In a democracy, the ability of the people to freely express their opinions is fundamental. Americans can express their opinions in many ways. They can write letters to newspapers. They can share their ideas in online forums on Facebook, blogs, and tweets. They can organize politically. They can vote. They can respond to opinion polls. Public opinion clearly plays an important role in our political system, just as it does in any democracy.
President Barack Obama and the Republicans in the U.S. House of Representatives found out how important public opinion was in July 2011 during the "debt-ceiling crisis." Congress must specifically authorize any increase in the federal debt, even if the larger debt is required by measures that Congress has already adopted. The debt ceiling has been raised dozens of times over the years without incident. In 2011, however, Republicans in the House seized upon the debt-ceiling vote as a way to force Democrats to agree to major federal budget cuts. If the debt ceiling had not been raised, someone would not have been paid - perhaps defense contractors, maybe Medicare providers, possibly even those who hold the federal debt itself in the form of Treasury obligations.
On July 31, at the very last moment, a deal was reached that provided for some budget cuts. The dispute had serious consequences. Many economists believed that uncertainty surrounding the vote slowed the recovery from the recession. Both the President and the Congress - especially congressional Republicans - suffered severe blows to their popularity. Obama's job approval rate dropped below 40 percent, his worst showing ever. Congressional job approval fell to 13 percent.
There is no doubt that public opinion can be powerful. The extent to which public opinion affects policymaking is not always so clear, however. For example, suppose that public opinion strongly supports a certain policy. If political leaders adopt that position, is it because they are responding to public opinion or because they share the public's beliefs? Also, political leaders themselves can shape public opinion to a degree.
DEFINING PUBLIC OPINION
There is no single public opinion, because there are many different "publics." In a nation of more than 315 million people, there may be innumerable gradations of opinion on an issue. What we do is describe the distribution of opinions about a particular question. Thus, we define public opinion as the aggregate of individual attitudes or beliefs shared by some portion of the adult population.
Typically, public opinion is distributed among several different positions, and the distribution of opinion can tell us how divided the public is on an issue and whether compromise is possible. When polls show that a large proportion of the American public appears to express the same view on an issue, we say that a consensus exists, at least at the moment the poll was taken. Issues on which the public holds widely differing attitudes result in divided opinion. Sometimes a poll shows a distribution of opinion indicating that most Americans either have no information about the issue or are not interested enough in the issue to formulate a position. Politicians may believe that lack of public knowledge of an issue gives them more room to maneuver, or they may be wary of taking any action for fear that opinion will crystallize after a crisis.
An interesting question arises as to when private opinion becomes public opinion. Everyone probably has a private opinion about the competence of the president, as well as private opinions about more personal concerns, such as the state of a neighbor's lawn. We say that private opinion becomes public opinion when the opinion is publicly expressed and concerns public issues. When someone's private opinion becomes so strong that the individual is willing to take action, then the opinion becomes public opinion. Many kinds of action are possible. An individual may go to the polls to vote for or against a candidate or an issue, participate in a demonstration, discuss the issue at work, speak out online, or participate in the political process in any one of a dozen other ways.
HOW PUBLIC OPINION IS FORMED: POLITICAL SOCIALIZATION
Most Americans are willing to express opinions on political issues when asked. How do people acquire these opinions and attitudes? Typically, views that are expressed as political opinions are acquired through the process of political socialization. By this, we mean that people acquire their political beliefs and values, often including their party identification, through relationships with their families, friends, and co-workers.
Models of Political Socialization
The most important early sources of political socialization are the family and the schools. Individuals' basic political orientations are formed in the family if other family members hold strong views. When the adults in a family view politics as relatively unimportant and describe themselves as independent voters or disaffected from the political system, however, children may receive very little political socialization.
In the past few decades, more and more sources of information about politics have become available to all Americans, especially to young people through the Web. Thus, although their basic outlook on the political system still may be formed by early family influences, young people are now exposed to many other sources of information about issues and values. This greater access to information may explain why young Americans are often more liberal than their parents on certain social issues, such as gay rights.
The Family and the Social Environment
Not only do our parents' political beliefs, values, and actions affect our opinions, but the family also links us to other factors that affect opinion, such as race, social class, educational environment, and religious beliefs. How do parents transmit their political beliefs to their offspring?
Studies suggest that the influence of parents is due to two factors: communication and receptivity. Parents communicate their feelings and preferences to their children constantly. Because children have such a strong need for parental approval, they are very receptive to their parents' views.
Children are less likely to influence their parents, because parents expect deference from their children. Nevertheless, other studies show that if children are exposed to political ideas at school and in the media, they will share these ideas with their parents, giving parents what some scholars call a "second chance" at political socialization.
Education as a Source of Political Socialization. From the early days of the republic, schools were perceived to be important transmitters of political information and attitudes. Children in the primary grades learn about their country mostly in patriotic ways. They learn about the Pilgrims, the flag, and some of the nation's presidents. They also learn to celebrate national holidays. In the middle grades, children learn additional historical facts and come to understand the structure of government and the functions of the president, judges, and Congress. By high school, students have a more complex understanding of the political system, may identify with a political party, and may take positions on issues.
Generally, education is closely linked to political participation. The more education a person receives, the more likely it is that the person will be interested in politics, be confident in his or her ability to understand political issues, and be an active participant in the political process.
Peers and Peer Group Influence. Once a child enters school, the child's friends become an important influence on behavior and attitudes. For children and for adults, friendships and associations in peer groups affect political attitudes. We must, however, separate the effects of peer group pressure on opinions and attitudes in general from the effects of peer group pressure on political opinions. For the most part, associations among peers are nonpolitical. Political attitudes are more likely to be shaped by peer groups when those groups are involved directly in political activities.
Individuals who join an interest group based on ethnic identity, for example, may find a common political bond through working for the group's civil liberties and rights. African American activist groups may consist of individuals who join together to support government programs that will aid the African American population. Members of a labor union may be strongly influenced to support certain pro-labor candidates.
Opinion Leaders' Influence. We are all influenced by those with whom we are closely associated or whom we hold in high regard - friends at school, family members and other relatives, and teachers. In a sense, these people are opinion leaders, but on an informal level. That is, their influence on our political views is not necessarily intentional or deliberate. We are also influenced by formal opinion leaders, such as presidents, lobbyists, congresspersons, news commentators, and religious leaders, who have as part of their jobs the task of swaying people's views.
The Impact of the Media
Clearly, the media - newspapers, television, radio, and Internet sources - strongly influence public opinion. This is because the media informs the public about the issues and events of our times and thus have an agenda-setting effect. To borrow from Bernard Cohen's classic statement about the media and public opinion, the media may not be successful in telling people what to think, but they are "stunningly successful in telling their audience what to think about."
Popularity of the Media. Today, many contend that the media's influence on public opinion has grown to equal that of the family. For example, in her analysis of the role played by the media in American politics, media scholar Doris A. Graber points out that high school students, when asked where they obtain information on which they base their views, mention the mass media far more than they mention their families, friends, and teachers. This trend may significantly alter the nature of the media's influence on public debate in the future.
The Impact of the New Media. The extent to which new forms of media have supplanted older ones - such as newspapers and the major broadcast networks - has been a major topic of discussion for several years. New forms include not only the World Wide Web, but also talk radio and cable television. Talk radio would seem to be a very dated medium, given that radio first become important early in the twentieth century. Between 1949 and 1987, however, the Federal Communications Commission enforced the Fairness Doctrine, which required radio and television to present controversial issues in a manner that was (in the commission's view) honest, equitable, and balanced. Modern conservative talk radio took off only after the Fairness Doctrine was abolished.
The impact of the various forms of new media appears to vary considerably. Talk radio and cable networks such as Fox News have given conservatives new methods for promoting their views and socializing their audiences. It is probable, however, that such media mostly strengthen the beliefs of those who are already conservative, rather than recruiting new members to the political right. A similar observation is often made about political blogs on the Internet, although in this medium liberals are at least as well represented as conservatives. Blogs appear to radicalize their readers, rather than turn conservatives into liberals or vice versa. Indeed, cable news, talk radio, and political blogs are widely blamed for the increased polarization that has characterized American politics in recent years.
The impact of social networking sites such as Facebook is more ambiguous. Facebook does have strongly political "pages" that, in effect, are political blogs. Many interactions on Facebook, however, are between members of peer groups, such as students who attend a particular school or individuals who work in the same profession. Such groups are likely to contain a variety of views than are groups explicitly organized around a political viewpoint. Facebook, in other words, may enhance peer group influence and even serve as a force for political moderation.
The Influence of Political Events
Generally, older Americans tend to be somewhat more conservative than younger Americans - particularly on social issues but also, to some extent, on economic issues. This effect probably occurs because older adults are likely to retain the social values that they learned at a younger age. The experience of marriage and raising a family also has a measurable conservatizing effect. Young people, especially today, are more liberal than their grandparents on social issues, such as on the rights of gay men and lesbians and on racial and gender equality. Nevertheless, a more important factor than a person's age is the impact of momentous political events that shape the political attitudes of an entire generation. When events produce such a long-lasting result, we refer to it as a generational effect (also called a cohort effect).
Historical Events. Voters who grew up in the 1930s during the Great Depression were likely to form lifelong attachments to the Democratic Party, the party of Franklin D. Roosevelt. In the 1960s and 1970s, the war in Vietnam, the Watergate break-in, and the subsequent presidential cover-up fostered widespread cynicism toward government. There is evidence that the years of economic prosperity under President Ronald Reagan during the 1980s led many young people to identify with the Republican Party. The high levels of support that younger voters have given to Barack Obama during his presidential campaigns may be good news for the Democratic Party in future years, even if it did not help the Democrats in the 2010 congressional elections.
The Political Mood. A number of political scientists believe they can make some broad generalizations about events that tug voters to the political right or the political left. One such proposition is that the public mood swings in a more liberal direction when the federal government is successful, especially in handling economic issues. Presumably, the identification of liberalism with active government means that the image of liberalism improves when the government is successful. Likewise, a trend is toward conservatism during periods of perceived government failure. The presidencies of Republican Gerald Ford (1974-1977) and Democrat Jimmy Carter (1977-1981) provide examples of failure. In those years, the government appeared to be incapable of addressing huge rises in prices, gasoline shortages, and other problems. The resulting trend toward conservatism culminated in the election of Republican Ronald Reagan (1981-1989) as president.
A second apparent tendency is that when the government seems to be introducing conservative policies - which usually means that a Republican is president - public sentiment drifts to the left. When governance is perceived to be liberal - which typically means there is a Democratic president - opinion moves to the right. This tendency and the success/failure dynamic just mentioned can reinforce each other or balance each other out. As an example, Reagan's presidency was both conservative and successful. As a result, the public mood, which was quite conservative at the time that Reagan was first elected, shifted back toward the center during his two terms in office. Eventually, voters picked Democrat Bill Clinton (1993-2001) as president.
POLITICAL PREFERENCES AND VOTING BEHAVIOR
A major indicator of voting behavior is, of course, party identification. In addition, a variety of socioeconomic and demographic factors also appear to influence political preferences. These factors include education, income and socioeconomic status, religion, race, gender, geographic region, and similar traits. People who share the same religion, occupation, or any other demographic trait are likely to influence one another and may also have common political concerns that follow from the common characteristic.
Other factors, such as perception of the candidates and issue preferences, are closely connected to the electoral process itself.
Party Identification and Demographic Influences
With the possible exception of race, party identification has been the most important determinant of voting behavior in national elections. Party affiliation is influenced by family and peer groups, by generational effects, by the media, and by the voter's assessment of candidates and issues.
In the middle to late 1960s, party attachment began to weaken. Whereas independent voters were only a little more than 20 percent of the eligible electorate during the 1950s, they constituted more than 30 percent of all voters by the mid-1990s, and their numbers have grown since that time. New voters are likely to identify themselves as independent voters, although they may be more ready to identify with one of the major parties by their mid-thirties. There is considerable debate among political scientists over whether those who call themselves independents are truly so: when asked, a majority say that they are "leaning" toward one party or the other.
Demographic influences reflect the individual's personal background and place in society. Some factors, such as race and (for most people) religion, have to do with the family into which a person was born. Others may be the result of choices made throughout an individual's life: place of residence, educational achievement, and profession. It is also clear that many of these factors are interrelated. People who have more education typically have higher incomes and may hold professional jobs. Similarly, children born into wealthier families are far more likely to complete college than are children from poor families.
Educational Achievement. In the past, having a college education tended to be associated with voting for Republicans. In recent years, however, this correlation has become weaker. In particular, individuals with a postgraduate education - more than a bachelor's degree - have become predominantly Democratic. Many people with postgraduate degrees are professionals, such as physicians, attorneys, and college instructors. Usually, a postgraduate degree is an occupational requirement for professionals.
Despite the recent popularity of the master of business administration (MBA) degree, businesspersons are more likely to have only a bachelor's degree or no degree at all. They are also much more likely to vote Republican.
Also, a higher percentage of voters with only a high school education voted Republican in the last four presidential elections, compared with the pattern in previous elections, in which that group of voters tended to favor Democrats more strongly.
Economic Status. Family income is a strong predictor of economic liberalism or conservatism. Those with low incomes tend to favor government action to benefit the poor or to promote economic equality. Those with high incomes tend to oppose government intervention in the economy or to support it only when it benefits business. On political issues, therefore, the traditional political spectrum (described in Ch. 1) is a useful tool. The rich often tend toward the right, and the poor often lean toward the left.
If we examine cultural as well as economic issues, however, the four-cornered ideological grid (Ch. 1) becomes important. It happens that upper-class voters are more likely to endorse cultural liberalism and lower-class individuals are more likely to favor cultural conservatism. Support for the right to have an abortion, for example, rises with income. It follows that libertarians - those who oppose government action on both economic and social issues - are concentrated among the wealthier members of the population. Those who favor government action both to promote traditional moral values and to promote economic equality - economic liberals, cultural conservatives - are concentrated among groups that are less well off.
That said, it remains generally true that the higher a person's income, the more likely that person will be to vote Republican. Manual laborers, factory workers, and especially union members are more likely to vote Democratic. There are no hard-and-fast rules, however. Many very poor individuals are devoted Republicans, just as some extremely wealthy people support the Democratic Party.
In recent years, some commentators have come to believe that a growing cultural gap separates well-educated members of the upper-middle class and less-well-educated working class individuals.
Religious Denomination. Traditionally, scholars have examined the impact of religion on political attitudes by dividing the population into such categories as Protestant, Catholic, and Jewish. In recent decades, however, such a breakdown has become less valuable as a means of predicting someone's political preferences. It is true that Jewish voters, as they were in the past, are notably more liberal than members of other groups, on both economic and cultural issues. Persons reporting no religion are very liberal on social issues but have mixed economic views.
Protestants and Catholics, however, have grown closer to each other politically in recent years. This represents something of a change - in the late 1800s and early 1900s, northern Protestants were distinctly more likely to vote Republican, and northern Catholics were more likely to vote Democratic. Even now, in a few parts of the country, Protestants and Catholics tend to line up against each other when choosing a political party.
Religious Commitment and Beliefs. Today, two factors turn out to be major predictors of political attitudes among members of the various Christian denominations. One factor is the degree of religious commitment, as measured by such actions as regular churchgoing. The other is the degree to which the voter adheres to religious beliefs that (depending on the denomination) can be called conservative, evangelical, or fundamentalist. High scores on either factor are associated with cultural conservatism on political issues - that is, with beliefs that place a high value on social order.
Voters who are more devout, regardless of their church affiliation, tend to vote Republican, while voters who are less devout are more often Democrats. In the 2012 presidential elections, for example, Protestants who attended church weekly gave 70 percent of their votes to Republican candidate Mitt Romney, compared with 55 percent of those who attended church less often. Among Catholics, there was a similar pattern. 57 percent of Catholics who attended church weekly voted for Romney, while 42 percent of Catholics who were not regular churchgoers voted for him. Exit polls following the 2010 congressional elections showed the same pattern. There is an exception to this trend: African Americans of all religious tendencies have been strongly supportive of Democrats.
The politics of Protestant Americans who can be identified as holding evangelical or fundamental beliefs deserve special attention. Actually, a majority of American Protestants can be characterized as evangelical. Not all are politically conservative. Some are politically liberal, such as former Democratic presidents Jimmy Carter and Bill Clinton. Fundamentalists are a subset of evangelicals who believe in a number of doctrines not held by all evangelicals. In particular, fundamentalists believe in biblical inerrancy - that is, that every word of the Bible is literally true. In politics, fundamentalists are notably more conservative than other evangelicals. Liberal fundamentalists are rare indeed.
Race and Ethnicity. Although African Americans, on average, are somewhat conservative on certain cultural issues, such as same-sex marriage and abortion, they tend to be more liberal than whites on social-welfare matters, civil liberties, and even foreign policy. African Americans voted principally for Republicans until Franklin Roosevelt's New Deal in the 1930s. Since then, they have largely identified with the Democratic Party. Indeed, Democratic presidential candidates have received, on average, more than 80 percent of the African American vote since 1956. Barack Obama's support among African Americans has been overwhelming.
Most Asian American groups lean toward the Democrats, although Vietnamese Americans are strongly Republican. Most Vietnamese Americans left Vietnam because of the Communist victory in the Vietnam War, and their strong anticommunism translates into conservative politics.
Muslim American immigrants and their descendants make up an interesting category. In 2000, a majority of Muslim Americans of Middle Eastern ancestry voted for Republican George W. Bush because they shared his cultural conservatism. By 2012, the issues of Muslim civil liberties and discrimination against Muslims had turned Islamic voters into one of the nation's most Democratic blocs.
The Hispanic Vote. The diversity among Hispanic Americans has resulted in differing political behavior. The majority of Hispanic Americans vote Democratic. Cuban Americans, however, are usually Republican. Most Cuban Americans left Cuba because of Fidel Castro's Communist regime. As in the example of the Vietnamese, anticommunism leads to political conservatism.
In 2004, Republican presidential candidate George W. Bush may have received almost 40 percent of the Hispanic vote. (A widely quoted survey put Bush's support at 44 percent, but that figure was almost certainly erroneous.) Since his days as Texas governor, Bush had envisioned creating a stronger long-term Republican coalition by adding Hispanics. Indeed, Latino voters appeared to show considerable sympathy for Bush's campaign appeals based on religious and family values and patriotism.
In 2008, however, Barack Obama won more than two-thirds of the Latino vote, a substantially larger number than his showing among non-Hispanic whites. Why did Hispanic support for the Republicans fall so greatly? In a word: immigration. Bush favored a comprehensive immigration reform that would have granted unauthorized immigrants (also known as illegal or undocumented immigrants) a path to citizenship. Most Republicans in Congress refused to support Bush on this issue and instead called for a hard line against unauthorized immigration. The harsh rhetoric of some Republicans on this issue convinced many Latinos that the Republicans were hostile to Hispanic interests.
The Gender Gap. Until the 1980s, there was little evidence that men's and women's political attitudes were very different. Following the election of Ronald Reagan in 1980, however, scholars began to detect a gender gap. A May 1983 Gallup poll revealed that men were more likely than women to approve of Reagan's job performance. The gender gap reappeared in subsequent elections, with women being more likely than men to support the Democratic candidate. The gender gap has continued in recent years. In 2012, 55 percent of women voted for Democrat Barack Obama, compared with 45 percent of men.
Women's attitudes also appear to differ from those of their male counterparts on a range of issues other than presidential preferences. They are much more likely than men to oppose capital punishment and the use of force abroad. Studies also have shown that women are more concerned about risks to the environment, more supportive of social welfare, and more in agreement with extending civil rights to gay men and lesbians than are men. In contrast, women were also more concerned than men about the security issues raised by the events of 9/11. This last fact may have pushed women in a more conservative direction, at least for a time.
Geographic Region. Finally, where you live can influence your political attitudes. The former solid Democratic South has now become the solid Republican South. There is a tendency today, at least in national elections, for the South, the Great Plains, and several of the Rocky Mountain states to favor the Republicans and for the West Coast and the Northeast to favor the Democrats. Perhaps more important than region is residence - urban, suburban, or rural. People in large cities tend to be liberal and Democratic. Those who live in smaller communities tend to be conservative and Republican.
This tendency appears to have grown stronger in recent years. Chicago's Barack Obama has done better than previous Democratic candidates in urban areas, but worse in the countryside. The urban-rural split even showed up in the 2012 Republican presidential primaries, in which former Massachusetts governor Mitt Romney, the eventual winner, swept the metropolitan vote but lost the small cities and rural areas to opponents such as former Pennsylvania senator Rick Santorum.
Election-Specific Factors
Factors such as perception of the candidates and issue preferences may have an effect on how people vote in particular elections. Candidates and issues can change greatly, and voting behavior can therefore change as well.
Perception of the Candidates. A candidate's image seems to be important in a voter's choice, especially for a presidential candidate. To some extent, voter attitudes toward candidates are based on emotions (such as trust) rather than on any judgment about experience or policy. In some years, voters have been attracted to a candidate who appeared to share their concerns and worries. In other years, voters have sought a candidate who appeared to have high integrity and honesty. Voters have been especially attracted to these candidates in elections that follow a major scandal, such as Richard Nixon's Watergate scandal (1972-1974) or Bill Clinton's sex scandal (1998-1999).
Issue Preferences. Issues make a difference in presidential and congressional elections. Although personality or image factors may be very persuasive, most voters have some notion of how the candidates differ on basic issues or at least know which candidates want a change in the direction of government policy.
Historically, economic concerns have been among the most powerful influences on public opinion. When the economy is doing well, it is very difficult for a challenger, especially at the presidential level, to defeat the incumbent. In contrast, inflation, unemployment, or high interest rates are likely to work to the disadvantage of the incumbent.
Perceptions of the 2012 Presidential Candidates. Economic conditions suggested a close presidential race in 2012, so perceptions of the candidates were important. By 2012, almost every voter had a firm opinion of Barack Obama. Mitt Romney was less well known. The Obama campaign sought, with some success, to paint Romney as an arrogant plutocrat, out of touch with ordinary people. A gaffe in which Romney seemed contemptuous of anyone not wealthy enough to pay income taxes aided this effort. At the first presidential debate in October, however, Romney largely dispelled this negative image, rescuing his campaign.
MEASURING PUBLIC OPINION
In a democracy, people express their opinions in a variety of ways, as mentioned in this chapter's introduction. One of the most common means of gathering and measuring public opinion on specific issues is through the use of opinion polls.
The History of Opinion Polls
During the 1800s, certain American newspapers and magazines spiced up their political coverage by conducting face-to-face polls or mail surveys of their readers' opinions. In the early twentieth century, the magazine Literary Digest further developed the technique of opinion polling by mailing large numbers of questionnaires to individuals, many of whom were its own subscribers, to determine their political opinions. From 1916 to 1932, more than 70 percent of the magazine's election predictions were accurate.
In 1936, however, the magazine predicted that Republican Alfred Landon would defeat Democrat Franklin D. Roosevelt in the presidential race. Landon won in only two states. A major problem was that in 1936, several years into the Great Depression, the Digest's subscribers were considerably wealthier than the average American. In other words, they did not accurately represent all of the voters in the U.S. population.
Several newcomers to the public opinion poll industry accurately predicted Roosevelt's landslide victory. These newcomers are still active in the poll-taking industry today: the Gallup poll of George Gallup and the Roper poll founded by Elmo Roper. Gallup and Roper, along with Archibald Crossley, developed the modern polling techniques of market research. Using personal interviews with small samples of selected voters (fewer than two thousand), they showed that they could predict with relative accuracy the behavior of the total voting population.
Sampling Techniques
How can interviewing fewer than two thousand voters tell us what tens of millions of voters will do? Clearly, it is necessary that the sample of individuals be representative of all voters in the population.
The Principle of Randomness. The most important principle in sampling, or poll taking, is randomness. Every person should have a known chance, and especially an equal chance, or being sampled. If sampling follows this principle, then a small sample should be representative of the whole group, both in demographic characteristics (age, religion, race, region, and the like) and in opinions. The ideal way to sample the voting population of the United States would be to put all voters names into a jar - or a computer file - and randomly sample, say, two thousand of them. Because this is too costly and inefficient, pollsters have developed other ways to obtain good samples. One technique is simply to choose a random selection of telephone numbers and interview the respective households. This technique produces a relatively accurate sample at a low cost.
To ensure that the random samples include respondents from relevant segments of the population - rural, urban, northeastern, southern, and the like - most survey organizations randomly choose, say, urban areas that they will consider as representative of all urban areas. Then they randomly select their respondents within those areas.
The Statistical Nature of Polling. Universally, when the results of an opinion poll are announced, the findings are reported as specific numbers. A poll might find, for example, that 10 percent of those surveyed approve of the job performance of Congress. Such precise figures can mislead you as to the essential nature of polling. In reality, it makes more sense to consider the results of a particular survey question as a range of numbers, not a single integer. That would mean that the question about Congress's job performance yielded an answer that fell somewhere between 7 percent and 13 percent. The reported figure of 10 percent is only the midpoint of the possible spread - the most probable result. If we had been able to question all members of the public, what are the chances that they would give Congress a 10 percent rating? The chances are not high. Even if the pollster in this case employed the best possible practices, the odds are better than 50-50 that the true answer is not exactly 10 percent, but some other number in the 7-13 percent range.
Sampling Error. Reputable polling firms report the margin of error associated with their results. The results of a carefully conducted poll that surveys a large number of respondents - say, two thousand people - might have a 95 percent chance of falling within a 3 percent margin of error. The 95 percent figure is an industry standard. What this means, however, is that the pollster believes that any given poll result has one chance in twenty of landing four or more percentage points away from the true answer, which we would get if we really could interview everyone. In the example of the question about Congress, there is one chance in twenty that the approval rating is actually below 7 percent or above 13 percent.
These variations are called sampling error. They follow from the fact that the poll taker is examining a sample and not the entire population. Sampling error is one reason that knowledgeable poll watchers disregard small variations in poll results. Gallup, for example, polls the public on its approval of the president's job performance every day. Such continuous polls are known as tracking polls. Let us say that on Monday, the tracking poll shows that the president has a 45 percent approval rating. On Tuesday, it is 47 percent, on Wednesday 46 percent, and on Thursday 43 percent. Was the president really more popular on Tuesday and Wednesday? Almost certainly not. These variations are simply artifacts of the polling process - so much "statistical noise."
The Difficulty of Obtaining Accurate Results
Reputable polling organizations devote a substantial amount of effort to ensuring that their samples are truly random. If they succeed, then the accuracy of their results should be limited only by sampling error. Unfortunately, obtaining a completely random sample of the population is difficult. Poll takers must consider that women are more likely to answer the telephone than men. Some kinds of people, such as students and low-income individuals, are relatively hard to contact. Despite their best efforts, pollsters may be unable to gather a pool or respondents that precisely mirrors the general public.
Weighting the Sample. Polling firms address this problem of obtaining a true random sample by weighting their samples. That is, they correct for differences between the sample and the public by adding extra "weight" to the responses of underrepresented groups. For example, 20 percent of the respondents in a survey might state that they are evangelical Christians. Based on other sources of information, however, the poll taker believes that the true share of evangelicals in the target population is 25 percent. Therefore, the responses of the evangelical respondents receive extra weight so that they make up 25 percent of the reported result.
It is relatively easy to correct a sample for well-known demographic characteristics, such as education, gender, race, religion, and geography. It is much harder, however, to correct for such characteristics as political ideology, partisan preference, and likelihood of voting. Difficult as it may be, each major polling organization tries to make such corrections. The formulae they employ are typically trade secrets and are never disclosed to the public or the press.
House Effects. One practical result of the use of secret in-house weighting schemes is that the results reported by one polling firm may differ from those reported by another in a systematic fashion. Pollster A might consistently rate the chances of Republican candidates as 2 percent higher than does pollster B. This could happen because pollster A thinks that turnout among Republican voters will be high, whereas pollster B predicts a lower turnout. Alternatively, pollster A may simply believe that there are more Republicans in the electorate. A consistent difference in polling results between firms is known as a house effect. (House here means the organization or firm, as when referring to casinos.) Generally, house effects are measured by comparing a firm's results with the average results of all other poll takers.
Some polling organizations have ties to one of the major political parties. Not surprisingly, many such firms have house effects that favor their parties. Such relationships do not always exist, however. Some party-linked pollsters exhibit no discernible house effect, and some firms that are known for their independence do demonstrate such an effect. Further, a house effect does not mean that a firm's results are in error. It could be noticing something important that its rivals have missed.
How Accurate Are the Results? Despite all of the practical difficulties involved in poll taking, the major polling organizations have a good record in predicting the outcome of presidential contests.
It is more difficult to conduct an accurate poll at the state level. It is even harder to maintain accuracy in primary elections, because voters find it much easier to switch allegiance from one of their party's candidates to another than to switch to a different party altogether.
Additional Problems with Polls
Public opinion polls are snapshots of opinions and preferences at a specific moment in time and as expressed in response to a specific question. Given that definition, it is fairly easy to understand situations in which the polls are wrong. For example, opinion polls leading up to the 1980 presidential elections showed President Jimmy Carter defeating challenger Ronald Reagan. Only a few analysts noted the large number of "undecided" respondents a week before the elections. Those voters shifted massively to Reagan at the last minute, and Reagan won the election.
Poll Questions. It makes sense to expect that the results of a poll will reflect the questions that are asked. Depending on what question is asked, voters could be said either to support a particular proposal or to oppose it. One of the problems with many polls is the yes/no answer format. For example, suppose that a poll question asks, "Do you support President Obama's intervention in Libya?" Respondents who have a complicated view of these events, as many people do, have no way of indicating their true position because "yes" and "no" are the only possible answers.
How a question is phrased can change the polling outcome dramatically. The Roper polling organization once asked a double-negative question that was very hard to understand: "Does it seem possible or does it seem impossible to you that the Nazi extermination of the Jews never happened?" The survey results showed that 20 percent of Americans seemed to doubt that the Holocaust ever occurred. When the Roper organization rephrased the question more clearly, the percentage of doubters dropped to less than 1 percent.
Respondents' answers are also influenced by the order in which questions are asked, by the possible answers from which the respondents are allowed to choose, and, in some cases, by their interaction with the interviewer. To a certain extent, people try to please the interviewer. They answer questions about which they have no information and avoid some answers to try to measure up to the interviewer's expectations.
Unscientific and Fraudulent Polls. A perennial issue is the promotion of surveys that are unscientific or even fraudulent. All too often, a magazine or Web site asks it readers to respond to a question - and then publishes the answers as if they were based on a scientifically chosen random sample. Other news media may then publicize the survey as if it were a poll taken by such reliable teams as Gallup, CBS and the New York Times, or the Wall Street Journal and NBC. Critical consumers should watch out for surveys with self-selected respondents and other types of skewed samples. These so-called polls may be used to deliberately mislead the public.
In recent years, polling experts have made a number of allegations of outright polling fraud. In these cases, so-called polling firms have been accused of reporting results without ever conducting any actual surveys. The numbers have been simply made up, out of thin air. After a challenge by polling experts, one such company ceased issuing polls and quietly disappeared. Another poll taker was sued by its largest customer, a major political blog. The blogger claimed that the pollster was making up its results. (The case was settled out of court.)
Push Polls. Some campaigns have used "push polls," in which the respondents are given misleading information in the questions asked to persuade them to vote against a candidate. Indeed, the practice has spread throughout all levels of U.S. politics - local, state, and federal. In 1996, in a random survey of forty-five candidates, researchers found that thirty-five of them claimed to have been victimized by negative push-polling techniques used by their opponents. Now even advocacy groups, as well as candidates for political offices, are using push polls. Push polling was prevalent during the campaigns in 2008, the congressional campaigns in 2010, and the 2012 presidential and congressional campaigns. Its use was widespread in campaigns for governorships and other state offices.
TECHNOLOGY AND OPINION POLLS
Public opinion polling is based on scientific principles, particularly the principle of randomness. Today, technological advances allow polls to be taken over the Internet, but serious questions have been raised about the ability of pollsters to obtain truly random samples using this medium. The same was not said long ago when another technological breakthrough changed public opinion polling - the telephone.
The Advent of Telephone Polling
During the 1970s, telephone polling began to predominate over in-person polling. Obviously, telephone polling is less expensive than sending interviewers to poll respondents in their homes. Additionally, telephone interviewers do not have to worry about safety problems, which is particularly important for interviewers working in high-crime areas. Finally, telephone interviews can be conducted relatively quickly. They allow politicians or the media to poll one evening and report the results the next day.
In recent years, many polling firms that rely on the telephone have begun employing automatic scripts read by a machine, the so-called robopolls. Polls of this type are much cheaper and easier to set up than traditional polls using live interviewers. Many observers also believe, however, that robopolls are substantially less accurate than traditional surveys.
Telephone Polling Problems. Somewhat ironically, the success of telephone polling has created major problems for the technique. The telemarketing industry in general has become so pervasive that people increasingly refuse to respond to telephone polls. More than 40 percent of households now use either caller ID or some other form of call screening. This has greatly reduced the number of households that polling organizations can reach.
For most telephone polls, the nonresponse rate has increased to as high as 80 percent. Such a high nonresponse rate undercuts confidence in the survey results. In most cases, polling only 20 percent of those on the list cannot lead to a random sample. Even more important for politicians is the fact that polling organizations are not required to report their response rates.
The Cell Phone Problem. An additional problem for telephone polling is the popularity of cell phones. Cellular telephone numbers are not always included in random-digit dialing programs and are not listed in telephone directories. Furthermore, pollsters cannot legally use robopolling to contact cell phone users. Also, individuals with cell phones may be located anywhere in the United States or the world, thus confounding attempts to reach people in a particular area. As more people, and especially younger Americans, choose to use only a cell phone and do not have a landline at all, polling accuracy is further reduced because these individuals are ignored by many polls.
A final problem for telephone pollsters that results from new technologies: many savvy Web users now rely on Skype and other voice-over-Internet services for their telephone calls. Skype allows users to see each other through the camera built into modern laptop computers, tablets, and cell phones. Groups of users can participate in videoconferences, viewing each other in separate windows on the screen. Skype's competitors include Ekiga, Empathy, Google Talk, and Linphone. Currently, no polling firm has discovered a way to survey voice-over-Internet users.
Enter Internet Polling
Obviously, Internet polling is not done on a one-on-one basis, because there is no voice communication. In spite of the potential problems, the Harris poll, a widely respected national polling organization, conducted online polls during the 1998 elections. Its election predictions were accurate in many states. Nonetheless, it made a serious error in one southern gubernatorial election. The Harris group subsequently refined its techniques and continues to conduct online polls. This organization believes that proper weighting of the results will achieve the equivalent of a random-sample poll.
How Representative Is the Internet? Public opinion experts, however, argue that the Harris poll procedure violates the mathematical basis of random sampling. Even so, the Internet population is looking more like the rest of America: almost as many women go online as men (76 percent of women and 80 percent of men), 71 percent of African American adults are online, and so are 68 percent of Hispanics (compared with 80 percent of non-Hispanic whites).
"Nonpolls" on the Internet. Even if organizations such as the Harris poll succeed in obtaining the equivalent of a random sample when polling on the Internet, another problem will remain: the proliferation of "nonpolls" on the Internet. Every media outlet that maintains a Web site allows users to submit their opinions. A variety of organizations and for-profit companies send so-called polls to individuals through e-mail. Mister Poll bills itself as the Internet's largest online polling database. Mister Poll allows you to create your own polls just for fun and to include them on your home page. Like many other polling sites, Mister Poll asks questions on various issues and solicits answers from those who log on to its Web site. Although the Mister Poll site states, "None of these polls is scientific," sites such as this one undercut the efforts of legitimate pollsters to use the Internet scientifically.
PUBLIC OPINION AND THE POLITICAL PROCESS
Public opinion affects the political process in many ways. Politicians, whether in office or in the midst of a campaign, see public opinion as important to their careers. The president, members of Congress, governors, and other elected officials realize that strong support by the public as expressed in opinion polls is a source of power in dealing with other politicians. It is far more difficult for a senator to say no to the president if the president is immensely popular and if polls show approval of the president's policies. Public opinion also helps political candidates identify voters' most important concerns and may help them shape their campaigns successfully.
Political Culture and Public Opinion
Americans are divided into a multitude of ethnic, religious, regional, and political subgroups. Given the diversity of American society and the wide range of opinions contained within it, how is it that the political process continues to function without being stalemated by conflict and dissension? One explanation is rooted in the concept of the American political culture, which can be described as a set of attitudes and ideas about the nation and the government. As discussed in Chapter 1, our political culture is widely shared by Americans of many different backgrounds. To some extent, it consists of symbols, such as the American flag, the Liberty Bell, and the Statue of Liberty. The elements of our political culture also include certain shared beliefs about the most important values in the American political system, including liberty, equality, and property.
Political Culture and Support for Our Political System. The political culture provides a general environment of support for the political system. If people share certain beliefs about the system and a reservoir of good feeling exists toward the institutions of government, the nation will be better able to weather periods of crisis. Such was the case after the 2000 presidential election when, for several weeks, it was not certain who the next president would be and how that determination would be made. At the time, some contended that the nation was facing a constitutional crisis. Certainly, in many nations this would have been true. In fact, however, the broad majority of Americans did not believe that the uncertain outcome of the elections had created a constitutional crisis. Polls taken during this time found that, on the contrary, most Americans were confident in our political system's ability to decide the issue peaceably and in a lawful manner.
Political Trust. The political culture also helps Americans evaluate their government's performance. Political trust, the degree to which individuals express trust in political institutions, has been measured by a variety of polling questions. One of these is whether the respondent is satisfied with "the way things are going in the United States." During the successful presidency of Republican Ronald Reagan, satisfaction levels rose from a fairly dismal 20 percent range to around 50 percent. Republican George H. W. Bush (1989-1993) enjoyed high levels of satisfaction until 1992, when rates fell back to the 20 percent range. This fall reflected the economic problems and other difficulties that handed the presidency to Democrat Bill Clinton.
Clinton's two terms appear to have been mostly a success, as satisfaction levels rose to around 60 percent. Under Republican George W. Bush (2001-2009), however, satisfaction levels slowly fell. In October 2008, at the peak of the crisis in the financial industry, satisfaction bottomed out at an unprecedented 7 percent. Thereafter, the rate leveled off at about 20 percent, which suggested ongoing dissatisfaction with the government.
Public Opinion about Government
A vital component of public opinion in the United States is the considerable ambivalence with which the public regards many major national institutions. Over the years, military and religious organizations have ranked highest. Note, however, the decline in confidence in churches in 2002 following sex-abuse allegations against a number of Catholic priests. The public has consistently had more confidence in the military than in any of the other institutions. In 2002 and 2003, confidence in the military soared even higher, most likely because of the military's role in the war on terrorism.
Confidence in Other Institutions. The United States Supreme Court and the public schools have scored well over time. Less confidence is expressed in television, organized labor, and big business.
In 1991, following a scandal involving congressional banking practices, confidence in Congress fell to 18 percent, but then recovered. From 2007 on, however, confidence in Congress almost disappeared. By 2010, the figure was down to 11 percent, an all-time low.
At times, popular confidence in all institutions may rise or fall, reflecting optimism or pessimism about the general state of the nation. For example, a Gallup poll in 2008 following the onset of the Great Recession showed that the level of national satisfaction with the state of the nation had dropped to 14 percent - the lowest rating since 1992.
The Most Important Problem. Although people may not have much confidence in government institutions, they nonetheless turn to government to solve what they perceive to be the major problems facing the country. The public tends to emphasize problems that are immediate and that have been the subject of many stories in the media. When coverage of a particular problem increases suddenly, the public is more likely to see that as the most important problem. Thus, the fluctuation in the "most important problem" may, at times, be attributed to media agenda setting. In other years, however, the nation's leading problem has been so obvious that media attention is a minor factor. The Great Recession, which began in December 2007, is one example.
Public Opinion and Policymaking
If public opinion is important for democracy, are policymakers really responsive to public opinion? A classic study by political scientists Benjamin I. Page and Robert Y. Shapiro in the early 1990s suggested that in fact the national government is somewhat responsive to the public's demands for action. In looking at changes in public opinion poll results over time, Page and Shapiro showed that when the public supports a policy change, the following occurs: policy changes in a direction consistent with the change in public opinion 43 percent of the time, policy changes in a direction opposite to the change in opinion 22 percent of the time, and policy does not change at all 33 percent of the time. More recently, scholars have contended that these studies overstate the effect of public opinion on public policy, in part because on many of the issues taken up by policymakers, the public has no clear opinions.
Setting Limits on Government Action. Although opinion polls cannot give exact guidance on what the government should do in a specific instance, the opinions measured in polls may set an informal limit on government action. For example, consider the highly controversial issue of abortion. Most Americans are moderates on this issue. They do not approve of abortion as a means of birth control, but they do feel that it should be available under certain circumstances. Yet sizable groups of people express very intense feelings both for and against legalized abortion. Many politicians also hold polarized beliefs on the topic of abortion. Political party platforms take firm positions on abortion, and members of Congress regularly make impassioned speeches on behalf of the right-to-life or freedom-of-choice positions. The strong words, however, are not matched with strong actions. At the national level, abortion policies have been roughly stable for years.
Politicians know that radical changes to abortion policies would offend the opinions of the moderate majority. In this example, as in many others, public opinion does not make public policy. Rather, it restrains officials from taking truly unpopular actions. If officials do act in the face of public opinion, the consequences will be determined at the ballot box.
The Public versus the Policymakers. To what degree should public opinion influence policymaking? It would appear that members of the public view this issue differently than policy leaders do. Polls indicate that whereas a majority of the public feel that public opinion should have a great deal of influence on policy, a majority of policy leaders hold the opposite position. Why would a majority of policy leaders not want to be strongly influenced by public opinion? One answer to this question is that public opinion polls can provide only a limited amount of guidance to policymakers.
The Limits of Polling. Policymakers cannot always be guided by opinion polls. In the end, politicians must make their own choices. When they do so, their choices necessarily involve trade-offs. If politicians vote for increased spending to improve education, for example, by necessity fewer resources are available for other worthy projects. Moreover, to make an informed policy choice requires an understanding not only of the policy area but also of the consequences of any given choice. Public opinion polls very rarely ensure that those polled have such information.
Finally, government decisions cannot be made simply by adding up individual desires. Politicians engage in "horse trading" with each other. Politicians also know that they cannot satisfy every desire of every constituent. Therefore, each politician attempts to maximize the net benefits to his or her constituents, while keeping within the limits of whatever the politician believes the government can afford.
A Policy Example: Contraception Insurance
Many policy issues could be used to demonstrate the limits of public opinion's effect on policymakers. One example is controversy that developed in early 2012: whether employers should be required to include coverage for birth control methods - contraception - in the insurance plans offered to their employees.
Under the health-care reform legislation adopted in 2010, it was left to the executive branch to define which medical services had to be covered by qualifying plans. In February 2012, the Obama administration ruled that religious bodies such as churches that opposed birth control in principle did not need to pay for health insurance plans that covered contraception. Institutions such as hospitals and schools that were owned or controlled by churches, however, would be required to provide coverage like all other employers.
The Controversy. The Obama ruling was denounced by religious bodies, including the Catholic Church, that oppose contraception and also control hospitals and colleges. The issue was quickly picked up by Republicans in Congress, who accused the administration of trampling on religious freedom. Defenders of the ruling argued in response that exempting religiously owned hospitals and schools from the mandate would mean that employers could force their own religious principles on their employees. In the end, the Obama administration came up with a compromise under which contraception would still be covered by insurance, but religiously controlled institutions would not have to pay for the coverage. Many Catholic hospitals and colleges accepted this plan. In May 2012, however, the Catholic bishops and a number of Catholic institutions sued to overturn it.
Public Opinion and the Controversy. What effect did public opinion have on this dispute? That is hard to say, because it is not clear what the opinion of the public actually was. Popular attitudes were determined overwhelmingly by whether the topic of the controversy was religious freedom or the right of women to reproductive health care. Consider two polls taken by CBS News/The New York Times 2012. In February, a poll asked whether the insurance plans of religiously affiliated institutions should be required to cover contraception. Respondents supported coverage by 61 percent to 31 percent. In a March poll, the question included the language "opt out based on religious and moral objections." This time, respondents agreed by 57 percent to 36 percent that the institutions ought to be able to opt out.
These differing survey results emphasize the important concept of framing. The public response to a particular issue may depend largely on how the question is framed - in this example, whether the issue is framed as a question of women's health rights or a question of religious freedom. Indeed, the contraception insurance issue soon became a political battle to determine which frame would predominate.
DID YOU KNOW?
-CNN reaches more than 2 billion people in more than one hundred countries.
-Britain had a major gender gap for much of the twentieth century - because women were much more likely than men to support the Conservative Party rather than the more left-wing Labour Party.
-More than 30 percent of people asked to participate in an opinion poll refuse to do so.
-A straw poll conducted by patrons of Harry's New York Bar in Paris has an almost unbroken record of predicting the outcomes of U.S. presidential contests.
-When Americans were asked if they thought race relations were good or bad in the United States, 68 percent said that they were "bad," but when asked about race relations in their own communities, 75 percent said that they were "good."
-In its "Cess Poll," an Iowa radio station assesses voters' preferences for presidential primary candidates by reading a candidate's name over the air, asking supporters of that candidate to flush their toilets, and measuring the resulting drop in water pressure.