In 1960, The American Voter, by Angus Campbell, Philip E. Converse, Warren E. Miller, and Donald E. Stokes, introduced the social-psychological approach to the study of voting. The core of the authors' theory was the mediating role that long-term psychological predispositions, particularly party attachment, play in individual voting decisions. In the decades since its publication, the book has spurred a wealth of scholarly research in this country and others. It also has generated questions of how well its findings have stood the tests of time. Now, almost fifty years later, a quartet of scholars has taken on the task of replicating The American Voter, focusing on the 2000 and 2004 presidential elections with an eye toward comparing today's American voter with that described a half-century ago. The end result is a well-crafted book that both updates and expands on the original foundational work.
In The American Voter Revisited, Michael S. Lewis-Beck, William G. Jacoby, Helmut Norpoth, and Herbert F. Weisberg retain much of the original authors' approach. Campbell and associates used survey data and focused on the 1952 and 1956 presidential elections. The new book uses National Elections Studies to examine the 2000 and 2004 presidential elections. The authors follow the same theory and book format as the original classic. Most chapters mirror themes and correspond to chapters in the 1960 book. The updated book deletes some chapters—for instance a chapter on agrarian political behavior was no longer necessary—to accommodate changes in election studies over time and also out of recognition of changing national priorities. It also takes advantage of more advanced statistical methods, for instance using logit rather than regression to capture the effect of partisan attitudes on vote decisions.
Notably different, the authors conclude each chapter in The American Voter Revisited with commentary and controversy sections. These sections outline advances in the study of voting behavior since the original book and also controversies in the field. For instance, in the chapter of the book outlining its theoretical orientation, the authors discuss alternative approaches to the study of voting behavior. The section provides a nice discussion of rational choice theory, as well as sociological, psychological, and historical institutional approaches. In each of the analytical chapters, the authors confront head-on controversies surrounding the approach used throughout the book. For instance, the authors address issues surrounding the use of open-ended questions or the reliability of likes and dislikes questions. In the chapter on partisan choice, the concluding section details the assumptions made when using partisan attitudes as determinants of the vote. It would require too much space to discuss the entirety of these sections. However, the discussions in the commentary and controversy sections are compelling and well done. These sections alone make this updated version of the original book worthwhile. As Philip E. Converse wrote in his foreword to American Voter Revisited, the sections are “invaluable source materials in and of themselves, and combined with the comparisons of raw findings, produce a sort of double prism in the time dimension that is at the very least novel, and to me thoroughly intriguing.”
The basic model employed in the 1960 and 2008 versions of the book stresses the role of attitudes in determining the vote (described on page 62 in The American Voter Revisited). The theory posits that a person's attitude toward political objects affects that person's vote. The authors focus on six political objects—the Democratic presidential nominee, the Republican presidential nominee, foreign policy issues, domestic policy issues, social group interests, and how well the parties manage the government. Individual voters evaluate each of these components either positively or negatively and that influences the vote choice. Using a series of NES open-ended questions, the authors develop two measures: one capturing whether the voter evaluated the object in a Republican, neutral, or Democratic way; and a second capturing intensity. To affect the vote outcome, the component must have both a strong effect on aggregated individual vote decisions and also work to the strong advantage of one candidate or the other.
The authors note that an individual's attitude, however, is not always based on current events or re-created every election. As did the earlier authors, Lewis-Beck and associates argue that political orientations are longstanding and that those longstanding political orientations influence political attitudes. Party identification is one long-term political orientation that has a large effect on a person's political attitudes, but it is not the only influence. In some national elections, party identification strongly influences vote outcomes, while in others, it is less of a factor. In the 2000 and 2004 elections, there was little deviation from long-term partisanship, the authors maintain.
Much of the discussion in the wake of the original American Voter was its conclusion that the American voter was largely ill-informed and the “mass electorate is not able either to appraise (government's) goals or the appropriateness of the means chosen to secure these goals.” Decades of research have supported the conclusion that American voters are relatively ignorant about government activities. Lewis-Beck and his colleagues do not provide a much rosier portrait of the American voter. They acknowledge the level of education in the United States is higher today than a half-century ago and that there is greater issue-voting and somewhat greater structure of issue-attitudes. Yet, they also find that the growth in political awareness is limited. “The average voter has at best partial acquaintance with the salient issues of the day … [A] substantial number of citizens undoubtedly experience difficulties during the process of ‘translating’ their own feelings about issues and policies into a choice between presidential candidates” (416).
At the same time, the current authors reject assertions that the original American Voter described voters as fools or their voting as “based on immutable party identification and superficial characteristics of the candidates, because voters were too unsophisticated to consider ideology and the issues” (ix). The original book and The American Voter Revisited instead portray voters as rational in their approach to politics, investing as little time or energy as necessary in informing themselves of political events. Although politics remain peripheral to citizens' daily concern, voters do acquire information about politics. In summing up the chapter on public policy and political preference, The American Voter Revisited points to recent research that indicates that a substantial number of citizens do vote correctly in terms of their policy interests (Richard R. Lau and David P. Redlawsk, Voting Correctly, 1997) and that policy leaders are responsive to the macro-polity (Robert S. Erikson, Michael B. MacKuen, and James A. Stimson, The Macro Polity, 2002).
In conclusion, The American Voter Revisited confronts the question of how the American voter of 2000 and 2004 compares with the voter of fifty years ago. Acknowledging the difficulty in answering such a broad question, the authors pose it in a more concise way: “Does the social-psychological explanation of presidential vote choice, developed in The American Voter and symbolized in its famous theory of the ‘funnel of causality,’ still hold?” The authors answer the question with a “yes.”
The American Voter Revisited is an honest and thorough replication of the original classic. The four authors of the updated book are diligent in their comprehensive examination of today's voter. They include detailed discussions of advances in the study of voting behavior and controversies in the literature. A series of appendixes compare tables and figures in the original book with those in the updated volume, as well as outlining problems with replication. In the end, the book shows that in a political environment that is more polarized than fifty years ago and in a vastly different media and social environment, much of what was revealed about that American voter in 1960 remains the same today.
