¹ For documentation of the decline of trust in government, see Miller, A., Brown, T., and Raine, A., ‘Social Conflict and Political Estrangement, 1958–1972’, unpublished paper delivered at the 1973 Midwest Political Science Association meeting.Google Scholar See also Citrin, J., Political Disaffection in America 1958–1968, unpublished Ph.D. dissertation, University of California, Berkeley 1972Google Scholar, and Harris, L., ‘US Confidence in Leaders Continues at Low Ebb’, Washington Post, 5 11 1973, p. A 7.Google Scholar

² For a discussion of these see Citrin, Political Disaffection, Chap. 2; Keniston, K., The Uncommitted(New York: Harcourt, Brace and World, 1965)Google Scholar, Appendix I; Schacht, R., Alienation (New York: Doubleday, 1970)Google Scholar; Finifter, A., ‘Dimensions of Alienation’, American Political Science Review, LXIV (1970), 389–410CrossRefGoogle Scholar; and Seeman, M., ‘Alienation and Engagement’, in Campbell, A. and Converse, P., eds., The Human Meaning of Social Change (New York: Russell Sage, 1972).Google Scholar The Marxist tradition, of course, defines alienation in objective terms. According to this perspective, someone could be alienated and yet feel satisfied and allegiant. Most contemporary researchers, however, define alienation in psychological terms and then explore the objective circumstances that give rise to it. For a good discussion of Marx's treatment of alienation see Schacht, Alienation.

³ Discontent with what is happening to society (or evaluations of governmental performance) and dissatisfaction with one's personal situation are not mutually exclusive, of course, and one objective of our research program is to identify the conditions under which unhappiness with the quality of one's own life leads to a rejection of the political system. We believe, however, that an association between life dissatisfaction and political alienation usually requires the mediation pf other political attitudes and identifications,

⁴ A compendium of such measures is available in Robinson, J. et al. , Measures of Social-Psychological Attitudes (Ann Arbor: Institute of Social Research Publications, 1968).Google Scholar

⁵ This assumes, of course, that partisans of the ‘out’ party believe that the leaders they support could win office under the existing ‘rules of the game’. In addition, when the ‘out’ party is explicitly ‘anti-system’, a relationship between party identification and a political alienation scale would be evidence for the measure's validity. And approval of the incumbents might accompany rejection of the regime if the politically alienated perceive those in office as themselves antagonistic to the underlying values and norms of the existing constitutional order.

⁶ The several ‘methods’ we have employed were all variants of paper-and-pencil procedures. Clearly, it would be desirable to develop alternative, unobtrusive methods for assessing attitudes toward the political system. See Campbell, D. and Fiske, D., ‘Convergent and Discriminant Validation by the Multitrait-Multimethod Matrix’, in Fishbein, M. ed., Readings in Attitude Theory and Measurement (New York: Wiley, 1967).Google Scholar

⁷ In addition, we excluded the adjective ‘democratic’ on the grounds that it might be ideologically biased. To many right-wingers, America is ‘a republic, not a democracy’. For them, calling the government ‘democratic’ might be an alienated rather than an allegiant response.

⁸ The judges were advanced graduate students of the University of California, Berkeley, or Stanford University.

⁹ For pairs of dichotomous items the coefficient employed was phi/phi-max. Otherwise it was Pearson's r.

¹⁰ The ratings ranged from a ‘worst’ score of ‘o’ to a ‘best’ of ‘12’ — each judge scoring it on a 3–point scale of 0–2. Items included in the PAI all had a judges’ rating of at least ‘6’.

¹¹ The formula employed was Spearman-Brown's with a correction for attenuation.

¹² For a summary of these studies, see Citrin, Political Disaffection, Chap. I,

¹³ The national data reported derive from the 1972 National Election Study conducted by the Center for Political Studies, University of Michigan. The data were made available by the Inter-University Consortium for Political Research.

¹⁴ Controlling for differences in the demographic, ideological and partisan composition of the two samples diminished but did not eliminate this variation in the level of political alienation and dissatisfaction.

¹⁵ For a useful discussion of these issues and of the role of relative deprivation in fostering political disaffection, see Gurr, T. R., Why Men Rebel (Princeton, N.J.: Princeton University Press, 1970).Google Scholar

¹⁶ We use the term interpretation in the sense of Hyman, H., Survey Design and Analysis (Glencoe, Ill.: The Free Press, 1957)Google Scholar, to refer to the process of identifying the intervening links between independent and dependent variables.

¹⁷ Thus, differences in the socioeconomic status of blacks and whites does not explain the greater alienation among blacks. Within racial groups, the relationships of social characteristics to PAI scores were similar. On the whole, however, there was less variation in attitudes toward the political system among blacks.

¹⁸ The items appearing in both studies are listed in Table 4 above.

¹⁹ See Citrin, Political Disaffection, Chap. 3, and Miller, W. and Miller, A., Confidence in Government, the Origin and Impact of the Decline, 1938–1972 (Ann Arbor, Mich.: Center for Political Studies, 1974).Google Scholar

²⁰ We do not report these data in the interests of conserving space. As an illustration of the difference, however, we note that 34 per cent of the BAS sample under 30 said they were not proud of the American system of government, compared to 16 per cent in this age group in the national sample.

²¹ These figures are based on the respondents in both the $15–20,000 and $20,000+ income groups.

²² Citrin, Political Disaffection, Chap. 4.

²³ These correlations are based on the responses of those who claimed they thought or spoke about these issues frequently.

²⁴ The sole exception to this generalization occurred among those earning less than $5,000 on the cost of living issue. However, among those earning less than $5,000 a year, the correlation between PAI and feelings about the unemployment issue was ·44; among those earning more than $15,000 it was ·49. Similarly, dissatisfaction with how the country was doing in controlling the cost of living and political alienation correlated ·27 among those dissatisfied with their own standard of living and ·32 among those who were economically satisfied.

²⁵ Among respondents who felt their local streets were almost always safe, 55.4 per cent who felt local government was ‘not really trying to make the streets safe’ had ‘high’ scores on the PAI, compared to 5 per cent of those who felt the local government was ‘trying hard’. Among respondents who felt local streets were ‘not safe at all’, the equivalent figures were 44.7 per cent and 6 percent.

²⁶ We asked respondents whether they had participated in a sit-in, a political boycott, a peaceful protest rally, or a protest ‘that turned violent’. Each respondent was given a protest participation score by summing the number of these kinds of activities in which he reported participating.