Public Opinion

Overview

• What is public opinion?

• Where does it come from?

• How do we measure it?

• What is the role of public opinion in politics?

What is Public Opinion?

• V.O. Key defined public opinion as reflecting “those opinions held by private persons which governments find it prudent to heed.”

• Another definition: the collective or aggregate opinions of the adult population.

The Origins of Public Opinion

• 

  Emerges in part from the political socialization process in childhood and early adolesence:

  • Family

  • School

  • Religion

  • Community

• Socialization helps form beliefs and values that shape our opinions later in life.

Influences Later in Life

• Opinions are also influenced by adult experiences:

  • Self-interest

    • Employment

    • Property ownership (stakeholding)

    • Parenthood, marriage

  • Higher education

    • Instills tolerance and other political values.

    • Increases political efficacy: the belief that one's actions can affect government policies.

  • Mass media

Tolerance Effects of Education

Measuring Public Opinion

• Before the 1930s, studies of public opinion relied on convenience samples or straw polls.

• 

  1930s: George Gallup developed the scientific approach to survey research still used today:

  • Use of random samples.

  • Efforts to reduce selection bias.

• Another consideration is measurement error: expressed opinions may not reflect underlying attitudes.

Random Sampling and Sampling Error

• In a random sample, each member of the population of interest has an equal chance of being surveyed.

• This rule guarantees that the sampling error—the error in polls due to using a sample rather than looking at the whole population—is as small as possible.

  • Bigger samples have less sampling error:

    • 500 respondents: ± 4.4%

    • 1000 respondents: ± 3.1%

Selection Bias

• If some members of the population are more likely to be surveyed than others, selection bias results.

• Sources of selection bias:

  • Sampling frame is not the population of interest.

    • Telephone polls omit people who don't have phones.

  • Nonresponse bias.

    • Some groups of people are less likely to respond to surveys than others.

  • Self-selection (opt-in), common in web polls

Measurement Error

• 

  Another common problem: survey questions may not accurately gauge interviewees' true attitudes:

  • Many opinions cannot be measured objectively and are hard to quantify.

  • Questions may lead to biased responses due to their wording.

  • Attitudes may be too complex for a single question.

Confusing Question Wording

• 

  In 1993 the American Jewish Committee commissioned a survey on public beliefs about the Holocaust when the U.S. Holocaust Memorial Museum opened in Washington.

• 22% of Americans said it was “possible… the Nazi extermination of the Jews never happened.”

Confusing Question Wording

• Question asked:

  • “Does it seem possible or does it seem impossible to you that the Nazi extermination of the Jews never happened?”

• Question was not clear or simple; it contains a double negative.

Confusing Question Wording

• Alternative question used in a subsequent poll:

  • “Do you doubt that the Holocaust actually happened, or not?”

• In this formulation of the question, 87% of respondents were certain it happened and only 9% said it did not (4% were uncertain).

• The full report, by David W. Moore and Frank Newport, is available in The Public Perspective (March/April 1994).

Value-Laden Wording

Sometimes these differences can emerge based solely on the terms used in the question:

Oversimplified Questions

• Some issues are too complex to be captured by a single survey question.

• Attitudes towards abortion are a classic example:

  • Many Americans are not neatly “pro-choice” or “pro-life” in all circumstances; instead they tend to fall in the middle.

  • Questions about the extremes lead to biased responses that don't reflect issue's complexity.

Abortion: A Complex Issue

Public Opinion and Government

• While many of the sampling and measurement issues can be overcome, there are other obstacles to “governing by public opinion”:

  • Voters are uninformed.

  • Voters lack ideological constraint.

  • Voters' attitudes are unpredictable.

Political Information and Rational Ignorance

• Voters will express opinions about things they know little about.

• A small fraction of the public knows “basic” information about politics like the identity of the Chief Justice of the United States or the Speaker of the House.

• Voters are rationally ignorant: they know little about politics because the benefits do not exceed the costs.

Americans Lack Political Knowledge

Exceptions to Rational Ignorance

• Higher education leads to more knowledge.

• Personal or group interest leads to acquiring knowledge about relevant topics: issue publics.

• Important events and media publicity may increase knowledge for a time.

Ideology and Unconstrained Beliefs

• Political elites (people who are deeply interested and involved in politics) tend to conceptualize the world using ideologies:

  • Liberals

  • Conservatives

• Ideologies help people shape the political universe and connect related ideas together.

• Most people (the mass public) don't think about politics in these terms.

Attitude Inconsistency

• Attitudes of the mass public often reflect logical contradictions:

  • Support for cutting government spending in the abstract, but opposition to cutting spending on specific programs.

  • Support for expanding government programs and cutting taxes at the same time.

  • Support for “free speech” but opposition to the exercise of free speech by unpopular groups.

Opinions and Governing

• Even when the public cares deeply about an issue, government action may still be unlikely:

  • Senators and representatives are responsible to their local constituents, whose opinions may differ from the public at large.

  • Elected officials may be more concerned about their supporters and single-issue voters than all of their constituents.

  • Elected officials may believe that voters have other, more important priorities.

Collective Rationality?

• Despite the shortcomings of the public, government policy does seem to respond to aggregate trends in opinion.

• While individual opinion often appears irrational, collective opinion seems to cancel out much of the noise: the “rational public.”