CHAPTER SIx • PublIC OPInIOn, POlITICAl SOCIAlIzATIOn, And THE MEdIA 127
Sampling Error
The difference between
a sample result and the
true result if the entire
population had been
interviewed.
Tracking Poll
A poll that is taken
continuously—sometimes
every day—to determine
how support for an issue
or candidate changes
over time.the best possible practices, the odds are better than 50-50 that the true answer is not 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 respon-
dents—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 a 5 percent chance 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 a 5 percent chance—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 rea-
son 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. Suppose 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 so much “statistical noise.”The Difficulty of Obtaining Accurate Results
Reputable polling organizations devote substantial 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 popula-
tion is difficult. 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.
Many poll takers now rely on automated scripts—“robocalls”—rather than live inter-
viewers. Under the law, however, robocallers are not allowed to dial cell phones. As a
result, automated surveys are less accurate than traditional ones. Pollsters currently have
no way at all to reach people who rely on Skype or similar systems for telephone service.
Finally, more than 85 percent of households contacted by typical polls refuse to be inter-
viewed. Despite their best efforts, pollsters may be unable to gather a pool of 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 sam-
ple 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 evangeli-
cal Christians. Based on other sources of information, 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/ethnicity, religion, and geography. It is much harder—and
more dangerous—to adjust for political ideology, partisan preference, or likelihood of vot-
ing. The formulas that firms use to weight their responses are typically trade secrets and
are never disclosed to the public.Copyright 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.