadian Provincial Politics (Scarborough: Prentice-Hall, 1972)Google Scholar. Regarding the Saskatchewan Liberal party in particular (which makes a substantial contribution to our sample of Prairie Liberals), see Smith, David E., Prairie Liberalism: The Liberal Party in Saskatchewan 1905–71 (Toronto: University of Toronto Press, 1975)Google Scholar, chaps. 8 and 9.

18 See especially LeDuc, Lawrence Jr. and White, Walter L., “The Role of Opposition in a One-Party System: The Case of Ontario,” this Journal 7 (1974), 86–100Google Scholar.

19 In particular, the theme of class conflict descriptions of political culture in British Columbia is consistent with this finding. See, for example, Martin Robin, “British Columbia: The Politics of Class Conflict,” Canadian Provincial Politics, 27–68; Gordon S. Galbraith, “British Columbia,” in Bellamy et al. (eds.), Provincial Political Systems, 63–75. On the absence of ideological conflict in the Atlantic provinces see J. Murray Beck, “The Party System in Nova Scotia: Tradition and Conservatism,” in Robin (ed.), Canadian Provincial Politics, 168–97; P. J. Fitzpatrick, “New Brunswick: The Politics of Pragmatism,” in Robin (ed.), Canadian Provincial Politics, 116–33; D. J. Bellamy, “The Atlantic Provinces,” in Bellamy et al. (eds.), Provincial Political Systems, 3–18.

20 That holding an “extreme” ideological position and the degree of importance ascribed to ideological positions would be related seems quite plausible. The existence and strength of such a relationship is, of course, an empirical question. On the need to investigate such relationships rather than merely assuming their existence see Putnam, Beliefs of Politicians, 49–63.

21 See, for example, Blake, Donald E., “The Measurement of Regionalism in Canadian Voting Patterns,” this Journal 5 (1972), 55–81Google Scholar.

22 For a detailed description of AID, see Sonquist, John A. et al. , Searching for Structure (Ann Arbor: Institute for Social Research, 1973)Google Scholar; and Sonquist, John A., Multivariate Model Building (Ann Arbor: Institute for Social Research, 1970)Google Scholar. Sonquist summarizes the operation of the AID programme thus: “The [AID] technique is a step-wise application of a one-way analysis of variance model. Its objective is to partition the sample into a series of non-overlapping sub-groups whose means explain more of the variation in the dependent variable than any other such set of subgroups” (20). By operating in this manner, AID avoids the linearity and additivity assumptions of conventional regression analyses, and provides information about possible interaction effects. AID has the added advantage of accepting predictor variables measured at nominal orordinal as well as interval levels. Although, strictly speaking, the dependent variable in an AID programme should be an interval scale, here the dependent variable is ideological self-perception, an ordinal measure. For purposes of the AID analysis, MLAs placing themselves on the left are scored I, moderately left 2, centre 3, moderately right 4, and right 5.

23 For purposes of the AID analysis these variables were measured as follows: religious affiliation—Protestant, Catholic, Other, None; income—$9,999 and under, $10,000–19,999, $20,000 and over; age—40 and under, 41–50, 51–60, 61 and over; urbanization—size of community of residence—100,000 and over (large city or metropolitan area), 10,000–99,999 (small city), 1,000–9,999 (town), less than 1,000 (village or rural area).

24 For a concise statement of the “brokerage” theory and references to relevant literature see Sniderman, Paul et al. , “Party Loyalty and Electoral Volatility: A Study of the Canadian Party System,” this Journal 7 (1974), 268–78Google Scholar.

25 Chi, “Class Voting in Canadian Politics”; Kornberg et al., “Political Elite and Mass Perceptions of Party Location in Issue Space.”

26 The reader should recall that the dependent variable is MLAs’ self-placement on the left-right continuum. Whether greater regional effects would emerge if MLAs had been requested to order the ideological positions of parties in their particular province is a moot point. Earlier research with elites and/or mass publics in Canada by Komberg et al. and Meisel involved the rank-ordering of parties along issue and left-right continua respectively. See Komberg et al., “Political Elite and Mass Perceptions,” 165–66, and Meisel, John, Working Papers on Canadian Politics (2nd ed., Montreal: McGill-Queen's University Press, 1975), 66Google Scholar.

27 See, for example, the suggestive analysis of the rise of the Union nationale by Pinard, Maurice in “Working Class Politics: An Interpretation of the Quebec Case,” Canadian Review of Sociology and Anthropology 7 (1970), 87–109CrossRefGoogle Scholar.

28 Preliminary analyses with the present data suggest that MLAs’ ideological selfperceptions have statistically significant independent effects on purposive and representational role orientations as well as the amount of time devoted to constituency service as opposed to other types of activities. Limitations of presently available data do not permit an in-depth analysis of ideology-behaviour relationships.