Published online by Cambridge University Press: 02 September 2013
Past voting turnout studies almost always have been static analyses. Usually they have described the relationship between participation rates and the demographic attributes, attitudes, and social experiences of members of the electorate. Since each such study ordinarily is based on a single cross-sectional survey or on statistics referring to the Election Day period alone, both turnout and its determinants are derived at the same point of time, only simultaneous correlations are possible, and the analyst cannot show how turnout is affected by temporally prior conditions.
The development of multi-wave panel interviews in modern social research permits the study of attitude change, decision-making, and action over time. By re-interviewing the same respondents at intervals, political sociologists already have discovered much about how voters decide their candidate choices during the course of an election campaign. A panel design permits such process analysis not only of candidate preference but also of turnout and non-voting.
1 My chapter in a book on Congressional voting behavior edited by William N. McPhee and scheduled for publication by The Free Press in 1959.
2 The principal publication is McPhee (ed.), ibid. Earlier findings appear in the Freeman and Showel articles and in the A.A.P.O.R. roundtable, all in The Public Opinion Quarterly, Vol. 15, pp. 703–714 and 805–807Google Scholar; and Vol. 17, pp. 288–292. See also William A. Glaser, “The Family and Voting Turnout,” forthcoming paper.
3 Berelson, Bernard et al. , Voting (Chicago, 1954), p. 31Google Scholar.
4 For example, in rough proportion to the relative prices of their various desired purchases, consumers tend to plan their shopping more soberly, and their expressed intentions more clearly predict their behavior. Katona, George, Psychological Analysis of Economic Behavior (New York: McGraw-Hill Book Company, 1951), pp. 82–83Google Scholar. For an act as serious as moving one's residence, the intention to move diverges greatly from the desire to move, and the former predicts mobility far more accurately than the latter. Rossi, Peter H., Why Families Move (Glencoe: The Free Press, 1955), Ch. VIGoogle Scholar. In the case of electoral participation, obviously people make no such clearcut distinction between “intention to vote” and “desire to vote.” All such findings corroborate the familiar principle in psychology that the more an intention corresponds to a central need of the organism, the less likely it will be forgotten. For a development of this idea, see Lewin, Kurt, Intention, Will and Need, in Rapaport, David (editor), Organization and Pathology of Thought (New York, Columbia University Press, 1951)CrossRefGoogle Scholar, Chapter 5. Our questionnaire items asking about intention did not remind the respondent about the barriers which might subsequently confront him, and voting turnout is the kind of action where the penalties for acting (resulting from neglect of some rival commitment) often exceed the penalties against abstention. On the weakness of questionnaire items which fail to confront the respondent with future dilemmas, see Dollard, John, “Under What Conditions Do Opinions Predict Behavior?”, Public Opinion Quarterly, Volume 12 (Winter 1948–1949), p. 632CrossRefGoogle Scholar.
5 My paper in Lazarsfeld, Paul F. et al. , (eds.), The Panel Method (Glencoe: The Free Press, 1959)Google Scholar.
6 The distribution of motivations, social stimuli, and role prescriptions among different social statuses in the Regional Panels data is discussed and documented more fully in William N. McPhee (ed.), op. cit.
Comments
No Comments have been published for this article.