¹ Elkins, David J., ‘The Nature of Belief Systems in Issue Publics’, paper presented at the annual meeting of the Canadian Political Science Association, Ottawa, Canada, 1982.Google Scholar

² Campbell, Angus, Converse, Philip E., Miller, Warren E. and Stokes, Donald E., The American Voter (New York: Wiley, 1960).Google Scholar

³ Miller, Warren E. and Stokes, Donald E., ‘Constituency Influence in Congress’, American Political Science Review, LVII (1963), 45–57CrossRefGoogle Scholar; Erikson, Robert S., Luttbeg, Norman R. and Tedin, Kent L., American Public Opinion: Its Origins, Content and Impact (New York: Wiley, 1980).Google Scholar

⁴ Simon, Herbert, ‘A Behavioral Model of Rational Choice’, Quarterly Journal of Economics, LXIX (1955), 99–108.CrossRefGoogle Scholar

⁵ Zaller, John, ‘Toward a Theory of the Survey Response’, paper presented at the annual meeting of the American Political Science Association, Washington, D.C., 1984.Google Scholar

⁶ For a thoughtful and nuanced layout of a hierarchical model of information processing, see Peffley, Mark A. and Hurwitz, Jon, American Journal of Political Science, XXIX (1985), 871–90.CrossRefGoogle Scholar

⁷ Converse, Philip E., ‘The Nature of Belief Systems in Mass Publics’, in Apter, David E., ed., Ideology and Discontent (Glencoe, Ill.: Free Press, 1964), pp. 206–61.Google Scholar

⁸ Weiner, Bernard, ‘A Cognitive (Attribution)–Emotion–Action Model of Motivated Behavior: An Analysis of Judgments of Help-giving’, Journal of Personality and Social Psychology, XXXIX (1980), 186–200.CrossRefGoogle Scholar

⁹ The data were made available by the Inter-University Consortium for Political and Social Research. The Center for Political Studies of the Institute for Social Research at the University of Michigan collected the data for the 1972 National Election Study under a grant from the National Science Foundation. Neither the consortium nor the original collectors of the data bear any responsibility for the analyses reported here.

¹⁰ We rely upon data from 1972, rather than from a more recent National Election Study, because subsequent NES surveys have not included all of the items required for our analysis.

¹¹ For each item used to assess preferences on racial policy, respondents who said that they were not interested in the issue or who said ‘don't know’ were rescored to the centre of the scale for that item.

¹² Brady, Henry E., ‘Factor Analysis for Ordinal Data which are not Interpersonally Comparable’ (Working Paper, Department of Government, Harvard University, 1986).Google Scholar

¹³ Duncan, Otis Dudley, Notes on Social Measurement (New York: Russell Sage Foundation, 1984).Google Scholar

¹⁴ Brady, , ‘Factor Analysis for Ordinal Data which are not Interpersonally Comparable’.Google Scholar

¹⁵ The analysis produced a second component with an eigenvalue slightly greater than the conventional critical value of 1·0. The eigenvalue of the first component, however, was more than twice that of the second, while the eigenvalue of the second component was only slightly larger than that of the third. The eigenvalues decreased gradually from that point. Moreover, the second component extracted was substantively indistinct from the first. For both of these reasons, the second component was discarded and only the first was employed in our analysis. A principal factors solution yielded almost identical results.

¹⁶ For another treatment of these items, different in methodology and in focus, see Sniderman, Paul M. with Hagen, Michael Gray, Race and Inequality (Chatham, N.J.: Chatham House, 1985).Google Scholar

¹⁷ People who said ‘don't know’ in response to a question were placed at the centre of the scale for that item.

¹⁸ This analysis also produced a second component with an eigenvalue slightly greater than the conventional critical value of 1·0. The eigenvalue of the first component, however, was substantially larger than that of the second, while the eigenvalue of the second component was only slightly larger than those of the third and fourth. Again, the eigenvalues declined gradually from there. Only the first component was employed in our analysis. A principal factors solution yielded almost identical results.

¹⁹ Sears, David O., ‘The Person-Positivity Bias’, Journal of Personality and Social Psychology, XLIV (1982), 233–50.Google Scholar

²⁰ People who responded ‘don't know’ or ‘haven't thought much about it’ were rescored to place them at the centre of the scale.

²¹ Brady, , ‘Factor Analysis for Ordinal Data which are not Interpersonally Comparable’.Google Scholar

²² Feldman, Stanley, ‘Economic Individualism and Mass Belief Systems’, paper prepared for delivery at the annual meeting of the Midwest Political Science Association, Cincinnati, Ohio, 1982Google Scholar; see also his ‘Economic Self-interest and Political Behavior’, American Journal of Political Science, XXVI (1982), 446–66.Google Scholar

²³ People who said ‘don't know’ in response to a question were placed at the centre of the scale for that item. A principal factors solution yielded almost identical results.

²⁴ SAS User's Guide, 1979 Edition (Raleigh, North Carolina: SAS Institute, 1979), p. 204.Google Scholar

²⁵ These calculations are based upon information from Figure 1 and Table 4.

²⁶ Note that our measures of racial policy preferences and of explanations for racial inequality are both computed from scores based on our principal components analyses. Thus the units of the two scales are similar and the unstandardized estimates of their effects on one another are comparable. Note as well that the two measures have standard deviations that are approximately equal.

²⁷ Evidence of robustness can be found by comparing Figure 2 to the estimates for the well educated in Figure 3.

²⁸ We should admit that we ourselves are surprised at this result for the well educated, for it is the well educated that we expected to find most prone to reason from general principle to specific policy preference. To discover why it is that even the well educated give so much weight to their racial policy preferences in arriving at their explanations for racial inequality is an important task for future research.

²⁹ It is the lack of a connection between our instrument – ideology – and policy preferences among the poorly educated that is responsible for grossly inflating the standard error of the estimated effect of policy preferences on explanations for inequality.

³⁰ Kinder, Donald R., ‘Diversity and Complexity in American Public Opinion’, in Finifter, Ada W., ed., Political Science: The State of the Discipline (Washington, D.C.: American Political Science Association, 1983), pp. 389–425Google Scholar; see also Levitin, Teresa E. and Miller, Warren E., ‘Ideological Interpretations of Presidential Elections’, American Political Science Review, LXXIII (1979), 751–71CrossRefGoogle Scholar; and Campbell, et al. , The American Voter.Google Scholar

³¹ Converse, , ‘The Nature of Belief Systems in Mass Publics’.Google Scholar

³² Converse, , ‘The Nature of Belief Systems in Mass Publics’.Google Scholar

³³ For an extended discussion of possible limits on our ability to distinguish cognitive and affective explanations, see Tetlock, Philip E. and Levi, Ariel, ‘Attribution Bias: On the Inconclusiveness of the Cognition–Motivation Debate’, Journal of Experimental Social Psychology, XIX (1982), 68–88.CrossRefGoogle Scholar

³⁴ Converse, Philip E., ‘Public Opinion and Voting Behavior’, in Greenstein, Fred I. and Polsby, Nelson W., eds, Handbook of Political Science, Volume 4 (Reading, Mass.: Addison-Wesley, 1975), pp. 75–169Google Scholar; and Converse, , ‘The Nature of Belief Systems in Mass Publics’.Google Scholar

³⁵ Converse, , ‘The Nature of Belief Systems in Mass Publics’Google Scholar; see also Achen, Christopher H., ‘Mass Political Attitudes and the Survey Response’, American Political Science Review, LXIX (1975), 1218–31.CrossRefGoogle Scholar

³⁶ Feldman, Stanley, ‘The Reliability and Stability of Policy Positions: Evidence from a Five-Wave Panel Study’, paper presented at the annual meeting of the American Political Science Association, New Orleans, Louisiana, 1985.Google Scholar

³⁷ Simon, Herbert A., ‘Human Nature in Politics: The Dialogue of Psychology with Political Science’, American Political Science Review, LXXIX (1985), 293–304, p. 301.CrossRefGoogle Scholar

³⁸ Zaller, , ‘Toward a Theory of The Survey Response’.Google Scholar

³⁹ Harman, Gilbert, ‘Positive versus Negative Undermining in Belief Revision’, in Kornblith, Hilary, ed., Naturalizing Epistemology (Cambridge, Mass.: Bradford, 1985), pp. 231–48.Google Scholar

⁴⁰ Brady, Henry E. and Sniderman, Paul M., ‘Attitude Attribution: A Group Basis for Political Reasoning’, American Political Science Review, LXXIX (1985), 1061–78CrossRefGoogle Scholar; Chubb, John E., Hagen, Michael G. and Sniderman, Paul M., ‘A Two-dimensional Theory of Ideology’ (Working Paper on Political Science, No, P-84–2, Hoover Institution, Stanford University, 1984).Google Scholar

⁴¹ Zajonc, R. B., ‘On the Primacy of Affect’, American Psychologist, XXXIX (1984), 117–23.CrossRefGoogle Scholar

⁴² For an analysis of some respects in which racial issues differ from many other issues in American politics, see Carmines, Edward G. and Stimson, James A., ‘The Two Faces of Issue Voting’, American Political Science Review, LXXIV (1980), 78–91.CrossRefGoogle Scholar