¹ The Survey Research Center has conducted studies of voting in the presidential elections of 1948, 1952, and 1956, as well as in the congressional elections of 1954. The sources of its data are samples of persons drawn by strict probability methods to represent the adult population of the United States living in private households. Statements of the sample design and sampling errors and copies of the questionnaires may be obtained from the Survey Research Center upon request.

² Key, V. O. Jr.,, and Munger, Frank, Social Determinism and Electoral Decision: The Case of Indiana. Unpublished manuscript, p. 21Google Scholar. If we assume that social characteristics are relatively constant through time yet vary across our several states, this point is implicit in the analysis of voting statistics by states in a number of other works. See, for example, Schattschneider, E. E., Party Government (New York, 1942), pp. 112–114Google Scholar, and Bean, Louis H., How to Predict Elections (New York, 1948), pp. 105–122Google Scholar.

³ The form of the four party questions was: “I'd like to ask you what you think are the good and bad points about the two parties. Is there anything in particular that you (like, don't like) about the (Democratic, Republican) Party? What is that?” The form of the four candidate questions was: “Now I'd like to ask you about the good and bad points of the two candidates for President. Is there anything in particular about (Stevenson, Eisenhower) that might make you want to vote (for him, against him)? What is that?”

⁴ The idea of a zero-point separating positive from negative, favorable from unfavorable, pro from anti, etc. has arisen a number of times in research on attitudes. See, for example, Thurstone, L. L. and Chave, E. J., The Measurement of Attitude (Chicago, 1929), pp. xi ff.Google Scholar; Krech, David and Crutchfield, Richard S., Theory and Problems of Social Psychology (New York, 1948), pp. 227–232CrossRefGoogle Scholar; and Green, Bert F., “Attitude Measurement,” in Lindzey, Gardner, ed., Handbook of Social Psychology (Cambridge, 1954), Vol. 1, pp. 357–358Google Scholar. For an application of Thurstone's ideas to the measurement of attitudes toward presidential candidates, see Beyle, Herman C., “A Scale for the Measurement of Attitude toward Candidates for Elective Office,” this Review, Vol. 26 (1932), pp. 527–544Google Scholar.

⁵ The procedure used here to confirm the location of the zero-point follows in some respects that developed by Guttman and Suchman. See Suchman, Edward A., “The Intensity Component in Attitude and Opinion Research,” in Stouffer, Samuel A. et al. , Measurement and Prediction, Vol. 4 of Studies in Social Psychology in World War II (Princeton, 1950), pp. 213–276Google Scholar; and Guttman, Louis and Suchman, Edward A., “Intensity and Zero Point for Attitude Analysis,” American Sociological Review, Vol. 12 (1947), pp. 57–67CrossRefGoogle Scholar.

⁶ The use of the term “group” here is intended to be quite inclusive. Some of the groups referred to could be thought to have a self-conscious existence; others were simply elements or categories within the population.

⁷ The findings of this section refer entirely to voters. Non-voters are excluded partly because we are interested in the actual vote and partly because the non-voter's post-election report of his preference seems peculiarly sensitive to the election returns.

⁸ For a precise description of the quantities which are compared in Figure 1, consult the note appended to this report.

⁹ For a more precise description of the methods used here, consult the note appended to this report.