¹ Support for the notion that the federal government is the most salient level of the political system can be found in Jennings, M. Kent and Zeigler, Harmon L., “The Salience of American State Politics,” American Political Science Review, 64 (June, 1970), 523–535CrossRefGoogle Scholar.

² A more complete conceptual and theoretical statement of the different dimensions of alienation is presented in Finifter, Ada W., “Dimensions of Political Alienation,” American Political Science Review, 64 (June, 1970), 389–410CrossRefGoogle Scholar, and Gamson, William A., Power and Discontent (Homewood, Illinois: Dorsey Press, 1968)Google Scholar.

³ For a discussion of the efficacy trend, see Converse, Philip E., “Change in the American Electorate,” in Campbell, Angus and Converse, Philip E., eds., The Human Meaning of Social Change (New York: Russell Sage Foundation, 1972)Google Scholar.

⁴ The three government responsiveness questions were:

1. How much do you feel that political parties help to make the government pay attention to what the people think: a good deal, some, or not much?

2. And how much do you feel that having elections makes the government pay attention to what people think: a good deal, some, or not much?

3. How much attention do you think most congressmen pay to the people who elect them when they decide what to do in Congress: a good deal, some, or not much? A percentage difference index was computed for each of the three questions by subtracting the percentage responding “not much” from the percentage answering “a good deal.” Positive values of the PDI therefore specify the degree of confidence, whereas negative values denote a lack of confidence. These PDI values are plotted in Figure 1 for whites and blacks.

It should be noted that these items indicate the respondent's confidence in elections, parties and congressmen to make government responsive and not the respondent's confidence, per se, in the institutions and leaders referred to by the questions. The interpretation of the items as indicators of “diffuse public support” was suggested by Dennis, Jack in “Support for the Institution of Elections by the Mass Public,” American Political Science Review, 64 (September, 1970), 819–835CrossRefGoogle Scholar.

⁵ Dennis, Jack, “Public Support for American National Political Institutions,” (paper presented at the Conference on Public Support for the Political System, Madison, Wisconsin, August 13–17, 1973)Google Scholar.

⁶ Gamson, , Power and Discontent, p. 51Google Scholar.

⁷ This statement is very straightforward; it is emphasized here because Citrin has ignored the strong correlation between the “change in form of government question” and political trust, choosing instead to call attention to only a few cells in the cross-tabulation of the two variables.

The reader should also be aware of an important difference between the trust in government index that Citrin uses and the trust in government scale used in “Political Issues and Trust in Government” and in this “Rejoinder.” The scale scores reflect the individual response pattern for the five trust in government items ordered according to the proportion of cynical responses. Individuals with no more than two missing data responses or scale errors were assigned the score that is the median value of the possible scale scores for that particular response pattern. Respondents with more than two errors or two missing data responses were excluded from the analysis. Citrin, however, simply formed an index by adding up responses to the five trust items, thereby ignoring the cumulative nature of the items. This method difference affects the overall distribution of the trust measure as indicated below for 1972:

^a Figures taken from Citrin's Table 2, this issue of the Review.

The index and scale scores are no doubt highly correlated but they are not identical. Citrin's analysis, therefore, cannot be viewed as a replication of “Political Issues and Trust in Government,” nor is it strictly comparable to the analysis presented in this “Rejoinder.”

The conceptual distinction between an additive index and a scale is well established in the empirical literature; for example, see Peak, Helen, “Problems of Objective Observation,” in Festinger, Leon and Katz, Daniel (eds.), Research Methods in the Behavioral Sciences (New York: Holt, Rinehay and Winston, 1953)Google Scholar, chapter 6. For a more recent statement of the response models underlying unidimensional scales, see Weisberg, Herbert F., “Dimensional Analysis of Legislative Roll Calls,” (Ph.D. dissertation, The University of Michigan, 1968)Google Scholar or Weisberg, Herbert F., “Dimensionland: Excursions into Spaces,” American Journal of Political Science (forthcoming, November, 1974)CrossRefGoogle Scholar.

⁸ This is the same item referred to by Citrin. The question was contributed to the 1972 CPS election study by J. Merrill Shanks.

⁹ The operational definition of the ideological orientations used here is the same as that used in “Political Issues and Trust in Government.”

¹⁰ For the complete relationship among the four variables (political efficacy, political trust, turnout and two-party vote), see Table 4 of Miller, Arthur H., Miller, Warren E., Raine, Alden S. and Brown, Thad A., “A Majority Party in Disarray: Policy Polarization in the 1972 Election,” (paper presented at the 1973 Annual Meeting of the American Political Science Association, New Orleans, Louisiana, September 4–8)Google Scholar.

¹¹ Miller, Miller, Raine and Brown, Table 4.

¹² For a detailed discussion of these data, see Miller, Miller, Raine and Brown, pp. 47–53. The concept of the normal vote is discussed in Converse, Philip E., “The Concept of a Normal Vote” in Campbell, Angus, Converse, Philip E., Miller, Warren E. and Stokes, Donald E., Elections and the Political Order (New York: John Wiley and Sons, 1966), pp. 9–39Google Scholar.

¹³ This relationship is discussed in Miller, Miller, Raine and Brown, p. 64.

¹⁴ Citrin, Jack, “Comment,” this issue of the Review, p. 976Google Scholar.

¹⁵ The political protest questions were worded as follows:

There are many possible ways for people to show their disapproval or disagreement with governmental policies and actions. I am going to describe three such ways. We would like to know which ones you approve of as ways of showing dissatisfaction with the government, and which ones you disapprove of.

1. How about taking part in protest meetings or marches that are permitted by the local authorities? Would you approve of taking part, disapprove, or would it depend on the circumstances?

2. How about refusing to obey a law which one thinks is unjust, if the person feels so strongly about it that he is willing to go to jail rather than obey the law? Would you approve of a person doing that, disapprove, or would it depend on the circumstances?

3. Suppose all other methods have failed and the person decides to try to stop the government from going about its usual activities with sit-ins, mass meetings, demonstrations, and things like that? Would you approve of that, disapprove, or would it depend on the circumstances?

¹⁶ Citrin, p. 982.

¹⁷ The average location of the Democratic party on the five issues of Vietnam, Urban Unrest, Campus Unrest, Protecting the Rights of the Accused, and Government Aid to Minorities was 3.49 in 1970 and 3.25 in 1972 for the total population. In 1970, Independents had placed the Democratic party, on the average, at 3.58 and in 1972 they located it at 3.22. Republicans placed the Democratic party at 3.20 in 1970 and 3.02 in 1972. For a further discussion of these shifts, see Miller, Miller, Raine and Brown, pp. 13–16.

¹⁸ Other factors, such as increased group identification among blacks, were substantially affecting black cynicism in 1972. Blacks who were not necessarily dissatisfied with the particular policy items analyzed were becoming distrusting for other reasons, thereby reducing the clarity and linearity of the policy dissatisfaction/cynicism relationship. The curvilinearity of the 1972 relationship between dissatisfaction with Democratic policy alternatives and trust partially reflects strength of party identification and the associated desire to replace Republican policies with Democratic policies. The curvilinearity also results, however, from perceived discrimination and patterns of social integration among blacks. For a discussion of these sources of black political cynicism, see Miller, Arthur H., Raine, Alden S. and Brown, Thad A., “Racial Trends in Political Estrangement, 1958–1972,” (unpublished paper, The University of Michigan, Center for Political Studies, 1974)Google Scholar.

¹⁹ The Nixon thermometer rating, rather than the performance evaluation item Citrin used in his Table 2, was used for the regression analysis because the performance question was only asked of half the 1972 sample and thus could not be correlated with all of the issue measures included in the regression analysis. It should be noted that the Nixon thermometer and the performance evaluation questions were strongly correlated (r = .68) for that half-sample that presented both items; and the zero-order correlations of each item with trust were nearly identical (.31 and .32, respectively). It is therefore unlikely that the regression results would have differed had the performance question been used in place of the thermometer.

²⁰ Citrin, , “Comment,” pp. 976–978Google Scholar.

²¹ Citrin, , “Comment,” p. 978Google Scholar.