Printer Friendly

Media Polls in American Politics.

Media Polls in American Politics Thomas E. Mann and Gary R. Orren, eds. (Washington, DC. The Brookings Institution, 1992) 172pp,

The publication of Media and Polls in American Politics is timely. Over the last dozen years the American people have been bombarded with media polls measuring public opinion on everything from the abortion debate to the changing role of the U.S. military abroad. Polls are now so numerous that they have become less meaningful: Americans now have less confidence, not only in the validity of the polls themselves, but also in the journalists and politicians who live and die by them. Media Polls, edited by Thomas E. Mann and Gary R. Orren of the Brookings Institution and Harvard's Kennedy School of Government respectively, provides an excellent reminder of the original purpose of polls: to provide accurate information so that citizens can make enlightened decisions regarding the public agenda.

Mann and Orren, however, use this volume to illustrate the negative effects of polls on public opinion: "They can set the news agenda and influence the coverage of political events in ways hostile to a constructive dialogue between citizens and their leaders." Well-written and easy to follow for both the layperson and the experienced journalist, the book is divided into seven short essays each covering different aspects of media polls. It is not, however, simply a collection of random topics; rather, each essay offers a chronology of the development of polls and comments on their evolving role in the democratic process. This format allows the reader a fuller appreciation of the subject. Mann and Orren pose two important questions that set the base of inquiry for the entire book: What are the strengths and weaknesses of media polls, and what has been the impact of media polls on U.S. politics? The editors choose papers that address each of these questions thoroughly and that offer examples of recent election outcomes to support their positions.

Three main themes emerge progressively from the collection: The impact of changing technology on media polls; polling techniques and ways in which polls can be misinterpreted; and finally, the role of polls in the democratic process. The first of these themes, the impact of technology, has played a crucial role in the development and evolution of polls. The two chapters devoted to it are among the most informative of the book.

Everett Ladd of the University of Connecticut and John Benson of the Roper Center examine the "enormous and steady" growth of media polling in the United States, which can be traced to the mid-nineteenth-century newspapers that canvassed political meetings to predict outcomes. It was not until the 1930s, however, that newspapers began to conduct polls based on systematic sampling. In the early 1970s news organizations began b use their own polls. The New York Times and CBS were the first of the major news organizations to form a partnership to conduct surveys. This move marked the beginning of a trend in media of independence from the main polling organization, Gallup, as well as of the many government organizations, public interest groups and politicians who conducted their own surveys. Today, while polls vary in size and audience, most are now run by the media.

Ladd and Benson focus at length on the findings of the Roper Center Survey conducted in 1989, which provided a complete overview of the use of media polls from 1970 to 1990. The main result of the Roper Study appears to be that there are more media polls to chase whatever topic is foremost in the news. Their conclusion points to two main implications: First, "horse-race" journalism becomes unavoidable, as the use of polls increases. Second, this increase has resulted in redundancy, "giving readers a chance to check one set of results against another to get a more complete view [of a given issue]." Throughout the book, simple but informative empirical data is provided to put these observations into perspective.

Kathleen Franckovic's essay "Technology and the Changing Landscape of Media Polls" expands on Ladd and Benson's analysis, offering a clear definition of how polls are conducted. She gives a brief, but enlightening description of each of the four most commonly used polls: exit polls, panel surveys, tracking polls and focus groups. Franckovic also explains how polling methods have moved away from long questionnaires and are now largely compiled through computer-assisted telephone interviews. She claims that this evolution has created a highly competitive environment in which journalists attempt to produce the most timely polls, often sacrificing quality in order to be the first with the news.

In light of the swift advancement of technology in this field, Franckovic poses a crucial question: How do news organizations satisfy their need to gather survey results quickly while utilizing reliable survey methods? This tradeoff facing news organizations is evident in the array of choices available. While the telephone is the simplest way to reach the American people, she explains, the exit poll is now the most common tool used to put elections into perspective. Indeed, development of the exit poll has been a positive development in the field: It gives news organizations the time to prepare evening interviews, and newspapers can allocate necessary space to election coverage. The result is that election outcomes are no longer covered solely by the pundits and candidates themselves, but are presented in a more objective framework in the effort to determine more accurately voters' decisions.

Technological advancements also have made it possible to process more data faster, and this ability has served to alter reporters' own understanding of public opinion. A proliferation of data tempts news organizations to accept less accurate information and to alter the timing of the news delivery. For example, after the 1984 campaign debate between President Ronald Reagan and the Democratic nominee Walter Mondale, CBS News correspondents would not comment on the outcome that evening; Franckovic asserts that they may have been unwilling to be contradicted by the polls. It was not until the next morning, when the polls showed clearly that Mondale had won, that the reporters would comment on the debate.

This first theme of Media Polls in American Politics delineates the basics of today's polling environment Ladd, Benson and Franckovic each warn to varying degrees that new technology and the speed at which polls are processed threaten to overwhelm the survey process. Data, especially those collected through "800"- and "900-numbers," are analyzed so quickly that unless results are properly reported, the national sentiment can easily be misinterpreted. In the future, news organizations must be more selective in their data collection to ensure that they are reporting news and not simply the results of their own polls.

The book's second theme, which is covered by Harry Brady, Gary Orren and New York Times reporter Michael Kagay discusses the nuts and bolts of polling, including the creation of survey questions and the proper structuring of a survey. In a section more suited to experienced pollsters, Kagay devotes more than 20 pages to statistical errors in polling -- a rather dry topic for most readers. The casual reader may find the references to Type I and Type II errors and the mathematical logic required to develop polls difficult to fathom unless he or she is familiar with introductory statistics.

The first of two essays covering data analysis proceeds from the discussion of a journalist's need for rapid reporting techniques. This demand for instant information clashes with the need for statistical caution in preparing a story: It is critical that those in the media understand that several types of statistical errors may occur in surveys. Brady and Orren explain, for example, that to determine public opinion on whether the federal deficit is out of control, the pollster simply finds the file on the economy, reads the section marked "deficit" and writes a questionnaire based on a standard summary of the topic. Yet most people do not have a clear understanding of such issues, and the views they have formed are often unlikely to yield a black-or-white response to a polling question.

Issue saliency must therefore be at the forefront of the pollster's thinking. A pollster must be able to recognize that political opinion can be expressed in different forms and can be shaped by many issues and influences. Proper interpretation of and reporting on polls therefore requires an understanding of the errors that can occur, and the dangers that exist should these basics be ignored. Even the best-designed polls can fall short of their intended goal. Moreover, in attempting to beat the competition, news organizations will often bias their polls using their own choice of wording, interpretation and presentation of findings; hence, the divergence of results. Michael Kagay points out that

the proliferation of national polls now makes the identification of

quirky or skewed results much easier. Nonetheless, the sophisticated

observer needs to be aware of the subtle differences between

equally well-conducted polls that can still lead to discrepancies in


This section of the book offers several excellent illustrations of how journalists can misinterpret public opinion. Kagay uses the 1991 Persian Gulf War to put this issue into perspective. For example, in the months prior to the outbreak of fighting, the media regularly tracked public opinion on the potential of hostilities: On 30 November 1991, the Washington Post asked respondents, "If Iraq does not withdraw from Kuwait, should the United States go to war with Iraq to force it out of Kuwait ... ?" with approximately 63 percent answering in favor. On 8 December a Wall Street journal NBC poll asked a slightly different question, "Would you favor or oppose the United States going to war against Iraq if Iraq does not withdraw its troops from Kuwait ... ?" with 54 percent answering in favor. Finally, a New York Times poll on 9 December asked, "If Iraq does not withdraw from Kuwait by January 15, 1992 do you think the United States should start military action against Iraq ... ?" with only 45 percent answering in favor. The cause of these seemingly contradictory responses was simply due to different options being presented to people and to events unfolding in the international arena, not due to differences or errors in polling methods or sampling procedures.

The final theme in the book addresses perhaps the most important question: What is the impact of media polls on democracy? Are they a help or a hindrance to the political process? It is obvious that media polls affect the public; this was clear from the cult-like following of election polls during the 1993 presidential race. However, as Mann and Orren remind us, both journalists and voters must understand the effects of polls on individual citizens, and the ways in which one can measure the true impact of polls when so many of their effects are indirect.

Media Polls stresses that the journalist who makes polling an integral part of his or her job cannot view the public as an inanimate object to be counted and studied. While the proliferation of polls and advanced methods of data analysis has made life easier for journalists, reporters have a responsibility to report the news accurately. Journalists must not become lazy simply because news stories are readily available. Michael Traugott and E.J. Dionne, Jr. cover these issues, in their discussion of the accuracy of polls, the problems of misinformation and the impact of surveys on the government and the democratic process. Though somewhat less easy to read than other sections of the book, Traugott provides an excellent example of how the use of polls can hamper the democratic process. Critics of polling often refer to the problem of the time difference between the East and West coasts when reporting on U.S. presidential elections -- the infamous example of which was the 1980 election, when President Carter conceded before the polls had closed on the West coast, once NBC had declared Reagan the victor. Reagan's margin of victory ended up being much greater than had been predicted because many West coast voters stayed away from the polls.

Dionne then offers two insights into polling's impact on the democratic process. First, polls create something out of nothing: People often do not have firm views on a given subject, and may give a polite answer when approached by a pollster -- one that might be different than the view they would articulate if given more time or forewarning. Second, polls can invite respondents to express private prejudices, which largely serve to taint the polling process. The error way be so great as to render some polls irrelevant. This leads to Media Polls' greatest contribution to the future of U.S. politics and the shaping of public opinion. Given the risks of inaccurate results, both the public and the media have a responsibility to ensure that polls help to report the news and not to create it.
COPYRIGHT 1993 Columbia University School of International Public Affairs
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Johnson, Stephen S.
Publication:Journal of International Affairs
Article Type:Book Review
Date:Jun 22, 1993
Previous Article:Telecommunications in Europe.
Next Article:Tales of Terror: Television News and the Construction of the Terrorist Threat.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters