This book deserves the attention of every scholar of American public opinion, voting, or elections. It should be among the required readings for every graduate course about American public opinion, voting, or elections. It is also worth considering as a central text for every upper-division undergraduate course on any of these topics. This is a recommendation that I am comfortable with, even though the findings, concepts, and underlying theories and models in the book are probably familiar to everyone who reads this review.

The American Voter Revisited is a topic-by-topic, table-by-table, and figure-by-figure replication of The American Voter (1960), with 2000 and 2004 American National Election Study data replacing the 1952 and 1956 data used by Angus Campbell and his colleagues. Sections and chapters repeat those in the earlier volume. Michael S. Lewis-Beck and his coauthors do not “try to do it better.” Their book is faithful to its replication design (a term often misapplied in the social sciences when subsequent studies of the same issue do a different analysis). Voter Revisited demonstrates that the variables and analyses used by Voter to describe the political attitudes, perceptions, and behavior of Americans in the 1950s are as analytically meaningful now as they were then. It establishes the continuing descriptive utility of that volume's core variables and the virtually unchanged relationships among them. The authors describe the data in approximately the same way (but with a significantly better writing style) and offer the same descriptive statements and conclusions about the citizenry presented.

Each chapter ends with a “Commentary and Controversy” section that indicates where subsequent research took the basic findings of the original chapter and assesses the scholarly impact of the original findings, whether and how the findings were criticized, and if the conclusions required significant modification. Typically, the authors also present some suggestions about what future research might be expected on the subject of the chapter. Overall, the book is an elegant and persuasive testimonial to the fundamental soundness of the earlier volume since, almost without exception, the patterns in the 2000 and 2004 data repeat what Campbell and his colleagues observed in their 1950s surveys.

This is a substantial achievement. The data-analysis manipulations required tremendous effort (peruse, for example, the analysis of open-ended data for the chapter on perceptions of the parties and the candidates), and the substance of the book is a significant contribution. In the first place, as Converse observes in the Foreword, the book has “obvious scientific merit of its own, addressing questions that have lingered for nearly half a century [about] how many and which of the 1950s findings were shaped by oddities [of the 1950s] as opposed to being … lasting features of electoral politics in the United States” (p.v). It is a needed confirmation because we continually cite a book of 50-year-old data to students (and citizens), who (especially the students) occasionally find it hard to accept such data as evidence of how citizens think and behave today. Intervening events, social changes, and population replacement justify questions about what still applies. After all, the youngest respondent in The American Voter's tables is 75 years of age today, and the median respondent died in about 1985. However, the updated book should quiet skeptics because it finds virtually identical patterns to those reported earlier, persuasive evidence that those authors got the fundamental facts right in 1960.

Second, this is a valuable “foundation textbook” to which we and others can refer when we want a comprehensive presentation of some proposition that is so fundamental that it is a part of what we all “just know” and, therefore, often not a subject in much recent literature. To be sure, there are many texts (often assigned in our undergraduate classes) with tables and figures that present basic factoids (the correlation between party identification and the vote) and paragraphs of text that tell the reader what they should believe the factoid demonstrates about the political life of Americans. However, there is arguably nothing to refer to if an inquiry also seeks an explanation that provides the logic that underlies the analysis and the conclusions about the data. The American Voter Revisited provides this essential conceptual and substantive assessment by explaining how the analysis was conducted, why something was done in a particular manner, and how it should be understood. That is, Lewis-Beck and his colleagues explain the replicated analysis as one would if this were an original assessment of the American voter. So unless you have recently read the earlier volume in its entirety, a cover-to-cover reading of the book under review almost seems fresh. For most students and interested nonacademics without our background, it will be a fresh and easier experience because of the good quality of the prose.

Finally, this book may help to ground some current research or even reinvigorate it (the latter is an announced intention of this project). Certainly, the former may be needed because it is not clear to me that all the citers of The American Voter read it. We often think we know what the foundational literature said and how the analysis was done, but anyone familiar with the earlier material is often aware of a gap between what is attributed to such books such and what the authors actually did and wrote. Sometimes we forget work and at other times we unknowingly repeat analyses that were done years ago. This book may re-establish an awareness of the foundations of the field and follow-up studies in a way that improves future research.

The current book does not revisit all the findings of Voter. Four chapters—“Election Laws and the Political Environment,” Agrarian Political Behavior,” “Population Movement,” and “Personality Factors in Voting Behavior” (respectively, Chapters 11, 15, 16, and 18)—are missing in this book, and the chapters on social class (13) and other social differences (17) are combined into a single new chapter. Few are likely to miss the chapters on agrarian political behavior and personality. On the other hand, a replication of Voter's election law and political environment chapter seems like an analysis that would draw attention, given the substantial post-AV literature on institutional effects. Arguably, these deleted analyses are peripheral to the central themes of the books, and their absence has no effect on the value of this book.

Like its template, The American Voter Revisited is not a study of elections but of voters; it stays close to the earlier approach. In doing so, Lewis-Beck, and his coauthors have given us a book to read and to refer to others as a definitive statement of our understanding of voters: what matters to them and what does not, and what categories and frameworks Americans use to view public affairs, the candidates, and the parties as they function as citizens.