Skip to main content Skip to main content

Democracy in America

Public Opinion: Voice of the People

Discover what public opinion is, how it varies and how it affects what the government does.

This program examines the power of public opinion to influence government policy, the increasing tendency of public officials to rely on polls, and the need to use many forms of feedback to get an accurate measure of public opinion.


Public Opinion: Voice of the People

Learning Objectives

After completing this session, you will be able to:

  • Define public opinion and discuss its major characteristics.
  • Discuss the role that public opinion has in the American political system.
  • Discuss the role of public opinion polls in politics and government.
  • Describe the factors important to carrying out an accurate poll, and the strength and weaknesses of data from polls.
  • Describe means, other than polling, for gauging public opinion.

Unit 11 examines the important role of public opinion in American democracy. The difficult concept of public opinion is illustrated by first exploring the political culture of the United States. The unit then offers an illustration of the powerful influence that public opinion can and often does have on decision-makers. The unit also elucidates the process of taking a poll, demonstrating the pitfalls that can befall those who do polling. Finally, the unit explores other avenues available to the public to express their opinion. The prevalence of polling often obscures these other powerful ways of expressing one’s views.

Most Americans share a common, or core, political culture, which is a set of values, beliefs, and traditions regarding politics and government. Our shared values include a general faith in representative democracy, in basic freedoms like freedom of speech, in basic rights such as voting, and in the principles of capitalism and private property.

While our shared values that make up our political culture remain intact, many Americans have lost confidence in their government and public officials. Beginning in the 1960s, an increasing proportion of Americans expressed a general distrust of government officials and a cynicism about politics and government. By 2000, almost two-thirds of Americans felt distant and disconnected from government. There are various possible reasons for this growing discontent with government and political processes including people’s frustration with the influence of moneyed interests in American politics and latent distrust of political leaders that stems from events such as the Vietnam War and Watergate. Despite these feelings of discontent and distrust, most Americans still believe that no other country provides the opportunities found in the United States.

Our core political values shape our attitudes. Political socialization is the process by which we learn our political values. There are several socializing agents that influence our political values including family, school, peers, mass media, and social groups. Children, for example, often adopt the party identification of their parents, especially if both parents strongly identify with the same party. In addition, primary and secondary schools familiarize young people with authority figures outside the family and teach respect for democratic processes. Personal and group attributes such as gender, race, and religion may also influence a person’s political values.

The process of political socialization is unique to each individual. However, some common attributes may be identified among people with similar social backgrounds. Socioeconomic status (SES) is a composite measure of an individual’s income, education, and occupation. While there are exceptions to the rule, individuals with higher incomes and education levels and more-professional occupations tend to be more politically conservative than those with lower incomes and education and with less-professional jobs. Individuals in higher SES brackets also tend to vote more than those with lower SES measures.

Public officials, political candidates, and media outlets have increasingly come to depend on public opinion polls to gauge public opinion. The selection of interviewees by method of random probability sampling, in which every person in the population theoretically has an equal chance of being selected, is essential to a good poll. But even with random probability sampling the best polls contain a margin of error, which means the results can vary by a defined probability depending on the size of the sample. For example, national samples of 1,500 respondents will usually be accurate within three percentage points, plus or minus.

Poorly designed polls provide inaccurate or misleading findings. In addition to inadequate sample size or lack of randomness in selecting interviewees, leading or biased questions can undermine the accuracy of poll results. For example, pollsters would likely get different aggregate responses to a question about “intrusive government programs” versus “necessary government programs.”

Polling technology has come a long way since the 1930s. But there are limits to how much polls can tell us. It is one thing to understand how a majority of people responds to a survey question, it is another thing to measure a person’s intensityof feelings. Some people may state an opinion or preference on an issue even though they don’t care deeply about it either way. Others may feel strongly about the issue, but the multiple choices offered don’t accurately reflect how deeply they care. Another factor to consider is issue salience, or how important the issue is to the public. Some issues that may be very important to most people may be addressed in the same poll with issues that most people care or know little about. Also, most polls only provide a snapshot picture of what the public feels on an issue at that time. Public opinion, however, is not always stable; as events change, so do attitudes.

The fact that public officials and political candidates use polls does not mean that their decision-making is guided solely by poll results. Instead, public officials use polls as one form of feedback on public opinion. They also try to gauge the intensity and saliency of the public’s policy preferences by talking to constituents, reading their mail, listening to interest groups, interpreting election results, and relying on their own political instincts and values. If they perceive that the public does not feel strongly about an issue or that opinions are unstable, they may try to shape public opinion through their leadership on an issue.

Using the Video: Pre-Viewing Activity and Discussion

Pre-Viewing Activity and Discussion (30 minutes)

Before viewing the video, discuss the following questions:

  • What is public opinion?
  • What role should public opinion play in a democracy?
  • What role does it play in democracy?
  • What are the consequences of relying heavily on polls to understand public opinion?

Using the Video: Watch the Video and Discuss

Watch the Video (30 minutes) and Discuss (30 minutes)

The video includes three segments:

1. A Change of Heart About Federalizing Airport Security

Shortly after the terrible events of September 11, 2001, the nation’s air traffic system came to a halt. But even after the airports reopened, millions of Americans remained afraid to fly as they believed the existing methods of screening for weapons and other security risks were inadequate. Congress immediately introduced legislation to improve airport security, and soon hearings in both the House and Senate were underway. Despite President Bush’s strong backing, the legislation stalled. The problem was that although everyone agreed on the need for more airport screeners and better equipment, there were deep disagreements over whether the airport screeners should be federal employees. Democrats wanted security personnel working directly for the government, where recruiting, pay, benefits, training, and supervision could be standardized and controlled. Republicans, on the other hand, felt that the federal government should set broad standards, but let private contractors actually employ and supervise the screeners. Clearly, the Republicans gave in to the Democrats’ demands when they realized that their position conflicted with public sentiment on the issue. On November 16, President Bush signed the bill.

Discussion Questions

  • Why did the Republicans eventually give in on the airport security bill?
  • How did political leaders know what the public wanted?
  • Is this example atypical of the power of public opinion?

2. The Voice of the People. Really?

Every four years, the presidential election brings with it a barrage of public opinion polls designed to gauge who’s up and who’s down, and which issues are in and which are out. The better polls use carefully crafted questions that are arranged in a precise order and a sample that accurately reflects the make-up of the larger population. But many Americans believe that a poll with more respondents will always be better than one with fewer respondents and few are sensitive to the precise wording necessary to conducting a good poll.

Discussion Questions


  • What was wrong with Mr. Perot’s poll?
  • What would he have needed to do to create a better poll?
  • What kinds of things should we, as consumers of polls, know about polling in order to evaluate their accuracy?

3. The Case of Vermont and Its Civil Unions Legislation

In the spring of 2000, the Vermont legislature passed a controversial bill that gave gay and lesbian couples the same legal protections as those afforded to married couples in the state. While Vermont was the first state in the U.S. to pass such legislation, the action was not without opposition. Several public opinion polls run prior to the vote showed that a majority of those surveyed did not favor passage of the legislation. In addition to polls showing that more Vermonters opposed than supported the bill, huge rallies and a letter-writing campaign organized by opponents showed that the intensity of feelings on the issue ran high. Vermont’s lawmakers clearly knew they were treading on dangerous political waters regarding this issue.

Discussion Questions


  • How did Vermont’s citizens express their views on civil unions?
  • Polls can show what position people may take on issues, but it does not do a good job of showing the intensity of those feelings. Given that, should they be ignored in favor of participation in rallies or letter-writing campaigns?
  • Given public opinion on this issue, did Senator MacDonald do the right thing?

Using the Video: Post-Viewing Activity and Discussion

Post-Viewing Activity and Discussion (30 minutes)

1. What Every Citizen Should Know About Polling But Is Afraid To Ask (20 minutes)

Scientific methods of public opinion polling were first developed in the 1930s. Before that, researchers and politicians attempted to gauge public opinion through things like chance encounters with citizens, straw polls (simply polling as many people as you can), and exit polls (questioning people as they leave the polling place). These methods were unscientific because no attempt was made to gather a representative sample of the whole polling population.
Today, professional pollsters have high confidence in their polls because they understand the science of polling and the art of interpreting poll results. However, not all polls are created equal. As the example of Ross Perot’s poll in TV Guide illustrates, some polls are haphazardly created and administered and can actually distort public opinion. The aim of this learning resource is to provide basic information about polling that can help all citizens become more knowledgeable consumers of polls. Can you think of other criteria that should be added to the list?

There are several things to look for in evaluating a given poll and its results:

  • Is the poll based on a random sample of the entire population? A truly random sample requires that everyone in the entire target population (usually all adults) have an equal chance of being interviewed. Typically, a “good” random sample size is somewhere between 1,000 and 1,500 persons who are selected in a “stratified” or multi-stage process where progressively smaller geographical units are randomly selected as sample areas. Most polls will describe their sample methodology in the fine print of their results. Those where the pollsters seek out respondents are likely to be scientific, while those that allow respondents to select themselves are usually not. Examples of unscientific polls include call-in polls, Internet polls, and mall surveys.
  • Are the questions understandable? The questions must be framed in basic language that people can understand. This means that pollsters avoid terms and jargon that require further explanation. If some explanation is needed, the pollsters should present it in a basic and balanced way. In addition, questions about obscure details of policy or politics must be avoided since respondents tend to give answers to questions on issues they know little or nothing about.
  • Are the questions asked fairly? Professional pollsters are always on the lookout to avoid question wording that can bias respondents’ answers. For example, good pollsters always eliminate leading adjectives, such as unpatriotic anti-war demonstrators, ungodly pro-choice activists, or the staunch Republican candidate.
  • Are the response categories offered sufficiently broad to capture people’s range of opinions? As the example of the Perot poll shows, if you offer respondents a simple either/or choice, you will get different (and usually much less accurate) results than if you offer several choice options. For example, when the Perot poll asked respondents whether or not they favored one deficit reduction proposal (For every dollar in tax increases there should be two dollars of spending cuts, with the savings earmarked to pay down the debt), 97 percent of respondents said yes. Yet when respondents were offered several debt-reduction options, support for the original Perot option dropped significantly.
  • Is the sampling error reported clearly? All polls have potential sampling errors, and the better ones will report their potential error rate up front. A 3 to 5 percent sampling error rate is considered reasonable, while polls with higher margins of error should be interpreted with caution. What does sampling error mean? Consider this: A poll with a 3 percent error rate means that it is 97 percent accurate. In other words, in a poll that registers high presidential public approval at 50 percent, we can be 97 percent sure that the actual rate of public approval is between 47 percent (50 minus 3 percent) and 53 percent (50 plus 3 percent). Obviously, if results of a poll are within 3 percentage points of each other, say one policy proposal being supported by 49 percent of those polled and opposed by 51 percent, then the results are considered “within the margin of error” and no reliable difference can be asserted.

2. What Are the Sources of Your Political Values? (10 minutes)

While the processes of political socialization are complex, we know that an individual’s political values and attitudes are shaped by a variety of factors including their parents, teachers, and peers, and their response to formative events such as war or political scandals. Think about your formative periods of political socialization, including the potential influence of parents, relatives, peers, teachers, and actual events on your political beliefs. To get things started, you might recall the context of your first “political awareness,” such as when your parents were discussing (or arguing about!) who they intended to vote for in a presidential election, when they first tuned into a military conflict on television, or their feelings about a major political event (e.g., presidential speech or action). What had the greatest influence on your current beliefs and opinions?

Using the Video: Homework

Read the following Readings from Unit 12 to prepare for next week’s session.

  • Introduction-Political Parties: Mobilizing Agents
  • Tocqueville, Democracy in America: “Parties in the United States”
  • Washington, “Letter to Thomas Jefferson”
  • Roosevelt, “Bull Moose Speech”
  • Piroth, “Selecting Presidential Nominees: The Evolution of the Current System and Prospects for Reform”

Read next week’s Topic Overview.

Using the Video: Classroom Applications

You may want to have your students do the post-viewing activities: What Every Citizen Should Know About Polling But Is Afraid To Ask and What Are the Sources of Your Political Values? They are provided for you as blackline masters in the Appendix of the print guide.


The Readings for Democracy in America unit 11 are available here for download as a PDF file. You’ll need a copy of Adobe Acrobat Reader to read the files. Acrobat Reader is available free for download from

Download Unit 11 Readings, Public Opinion: Voice of the People

  • Introduction—Public Opinion: Voice of the People
  • Tocqueville, Democracy in America: “Political Associations in the United States”
  • Paine, “Common Sense”
  • Federalist Papers: “Federalist No. 10”
  • Hahn, “Student Views of Democracy: The Good News and Bad News”


  1. What did Tocqueville suggest was the constitutive element of liberty?
  2. How did Paine distinguish between society and government?
  3. Within whom does the freedom of the state reside in a republic, according to Paine? In what way is it determined?
  4. How did Tocqueville explain the prominence of social organizations within the United States?

Introduction—Public Opinion: Voice of the People

President George Bush derisively dismissed President William Clinton’s administration as governing by polls. While there is no evidence that the Bush administration was any less interested in the polls than previous administrations were, this claim does reveal a public perception of the use and abuse of polls. While everyone would agree that government should do what citizens want it to do, if administrations follow the dictates of the polls too slavishly they appear to lack leadership. Tocqueville explained that above the government’s institutions, “and beyond all these characteristic forms, there is a sovereign power, that of the people, which may destroy or modify them at its pleasure.” The many ways that the sovereign people influenced the government was a central concern for Tocqueville. “It remains to be shown in what manner this power, superior to the laws, acts; what are its instincts and its passions, what the secret springs that retard, accelerate, or direct its irresistible course, what the effects of its unbounded authority, and what the destiny that is reserved for it” (179). Here at the beginning of the twenty-first century, the power and presence of public opinion polls challenges constitutional government and, in connection with the growth of mass culture, threatens individualism and difference.

The Constitution creates and limits the institutions of government. The reliance by the government on opinion polls for support and legitimacy undermines the role of the Constitution in creating governmental legitimacy by giving the branches the power to do whatever the people will allow. For example, in the twentieth century, the executive branch has become the branch of government most involved in war-making. This flies in the face of the constitutional grants of war-making power, most explicitly the power of Congress to declare war. This function is now routinely performed by the presidency. This fundamental constitutional change occurred without a change in the actual document.

Tocqueville noticed that in America there was a tendency for people to look to mass culture for opinion. The pressures of equality, Tocqueville believed, would make authority less appealing to Americans to such a degree that that they would be less willing to take direction from local authorities. He believed, similarly to Madison’s account in “Federalist No. 10,” that local communities and differences would become less important and less valued as the attention of American citizens was commanded by the national public. This identification with the national government, Tocqueville and Madison believed, would reduce the power and saliency of local identifications and reduce the political activity of American citizens. The production of national polls can emphasize this tendency to look to an abstract national identification for guidance and authority.

The readings for this chapter cover several aspects of this problem. The central challenge to democracy foreseen by Tocqueville and institutionalized by Madison constitutes the core reading. Caroline Hahn’s article on the surveys of school children reveals some interesting opinions and beliefs of school children in an example of an interesting opinion survey. These readings contribute to any attempt to understand the destiny of the people.

Web-based Resources   The National Election Study, conducted by the University of Michigan, offers polls past and present for analysis, data sets for analyzing poll results, and sets of poll questions that can be used to create your own polls. Gallup Organization offers an online program for students and special polling features organized by contemporary issue areas.

http://www.ropercenter.uconn.eduThe Roper Center for Public Opinion Research offers archival data on past polls and poll results, reports on contemporary public opinion, and a “Public Opinion Matters” link to data and reports organized in broad issue areas.

http://www.people-press.orgThe Pew Research Center for the People & the Press provides study features on contemporary topics of domestic and international interest. The site also offers quick links to other recent poll results compiled by dozens of polling organizations.