Saylor.org's Comparative Politics/Understanding Diverse Populations and Public Opinion
by David L. Paletz, Diana Owen, and Timothy E. Cook
It has become a ritual for Americans to spend the evening of presidential elections gathered in front of their televisions, or more recently, their computer screens, to follow the voting returns as they are announced state by state. Election Night 2000 began like any other since the late 1960s, when the television networks began using exit polls of voters taken as they leave the polling place to predict the winner. Election Night coverage is driven by anchors making projections about which candidate will win each state’s electoral votes. Typically, news organizations have a good sense of who will be the next president of the United States based on exit polls by late afternoon, although they hold off on making a prediction until later in the evening.Robert S. Erikson and Kent L. Tedin, American Public Opinion, 8th ed. (New York: Longman, 2011).
The 2000 presidential election was the closest in many decades. There was much uncertainty about whether Republican George W. Bush or Democrat Al Gore would emerge victorious. As Election Night unfolded, it became clear that the outcome would be decided by Florida’s twenty-five electoral votes.James W. Ceaser and Andrew E. Busch, The Perfect Tie (Lanham, MD: Rowman & Littlefield, 2001).
Network and cable news anchors discussed the closeness of the election and told the public to anticipate a long and interesting evening—a statement that proved prescient. By 8 p.m., exit polls indicated that Al Gore was leading the state of Florida, prompting television news organizations to speculate that Gore would be headed to the White House. CBS News anchor Dan Rather observed on air, “Now, remember, Florida is the state where Jeb Bush, the brother of George Bush, is the governor, and you can bet that Governor Bush will be madder than a rained-on rooster that his brother, the governor, wasn’t able to carry this state for him.”Dan Rather, CBS Evening News, Election Night Coverage, November 7, 2000, 8:15 p.m.
Three hours later, the networks began to rescind the call of Florida for Gore when it became evident that data from exit polls conflicted with actual returns from voting precincts. Network anchors reported that Florida’s electoral votes were still up for grabs until Fox News called Florida for Bush at 2:16 a.m.; ABC, CBS, and NBC quickly followed suit. With the media’s proclamation of Bush as the winner, Gore phoned Bush and conceded the election. Gore then departed from his hotel suite in Nashville to make his concession speech in front of his supporters. While Gore was en route, the press once again changed their position, stating that the election was too close to call. Gore returned to his hotel, as the media’s Election Night prediction of a Bush victory lasted all of ninety minutes.
Television news was not the only media source to prematurely call the election. Print newspapers, including the New York Post, the Miami Herald, and the San Francisco Chronicle, ran headlines declaring Bush the winner. The New York Times released 100,000 newspapers stating that Bush “appears to have won.” For an hour, the New York Times website proclaimed, “Bush Captures the White House.”Diana Owen, “Media Mayhem,” in Overtime!, ed. Larry J. Sabato (New York: Longman, 2002), 123–56.
The 2000 election was not decided on Election Night—November 7. Instead, a recount of the votes in Florida was undertaken in an attempt to determine the winner. The recount was halted by the US Supreme Court on December 12, 2000, and George W. Bush was sworn in as president on January 20, 2001.
The Election Night 2000 media debacle illustrates a number of points relevant to this chapter. Polling is an integral element of American politics. Polls shape the way that news organizations frame their stories and convey information to the public. In fact, many news organizations have in-house polling operations or collaborate with polling firms to have public opinion data constantly available. Poll results allow the media to convey information to the public in a concise and authoritative manner. Polls can provide guidance to decision makers about election outcomes and policy debates. However, poll results are not always accurate, as was the case with the exit polls in the 2000 presidential election, and they can misrepresent public sentiment. Therefore, it is important for people to be savvy consumers of opinion polls.
What Is Public Opinion?
Public opinion is one of the most frequently evoked terms in American politics. At the most basic level, public opinionPeople’s collective preferences on matters related to government and politics. represents people’s collective preferences on matters related to government and politics. However, public opinion is a complex phenomenon, and scholars have developed a variety of interpretations of what public opinion means. One perspective holds that individual opinions matter; therefore, the opinions of the majority should be weighed more heavily than opinions of the minority when leaders make decisions. A contrasting view maintains that public opinion is controlled by organized groups, government leaders, and media elites. The opinions of those in positions of power or who have access to those in power carry the most weight.
Public opinion is often made concrete through questions asked on polls. Politicians routinely cite public opinion polls to justify their support of or opposition to public policies. Candidates use public opinion strategically to establish themselves as front-runners or underdogs in campaigns. Interest groups and political parties use public opinion polls to promote their causes. The mass media incorporate reports of public opinion into news story about government and politics.
Defining Public Opinion
What exactly is public opinion? Scholars do not agree on a single definition of public opinion. The concept means different things depending on how one defines “the public” and assumptions about whose opinion should or does count the most—individuals, groups, or elites.
Most simply, the public can be thought of as people who share something in common, such as a connection to a government and a society that is confronted by particular issues that form the bases of public policies. Not all people have the same connection to issues. Some people are part of the attentive publicPeople who pay close attention to government and politics in general. who pay close attention to government and politics in general. Other individuals are members of issue publicsPeople who focus on particular public policy debates and ignore other issues. who focus on particular public policy debates, such as abortion or defense spending, and ignore others.
They may focus on a policy that has personal relevance; a health-care activist, for example, may have a close relative or friend who suffers from a prolonged medical problem. Some members of the public have little interest in politics or issues, and their interests may not be represented.
An opinion is the position—favorable, unfavorable, neutral, or undecided—people take on a particular issue, policy, action, or leader. Opinions are not facts; they are expressions of people’s feelings about a specific political object. Pollsters seeking people’s opinions often say to respondents as they administer a survey, “there are expressions of people’s feelings about a specific political object. Pollsters seeking people’s opinions often say to respondents as they administer a survey, “there are no right or wrong answers; it’s your thoughts that count.” Opinions are related to but not the same as attitudes, or persistent, general orientations toward people, groups, or institutions. Attitudes often shape opinions. For example, people who hold attitudes strongly in favor of racial equality support public policies designed to limit discrimination in housing and employment.
Public opinion can be defined most generically as the sum of many individual opinions. More specific notions of public opinion place greater weight on individual, majority, group, or elite opinion when considering policy decisions.
Equality of Individual Opinions
Public opinion can be viewed as the collection of individual opinions, where all opinions deserve equal treatment regardless of whether the individuals expressing them are knowledgeable about an issue or not. Thus, public opinion is the aggregation of preferences of people from all segments of society. The use of public opinion polls to gauge what people are thinking underlies this view.
By asking questions of a sample of people who are representative of the US population, pollsters contend they can assess the American public’s mood.
People who favor this perspective on public opinion believe that government officials should take into account both majority and minority views when making policy.
Another perspective maintains that public opinion is the opinion held by the most people on an issue. In a democracy, the opinions of the majority are the ones that should count the most and should guide government leaders’ decision making. The opinions of the minority are less important than those of the majority. This view of public opinion is consistent with the idea of popular election in that every citizen is entitled to an opinion—in essence a vote—on a particular issue, policy, or leader. In the end, the position that is taken by the most people—in other words, the position that receives the most votes—is the one that should be adopted by policymakers.
Rarely, if ever, does the public hold a single unified opinion. There is often significant disagreement in the public’s preferences, and clear majority opinions do not emerge. This situation poses a challenge for leaders looking to translate these preferences into policies. In 2005, Congress was wrestling with the issue of providing funding for stem cell research to seek new medical cures. Opinion polls indicated that a majority of the public (56 percent) favored stem cell research. However, views differed markedly among particular groups who formed important political constituencies for members. White evangelical Protestants opposed stem cell research (58 percent), arguing the need to protect human embryos, while mainline Protestants (69 percent) and Catholics supported research (63 percent).
Public Debate among Groups
Some scholars contend that public opinion emerges from public debate among groups rather than from individual opinions.
Political parties, interest groups, trade associations, nonprofit organizations, trade unions, and corporations will articulate positions and front public discussion of issues in which they have a stake. Groups representing opposing viewpoints often find themselves in a position to define social problems. While individuals often find it difficult to make their views known and have them taken seriously, organized groups have the resources, such as lobbyists and funding to administer polls and pay for advertising, as well as the ability to attract the attention of policymakers and the mass media. Social media have made it easier for groups without significant resources to publicize their opinions by using Facebook groups and other platforms.
Groups work hard to frame issue debates to their advantage. They often will gauge public preferences and use this information when devising media tactics to gain support for their positions.
Opposing groups will present competing public opinion poll data in an effort to influence decision makers and the press. In 1997, the United States’ participation in a summit in Kyoto, Japan, where nations signed a climate-control treaty, sparked a barrage of media stories on the issue of global warming and the potential for deadly gasses to induce climate change. Most Americans believed then that global warming existed and that steps should be taken to combat the problem.
Groups such as the Environmental Defense Fund, Greenpeace, and the Sierra Club who favor government-imposed regulations on fossil-fuel companies and automobile manufacturers to curb pollution cited opinion poll data showing that over 70 percent of the public agreed with these actions. Organizations representing industry interests, such as the now-defunct Global Climate Coalition, used opinion polls indicating that the public was reluctant to sacrifice jobs or curb their personal energy use to stop global warming.
The debate in the media among competing groups influenced public opinion over the following decade. There was a massive shift in opinion, as only 52 percent believed that global warming was a problem in 2010.
Politicians, pollsters, policy specialists, activists, and journalists have assumed the position of opinion leaders who shape, create, and interpret public opinion. These political elites are devoted to following public affairs—it’s their job.
Noted journalist and social commentator Walter Lippmann observed that average people have neither the time nor the inclination to handle the impossible task of keeping up with the myriad issues that confront the nation. They do not have the opportunity to directly experience most political events and must rely on second-hand accounts conveyed by elites primarily through mass media. In Lippmann’s view, public opinion is best managed by specialists who have the knowledge and capabilities to promote policies. Thus, elite opinion, and not the views of average citizens, should count the most.
The mass media rely heavily on the opinions of government elites, especially when covering foreign policy and domestic issues, such as the economy and employment. The breadth of news coverage about foreign affairs is constrained to reflect the range of viewpoints expressed by officials such as members of Congress who are debating the issues. The voices of average Americans are much less prominent in news coverage.
As political scientist V. O. Key stated, “The voice of the people is but an echo.”
Elite opinion is increasingly articulated by punditsPeople who offer their opinion or commentary on political issues through the media. who offer their opinion or commentary on political issues. College professors, business and labor leaders, lobbyists, public relations representatives, and pollsters are typical pundits who provide expert opinion. Some pundits represent distinctly partisan or ideological viewpoints and use public opinion data selectively to support these positions. Pundits can establish their credentials as experts on governmental affairs and politics through their frequent media appearances as “talking heads” on cable television programs such as CNN, MSNBC, and Fox News.
Democracy and Public Opinion
Political scientist Harold Lasswell once noted, “The open interplay of opinion and policy is the distinguishing mark of popular rule.”
Public opinion plays a number of important roles in a representative democracy. Leaders can take public opinion into account when making laws and formulating policy. Public opinion can act as a check on leadership, as the members of the public can express their dissatisfaction with politicians who refuse to take their opinions into account and vote them out of office.
Public Opinion and Public Policy
One purpose of public opinion in a democracy is to inform public policymaking. Opinion polls provide a mechanism for succinctly presenting the views of the mass public to government leaders who are making decisions that will affect society. Leaders often monitor the public pulse when making policy decisions, especially when they face an election campaign.
Perspectives about the relationship of public opinion to policymaking differ vastly. On the one hand, scholars and political practitioners believe that public policy should be guided by public opinion so that the will of the people is respected. Public opinion polls are essential to ensuring democratic governance. Political scientist Sidney Verba supports this view: “Surveys produce just what democracy is supposed to produce—equal representation of all citizens. The sample survey is rigorously egalitarian; it is designed so that each citizen has an equal chance to participate and an equal voice when participating.”
From this perspective, members of Congress, state legislators, and local officials should pay close attention to the public’s views when making laws.
Others disagree with the notion that leaders should pay close attention to public opinion when making decisions. They point out that many members of the public are uniformed about issues, and the opinions they record on polls are not carefully reasoned. Journalist and scholar Walter Lippmann noted that governing by popularity is not always best. Instead, public policy should be grounded in sound principles supported by experts; decision making should not simply be the result of popular will. This view is consistent with the belief that the country is being run by pollsters and their followers and not by leaders with integrity and principle. As an editorial in the Wall Street Journal lamented, “Spend too much time following polls and you simply forget how to lead, especially when it matters.”
Some scholars take issue with Verba’s assessment of the egalitarian nature of polls in democracy and argue that minority opinion is not given sufficient weight. Certain people, such as individuals with few economic resources, have a difficult time getting their views recognized. Pollsters may not reach these people because they do not have regular telephone or Internet service, or they do not have the time or inclination to answer questionnaires.
Public opinion, especially as measured by polls, is a quick take on the public pulse. It often does not require that members of the public have much knowledge about politicians, government, or policies; they merely must be willing to state whatever views pop into their heads. Public opinion polls often deal with issues and abstract ideas that people have not thought much about.
Public judgment, in contrast, is a special type of public opinion where people carefully consider the important issues of the day, contemplate the range of viewpoints, and weigh the consequences of policy prescriptions. Rather than stating positions off the top of their heads, public judgment requires people to be knowledgeable about an issue and debate the merits of policies before arriving at an informed opinion. For example, public opinion polls conducted by the Pew Research Center in 2011 indicate that the public favors tougher immigration laws and better enforcement of border security. However, when people exercise public judgment and consider the consequences of immigration policy, such as the moral issues related to the welfare of children of illegal immigrants, they support more generous policies.
Public judgment is not easily achieved, but it offers an important counterbalance to the domination of elite opinion in the policy sphere.
Deliberative polling is a technique that provides members of the public with the opportunity to think carefully about issues and their relationship to public policy. It attempts to deal with the fact that many people know little about issues because they lack the time to acquire information. Deliberative polling was pioneered in 1988 and has been used around the world to gauge opinion. The Center for Deliberative Democracy at Stanford University applies and studies the use of deliberative polling.
A random, representative sample of people is first polled about their positions on targeted issues. They are then brought together for a few days to discuss a particular issue in detail. The participants are provided with briefing materials, engage in a dialogue with experts on specific topics, and discuss their views in small groups led by trained moderators. The deliberations are shared with the general public through television broadcasts. The participants are polled again after they have deliberated to determine if their opinions have changed. Scholars believe that deliberative polls represent the opinions the public would hold on issues if they had the opportunity to exercise public judgment and carefully consider their options. After deliberating on an issue, members of the public frequently shift positions.
For example, people participating in a deliberative polling experiment in Texas shifted their views on the use of wind power from 54 percent to 84 percent in favor. As a result, political leaders heeded the views of Texas’s population, and the state went from last to first in the use of wind power.
Political scientist James Fishkin, who pioneered deliberative polling, observes, “The Public is very smart if you give it a chance. If people think their voice actually matters, they’ll do the hard work, really study their briefing books, ask the experts smart questions and then make tough decisions.”
Polling the Public
Public opinion polling has a long history in the United States. Polls are ubiquitous in American political life. In 2007, there were nearly 5,500 polling organizations in the United States, an increase of over 1,500 organizations in ten years.
Every day the public is polled about topics ranging from their views about taxes and the federal budget, their opinions about the environment and global warming, and whether or not a principal has the right to prevent students from bringing their lunches to school.
Polls vary greatly in terms of their quality, content, and purpose. Reliable public opinion data are gathered by reputable commercial polling organizations, nonpartisan think tanks, academic institutions, government agencies, and news organizations. Misleading information about public opinion can result from quick polls that do not employ appropriate data-gathering techniques.
History of Opinion Polling
Public opinion polls date back to the early days of the American republic. From the outset, polls were linked closely with newspapers. The Harrisburg Pennsylvanian conducted the first informal “straw pollAn informal poll, often used to gauge opinions about candidates, that is administered haphazardly and without attention to proper sampling methods.” in 1824 that incorrectly predicted Andrew Jackson as the winner over John Quincy Adams in the presidential election.
Early straw polls were administered haphazardly and without concern for drawing a proper sample, often right after people had cast their ballots in elections or even when they were checking into a hotel. They were notoriously inaccurate, yet they became a popular feature of newspapers and magazines, which treated poll data as a source of news much like today.
Straw polls were sponsored by publishers as a gimmick to attract readers who would fill out mail-in ballots that included subscription offers. Over eighty straw polls were conducted during the 1924 presidential election, six of which were national polls. Newspapers also conducted polls on pressing issues of the day, such as whether or not people favored Prohibition, the constitutional ban on alcohol. Coverage of these polls in print publications generated thousands of column inches.
By the 1920s, market researchers had begun to use scientific polls that employed samples representative of the population to ascertain consumer product preferences. They used polls to discover everything from what kinds of magazine stories readers enjoyed most to what automobiles people preferred.
Commercial pollsters applied market research techniques to determine what candidates voters favored, how satisfied the public was with the way the president was doing his job, and how people felt about the pressing issues of the day.
Despite these advances, magazines and newspapers continued to use unscientific straw polls, which were less expensive to administer and contributed to the profitability of the publication. The problems associated with straw polls came to a head in the 1936 presidential election when the Literary Digest, a popular magazine with a large circulation, incorrectly predicted the presidential election outcome, prompting the public to lose faith in polls. For a time after the Literary Digest debacle, newspapers shied away from highlighting polls in their coverage.
The Literary Digest Poll
In polling, more subjects does not necessarily yield better results. This lesson was learned all too well by the Literary Digest in 1936. Founded in 1890, the Literary Digest was a venerable general interest magazine that catered to an educated, well-off clientele. In 1916, the magazine initiated a presidential election poll that became a popular feature. Subscribers mailed in sample ballots indicating their preference in the election. The poll correctly predicted that Woodrow Wilson would be the winner, and the magazine’s poll went on to successfully call the next four elections. Newspapers gave substantial coverage to the poll, which drove up the magazine’s readership. In 1932, James A. Farley, chairman of the Democratic National Committee, was widely quoted as saying, “Any sane person cannot escape the implication of such a gigantic sampling of popular opinion as is embraced in the Literary Digest straw vote.… It is a Poll fairly and correctly conducted.”
The magazine set out to launch its most ambitious poll ever in 1936. Over 10 million postcards were mailed to Literary Digest subscribers, people on automobile registration lists, and names in telephone directories, of which 2.4 million were returned. The Literary Digest issued its predictions in an article boasting that the figures represented the opinions of “more than one in every five voters polled in our country” scattered throughout the forty-eight states. The results indicated that Republican candidate Alfred Landon would defeat Franklin Roosevelt, receive 57 percent of the popular vote, and carry thirty-two states in the Electoral College. Roosevelt won by a landslide, commanding 61 percent of the popular vote and winning in all but two states.
While the magazine made no claims of infallibility, its methodology was heavily flawed. The sample was biased toward Republican-leaning voters who could afford telephone service, cars, and magazine subscriptions. The volunteers who tabulated the results were not carefully trained, which introduced additional error into the calculations. The backlash from the errant results was monumental. The Literary Digest went bankrupt, and the public’s faith in polls was shattered.
Commercial pollsters using scientific techniques correctly predicted that Roosevelt would defeat Landon in the 1936 election. These pollsters conduct polls for clients for a profit. The Gallup Poll administered personal interviews with a quota sampleA method of selecting survey participants that involves choosing subjects on the basis of their fitting into particular demographic categories, such as sex and age groups. of people who fit into particular demographic categories, such as sex and age groups. Gallup correctly predicted the winners of the 1940 and 1944 presidential contests. However, during the 1948 election, three major pollsters—Gallup, Roper, and Crossley all incorrectly predicted that Republican presidential candidate Thomas Dewey would defeat Democratic candidate Harry Truman. The quota sampling method used by these pollsters was problematic and was replaced by probability samplingA method of selecting survey participants at random., in which subjects are randomly selected to take part in a poll.
Dewey Defeats Truman
The 1948 presidential election did not start off well for Democratic candidate Harry S. Truman. As vice president, Truman was sworn in as president when Franklin Roosevelt died less than three months into his fourth term. Truman was forced to deal with a variety of controversial issues, including the decision to drop atomic bombs on Hiroshima and Nagasaki, which he believed would end World War II in the Pacific. Newspapers labeled Truman a “little man,” a tag that resonated with the public who contrasted him unfavorably to the larger-than-life Roosevelt.
The Democrats were highly factionalized when they met in Philadelphia for their national nominating convention. They attempted unsuccessfully to recruit popular war hero Dwight D. Eisenhower to be their candidate. When the convention adopted a strong civil rights platform, Southern delegations bolted and nominated their own candidate, Strom Thurmond of South Carolina. Liberals who disapproved of Truman’s policies formed the Progressive Party and nominated Henry Wallace of Iowa as their candidate. In the end, Truman became the nominee with Senator Alben Barkeley of Kentucky as his running mate. The pair was faced with an unenthusiastic constituency.
In contrast, the Republican Party united behind Thomas E. Dewey, the popular governor of New York. Dewey had been the Republicans’ candidate in the 1944 presidential campaign, and had come close to Roosevelt in the popular vote. California Governor Earl Warren, future chief justice of the Supreme Court, was the vice presidential candidate.
Pollsters and the press anticipated that Dewey would win by a landslide. On September 9, 1948, nearly two months prior to the election, noted pollster Elmo Roper declared that there would be no more Roper Polls predicting the outcome: “My whole inclination is to predict the election of Thomas E. Dewey by a heavy margin and devote my time and efforts to other things.”
Normally, incumbents such as President Truman run low-key campaigns, and challengers such as Governor Dewey work hard to win. Dewey campaigned like a front-runner, remaining aloof and dignified while avoiding discussions of controversial issues. Roles were reversed in the 1948 presidential campaign. Truman, realizing he had nothing to lose, launched an aggressive “Whistle Stop” campaign. Traveling in a special Pullman railroad car nicknamed the Ferdinand Magellan, after the explorer who circumnavigated the world, Truman covered 32,000 miles and gave 355 rousing speeches. At each stop, Truman would introduce his family to the crowd, answer questions, and shake hands. As he fought his way through the campaign, he acquired the slogan “Give ‘em hell, Harry!”
Even as Truman’s campaign picked up steam and polls showed the gap between the candidates was closing, the press refused to concede that he could win. Newsweek polled fifty political journalists a month before the campaign, and all of them stated that Dewey would win. Truman had the support of only 15 percent of the nation’s newspapers.
By Election Day, polls indicated that Truman might pull an upset, but journalists stuck to their story that Dewey would win by a landslide. Reports filtered in throughout Election Night that Truman was leading in the popular vote, but the press continued to report that he could not emerge victorious. The Chicago Tribune was so certain that Truman would lose, the headline of the early edition proclaimed “DEWEY DEFEATS TRUMAN.” The paper had already been delivered, and the Tribune dispatched employees to retrieve the papers from newsstands and homes, but many remained in the hands of readers. Traveling by train from his home state of Missouri to Washington, DC, Truman made a brief stop in St. Louis, where he was presented with one of the papers bearing the infamous headline. Truman held up the paper and quipped, “This is for the books.”
Survey research organizations associated with academic institutions emerged in the 1940s with the establishment of the National Opinion Research Center (NORC) at the University of Chicago and the Survey Research Center (SRC) at the University of Michigan. These organizations and others like them, such as the Roper Center at the University of Connecticut, field and archive detailed surveys that provide researchers with a wealth of data to use in studies to gain a deeper understanding of the public’s political attitudes and behavior. Nonpartisan survey research organizations, such as the Pew Research Center and the Field Poll in California, provide data to news organizations and academics. Commercial pollsters, including Gallup and IBOPE Zogby International, provide polling services to clients and also share their data with the press, scholars, and the public through their websites.
Types of Polls
The amount of polling data available today from commercial polling firms, academic survey research organizations, campaign organizations, trade associations, interest groups, media outlets, and online sources is almost overwhelming. There are great variations in the type and quality of polling data. A public opinion survey fielded by a reputable organization using proper social scientific techniques differs greatly from a quick poll consisting of one or two questions administered online to whoever is willing to take it.
Questionnaires used to measure public opinion include a variety of question types. Closed-ended questions provide respondents with a fixed number of options about a topic from which they can choose the one that best fits their position. provide respondents with a fixed number of options about a topic from which they can choose the one that best fits their position. A closed-ended question frequently asked to gauge people’s feelings about the direction in which the country is headed is “Generally speaking, would you say things in this country are heading in the right direction, or are they off on the wrong track?” Respondents must select one of the options: the right direction or the wrong track. Closed-ended questions are easier and less time-consuming to analyze, although they limit the respondent’s ability to express their opinions to the choices offered by the researcher. Open-ended questions do not provide fixed options but instead allow respondents to reply to a question in their own words. This type of question elicits more information from respondents and can be useful in gaining insight into sensitive topics. The drawbacks of open-ended questions are that people may not want to take the time to answer them and they are more time-consuming for pollsters to analyze. An open-ended question about the direction in which the country is headed would ask people to express their own views in response to the question “How do you think things are going in this country?”
Most polls provide snapshots of people’s opinions at a particular point in time. Other polls track opinions over time in order to determine if people’s views remain stable or change. In rare cases, studies have tracked the opinions of the same groups of people over years, even decades. The views of the women who attended Bennington College in the 1930s were tracked through the 1980s. The study revealed that the college experience changed some of the women’s attitudes and that the views acquired in college remained stable over time.
Polls and Surveys
The terms “poll” and “survey” often are used interchangeably, yet there are distinctions between them. A public opinion poll is typically conducted by a commercial organization working for a profit. A poll generally consists of a short questionnaire administered over a brief period of time to a sample of between six hundred and fifteen hundred people. A survey most often is conducted by academic or government researchers. Surveys consist of longer questionnaires designed to examine the foundations and consequences of opinions in some detail. Researchers may administer the survey to thousands of subjects interviewed over an extended period of time.
Scientific polls and surveys are considered to be the gold standard for measuring public opinion. They adhere to established procedures that help ensure the accuracy of their results, which includes using proper techniques for drawing a sample and designing questions. Scientific polls and surveys are administered to a sample of people who are representative of a larger population. The sample is drawn using probability sampling, meaning that each person in the population has a chance of being included in the sample. It is possible to get an accurate accounting of public opinion with a relatively small sample. A representative sample of twelve hundred people can accurately reflect the public opinion of the entire population of the United States. On the other hand, large samples that are not representative may not reflect public opinion accurately at all. Question wording is another important consideration when measuring public opinion. Questions need to be clearly stated, and they should not lead the respondent to choose one answer over another. A poorly worded question can be misunderstood by the respondent and ultimately can misrepresent the public’s viewpoints. Answer options that do not provide the public with clear alternatives also are problematic.
There are many ways in which polls and surveys can be administered, including through face-to-face interviews, telephone interviews, mail questionnaires, and online questionnaires. Each of these methods has pros and cons. Face-to-face interviews are advantageous for administering long, complicated surveys, yet they are costly and subjects may be reluctant to talk to a stranger about their opinions. Telephone interviews are relatively easy to administer, but getting a representative sample has become more difficult as many polling organizations rely on landline telephone directories to recruit respondents, and people increasingly are relying on cell phones. Young people are not well represented in landline polls.
Mail questionnaires are a low-cost method that allows subjects privacy when answering questions, which can yield more accurate results. However, mail surveys often suffer from low response rate, as people simply opt out because the questionnaire is self-administered.
Online polls have become a more popular option in recent years as the majority of the public has access to the Internet. According to the Pew Internet and American Life Project, 79 percent of American adults were online in May 2010. Studies indicate that online polls are no less reliable than other forms of polling. They have the advantage of being cost-effective, and allowing respondents privacy when answering questions. Online polls also provide opportunities for innovation, such as getting reactions to video clips of campaign ads. The limitation of online polls is that it is more difficult to get a representative sample using the
Internet than with some traditional methods, because not all of the public is online. Also, online surveys are self-administered, and people can drop out before they are completed, especially if the questionnaire is lengthy.
Exit polls are face-to-face interviews with voters taken as they leave the voting booth to determine their candidate preference in the election and their positions on issues. They are fielded in a small number of voting precincts with states with the goal of acquiring representative data. They are used to predict the outcomes of elections and to determine the characteristics of voters who supported particular candidates. Exit poll data can reveal, for example, who female, Latino, Republican voters favored in an election campaign.
Exit polls are a major component of the media’s Election Night coverage. Until 1992, each news network had its own in-house exit polling operation. To cut costs, an exit poll consortium, Voter News Service (VNS), was formed to provide data to all the major networks. VNS released the exit poll data that prompted the networks to prematurely declare the results of the 2000 presidential election, and the organization subsequently was disbanded. Exit poll data in the 2008 presidential election and 2010 midterm elections were provided to major television news organizations and the Associated Press by the National Election Exit Polls conducted by Edison Research.
News organizations use exit polls to declare a winner, sometimes when few of the actual returns from the voting precincts have been recorded. This practice has raised concerns, especially since the major television networks all rely on exit poll data from the same source—the National Election Exit Poll. While exit polls are often accurate, if the sample of voters is unrepresentative of the population, the survey questions are poorly written, or interviewers are not trained to properly administer the poll, the results can be wrong, as was the case in the 2000 presidential election.
Some scholars allege that media reports of exit polls can depress election turnout. When the media declare the winner in a presidential election on the basis of exit polls before the voting booths have closed across the country, people who have not yet voted may decide not turn out. Network television newscasts declared Ronald Reagan the winner of the 1980 presidential election on the basis of exit polls hours before the voting booths had closed on the West Coast. A controversy ensued around the allegation that West Coast voters were discouraged from casting a ballot because they felt their vote was irrelevant. The networks agreed voluntarily to refrain from declaring a winner in elections until after all the polls have closed nationwide—an agreement that has not always been followed.
A quick poll usually consists of one or two questions that are posted to a website, blog, discussion board, social media platform, or podcast. Quick polls have become standard features of websites of news organizations, political leaders, issue advocacy groups, political parties, candidates, bloggers, and even average citizens. They can be distributed through website sidebars, e-mail links, Facebook postings, and Twitter feeds. There are many platforms available that make it easy for just about anyone to field a quick poll. Quick polls also can be administered through robo-polling-administering automated polls by phone using a recorded voice to ask the question and requiring respondents to answer by pressing the touch pad on their telephone.
Quick polls do not conform to the established protocols for conducting scientific polls, and they generally are not reliable indicators of public opinion. They often use an unscientific convenience sample of people who may see the poll posted online or have the link sent to them through e-mail. Most respondents to quick polls are self-selected, and they may have a strong interest in the topic. Often it is possible for people to register their views more than once, which can bias the outcome of the poll. Quick polls may generate many responses, but the results can be wildly inaccurate. In addition, quick poll questions can be designed in a way that elicits a particular response that is then used to promote a particular position. For example, a quick poll might seek to find support for bike lanes in cities by stating, “Seven out of ten Americans favor designating bike lanes in major cities. Do you favor or oppose designating bike lanes in your city?”
Quick polls can be a fun way to generate interest in political affairs. People can express their views easily, and they often get immediate feedback about where they stand compared to others. The results of quick polls often are revealed in visually appealing graphics. Reporters and bloggers use the results of quick polls to generate story lines and supplement the text of their pieces. However, quick polls can be misused when the results are interpreted as if they truly reflect public opinion rather than the views of the people who chose to take them.
A push poll is a marketing technique used by political campaigns and issue advocacy groups to influence the opinions of respondents. Despite their name, push polls are not legitimate public opinion polls. They are a form of advertising masquerading in the form of an opinion survey. No one collects or analyzes data from a push poll. However, push polls can influence vote choice in campaigns by incorporating negative attacks on a candidate into the questions asked or associating a candidate with a particular issue position which may or may not be accurate.
Push polls were used against Republican candidate John McCain during the 2000 presidential primary. Voters in South Carolina were asked questions like “Would you be more or less likely to vote for John McCain for president if you knew he had fathered an illegitimate black child?” Push polls were used to target Democratic candidate Barack Obama in the 2008 presidential campaign. Voters in Ohio received phone calls from Opinion Access Corporation asking if they would be more or less likely to vote for Barack Obama if they knew that he had voted to let convicted child sex offenders out early.
While these allegations were untrue or taken out of context, the information was spread to voters. Push polls have been outlawed in certain states and they have been condemned by the American Association of Public Opinion Researchers (AAPOR), the organization that upholds standards for polling and survey research.
Other Ways of Measuring Public Opinion
There are a variety of ways of measuring public opinion aside from polls. The different sides of an argument expressed in public debates or at a community meeting reflect public opinion. The positions taken in letters to the editor, blog and social media posts, and the comments in response to news stories and editorials are all indicators of public sentiment. The commentary that people post in response to news stories can provide a rich source of information about public opinion, especially when people take the issue seriously and are respectful when expressing their views. This commentary also can be careless and vitriolic, as people resort to personal attacks or post quick reactions to complex issues.
Focus groups have been used for over eighty years to ascertain people’s attitudes, beliefs, and opinions about politics within a group setting. A facilitator asks questions of a group of between eight and twelve people who can engage in a conversation about the topic. Focus groups not only are useful for gaining in-depth insights into what individuals think but also aid in understanding the group dynamics behind public opinion. Focus groups can reveal when people feel comfortable expressing their beliefs, when they will confront others about their views, when they will withdraw from a discussion, and when they are influenced by the opinions of others.
Focus groups have been used to allow college students to reveal their views about government and their role in a democratic polity. Talking with students in a group setting, researchers discovered that young people are more interested and engaged in politics than survey-based studies indicate, and that they are thinking creatively about ways to become involved, especially using social media.
Focus groups are used extensively in election campaigns to determine what voters are thinking about and which candidates they prefer.
 James A. Stimson, Public Opinion in America, 2nd ed. (Boulder, CO: Westview, 1999).
 Carroll J. Glynn, Susan Herbst, Garrett J. O’Keefe, and Robert Y. Shapiro, Public Opinion (Boulder, CO: Westview, 1999).
 Susan Herbst, Numbered Voices (Chicago: University of Chicago Press, 1993).
 Pew Research Center for the People & the Press, “More See Benefits of Stem Cell Research” (Washington, DC: Pew Research Center, May 23, 2005).
 Carroll J. Glynn, Susan Herbst, Garrett J. O’Keefe, and Robert Y. Shapiro, Public Opinion (Boulder, CO: Westview, 1999).
 Ken Kollman, Outside Lobbying (Princeton, NJ: Princeton University Press, 1999).
 Jon A. Krosnick, Penny S. Visser, and Allyson L. Holbrook, “American Opinion on Global Warming,” Resources no. 133 (Fall 1998): 5–9.
 Glynn R. Wilson, “Global Warming: Competing Ideas and Interest Groups,” Public Opinion Project, May 2, 1998, accessed June 19, 2005, http://www.southerner.net/fast/pocompet.html.
 Christopher R. Borick, Erick Lachapelle, and Barry G. Rabe, “Climate Compared: Public Opinion on Climate Change in the United States and Canada,” Issues in Governance Studies, no. 39, April 2011, accessed April 11, 2011, http://www.brookings.edu/~/media/Files/rc/papers/2011/04_climate_change_opinion/04_climate_change_opinion.pdf.
 John Zaller, The Nature and Origins of Mass Opinion (New York: Cambridge, 1992).
 W. Lance Bennett, Regina C. Lawrence, and Steven Livingston, When the Press Fails (Chicago, IL: University of Chicago Press, 2007).
 V. O. Key Jr., Public Opinion and American Democracy (New York: Alfred A. Knopf, 1961).
 Harold D. Lasswell, Democracy through Public Opinion (Menasha, WI: George Banta Publishing Company, 1941), 15.
 As cited in Scott Keeter, “Public Opinion Polling and Its Problems,” in Political Polling in the Digital Age, ed. Kirby Goidel (Baton Rouge: Louisiana State University Press, 2011), 28.
 Kenneth F. Warren, In Defense of Public Opinion Polling (Boulder, CO: Westview Press, 2004), 6.
 Daniel Yankelovich, Coming to Public Judgment (Syracuse, NY: Syracuse University Press, 1991).
 James S. Fishkin, When the People Speak (New York: Oxford University Press, 2009).
 Joe Klein, “How Can a Democracy Solve Tough Problems?,” Time, September 2, 2010, accessed June 6, 2011, http://www.time.com/time/politics/article/0,8599,2015481,00.html.
 Kirby Goidel, “Public Opinion Polling in a Digital Age: Meaning and Measurement,” in Political Polling in the Digital Age, ed. Kirby Goidel (Baton Rouge: Louisiana State University, 2011), 11–27.
 Robert S. Erikson and Kent L. Tedin, American Public Opinion, 8th ed. (New York: Longman, 2011).
 Robert S. Erikson and Ken L. Tedin, American Public Opinion, 8th ed. (New York: Longman, 2011).
 Kathleen Morgan Drowne, The 1920s (Westport, CT: Greenwood Press, 2004).
 Barbara A. Bardes and Robert W. Oldendick, Public Opinion: Measuring the American Mind (Belmont, CA: Wadsworth/Thompson Learning, 2006).
 Elmo Roper as quoted in David McCullough, Truman (New York: Simon & Schuster, 1992), 657.
 American Treasures of the Library of Congress, “Dewey Defeats Truman” (Washington, DC: Library of Congress, American Memory Collection, 2003).
 David McCullough, Truman (New York: Simon & Schuster, 1992).
 Duane F. Alwin, Ronald L. Cohen, and Theodore M. Newcomb, Political Attitudes over the Life Span: The Bennington Women after Fifty Years (Madison: University of Wisconsin Press, 1991).
 Michael W. Traugott and Paul J. Lavrakas, The Voter’s Guide to Election Polls, 2nd ed. (New York: Chatham House, 2000).
 Scott Keeter, “Public Opinion Polling and Its Problems,” in Political Polling in the Digital Age, ed. Kirby Goidel (Baton Rouge: Louisiana State University Press, 2011), 28–53.
 Michael W. Traugott and Paul J. Lavrakas, The Voter’s Guide to Election Polls, 2nd ed. (New York: Chatham House, 2000).
 Scott Keeter, “Public Opinion Polling and Its Problems,” in Political Polling in the Digital Age, ed. Kirby Goidel (Baton Rouge: Louisiana State University Press, 2011), 28–53.
 Mark Blumenthal, “The Case for Robo-Pollsters,” National Journal, September 14, 2009, accessed April 10, 2011, http://www.nationaljournal.com/njonline/the-case-for-robo-pollsters-20090914.
 Sam Stein, “Nasty Anti-Obama Push Poll Launched in Ohio,” Huffington Post, September 11, 2008, accessed June 6, 2011, http://www.huffingtonpost.com/2008/09/11/nasty-anti-obama-push-pol_n_125607.html.
 David W. Stewart, Prem N. Shamdasani, and Dennis W. Rook, Focus Groups: Theory and Practice, 2nd ed. (Thousand Oaks, CA: Sage Publications, 2007).
 Nicholas V. Longo and Ross P. Meyer, College Students and Politics: A Literature Review, Circle Working Paper 46 (College Park, MD: Center for Information and Research on Civic Learning and Engagement, May, 2006).