¹ It might be possible to argue that, politically speaking, only aggregate change is relevant. A simple example dispels this notion. Consider two populations in which aggregate opinion changed not at all, one in which there was no individual change whatsoever and one in which an enormous amount of individual-level ‘cross-cutting’ or ‘cancelling’ change occurred. Surely the politics of these two populations would be different, especially if the relationship between, say, class and attitude were turned topsy-turvy by the individual changes. An aggregate approach to much of the same data analyzed here is given in our ‘Continuity and Change in Political Orientations: A Longitudinal Study of Two Generations’, American Political Science Review, LXIX (1975), 1316–35.Google Scholar

² Sears, David O., Political Attitudes Through the Life Cycle (San Francisco: W. H. Freeman, forthcoming).Google Scholar Our thinking has been much influenced by this work and we draw freely on it throughout this article.

³ For statements supporting this position from a political scientist see Davis, James C., ‘The Family's Role in Political Socialization’, Annals of the American Academy of Political and Social Science, CCCLXI (1965), 10–19CrossRefGoogle Scholar, and his ‘Political Socialization: From Womb to Childhood’, in Renshon, Stanley, ed., Handbook of Political Socialization: Theory and Research (New York: Free Press, 1977).Google Scholar See also various selections in Schwartz, David C. and Schwartz, Sandra Kenyon, eds., New Directions in Political Socialization (New York: Free Press, 1975).Google Scholar

⁴ See Sears, , Political Attitudes Through the Life Cycle, Chap. 10.Google Scholar

⁵ The classic work (from 1928) remains Karl Mannheim, ‘The Problem of Generations’, in Kecskemeti, P.. ed., Essays on the Sociology of Knowledge (London: Routledge and Kegan Paul, 1952).Google Scholar Among the many works treating the concept of generations and the related phenomena of life-cycle and period effects see Riley, Matilda White, ‘Aging and Cohort Succession: Interpretations and Misinterpretations’, Public Opinion Quarterly, XXXVII (1973), 35–49CrossRefGoogle Scholar; Bengtson, Vern L. and Cutler, Neal E., ‘Generations and Intergenerational Relations: Perspectives on Time, Age Group, and Social Change’, in Binstock, Robert and Shanah, Ethel, eds., The Handbook of Aging and Social Sciences (New York: Van Nostrand-Reinhold, 1976)Google Scholar; Glenn, Norval D., ‘Aging and Conservatism’, Annals of the American Academy of Political and Social Science, CDXV (1974), 176–86CrossRefGoogle Scholar; and Journal of Social Issues, XXX (1974), entire issues of Nos. 2 and 3.Google Scholar

⁶ The milder version, with respect to post-Second World War birth cohorts is advanced by Inglehart, Ronald, ‘The Silent Revolution in Europe: Intergenerational Change in Post-Industrial Societies’, American Political Science Review, LXV (1971), 991–1017.CrossRefGoogle Scholar

⁷ Some notable exceptions beginning at the pre-adult levels include Newcomb, Theodore et al. , Persistence and Change: Bennington College and Its Students after Twenty-Five Years (New York: Wiley, 1967)Google Scholar; Block, Jack, Lives Through Time (Berkeley, Calif.: Bancroft Books, 1971)Google Scholar; Himmelweit, Hilde T. and Swift, Betty, ‘Adolescent and Adult Authoritarianism Re-Examined: Its Organization and Stabilization over Time’, European Journal of Social Psychology, 1 (1971), 357–84CrossRefGoogle Scholar; Himmelweit, H. T. and Swift, B., Social and Personality Factors in the Development of Adult Attitudes Toward Self and Society; An Eleven-year Follow-up Study of Adolescent Boys (Report to the Social Science Research Council, London, 1971)Google Scholar; and Greeley, Andrew M. and Spaeth, Joe L., ‘Political Change Among College Alumni’, Sociology of Education, XLIII (1970), 106–13.CrossRefGoogle Scholar

⁸ Jennings, M. Kent and Niemi, Richard G., The Political Character of Adolescence (Princeton, N.J.: Princeton University Press, 1974).Google Scholar

⁹ In American, as opposed to British, usage, ‘youths’ has no particular negative connotations.

¹⁰ Mannheim, , ‘The Problem of Generations’Google Scholar; Lambert, T. Allen, ‘Generations and Change: Toward a Theory of Generations as a Force in Historical Processes’, Youth and Society, IV (1972), 21–46CrossRefGoogle Scholar; Rintala, Marvin, ‘Generations: Political Generations’, in Encyclopedia of the Social Sciences, Vol. 6 (New York: Macmillan and Free Press, 1968).Google Scholar

¹¹ Erikson, Erik H., Identity: Youth and Crisis (New York: Norton, 1968)Google Scholar; Keniston, Kenneth, The Young Radicals (New York: Harcourt, Brace, 1968).Google Scholar

¹² It should be pointed out that the results below are not an artifact of the differing educational attainments of the youths and parents. The youths, of course, are relatively more educated than their parents. If it were the case that stability was strongly related to education, then parent-youth contrasts would to some extent be a function of the education difference rather than being life cycle or generational in origin. To check for this, the stability data reported below were also generated for each of three sub-groups in the younger generation: those who had no college training, those who matriculated but did not receive a four-year degree, and those with at least a college degree. There are occasional large differences in the stability figures, but there is no consistent pattern of greater or lesser stability in one group. Thus, while a complete analysis of the relationship between education and change still awaits us, the results here are not artificially due to the greater education of the youths.

¹³ Some of the change is maturational and represents accretions in learning, but some of the apparent change is also simply measurement error, especially in terms of trying to uncover attributes which may simply not be present among younger children. See Hess, Robert D. and Torney, Judith V., The Development of Political Attitudes in Children (Chicago: Aldine, 1967)Google Scholar, Appendix A: Vaillancourt, Pauline, ‘Stability of Children's Survey Responses’, Public Opinion Quarterly, XXXVII (1973), 373–87CrossRefGoogle Scholar; Vaillancourt, Pauline and Niemi, Richard G., ‘Children's Party Choices’Google Scholar, and Sigel, Robert S. and Brookes, Marilyn, ‘Becoming Critical about Polities’, both in Richard G. Niemi and Associates, The Politics of Future Citizens (San Francisco: JosseyBass, 1974)Google Scholar; Zellman, Gail L.. ‘Sex Roles and Political Socialization’ (Ph.D. dissertation, University of California at Los Angeles, 1973)Google Scholar; Chaffee, Steven H., Ward, L. Scott and Tipton, Leonard P., ‘Mass Communication and Political Socialization’, in Dennis, Jack, ed., Socialization to Politics (New York: Wiley. 1973)Google Scholar; and Sears, David, ‘The Problem of Generations: A Social Psychologist's Perspective’ (paper presented at the annual meeting of the American Political Science Association, San Francisco, 1975).Google Scholar

¹⁴ Weisberg's prescription seems especially apropos here. We will find several instances in which these two measures yield very different values. The following table illustrates how this can happen:

In this instance, tau-b will be considerably less than 1·0, the exact value depending on the marginals. This low to moderate correlation alerts us to the fact that 1965 responses are not an especially good predictor of 1973 responses; most individuals said ‘Yes’ in 1973 regardless of their 1965 opinion. If we are looking for exact correspondence between 1965 and 1973, tau-b correctly tells us that that correspondence is not too high. But the response pattern represented in the table may make a great deal of sense. If there were strong period effects favoring the ‘Yes’ response, this is exactly the kind of pattern we would expect to find. Gamma is especially sensitive to this type of change, and the perfect gamma in the above table calls attention to this pattern in which change is concentrated in one 1965 response category. Thus the use of two statistics emphasizes the twin aspects of the table – the low predictability and the highly concentrated change. Reliance on only one statistic would not capture both elements. See Weisberg, Herbert F., ‘Models of Statistical Relationship’, American Political Science Review, LXVIII (1974), 1638–55.CrossRefGoogle Scholar See also Wilson, Thomas P., ‘Measures of Association for Bivariate Hypotheses’, in Blalock, Hubert M. Jr., ed., Measurement in the Social Sciences (Chicago: Aldine, 1974).Google Scholar

¹⁵ The ·30 value may still seem low. But there are two major considerations to keep in mind. First is the fact that this is an eight-year panel. For almost any attitude and probably most non-attitudinal orientations it is unreasonable to expect extremely high correlations, such as ·8 or ·9, over this long a period. Second, there is the matter of question reliability. Clearly the measures used by political scientists should in some way be corrected for unreliability. But so far there is little agreement about how best to do so or just what the magnitude of such correction factors would be.

¹⁶ With few exceptions the specific measures used can be found in our report on the 1965 study, Jennings, and Niemi, , Political Character of Adolescence.Google Scholar It should be pointed out that the combined measures used in this paper are generally additive indexes rather than Guttman scales as in 1965. However, we have rescored the repeated measures so that all variables reported in this paper are scored identically for 1965 and 1973. Complete specification and justification for our procedures will be provided in a future monograph. One new measure used here is the self-confidence measure of personal efficacy. Like the opinion strength measure, it is an index based on three forced-choice items.

¹⁷ Retrospective data indicate that more politically-interested parents breed more politically-interested off spring. But the correlations between parental self reports and adolescent offspring self reports are very weak. See Jennings, and Niemi, , Political Character of Adolescence, Chap. 6.Google Scholar

¹⁸ For magazines the summary figures for the young are tau-b = ·27, gamma = ·44, gross stability = 55 per cent; for the parents, ·42, ·68, and 65 per cent. For radio the figures for the young are ·11, ·15, 37 per cent; for the parents, ·23, ·36, 46 per cent. An analysis of some over-time linkages between media behavior and other political orientations for the youth portion of the panel is contained in Chaffee, Steven H.et al., ‘Political Socialization in Mass Communications’, in Renshon, Handbook of Political Socialization.Google Scholar

¹⁹ Balch, George I., ‘Multiple Indicators in Survey Research: The Concept “Sense of Efficacy”’, Political Methodology, 1 (1974), 1–44.Google Scholar

²⁰ Converse, Philip E., ‘The Nature of Belief Systems in Mass Publics’, in Apter, David, ed., Ideology and Discontent (New York: Free Press, 1964)Google Scholar; Jennings, and Niemi, , Political Character of Adolescence, Chap. 4.Google Scholar

²¹ Niemi, Richard G. and Weisberg, Herbert F., eds., Controversies in American Voting behavior (San Francisco: W. H. Freeman, 1976), Chap. 17.Google Scholar

²² Some initial work dated the decline in intensity from the end of the Second World War and rendered a generational interpretation. More recent work with different analytic approaches and larger data bases argues that the age gain in intensity lasted well up into the 1960s and that the change since then primarily reflects period effects. On the first view see Abramson, Paul R., ‘Generational Change and the Decline of Party Identification in America’, American Political Science Review, LXX (1976), 469–78.CrossRefGoogle Scholar On the second view see Knoke, David and Hout, Michael, ‘Social Demographic Factors in American Political Party Affiliations, 1952–1972’, American Sociological Review, XXXIX (1973), 242–58Google Scholar; and for a much stronger statement, Converse, Philip E., The Dynamics of Party Support (Beverly Hills, Calif.: Sage, 1976).Google Scholar

²³ The accompanying 1964–68 correlations for the young are tau-b = ·31, gamma = ·58; for the parents, ·43 and ·74. The 1968–72 figures for the young are ·46 and ·70; for the parents, ·51 and ·78.

²⁴ Converse, , ‘The Nature of Belief Systems’Google Scholar, and Converse, , ‘Attitudes and Non-Attitudes: Continuation of a Dialogue’, in Tufte, Edward R., ed., The Quantitative Analysis of Social Problems (Reading, Mass.: Addison-Wesley, 1970), pp. 168–89.Google Scholar

²⁵ E.g. Pomper, Gerald M., ‘From Confusion to Clarity: Issues and the American Voters, 1956–68’, American Political Science Review, LXXI (1977), 31–43Google Scholar; Norman H. Nie with Andersen, Kristi, ‘Mass Belief Systems Revisited: Political Change and Attitude Structure’, Journal of Politics, XXXVI (1974), 549–91Google Scholar; Achen, Christopher H., ‘Mass Political Attitudes and the Survey Response’, American Political Science Review, LXIX (1975), 1218–31CrossRefGoogle Scholar; Converse, Philip E., ‘Public Opinion and Voting behavior’, in Greenslein, Fred I. and Polsby, Nelson W., eds., Handbook of Political Science, Vol. 4 (Reading, Mass.: Addison-Wesley, 1975).Google Scholar

²⁶ Again, we are indebted to Sears, , Political Attitudes Through the Life Cycle, Chap. 12.Google Scholar

²⁷ This includes a small percentage who said they had an opinion but could not say which they favored.

²⁸ Jennings, and Niemi, , ‘Continuity and Change’.Google Scholar

²⁹ Miller, Arthur H., ‘Political Issues and Trust in Government: 1964–1970’, American Political Science Review, LXVIII (1974), 951–72CrossRefGoogle Scholar; and Miller, Arthur H., Brown, Thad A., and Raine, Alden S., ‘Social Conflict and Political Estrangement, 1958–1972’ (paper presented at the annual meeting of the Midwest Political Science Association, Chicago, 1973).Google Scholar

³⁰ Jennings, and Niemi, , ‘Continuity and Change’.Google Scholar

³¹ Results of a recent cohort analysis, using some of the same orientations examined here, also support this conclusion. See Searing, Donald, Wright, Gerald and Rabinowitz, George, ‘The Primacy Principle: Attitude Change and Political Socialization’, British Journal of Political Science, VI (1976), 83–113.CrossRefGoogle Scholar

³² Converse, Philip E., ‘Attitudes and Non-Attitudes’.Google Scholar