¹ The literature on political participation is voluminous but very little of it refers to Canada. For specifically Canadian items, see Dion, L., “Participating in the Political Process,” Queen's Quarterly, 75, no. 3 (Autumn 1968), 432–47Google Scholar, and Scarrow, Howard A., “Patterns of Voter Turnout in Canada,” Midwest Journal of Political Science, V (Nov. 1961), 351–65CrossRefGoogle Scholar. Other literature on Canadian voting is mainly concerned with the partisan direction of voting rather than with turnout itself. However, some relevant information is to be found in such sources as Meisel, J., The Canadian General Election of 1957 (Toronto, 1962)Google Scholar, and Papers on the 1962 Election (Toronto, 1964), and Alford, R. R., Party and Society (Chicago, 1963).Google Scholar

For non-Canadian data the most concise and valuable source is Milbrath, Lester, Political Participation (Chicago, 1965)Google Scholar. This source also contains an extensive bibliography covering virtually all of the literature on participation to 1965. Several other major authors in political science have discussed political participation. In particular one should be aware of work by Lane, Robert, particularly Political Life (Glencoe, Ill., 1959)Google Scholar, and Political Ideology (New York, 1962), and by Dahl, Robert, particularly Who Governs? (New Haven, 1961)Google Scholar. Angus Campbell has published a number of relevant articles and much of the work of the Survey Research Center of the University of Michigan is in some way relevant. There is also a large European school of literature on political participation. Its most prominent voice is that of Stein Rokkan. See, for example, his “The Comparative Study of Political Participation,” in Ranney, Austin, ed., Essays on the Behavioral Study of Politics (Urbana, Ill., 1962), 47–90.Google Scholar

² Adapted from Milbrath, Political Participation, 18. The description of the hierarchy carries the implication that all or parts of it form Guttman scales. To our knowledge no participation data has been checked for scaleability. The particular survey we are reporting on here did not pick up enough respondents with high levels of activity to make any overall test for scaleability possible. Moreover, the concept of “dimensions” of a political activity introduced below suggests that different forms of participation are qualitatively as well as quantitatively different. If this is so, the hierarchy is not likely to form a single Guttman scale but rather a series of them.

³ This model has much in common with that proposed by Milbrath in Political Participation, 28 ff, but this diagram places more emphasis on socio-economic and political resources and centralizes the role of political interest as a final determinant of participation.

⁴ More and more sources are stressing the significance of political socialization and the literature on the subject is too voluminous to be referred to in a footnote. Several analysts of political participation have also recognized the central position which socialization must be accorded in participatory models. See, for example, Allardt, E. and Pesonen, P., “Citizen Participation in Political Life in Finland,” International Social Science Journal, XII, 1 (1960), 27–39.Google Scholar

⁵ Many writers have recognized the importance of resources for participation. See, in particular, Dahl, Who Governs? 270–304. Many analysts, however, confine themselves to a discussion of the effect of socio-economic status on political participation without adequately discussing the effect of the resources which go with a given socio-economic status.

⁶ Much work has been done on the effect of personality variables – in particular the sense of efficacy a person exhibits – on political participation. On the most basic personality variables see Lane, Political Life, Davies, James, Human Nature in Politics (New York, 1963)Google Scholar, and Lasswell's, Harold venerable but dubious Psychopathology and Politics (Chicago, 1930)Google Scholar. The concept of political efficacy was first developed extensively by the Survey Research Center at the University of Michigan. See in particular, Campbell, A.et al., The American Voter (New York, 1960)Google Scholar, and Campbell, A.et al., The Voter Decides (Evanston, Ill., 1954)Google Scholar. The literature on efficacy is quite vast and the reader is referred to the bibliography in Milbrath, Political Participation, for further sources. Recently N. H. Nie, K. Prewitt, and C. B. Powell have conducted a further investigation of, among other things, the effect of efficacy on participation. Their results, like ours, downgrade its influence. Nie, et al., “Social Structure and Political Participation,” American Journal of Political Science, V, 63 (June and Sept. 1969)Google Scholar. The construction of efficacy scales themselves is also in some ways dubious. See Daudt, H., Constance Maissen, E. Van der, and Mokken, R. J., “Political Efficacy: A Further Exploration,” mimeo., paper presented to the International Political Science Association, Brussels, Sept. 1967.Google Scholar

⁷ Most writers on political participation have been either sociologists or social psychologists and consequently have paid relatively little attention to political resources. It can, however, he suggested that to look at political efficacy is to look at a person's perceptions of the political resources available to him. On this latter point Almond, G. and Verba, S., The Civic Culture (Princeton, 1963)CrossRefGoogle Scholar, passim, may be useful. More generally on the effects of political factors, see Dahl, Who Governs? or Lane, Political Life. As well, much of the literature on parties and pressure groups bears incidentally on the ways in which the political system attempts to contact the public and vice versa.

⁸ Nie et al., “Social Structure and Political Participation,” report a similar finding using the Almond and Verba five-nations data. They call the final variable “political attentiveness” and it consists both of being interested in and paying attention to politics. Ours consists of interest alone as determined by the question: “How much interest do you generally have in what is going on in politics – a good deal, some, or not much?”

⁹ The difference is parallel to that suggested for whole societies by Almond and Verba in The Civic Culture. The gladiatorial or transitional participant would be a member of their participant culture while the passive citizen would fit into their subject culture.

¹⁰ Political Participation, 9–13.

¹¹ Lane, Political Life, 163–81 and passim.

¹² For comparative analysis see Almond and Verba, The Civic Culture; Nie et al., “Social Structure and Political Participation”; Rokkan, S., “Approaches to the Study of Political Participation”, Acta Sociologica, VI (fasc. 1–2, 1962), 1–8CrossRefGoogle Scholar, and “The Comparative Study of Political Participation,” 47–90. The bibliography in Milbrath, Political Participation, contains many other studies useful for comparative purposes.

¹³ Data for Table I and for all the other tables in this study were taken from a sample survey of some 2,100 Canadian voters drawn from voters’ lists and interviewed in mid-winter following the 1965 election. The sample was a modified cluster sample formed by choosing 121 constituencies on a basis proportional to their size and giving increased probability of selection to all areas outside metropolitan Toronto and Montreal. Within each constituency seven polling divisions were chosen on a systematic random basis, and within each polling division four voters were chosen at random and interviewed for about one hour in their homes by a commercial polling agency.

The survey was conceived by Professor John Meisel of Queen's University and he was joined by Professor Philip Converse of the University of Michigan, Professor Mildred Schwartz of the University of Illinois, Professor Maurice Pinard of McGill University, and Professor Peter Regenstreif of the University of Rochester, in drawing up the questionnaire and in designing the project. The research was supported by the Canada Council, the Laidlaw Foundation, and the Committee on Election Expenses; analysis was carried out at the Queen's University Computing Centre.

¹⁴ The campaign activity index is a Canadian variation of an idea suggested in Milbrath, Political Participation, 155–6. Its items consist of: voting and reading about politics, each weighted singly; trying to convince someone how to vote, doubly weighted; and belonging to a political party or working during an election campaign, triply weighted. The weight factors were arbitrarily chosen to reflect the “difficulty” of the activity in terms of its dimensions. The index itself can take the three values, high, medium and low, and the survey under discussion here indicated that 4 per cent of Canadians had a high CAI rating, 22 per cent a medium rating, and 73 per cent a low one. Very roughly these could be pictured as corresponding to the gladiatorial, transitional, and passive levels on our hierarchy of participation.

Class in this table is from a class self-identification question. The results are not significantly different if an objective measure of class is used.

¹⁵ For an expanded version of Table I, the chi-squared value is 70.5 with 21 df. The probability of such a figure occurring by chance is less than 0.001. However, the tau-beta measure of rank order correlation for the table is only 0.12. This is not particularly high relative to some of the findings which will appear later in this paper, indicating that class is not the most important determinant of political participation. On tau-beta, see Blalock, H. M., Social Statistics (New York, 1960), 319–24Google Scholar. The value of tau varies from —1 to 1 and its absolute value is an indication of the strength of an ordinal correlation.

¹⁶ The variable “voting” in this table is derived from the question, “In federal (provincial) elections since you have been old enough to vote would you say you have voted in all of them, most of them, some of them, or none?” The correlations using this question are similar to those we obtain by simply asking whether or not the respondent voted in the latest election.

¹⁷ Personal Influence (Glencoe, Ill., 1955).

¹⁸ Scarrow, “Patterns of Voter Turnout in Canada,” 351–65. Note, however, that in Table v of this paper, Quebeckers claim a slightly higher turnout rate at federal than at provincial elections. This may be because our survey dealt mainly with federal politics but this provides another demonstration that there is often a gap between what people say they do and what they actually do.

¹⁹ Agrarian Socialism (Berkeley, 1950), passim.

²⁰ This point is made in Campbell et al., The American Voter, 402–41. See also Rokkan, S. and Valen, H., “The Mobilization of the Periphery: Data on Turnout, Membership and Party Recruitment in Norway,” Acta Sociologica, VI (fasc. 1–2, 1962), 111–58CrossRefGoogle Scholar, for some important qualifications in another setting.

²¹ See Nie et al., “Social Structure and Political Participation.” Our findings are not so sharply defined as are theirs.

²² Party identification is based on the question, “Generally speaking, do you usually think of yourself as Conservative, Liberal, Social Credit, Créditiste, NDP, or what?”

²³ It should be pointed out that this point does not go undebated in the literature. For years it was asserted that the lack of a firm party identification led to non-voting and was a correlate of apathy. However, recent re-analysis of voting data indicated that this assertion could be questioned. See Key, V. O., The Responsible Electorate (Cambridge, Mass., 1966)CrossRefGoogle Scholar, and Daudt, H., Floating Voters and the Floating Vote (Leiden, 1961)Google Scholar. It should be reemphasized that my assertion is that a particular party identification at the time is an important correlate of participation, not that a firm and unswerving devotion to one party forever is necessary. In this sense my point does not conflict with that of Key and Daudt.

²⁴ As determined by asking, “How strongly (name of party) do you generally feel – very strongly, fairly strongly, or not very strongly?” Some 23 per cent of the sample were very strong identifiers, 41 per cent were fairly strong, 15 per cent not very strong, and 21 per cent had no party identification or did not answer.

²⁵ The question asked was, “One of the things we are interested in is the differences which exist between our federal political parties. Considering everything the parties stand for, would you say that there is a good deal of difference between the parties, some differences, or not much difference?”

²⁶ The question was, “In your opinion, do you think it makes a great deal of difference, some difference, or no difference which political party runs this country?” The question was placed in the interview about five minutes after the ideological difference question.

²⁷ The efficacy scale is made up of four items. Respondents were asked to agree or disagree with each of the following statements:

(a) I don't think the government cares much what people like me think.

(b) Sometimes politics and government seem so complicated that a person like me can't really understand what's going on.

(c) People like me don't have any say about what the government does.

(d) Generally, those elected to parliament soon lose touch with the people.

The items were then combined into a three-part scale whose marginal distribution was: “low,” 26 per cent, “medium,” 46 per cent, “high,” 29 per cent. Unfortunately the original items were not rotated to avoid response set bias.

²⁸ Cf. Nie et al., “Social Structure and Political Participation,” part 2. They found that if participation was correlated with high socio-economic status then it was also significantly, but not highly, correlated with efficacy. However, an entirely separate group of respondents appeared to participate in conjunction with organization memberships and they did not show high efficacy-participation correlations. The tau-beta for Table VII is only 0.113 so one should not be overly enthusiastic about interpreting its results.

²⁹ Lane, Political Life, has this as a recurring theme as does Davies, Human Nature in Politics.

³⁰ See Milbrath, Political Participation, 48–90, for a compendium on such findings. See also Lane, Political Life.

³¹ Dahl, R., A Preface to Democratic Theory (Chicago, 1956)Google Scholar, and Dahl, R. and Lind-Blom, E., Politics, Economics and Welfare (New York, 1953).Google Scholar

³² Satisfaction was determined by asking, “During the last few years has your financial situation been getting better, getting worse, or has it stayed the same?” The marginal responses were: getting better, 40.1 per cent; getting worse, 14.1 per cent; stayed about the same, 44.8 per cent.

³³ The level of interest was determined by asking, “How much interest do you generally have in what is going on in politics – a good deal, some, or not much?” The tau-beta correlation coefficient for Table VIII is 0.273. It will be recalled that the same coefficient for the efficacy-campaign activity table was only 0.113.

³⁴ Blalock, Social Statistics, 336.

³⁵ Comparative data is found in Almond and Verba, The Civic Culture; Milbrath, Political Participation; and Lane, Political Life.

³⁶ Himmelstrand, Ulf, “A Theoretical and Empirical Approach to Depoliticization and Political Involvement,” Acta Sociologica, VI (fasc. 1–2, 1962), 83–111CrossRefGoogle Scholar, provides a similar analysis in the Swedish setting and arrives at similar conclusions.

It should be noted, however, that political systems may retain a considerable propensity for converting apolitical politics into highly political politics if appropriate issues appear. In the early 1960s the United States showed many of the symptoms of apolitical politics as the “image” of the leader became the main determinant of much of what went on in the political system. In 1970 we can hardly accuse American politics of being apolitical. Under the pressure of Vietnam, civil rights issues, poverty, and economic problems American politics has once again become highly politicized and polarized. It may well be that politics moves cyclically from being apolitical (at which time those in power describe conditions as normal) to being highly polarized. This is a question which requires treatment in much greater depth than can be given here.

³⁷ Key, , Public Opinion (New York, 1961)Google Scholar, expresses a similar viewpoint in his concluding chapter.