In lieu of an abstract, here is a brief excerpt of the content:

Reviewed by:
  • Collective Preferences in Democratic Politics: Opinion Surveys and the Will of the People
  • Amy Fried
Collective Preferences in Democratic Politics: Opinion Surveys and the Will of the People. By Scott L. Althaus. Cambridge, UK: Cambridge University Press, 2003; pp 370. $75.00 cloth; $28.00 paper.

Invoking "the people" is an old rhetorical approach in politics. In democracies and other political systems, arguing that the public is on one's side serves to give an argument greater legitimacy. Furthermore, in the United [End Page 524] States, a place of perceived social equality, references to what Americans think may have an enhanced effect. As Tocqueville wrote, "The nearer men are to a common level of uniformity, the less they are inclined to believe blindly in any man of any class. But they are readier to trust the mass, and public opinion becomes more and more mistress of the world." In our time, we turn to polls to learn what the people think. Whether or not we would wish citizens (and politicians) to follow surveys, these numerical indicators suggest that public opinion can be discerned.

In Collective Preferences in Democratic Politics, Scott L. Althaus is deeply concerned with the quality of public opinion and the ways that surveys construct citizens' views. Althaus, an associate professor of speech communication and political science at the University of Illinois, Urbana-Champaign, brings to his task a vibrant multidisciplinary perspective. The book provides a systematic, rigorous, and careful empirical analysis that disputes scholars who have concluded that holding minimal information can nonetheless lead to reasonable political assessments.

Althaus examines the differences political knowledge makes. One impact is on overall reports of public opinion. People who say "don't know" when faced with a survey question are different, in other ways, from people who choose one of the survey options. Those who answer questions are more apt to be male, wealthier, better educated, and more partisan than those who do not, and their views have a disproportionate impact on survey results. The politically advantaged thus speak with a louder voice than is fair.

Furthermore, using a method of simulating the impact of knowledge on political views, Althaus concludes that, in many cases, having knowledge affects individuals' assessments. The core method involves simulating the impact of having knowledge on political views, by comparing actually surveyed opinion to the opinions that would have been held by the population if everyone were well-informed, or "fully informed." By using multiple variables, Althaus takes demographic variations into account, and finds large and minimal gaps between simulated and surveyed opinion. These gaps, or "information effects," vary by issue and over time. For example, Althaus finds a small "information effect" in views toward job discrimination toward gays in 1988, a larger one in 1992, and a little one in 1996.

But Althaus's book is more than a creative and methodical analysis of the effect of knowledge on responses. Althaus grounds his empirical analysis in democratic theory as well as contemporary debates about the meaning of public opinion and the usefulness of surveys for understanding public views. Collective Preferences also does well in criticizing how pollsters and journalists use polls and provides food for thought on the proper place of survey information in politicians' decisions. [End Page 525]

While the book is a significant contribution to research on citizen knowledge and public opinion more broadly, it has certain limits and problems, some of which are conceptual and some of which are related to the use of particular data. For one, Althaus relies on an operationalization of "fully informed" opinion, which is rather general. A person is "fully informed" if she can do such things as identify the office held by certain politicians and the issue stances of the major parties. This approach does not capture whether the person answering the question is informed about the subject of the questions, nor does it take into account the tendency for social group members to be especially cognizant of issues that affect their group. As Michael Delli Carpini and Scott Keeter note in What Americans Know about Politics and Why It Matters, "[M]ost citizens are political generalists, meaning that those who are knowledgeable...

pdf