1 Milbrath, Lester W. and Goel, M. L., Political Participation: How and Why Do People Get Involved in Politics? (2nd ed.; Chicago: Rand McNally, 1977), 57–58Google Scholar. The research on efficacy, trust and alienation in the United States is comprehensively reviewed in Wright, James D., “Political Disaffection,” in Long, S. L. (ed.). The Handbook of Political Behavior, Vol. 4 (New York: Plenum Press, 1981)Google Scholar, chap. I. For Canadian studies on political efficacy and/or trust, see Manzer, Ronald, Canada: A Socio-Political Report (Toronto: McGraw-Hill Ryerson, 1974), 304–23Google Scholar; LeDuc, Lawrence Jr, “Measuring the Sense of Political Efficacy in Canada: Problems of Measurement Equivalence,” Comparative Political Studies 8 (1976), 490–500CrossRefGoogle Scholar; Sinclair, Peter, “Political Powerlessness and Sociodemographic Status in Canada,” Canadian Review of Sociology and Anthropology 16 (1979), 125–35CrossRefGoogle Scholar; Grabb, Edward G., “Relative Centrality and Political Isolation: Canadian Dimensions,” Canadian Review of Sociology and Anthropology 16 (1979), 343–55CrossRefGoogle Scholar. and “Class, Conformity and Political Powerlessness,” Canadian Review of Sociology and Anthropology 18 (1981), 362–69Google Scholar; Simeon, Richard and Elkins, David J., “Provincial Political Cultures in Canada,” in Elkins, D. J., Simeon, R., et al.. Small Worlds: Provinces and Parties in Canadian Political Life (Toronto: Methuen, 1980), 31–76Google Scholar: Brown, Steven D. and McMenemy, John, “Generality or Specificity in Political Orientations: A Case Study,” this JOURNAL 15 (1982), 365–76Google Scholar: Kornberg, Allan. Mishler, William and Clarke, Harold D., Representative Democracy in the Canadian Provinces (Toronto: Prentice-Hall of Canada, 1982), 78–84Google Scholar: Clarke, Harold D., Kornberg, Allan and Stewart, Marianne C., “Parliament and Political Support in Canada.” American Political Science Review 78 (1984), 452–69CrossRefGoogle Scholar; Apostle, Richard, Kasdan, Leonard and Hanson, Arthur, “Political Efficacy and Political Activity Among Fishermen in Southwest Nova Scotia: A Research Note,” Journal of Canadian Studies 19 (1984), 157–65CrossRefGoogle Scholar; Johnston, Richard, Public Opinion and Public Policy in Canada, Studies of the Royal Commission on the Economic Union and Development Prospects for Canada, Vol. 35 (Toronto: University of Toronto Press, 1986), 23–24.Google Scholar

2 Almond, Gabriel A. and Verba, Sidney, The Civic Culture (Princeton: Princeton University Press, 1963), 257CrossRefGoogle Scholar. See also Campbell, Angus, Gurin, Gerald and Miller, Warren E., The Voter Decides (Evanston, III.: Row, Peterson, 1954), 187.Google Scholar

3 For example, see Balch, George I., “Multiple Indicators in Survey Research: The Concept ‘Sense of Political Efficacy,’” Political Methodology 1 (1974), 1–43Google Scholar; Craig, Stephen C., “Efficacy, Trust, and Political Behavior: An Attempt to Resolve a Lingering Conceptual Dilemma,” American Politics Quarterly 7 (1979), 225–39CrossRefGoogle Scholar; Craig, Stephen C. and Maggiotto, Michael A.. “Measuring Political Efficacy,” Political Methodology 8 (1982), 85–109Google Scholar;Acock, Alan, Clarke, Harold D. and Stewart, Marianne C., “A New Model for Old Measures: A Covariance Structure Analysis of Political Efficacy,” Journal of Politics 47 (1985), 1062–84.CrossRefGoogle Scholar

4 See, especially, Balch, “Multiple Indicators,” 23–28; Acock etal., “A New Model.”

5 Craig, “Efficacy, Trust,” 226.

6 Using data from the 1974 Canadian National Election Study, Kornberg, Clarke and LeDuc found that people who strongly identified with the Liberal party, then in power in Ottawa, were more likely to score higher on a feeling thermometer toward the federal government, than were supporters of other parties. They interpreted these findings by supposing that the “average citizens” in their study probably did not distinguish between “regime” and the “government of the day.” especially since the Liberal party had “formed the national government for 34 of the last 40 years.” See Kornberg, Allan, Clarke, Harold D. and LeDuc, Lawrence, “Some Correlates of Regime Support in Canada,” British Journal of Political Science 8 (1978), 201, 208CrossRefGoogle Scholar: see also Simeon and Elkins, Small Worlds, 58, 62. In his 1986 book. Public Opinion. Johnston subsumed efficacy under the rubric of “political support,” along with a wide variety of cognate variables. However, he did not take respondents' party attachments into account in his efforts to explain political support. In the United States, Aberbach found some evidence in 1958 and especially in 1964 for the partisan effect on the trust variable. These effects were demonstrated without benefit of statistical controls. Also in the United States, studies by Miller on political cynicism and by Wright on efficacy and trust have failed to find any effects for partisanship. See Aberbach, Joel D., “Alienation and Political Behavior,” American Political Science Review 63 (1969). 94CrossRefGoogle Scholar; Miller, Arthur H., “Political Issues and Trust in Government: 1964–1970,” American Political Science Review 68 (1974), 964CrossRefGoogle Scholar; Wright, James D., The Dissent of the Governed: Alienation and Democracy in America (New York: Academic Press, 1976), 153–55.Google Scholar

7 Citrin, Jack, “Comment: The Political Relevance of Trust in Government,” American Political Science Review 68 (1974), 976–977CrossRefGoogle Scholar. Compare Jack Citrin, McClosky, Herbert, Shanks, J. Merrill and Sniderman, Paul M., “Personal and Political Sources of Political Alienation,” British Journal of Political Science 5 (1975), 10–12.Google Scholar

8 See, for example, Cole, Richard L., “Toward a Model of Political Trust: A Causal Analysis,” American Journal of Political Science 17 (1973), 813; Balch, “Multiple Indicators,” 19–23: Craig, “Efficacy, Trust,” 234–36; compare, Craig and Maggiotto, “Measuring Political Efficacy,” 96–98. Although our hypothesis assumes that perceived system responsiveness and political trust speak to the same condition, other studies have capitalized on the relative independence of efficacy and trust. See, for example: Simeon and Elkins, Small Worlds, 42–46CrossRefGoogle Scholar; Gamson, William A., “Discontent and Trust,” in his Power and Discontent (Homewood, III.: Dorsey Press, 1968), 48Google Scholar; Paige, Jeffery M., “Political Orientation and Riot Participation,” American Sociological Review 36 (1971), 812.CrossRefGoogle ScholarPubMed

9 Citrin, “Comment,” 976, 977.

10 Coleman and Davis, “Structural Contexts,” 190–94: Craig and Maggiotto, “Measuring Personal Efficacy,” 86.

11 Compare, Kornberg, Clarke and LeDuc, “Some Correlates,” 201, 208.

12 These might be reflected, for example, in the dramatic changes in the parties' standings in the polls during the month prior to election day. See B. J. Kay, S. D. Brown, J. E. Curtis, R. D. Lambert and J. M. Wilson. “The Character of Electoral Change: A Preliminary Report from the 1984 National Election Study,” paper presented at the Annual Meetings of the Canadian Political Science Association, Montreal, 1985, Figure I; Frizzell, Alan and Westell, Anthony, The Canadian General Election of 1984: Politicians, Parties, Press and Polls (Ottawa: Carleton University Press, 1985), chap. 4. On the relationship of party identification to efficacy and trust, see Simeon and Elkins, “Provincial Political Cultures,” 64–66.Google Scholar

13 Craig and Maggiotto, “Measuring Personal Efficacy.” 87; Kornberg et al., Representative Democracy, 82.

14 See, for example. Milbrath and Goel. Political Participation. 57–66: Abramson, Paul R. and Aldrich, John H.. “The Decline of Electoral Participation in America.” American Political Science Review 76 (1982), 502–21CrossRefGoogle Scholar: Guterbock, Thomas M. and London, Bruce, “Race, Political Orientation, and Participation: An Empirical Test of Four Competing Theories,” American Sociological Review 48 (1983), 439–53.CrossRefGoogle Scholar

15 Lambert, R. D., Brown, S. D., Curtis, J. E., Kay, B. J. and Wilson, J. M.. 1984 Canadian National Election Study Codebook (Waterloo. Ont., 1986).Google Scholar

16 We wished to establish that our findings were not distorted by the practice of substituting the mean value for three items when the fourth item had produced a missing value. We therefore repeated all of the analyses for each of the trust scales only for those respondents who had answered all four items comprising the scale. These analyses yielded essentially the same pattern of results described in this study.

17 The partisanship variable was constructed from party identification rather than how people voted so that we could test the predictions involving strength of party identification. We performed supplementary analyses, however, in which respondents' federal votes were used to define partisanship. This alternative version of the partisanship variable was significantly associated with system responsiveness (F=5.65; p<.05), but not with political trust (F=0.32). Neither was it related with political competence (F=0.05). It does not appear from these findings that we have sacrificed stronger effects in order to test predictions involving strength of party identification.

18 We wondered whether the federal predictors would also be associated with the provincially-defined dependent measures. The effects for federal partisanship in these cross-over analyses were F=7.47 (p<.01) for competence, F=30.95 (p<.001) for responsiveness and F=47.87(p<.001) for trust. The Fs for the MP predictor were 2.22 (ns), 3.61 (ns)and 10.50 (p<.01), respectively. Once again, respondents who identified with their MPs' parties trusted their (provincial) governments less than respondents who identified with one of the other parties. In still other analyses, we observed that inclusion of the federal partisanship predictors along with the provincial predictors did not seriously impair the effects produced by the latter variables. To complete the cross-over picture, the provincial partisan measure failed as a predictor of the three federally-defined dependent variables.

19 In supplementary analyses, we checked for the effects of gender and age (coded 18–29; 30–39; 40–55; 56+) at both levels of government. First, neither variable affected the pattern of findings reported here. Second, gender had a significant effect on competence at the federal level (F=6.45; p<.01), but not on responsiveness or trust. Women felt less politically competent than men. When we analyzed the provincial-level variables for the national sample, gender was significantly associated with competence (F=6.12: p<.01) and trust (F=4.50; p<.05), but not with responsiveness. While women were lower on political competence and trust than men, they were higher on perceived responsiveness. Third, age was a significant predictor for competence at the federal level (F=7.26; p<.001), but it was inconsequential for the other two measures. In general, the oldest respondents tended to feel less effective. For the provincial-level variables, age was significantly associated with competence (F=7.55; p<.001) and trust (F=3.39; p<.05), but not with responsiveness. Older respondents tended to feel less competent and expressed more trust. On the relationships of demographic variables, especially gender and age, to efficacy and/or trust, see, for example, Cole, “Toward a Model of Political Trust,” 813; Sinclair. “Political Powerlessness.” 130–31; Kornberg, Mishler and Clarke, Representative Democracy, 82.

20 The three tables discussed here may be obtained from the authors.

21 For example, as discussed in Craig and Maggiotto, “Measuring Personal Efficacy,” 86.

22 Regional effects on efficacy and trust and differences between provincial and federal levels of these variables is the subject matter of a separate study.

23 We calculated partial correlations between the partisanship and dependent variables, controlling for education, strength of party identification, attention to politics, age, sex (provincially) and MP (federally). The correlations were calculated at the federal and the provincial levels with the variables appropriate to each level, and in spite of the fact that partisanship, sex and MP are binary variables. Federal partisanship was significant for responsiveness (r=.064: p<.001) and t rust (r=.037: p<.05), although its impact in both cases was obviously weak. Provincial partisanship, on the other hand, was significant at p<.001 for responsiveness (r= .218), trust (r= .249) and competence (r=.115). Provincially, nine of the ten correlations for responsiveness were significant at p<.05 or better. Only the figure for Prince Edward Island (r=-.093) failed to reach significance. As well, nine of the partial correlations for provincial trust were significant at p<.05 or better, including correlations of .274 in Quebec, .308 in New Brunswick and .342 in British Columbia. Only the partial correlation for PEI fell short of significance (r= .155).

24 The proximity principle is discussed in Brown and McMenemy, “Generality or Specificity,” 369–72.

25 On this point, we analyzed data from the 1974 Canadian National Election Study because the interviews were conducted following an election in which the incumbent Liberals were returned to power. The four trust items did not appear in this survey: as well, the four efficacy items were administered to only a half sample. Partisanship produced significant effects for perceived system responsiveness (F= 16.42, p<.001) and political competence (F=5.44, p<.05). When we constructed the partisanship variable in terms of respondents' reported vote in the 1974 election instead of their party identification, the effect was significant for responsiveness (F= 12.98, p<.001), but not for competence (F= 1.52). Analyses of the merged 1965 and 1968 data are presented in Simeon and Elkins, Small Worlds, 58, 62.

26 We assessed the effects of the partisanship variable in supplementary analyses, with “months since the last provincial election” as one of the control variables. Partisanship was unimpaired as a predictor of the three dependent variables. Since there were significant interactions between partisanship and months for responsiveness (F=3.66; p<.05) and for trust (F=4.03; p<.01), we constructed an interaction index from these two predictors. In separate analyses using this index, with controls, there appeared to be curvilinear relationships between time, on the one hand, and competence and trust, on the other, within levels of the partisanship variable. The responsiveness and trust scores were lowest for the most recent (1983) and the most remote (1981) provincial elections.

27 Kornberg, Clarke and LeDuc found that reportedinteraction with MPs and regime support were negatively associated. They speculated that this unexpected finding might be due to often unsatisfactory or mistaken attempts on the part of constituents to obtain the assistance of their MPs. See Kornberg et al., “Some Correlates,” 208, 209.

28 See Easton, David, “A Re-assessment of the Concept of Political Support,” British Journal of Political Science 5 (1975), 435–57; also Kornberg, Clarke and LeDuc, “Some Correlates,” 201, 208, and footnote 10 above.CrossRefGoogle Scholar

29 For example, see Manzer, Socio-Political Report, 322.

30 Noelle-Neumann, Elisabeth, The Spiral of Silence: Public Opinion—Our Social Skin (Chicago: University of Chicago Press, 1984), 1–8.Google Scholar

31 Ibid., 23–33.

32 Ibid., 170–73.