Why Is Turnout Down?

Abstract

The monotonic decline in turnout in presidential elections since 1960 is the subject of this analysis of survey data. After some common explanations for this decline were rejected, it was discovered that the decline occurred mainly among low-income and low-education whites. Two explanatory hypotheses were examined, but appropriate data for testing them were unavailable; however, in the 1970s nonvoters were more likely than voters at all income levels to express dissatisfaction with the political system. Nonvoting whites are not always a Democratic group, and their voting behavior is unpredictable over time. Their failure to vote may have an especially significant impact on Democratic party policies, and implies that palliatives like reform of voter registration laws may not have the desired effect. Howard Reiter is Assistant Professor of Political Science at the University of Connecticut at Storrs. Financial assistance was provided by the University of Connecticut Research Foundation, and technical assistance was provided by the staff of the Social Science Data Center and the Roper Center of the University of Connecticut. The author is indebted to his colleagues Fred Kort and Everett Carll Ladd, Jr., for their advice. Anonymous referees were also helpful. Public Opinion Quarterly ?) 1979 by The Trustees of Columbia University Published by Elsevier North-HoDland, Inc. 0033-362X/79/0043-297/$1.75 This content downloaded from 207.46.13.28 on Wed, 31 Aug 2016 05:28:40 UTC All use subject to http://about.jstor.org/terms 298 HOWARD L. REITER with this decline, and to speculate on possible causes and effects. We should expect no simple answers, and in this relatively brief essay we will present some initial findings that demand further analysis at greater length. Because some of the variables we will investigate are not accessible in aggregate form, we are using the surveys conducted by the Survey Research Center of the Center for Political Studies of the University of Michigan.' SRC/CPS turnout is well above that reported by the Census Bureau's aggregate data, but the trend over time is parallel to the aggregate data trend, and as we examine groups in the population we must assume that the degree of overreporting is comparable among all of them. The political participation literature is filled with factors associated with high or low turnout. Some are demographic: nonvoters include disproportionate numbers of nonwhites, women, Protestants, southerners, the young and the old, rural dwellers, those of low education, income, and occupational status, and workers who do not belong to labor unions. Psychological factors come into play, too, as nonvoters are said to be disproportionately low in their sense of political efficacy, personal competence, political involvement (including media usage), and trust in government. Political factors associated with nonvoting include nonpartisanship, weak preference for one candidate over the other, and expectation of a landslide in which one's vote is especially unlikely to make a difference (Lane, 1959:46-52; Milbrath, 1965:58-62). We will examine all these factors2 except for registration requirements, because the SRC/CPS sample does not represent all states. For reasons to be discussed below, such laws probably did not affect the decline in turnout, and the most careful study that we have of the effects of these registration laws produces conclusions not far from our own (Rosenstone and Wolfinger, 1978:22-45). The first observation to be made is that blacks were not part of this trend, as Figure 1 demonstrates. The dramatic-rise in the turnout of Southern blacks from 1960 to 1968 was clearly the result of deliberate political mobilization and the destruction of the many barriers to the franchise, but even in the North, black turnout held fairly steady. As we have reported elsewhere, the period under scrutiny was one in which blacks became much more interested in campaigns than they had been earlier (Reiter, 1975). Whites in both regions experienced roughly parallel declines, with a rise only in the turnout of southern I The data were made available by the Inter-University Consortium for Political Research, and neither the SRC/CPS nor the ICPR bears any responsibility for the present analysis or interpretation. 2 Limitations of space preclude discussion of our operationalization of these variables, but the author will be happy to respond to queries about it. This content downloaded from 207.46.13.28 on Wed, 31 Aug 2016 05:28:40 UTC All use subject to http://about.jstor.org/terms WHY IS TURNOUT DOWN? 299