• Poll
• Polling
• Surveys

Opinion poll

Margins of error

All polls have a margin of error, which is a function of the number of people polled. The margin of error reflects the effects of chance in the sampling process, but does not reflect other sources of error, such as measurement error or errors in data processing. A poll with a random sample of 500 people has margin of sampling error of 4.5% for the estimated percentage of the whole population. A 4.5% margin of error means that 95% of the time the procedure used would give an estimate within 4.5% of the percentage to be estimated.

Wording of questions

It is well established that the wording of the questions, and the order in which they are asked, can influence results of polls. Thus comparisons between polls often boil down to the wording of the question. One way to minimize this effect is to ask the same set of questions over time, in order to track changes in opinion.

Nonrepresentative samples

Another source of error is the deliberate or accidental use of nonrepresentative samples. For example, when home telephones were rare, telephone sampling had a built-in error because most of those surveyed were well-to-do.

Polling organizations

There are many polling organizations. The most famous is the Gallup Poll, created by George Gallup.

Other major polling organizations in the United States are

Quinnipiac Polls, run by the University of Connecticut, and started as a student project.

polls conducted by the Pew Charitable Trusts, concentrating on media and political beliefs.

The Harris Poll

Nielsen Ratings, virtually always for television.

The best-known failure of opinion polling to date in the United States is the polls that said Thomas Dewey would defeat Harry S. Truman in the 1948 Presidential election.

See also Exit poll, confidence interval.
Source | Copyright