Don't Worry, We've Got Your Number
Washington Post Staff Writer
Sunday, January 10, 1999; Page C4
Though there have been stacks of polls on the Clinton scandal these past months, no media pollster has ever called you, right? At least, no pollster has called you yet.
I know the feeling. I haven't gotten a call, either. But every year, about 20 million Americans participate in opinion surveys, one-fourth of these in surveys sponsored by the federal government (and this, of course, does not count that ultimate survey, the decennial census). In 1998, interviewers working on behalf of The Washington Post talked to about 25,000 people.
So why haven't you and your opinions on impeachment and infidelity made the cut? The simplest answer is that the people who conduct each poll need only contact a very small proportion of Americans to represent the opinions of the whole country. A typical Post survey includes about 1,000 people.
Can the opinions of just 1,000 accurately reflect those of millions of Americans? Yes – according to statistical and probability theory – if the right methods are used to choose the sample of people. Stripped of its math-o-magic, sampling the population is like testing the temperature of a pot of soup – you don't need to eat the whole bowl, just stir it up and slurp a spoonful or two . . . or 1,000, if you're sampling national attitudes.
One key element of probability sampling is that poll respondents must be chosen randomly. Pollsters can't just survey their 1,000 closest friends or the people logged in to their statistics chat room. Many accomplish this by using computers to generate random phone numbers, which ensures that people with unlisted numbers are included. Each adult in the population should also have an equal chance of being included in the poll.
The resulting sample should closely mirror the diversity of the country, including people from all parts of the nation, of varying ages, racial backgrounds and party affiliations.
Polls aren't always a perfect measure of national opinion. There is some error associated with sampling, as well as some caused by such factors as the occasional poorly worded question or large numbers of respondents who refuse to answer. Some of this error is measurable, some is not. Polling is part science, part art.
Those caveats aside, well-designed polls work. You don't have to take my word for it: Every four years, pollsters predict support for presidential candidates, and then an election comes along and proves them right or wrong. With a few memorable exceptions (President Thomas Dewey, for example, who existed only in the minds of pollsters), most polls are close to the mark. The National Council on Public Polls reported that the average error of the major polls conducted in the days immediately prior to the 1996 election was 1.7 percent. Not bad.
If you are at least 18 years old, reside in the lower 48 states and live in a home with a phone, there's a chance you could be contacted for the next Washington Post poll. So, if you want to be polled, keep the phone on the hook and don't hang up on strangers. We might need to reach you.
Claudia Deane is assistant director of polling for The Washington Post.
© Copyright 1999 The Washington Post Company