William_T._Bianco,_David_T._Canon]_American_Polit

Measuring public opinion 215

although polls taken early in a presidential campaign (such as a year in advance of the

election) are poor predictors of the ultimate outcome on Election Day, polls taken at

the beginning of the official presidential race, when both parties’ nominees are known,

can provide very good predictions of who will win the election and how many votes he

or she will receive.^36

An alternate technique for measuring public opinion uses focus groups, which are

small groups of people interviewed in a group setting. Because focus groups allow

respondents to answer questions in their own words rather than being restricted to a

few options in a survey question, they can provide deep insights into why people hold

the opinions they do. Candidates sometimes use focus groups to test campaign appeals

or fine-tune their messages. However, because of their small size, focus groups cannot

be used to draw conclusions about public opinion across the entire country.

Large-scale surveys such as the American National Election Study (ANES), which

is conducted every election year (typically both before and after the general election),

use various types of questions to measure citizens’ opinions. In presidential election

years, participants in the ANES are first asked whether they voted for president. If they

say they did, they are asked which candidate they voted for: a major-party candidate

(for example, Hillary Clinton or Donald Trump in 2016), an independent candidate, or

some other candidate.

Another kind of survey question measures people’s preferences using an issue scale.

For a range of topics, two opposing statements are given and respondents are asked to

45 state- and national-
level polls of the presidential race
were released in the last two weeks
of the 2016 campaign.
Source: Pollster.com

DID YOU KNOW?

On average, people should be

more skeptical when they see

numbers. They should be more

willing to play around with the

data themselves.

—Nate Silver, FiveThirtyEight

.com

pick the statement, or one of the choices between the two extremes, that comes closest

to their views. As we discuss later, on questions such as these, most Americans pick

positions in the middle of these scales.

A typical survey like the GSS or the ANES will ask several hundred questions related

to issues and candidates, along with additional questions that elicit personal information

such as a respondent’s age, education, marital status, and other factors. Some surveys

conducted by media sources, candidates, or political parties are shorter, focusing on

voter evaluations of the candidates and the reasons for these evaluations. In the main,

the length of a survey reflects the fact that people have limited attention spans, so there is

a trade-off between learning more about each respondent’s opinions and getting him or

her to agree to be surveyed in the first place. Shorter surveys are also less expensive.

An important development in recent years is the use of techniques to combine

different surveys into a more powerful measure of public opinion. These techniques

weigh data from different polls based on their accuracy in previous elections and

combine state- and national-level results to generate predictions—thus, if a poll finds

that candidate A is ahead in state 1 and polls have found that the winner in state 1 will

also win state 2, then the results from state 1 will be used to predict the result in state 2.

These techniques generate powerful predictions of election returns—for example,

Nate Silver of FiveThirtyEight.com was able to correctly predict the winner of 49 out of

50 states in the 2008 and 2012 presidential elections and came the closest of all political

analysts in predicting Trump’s upset victory in 2016.

Problems in Measuring Public Opinion

Although Nate Silver came close, virtually all major political pollsters failed to

predict Trump’s 2016 victory. While this result is exceptional—most of the time, polls

are extremely accurate predictors of election outcomes—it highlights the difficulty

of measuring public opinion. The problems begin with determining and gathering an

appropriate, representative sample of subjects. These difficulties are compounded

by issues pertaining to the wording of the questions and further complicated by

the very nature of public opinion itself. Consequently, survey results must be read

carefully, taking into account who is being surveyed, when they are being surveyed,

what opinions people are being asked about, and what mechanism is used to ask

survey questions.

Issues with Survey Methods Building a random sample of individuals is not an

easy task. One tactic is to choose households at random from census data and send

interviewers out for face-to-face meetings. Another is to contact people by telephone

using random digit dialing, which allows surveyors to find people who have unlisted

phone numbers or who only use a cell phone. In theory, each of these techniques

produces a random sample, but in practice, they both may deviate from this ideal.

Face-to-face interviews of people in their homes during work hours won’t include

adults who work during the day and are not home, and is very expensive. Even with

phone surveys, many people refuse to participate. Pollsters must adjust their survey

results to account for these and other potential biases. Sometimes these adjustments

successfully predict results, and sometimes they don’t. For example, in the 2016

presidential race, pollsters did a relatively poor job (compared to recent elections) of

predicting who was likely to vote. Most used a technique that relied on respondents’

past turnout decisions to predict their behavior in 2016. Usually this technique works

very well, but in 2016 it overestimated turnout by African Americans based on their

high turnout rates in 2008 and 2012. The 2008 and 2012 numbers most likely were

90% of people

contacted by phone to participate in

a political survey either don’t answer

or refuse to participate.

Source: Pew Research Center

DID YOU KNOW?

random sample

A subsection of a population chosen

to participate in a survey through

a selection process in which every

member of the population has an

equal chance of being chosen. This

kind of sampling improves the

accuracy of public-opinion data.

Full_07_APT_64431_ch06_198-231.indd 215 16/11/18 1:38 PM