¹ While the motivations involved are unknown, renunciation of nationality appears to have strong content validity as a “hard” indicator of political estrangement. Formally recorded renunciations of nationality have increased consistently over the last two decades, from 149 in 1950 to 679 in 1968. (See Annual Reports, Immigration and Naturalization Service, United States Department of Justice.) This represents almost a quadrupling of the rate, from .15 per 100,000 voting age population in 1950 to .57 per 100,000 voting age population in 1968.

² Almond, Gabriel and Verba, Sidney, The Civic Culture (Princeton: Princeton University Press, 1963)CrossRefGoogle Scholar.

³ Easton, David, A Systems Analysis of Political Life (New York: John Wiley, 1965), p. 220Google Scholar.

⁴ For an example of Easton's own concern with the operationalization of “support,” see Easton, David and Dennis, Jack, “The Child's Acquisition of Regime Norms: Political Efficacy,” this Review, LXI (03, 1967), 25–38Google Scholar. References to other empirical studies of support and alienation will be found in the footnotes to the present article; an extensive list of earlier studies is found in f. 4 of Easton and Dennis, op. cit.

⁵ See, for example, Seeman, Melvin, “On the Meaning of Alienation,” American Sociological Review, 24 (12, 1959), 783–791CrossRefGoogle Scholar; Dean, Dwight, “Alienation: Its Meaning and Measurement,” American Sociological Review, 26 (1961), 753–758CrossRefGoogle Scholar; Feuer, Lewis S., “What Is Alienation: The Career of a Concept,” New Politics, 1 (Spring, 1962), 116–134Google Scholar; Neal, Arthur G. and Rettig, Salomon, “Dimensions of Alienation among Manual and Non-Manual Workers,” American Sociological Review, 28 (08, 1963), 599–608CrossRefGoogle Scholar; Keniston, Kenneth, The Uncommitted (New York: Dell, 1965), esp. pp. 449–476Google Scholar; and Browning, Charles J., et al., “On the Meaning of Alienation,” American Sociological Review, 26 (1961), 780–781Google Scholar.

⁶ See, especially, Seeman, op. cit., and Neal and Rettig, op. cit. In his stimulating discussion, Keniston uses the term “mode” to express whether alienation is manifest as an attempt to change society (alloplastic) or is, rather, directed inwardly toward change in the individual himself (autoplastic). Keniston, op. cit. Although the behavioral expression of alienation is of obvious relevance to a theory of political alienation, it seems to me that it does not need to be made part of the definition of the concept itself but might be treated as a separate variable whose relation to the attitude of alienation is conditioned by other variables such as the opportunities for social change present in the system, or certain psychological characteristics of the individual.

⁷ The concepts of “powerlessness,” “meaninglessness,” and “isolation,” used here are based on Seeman's discussion of their generic types; the present discussion specifies how these modes of alienation relate to political institutions. The concept of “normlessness” developed here departs from the meaning attached to this term by Seeman. In addition to these four types, Seeman also discusses a type of alienation identified as “self-estrangement.” The sense in which this type of alienation can be said to have an institutional referent is unclear.

⁸ See especially the work emanating from the University of Michigan studies, most notably Campbell, Angus, Converse, Philip E., Miller, Warren B. and Stokes, Donald E., The American Voter (New York: John Wiley, 1960)Google Scholar.

⁹ Lane, Robert E., Political Ideology (New York: The Free Press, 1962), p. 162Google Scholar.

¹⁰ See Neal and Rettig, op. cit.; the same authors' “On the Multidimensionality of Alienation,” American Sociological Review, 32 (02, 1967), 54–64CrossRefGoogle Scholar; and Struening, Elmer L. and Richardson, Arthur H., “A Factor Analytic Exploration of the Alienation. Anomie and Authoritarianism Domain,” American Sociological Review, 30 (1965), 768–776CrossRefGoogle Scholar.

¹¹ See for example, Agger, Robert E., Goldstein, Marshall N., and Pearl, Stanley A., “Political Cynicism: Measurement and Meaning,” Journal of Politics, 23 (08, 1961)CrossRefGoogle Scholar, and Janda, Kenneth, “A Comparative Study of Political Alienation and Voting Behavior in Three Suburban Communities,” in Studies in History and the Social Sciences (Normal: Illinois State University, 1965), pp. 53–68Google Scholar. However, in a recent publication, Aberbach reports that factor analysis confirmed the dimensionality of several different measures of alienation, including two measures of political alienation. He does not report the factor loadings, although he describes them as “strong.” His data support the utility of considering specific types of political alienation separately. See Aberbach, Joel, “Alienation and Political Behavior,” this Review, LXIII (03, 1969), 86–99Google Scholar. Two recent articles dealing with support for particular political institutions also use factor analytic techniques to dimensionalize attitudes. See Dennis, Jack, “Support for the Party System by the Mass Public,” this Review, LX (09, 1966), 600–615Google Scholar and Boynton, G. R., Patterson, S.C. and Hedlund, R. D., “The Structure of Public Support for Legislative Institutions,” Midwest Journal of Political Science, XII (05, 1968), 163–180CrossRefGoogle Scholar.

¹² See Almond and Verba, op. cit., pp. 519–523. The data were made available by the Inter-University Consortium for Political Research. Neither Professors Almond nor Verba nor the Consortium bear any responsibility for the analyses or interpretations presented here.

¹³ Borgatta, Edgar F., “On Analyzing Correlation Matrices: Some New Emphases,” Public Opinion Quarterly, XXII (Winter, 1958–1959), p. 523Google Scholar.

¹⁴ For a comparison of the components and factor models see Harman, Harry H., Modern Factor Analysis (Chicago: The University of Chicago Press, 1967), Chapter 8Google Scholar.

¹⁵ Questions 24, 28, 27, 23, 15, 16, 72f and 14 compose one strong cluster, each having a correlation of .3 or higher with at least one other variable in the cluster, and Questions 37b, 35, 34, 37a, 32b, 31b, 51, and 52 form a weaker but identifiable second cluster.

¹⁶ See Thurstone, L. L., Multiple Factor Analysis (Chicago: University of Chicago Press, 1947), p. 335Google Scholar, and Guilford, J. P., Psychometric Methods (2nd ed.; New York: McGraw-Hill, 1954), p. 508Google Scholar.

¹⁷ Specifically, the results of the two-component, twenty-six item solution demonstrated that (a) Qs. 45 and 72a had weak loadings on both components, and (b) Q. 46 loaded negatively on its dominant component. To investigate further the structure of the items, Q?. 46 and 72a were eliminated, and two scales were formed using a cluster scoring procedure. Item analyses (based on response choice-total scale biserial correlations) indicated that Qs. 46, 47, and 48 were only weakly related to the dominant content of their scales. Since all three of these questions concerned attitudes toward election campaigns (as did Q. 45, which was eliminated because of its weak loadings), it seemed reasonable to conclude that the domain tapped by these questions was essentially different from that of the other items and all were eliminated.

¹⁸ Consistent with this interpretation, Almond and Verba report that these questions were intended to measure the “qualities our respondents imputed to the executive side of government.” Op. cit., p. 106. At a different point in their analysis, however (see especially Chapter 8), they use these questions as measures of “administrative competence,” which might be construed as a subtype of powerlessness. There is clearly some conceptual overlap here. Although a sense of departure from community norms is being used in the present study to define “perceived normlessness,” it is obvious that, when the particular norm in question refers to the role the individual should play in his interactions with political decision-makers, assessments of norm deviations may also reflect feelings of powerlessness. That items 34, 35, 37a, and 37b have quite low loadings on the factor that is here called “powerlessness” increases confidence that their interpretation in terms of role expectations and norm deviations is more adequate than an interpretation using the powerlessness concept.

¹⁹ Almond and Verba, op. cit., p. 102.

²⁰ See Agger, et al., op. cit., p. 479, for a list of the questions used.

²¹ See especially Merton, Robert K., Social Theory and Social Structure (rev. ed.; New York: The Free Press, 1957), pp. 131–194Google Scholar. Another related concept inspired by Durkheim's analysis of anomie has recently been discussed by Inkeles, who develops the concept of “political anomie” along lines which are strikingly similar to those defining “perceived political normlessness.” Inkeles defines political anomie as the individual's feeling that others are not complying with the rules of society, and operationalizes it as “the individual's perception of the extent to which politicians and government officials pay attention to the common man, servo the public rather than their own careers, and keep their campaign promises after the election.” Inkeles, Alex, “Participant Citizenship in Six Developing Countries,” this Review, LXIII (12, 1969), p. 1125Google Scholar. Emphasis in original. The content of the first and third of Inkeles' defining questions is essentially the same as Qs. 18a and 72d of the perceived political normlessness scale.

²² The Hoyt method was used. See Hoyt, Cyril J. and Stunkard, Clayton L., “Estimation of Test Reliability for Unrestricted Item Scoring Methods,” Educational and Psychological Measurement, 12 (1952), 756–758CrossRefGoogle Scholar.

²³ Since multivariate techniques assume a constant N, some procedures had to be adopted to eliminate missing data. The entire sample of 970 was used for the component analyses reported above, with “Not Ascertained” or “Don't Know” cases coded the middle position of continuous variables or the modal position of dichotomies. Once scale composition was determined, cases were excluded from the analysis if more than one-third of the questions comprising either of the dependent variables had missing responses or if information was missing on a major socioeconomic status variable, i.e., occupation or income, since these were known to have important effects on alienation (there were no missing data for education, place of birth, or race). The two-thirds present criterion for the dependent variables was somewhat arbitrary. It seemed likely that with more than one-third of the items missing, there would be a relatively high probability of misrepresenting the respondent's true attitude. Conversely, a more stringent criterion would have eliminated a great many more cases. These procedures resulted in 119 exclusions, accounted for by 102 individuals. An analysis designed to ascertain the effects of excluding these 102 cases on the representativeness of the sample indicated that the revised sample generally corresponded more closely than the original to a number of census distributions. This was because missing data cases tended to have demographic characteristics similar to those which vere overrepresented in the original sample. This finding and the great, likelihood of unreliability in attempting to score the missing data cases were primary factors in the decision to remove these cases from the analysis. For a more detailed discussion of some problems raised by missing data and a consideration of alternative methods of handling it, as well as a detailed description of the procedures followed in the elimination of cases reported here, see Finifter, Ada W., Dimensions of Political Alienation: A Multivariate Analysis, Unpublished Ph.D. dissertation, University of Wisconsin, 1967, pp. 60–70Google Scholar.

²⁴ Bivariate plots between each predictor variable and each alienation scale showed no significant departures from linearity. The relationship between age and powerlessness was the only one which was very slightly curvilinear, with the youngest age group (18–25) having as high a mean powerlessness score as the 41–50 age group, but not nearly as high as that of the oldest group. This result confirms neither previous findings of a linear relationship between these variables, reported by several investigators, nor those findings indicating extreme curvilinearity, with the youngest being more alienated than any other age group. See Thompson, W. E. and Horton, J. E., “Political Alienation as a Force in Political Action,” Social Forces, 38 (03, 1960), 190–195CrossRefGoogle Scholar.

²⁵ For example, see Agger, et al., op. cit., 484–487; Middleton, Russell, “Alienation, Race, and Education,” American Sociological Review, 28 (12, 1963), 973–977CrossRefGoogle Scholar; Haer, John, “Social Stratification in Relation to Attitudes Toward Sources of Power,” Social Forces, 35 (12, 1956), 137–142CrossRefGoogle Scholar; and Quinney, Richard, “Political Conservatism, Alienation, and Fatalism,” Sociometry, 27 (09, 1964), 372–381CrossRefGoogle Scholar. My review of previous studies intentionally omits those using Srole's anomia scale as a measure of alienation. The heavy emphasis on personal life situations in the Srole scale suggests that it may largely measure personal disorganization or maladjustment rather than attitudes toward aspects of social structure. The two may be related, but this is an empirical rather than definitional question, and the indiscriminate use of the Srole scale as a de facto measure of “alienation” serves mainly to confuse these concepts. Furthermore, there is some empirical evidence that aDomia and some measures of alienation are independent dimensions. See the articles cited in note 10.

²⁶ Campbell, Angus, Gurin, Gerald, and Miller, Warren E., The Voter Decides (Evanston: Row, Peterson and Company, 1954), pp. 190–192Google Scholar; Campbell, et al., The American Voter, op. cit., pp. 516–520.

²⁷ See, for example, Middleton, op. cit. and Campbell, Gurin, and Miller, op. cit.

²⁸ Almond and Verba, op. cit., esp. Chapters 1 and 8.

²⁹ The ranking of ethnic prestige follows Lenski, Gerhard, “Status Crystallization: A Non-Vertical Dimension of Social Status,” American Sociological Review, 19 (1954), p. 407CrossRefGoogle Scholar.

³⁰ Litt found that familiarity with Boston politics (as measured by length of residency in that city) increases political cynicism. Litt, Edgar, “Political Cynicism and Political Futility,” Journal of Politics, 25 (05 1963), 312–323CrossRefGoogle Scholar.

³¹ For descriptions of attitudes toward authority in different political systems see Almond and Verba, op. cit., esp. Chapters 4 and 14; Wylie, Lawrence, Village in the Vaucluse (New York: Harper and Row, 1964), Chapter 10Google Scholar; and Easton, David and Hess, Robert D., “The Child's Political World,” Midwest Journal of Political Science, 6 (1962), 229–246CrossRefGoogle Scholar. Easton and Hess comment particularly on the idealization of authority encouraged in American socialization practices.

³² Then, too, it is possible that the foreign-born may be more hesitant to make statements critical of American government to an American interviewer. Since the items in the normlessness scale tend to focus on government officials they may be more threatening in this respect than the items in the powerlessness scale, which may be perceived more in terms of the respondent's own shortcomings.

³³ Although a significant correlation between political participation and powerlessness was expected on theoretical grounds, its magnitude prompted re-examination of the scales to explore the possibility of a spurious effect caused by similarity of some of their questions. Questions 24 and 28 of the powerlessness scale asked for the likelihood that the respondent would actually try to do something about a national law or local regulation that he thought unjust. Questions 25 and 29, used in the participation scale, asked if the respondent had actually ever done anything to try to influence a national or local decision. The first set is an almost classic operationalization of the definition of an attitude as a predisposition to behavior. Questions 25 and 29, however, ask for reports on actual behavior. Thus, the four questions are legitimately placed in their respective scales of attitude toward the political system and actual participation in it. Nevertheless, in order to avoid any possibility of spuriousness, the powerlessness scale was reconstructed without questions 24 and 28. The reliability of this new scale dropped from .77 to .71 because of the loss of the two items, which had been very highly correlated with the other items of the scale. However, contrary to the spuriousness hypothesis, the correlation between the participation scale and the new powerlessness scale dropped only to −.60 (from −.64). Since the decrease in actual prediction is less than the theoretically possible decrease due to loss of reliability alone, the diminution of the correlation is probably due solely to the decrease in reliability rather than to the omission of these two particular items. If, in contrast, the decrease in correlation had been larger than warranted by the decrease in reliability, the spuriousness hypothesis would have been more plausible. Furthermore, the correlations between these particular items of the two scales were not large enough to have caused the high correlation between the scales themselves, especially in view of the fact that they comprise only 2/7 of the participation scale and 2/11 of the powerlessness scale.

In contrast, the correlation between items 25 and 29 of the participation scale is .47 and between items 24 and 28 of the powerlessness scale .55. This analysis demonstrated that the items contributed far more to the reliability of their respective scales than to any spurious correlation between the scales. On the basis of this evidence the original powerlessness scale was retained because of its higher reliability.

³⁴ Rosenberg, Morris, “Misanthropy and Political Ideology,” American Sociological Review, 21 (12, 1956), p. 690CrossRefGoogle Scholar. The reliability of this scale in the current project was .58.

³⁵ Campbell, et al., The American Voter, op. cit., pp. 479–481.

³⁶ Horton, John E. and Thompson, Wayne E., “Powerlessness and Political Negativism: A Study of Defeated Local Referendums,” American Journal of Sociology, 67 (1962), pp. 491–492CrossRefGoogle Scholar; and Thompson, W. E. and Horton, J. E., “Political Alienation as a Force in Political Action,” Social Forces, 38 (03, 1960), 190–195CrossRefGoogle Scholar.

³⁷ Janda, op. cit.

³⁸ Aberbach, op. cit.

³⁹ Agger, et al., op. cit., pp. 495–497.

⁴⁰ Douvan, Elizabeth, “The Sense of Effectiveness and Response to Public Issues,” Journal of Social Psychology, 47 (1958), 111–126CrossRefGoogle Scholar.

⁴¹ Rose, Arnold M., “Attitudinal Correlates of Social Participation,” Social Forces, 37 (1959), 202–206CrossRefGoogle Scholar; by the same author, “Alienation and Participation: A Comparison of Group Leaders and the ‘Mass’,” American Sociological Review, 27 (12, 1962), 834–838CrossRefGoogle Scholar; and Almond and Verba, op. cit., pp. 307–322.

⁴² Dahl, Robert A., Who Governs? (New Haven: Yale University Press, 1961), Chapter 26Google Scholar.

⁴³ For a pioneering application of methods of analyzing reciprocal causation, see Duncan, Otis Dudley, Haller, Archibald O., and Portes, Alejandro, “Peer Influences on Aspirations: A Reinterpretation,” American Journal of Sociology, 74 (09, 1968), 119–137CrossRefGoogle ScholarPubMed.

⁴⁴ The particular program used was “BMD02R, Stepwise Regression.” For a description of this program, see Dixon, W. J. (ed.), BMD Biomedical Computer Programs, revised edition (Los Angeles: Health Sciences Computing Facility, Department of Preventive Medicine and School of Health, School of Medicine, UCLA, 09 1, 1965), pp. 233–257Google Scholar.

⁴⁵ On the other hand, this very inter-relatedness may lead to difficulties in obtaining stable estimates of regression coefficients. As Farrar and Glauber suggest, the problem of multicollinearity is most usefully viewed as one of degree of severity rather than as one of categorical existence. The size of the correlation coefficients among the independent variables in the present study does not appear to constitute a serious problem. See Farrar, Donald E. and Glauber, Robert R., “Multicollinearity in Regression Analysis: The Problem Revisited,” Review of Economics and Statistics, 49 (1967), 92–107CrossRefGoogle Scholar.

⁴⁶ See, for example, Easton, David and Dennis, Jack, “The Child's Image of Government,” The Annals of the American Academy of Political and Social Science, 361 (09, 1965), 40–57CrossRefGoogle Scholar; Greenstein, Fred I., Children and Politics (New Haven: Yale University Press, 1965)Google Scholar; and Hess, Robert D. and Torney, Judith V., The Development of Political Attitudes in Children (Chicago: Aldine Publishing Company, 1967)Google Scholar.

⁴⁷ Verba suggests that attitudes toward specific political issues may develop before more coherent or inclusive ideologies. See Verba, Sidney, Small Groups and Political Behavior (Princeton: Princeton University Press, 1961), pp. 41–42Google Scholar.

⁴⁸ See Robinson, W. S., “The Motivational Structure of Political Participation,” American Sociological Review, 19 (1952), 151–156CrossRefGoogle Scholar and Finifter, Bernard M.. “Styles of Participation in Political Life: Differential Multivariate Prediction with and without Allowance for Interaction Effects,” paper presented to the Midwest Society of Multivariate Experimental Psychology, Chicago, 05, 1968Google Scholar.

⁴⁹ Results of an analysis of interaction effects provide some further support for the suggestion that political participation may increase perceived normlessness. The analysis identified a small but theoretically interesting sub-group (N = 56) of white, mobile, college-educated people with high faith in people among whom higher political participation is associated with higher alienation: Ada W. Finifter, op. cit., pp. 144–151.

⁵⁰ Berelson, Bernard, Lazarsfeld, Paul F., and McPhee, William N., Voting (Chicago: University of Chicago Press, 1954), p. 316Google Scholar.

⁵¹ Geertz, Clifford, “Ritual and Social Change,” in Demerath, N. J. III, and Peterson, Richard A. (eds.), System, Change, and Conflict (New York: The Free Press, 1067), p. 233Google Scholar.

⁵² See, for example, Coser, Lewis, The Functions of Social Conflict (New York: The Free Press, 1956), pp. 15–31Google Scholar; and Hield, Wayne, “The Study of Change in Social Science,” British Journal of Sociology (03, 1964), 1–11Google Scholar.

⁵³ From the Greek “orthos,” to straighten, correct, or make right, as in orthopsychiatry, orthopedics, etc. As an analytic concept, “orthofunctional” is not intended to increase the explanatory power of a functional analysis. See Gregor, A. James, “Political Science and the Uses of Functional Analysis,” this Review, LXII (06, 1968), 425–439Google Scholar.

⁵⁴ See Riesman, David and Glazer, Nathan, “Criteria for Political Apathy,” in Gouldner, A. W. (ed.), Studies in Leadership (New York: Harper and Brothers, 1950), pp. 505–559Google Scholar, for this characterization of most activities used by social scientists as measures of political participation. The current “participation revolution” makes this characterization even more telling today than it was in 1950.

⁵⁵ Some substantiating evidence for this hypothesis lies in Ransford's finding that among Negroes in the Watts area, infrequent contact with whites was related to willingness to use violence under the two conditions of a high sense of powerlessness and a high dissatisfaction with treatment received as Negroes. See Ransford, H. Edward, “Isolation, Powerlessness, and Violence: A Study of Attitudes and Participation in the Watts Riot,” American Journal of Sociology, 73 (03, 1968), 581–591CrossRefGoogle ScholarPubMed.

⁵⁶ Campbell, et al., The American Voter, pp. 516–518.

⁵⁷ See Smelser, Neil J., Theory of Collective Behavior (New York: The Free Press of Glencoe, 1963)CrossRefGoogle Scholar.

⁵⁸ For some interesting ideas on the subject of change in basic political attitudes, see Gamson, William A., Power and Discontent (Homewood, Illinois: The Dorsey Press, 1968), esp. pp. 172–183Google Scholar.