¹ The collection of data from a national sample of the electorate around the 1964 election was made possible by a grant to the Survey Research Center of the University of Michigan from the Carnegie Corporation of New York, which had also supported the 1952 election study.

² The most fertile elaboration of this classic script is of course contained in Downs, Anthony, An Economic Theory of Democracy (New York, 1957)Google Scholar.

³ The New York Times, July 19, 1964.

⁴ This datum is not as absurd as it might appear if the reader has failed to grasp the import of the preceding text. That is, in 1952 it was the most intense and ideologically “pure” Republicans who tended to prefer Taft to Eisenhower, much as 12 years later their counterparts chose Goldwater over the other Republican alternatives. It was the less ideologically committed (either by persuasion or by lack of ideological sensitivity) who were more satisfied with the Eisenhower candidature. The erstwhile Taft supporters did not perversely turn out at higher rates because they were disappointed in the convention choice, but because their striking commitment to Republicanism compelled them to more ardent support of its candidate whatever his ideological position.

⁵ See “The Concept of a ‘Normal Vote,’” ch. 1 in Campbell, A., Converse, P., Miller, W. and Stokes, D., Elections and the Political Order (New York, 1965.Google Scholar)

⁶ In our estimation, some challengers of this description have been prematurely discouraged from competition by poll results which might well have changed radically with greater exposure.

⁷ Lodge's strong grass-roots popularity was one of the untold stories of the 1960 election, when he ran for vice-president. Well-known for his televised confrontations with the Russian delegation in the United Nations, he was far and away the most widely recognized and warmly regarded first-time vice-presidential candidate in the elections we have studied. Given the tarnish which seems to accompany second efforts at the presidency in American elections and which would undoubtedly have hurt Nixon, it may well be that Lodge, had he been acceptable to the Republican Party leadership, could have pushed Lyndon Johnson to a closer race than any other of the Republican hopefuls.

⁸ While these rates may sound mountainous, it should be remembered that the expected defection rates for most of these groups are rather low—in the vicinity of 10 percent. Nonetheless, 40 percent of the Rockefeller Republicans in our sample voted for Johnson.

⁹ We have examined this possibility in some seriousness simply because often in the past we have found public perceptions of party differences on major issues totally confused and muddy. Even on issues where the politically sophisticated see marked party differences, general public inattention and the ambiguities which politicians exploit to blur the edges of their positions combine to produce either lack of recognition of differences, or very conflicting impressions of what those differences are at any given point. See Campbell, et al. , The American Voter (New York, 1960), pp. 179ffGoogle Scholar.

¹⁰ The statistic is such that if all citizens in the sample agreed that the Democrats represented one side of the issue and the Republicans the other, the figure would be 1.00 (perfect association). A figure of .00 represents the case of no aggregate association whatever.

¹¹ Converse, Philip E., “A Major Political Realignment in the South?” in Sindler, Allan P., ed., Change in the Contemporary South (Durham, N. C., Duke University Press, 1963)Google Scholar.

¹² These religious effects were described in Converse, et al., “Stability and Change in 1960: a Reinstating Election,” this Review, Vol. 55 (06, 1961), pp. 269–80Google Scholar.

¹³ In our data, expressions of party loyalty from the South which had been slowly losing Democratic strength throughout the 1950s show a sudden rebound in 1964. However, all of the rebound can be traced to Southern Negroes; the downward trend among Southern whites continued and at about the same pace.

¹⁴ Take, for example, the charge hung on Goldwater by Democrats and some Republicans that he was “impulsive.” This allegation reverberated in the public and came to make up one of our largest single categories of negative references to Goldwater. “Impulsiveness” is a personality trait that on one hand might have been less plausible for some other right-wing leader. Yet the charge took roots and began to flourish with respect to a cluster of policies that Goldwater shared with other Republican leaders of similar persuasions. It seems quite arbitrary to decide that it is exclusively either the person or the policy which is “impulsive.”

¹⁵ Undoubtedly, for such an observer, letters were not weighted equally in his impressions as to how opinion stood: some were more cogent than others, some were more distressed, and so on. But as a rough first approximation, one can imagine that what registered as “public opinion” on a particular issue in the mind of such an observer was closely related to the simple frequency of letters pro and con.

¹⁶ Data on letters to the news media are not presented graphically, in part because the inequality is so complete that there is little one can discriminate in the figure. The Gini index of concentration for the newspaper and magazine letters is .99. See Alker, H. and Russett, B., “On Measuring Inequality,” Behavioral Science, Vol. 9, No. 3 (07, 1964), pp. 207–18CrossRefGoogle ScholarPubMed.

¹⁷ We wish to stress that it remains a crude approximation, in part because we do not know, letter by letter, what political opinions the respondent was expressing. Conceivably in many cases they lay outside the range of any of our items. But the exercise is worth completing in part because it is likely that our hypothetical observer generalizes beyond the specific content of letters (“if ultra-conservative opinion on issue x is running about 30 percent, then it is likely that ultra-conservative opinion on issue y would run about the same level if something made that issue salient”); and in part because the systematic lines of displacement of “letter opinion” from “public opinion” in the mass electoral sense are undoubtedly valid in their general direction, whatever the details.

¹⁸ The wordings of the issue items involved in Figure 2(c) and (d) were as follows:

(For 2c) “Some people are afraid the government in Washington is getting too powerful for the good of the country and the individual person. Others feel that the government in Washington has not gotten too strong for the good of the country…. What is your feeling?”

(For 2d) “Some people think our government should sit down and talk to the leaders of the Communist countries and try to settle our differences, while others think we should refuse to have anything to do with them…. What do you think?”

Figure 2(e) is based on a set of questions that asked people to indicate their affective reactions toward a variety of groups, including “conservatives” and “liberals.” The scores for the figure are based on the difference in reaction to the two stimuli.

¹⁹ It is likely that this contingent is roughly coterminous with that 40–50 percent of the American electorate which we have described elsewhere as having no impression as to what such terms as “conservative” and “liberal” mean. See Converse, Philip E., “The Nature of Belief Systems in Mass Publics,” in Apter, David E., ed., Ideology and Discontent (New York, 1964), pp. 206–61Google Scholar. The data presented there were gathered in 1960. In the 1964 study we collected the same data on recognition of ideological terms, thinking that perhaps the nature of the Goldwater campaign might render these terms and meanings more salient to a wider public The data show that it did not.