THE VIRGINIAN-PILOT
Copyright (c) 1996, Landmark Communications, Inc.
DATE: Sunday, September 15, 1996 TAG: 9609130708
SECTION: COMMENTARY PAGE: J1 EDITION: FINAL
SOURCE: BY TONY WHARTON, STAFF WRITER
LENGTH: 89 lines
The American public opinion poll as we know it is 60 years old this fall, and it is thriving.
About 20 million poll interviews are conducted every year, on every imaginable topic, from sex to eating habits to politics. You stand a good chance of being interviewed for a poll several times in your life.
You stand an even better chance of hearing the results of numerous polls during a political campaign. What does it all mean? How can you use it?
``Some of this is good information, and some of this is bad information,'' said Michael W. Traugott, a political science professor at the University of Michigan in Ann Arbor and co-author of a primer on the subject, ``The Voter's Guide to Election Polls.''
Polls have been criticized as superficial snapshots of an electorate that may not have thought through the issues before they're asked a question by a pollster. Experiments have shown that people are willing to give opinions on topics which are actually fictitious, invented for the occasion.
James S. Fishkin, a professor at the University of Texas in Austin, said these quick, insubstantial opinions can have a lasting influence.
``When the fleeting and volatile nature of many polls is combined with the mass media, nonexistent opinions may take on a life of their own,'' he wrote in ``The Voice of the People.'' ``Poll results are reported in the media and are bounced back by the public in further polls, regardless of whether the initial results had any substantial thought behind them.''
Traugott, who conducted polls himself for the Detroit News, thinks they can be valuable.
``The great service of the polls is that they give the electorate an opportunity for independent voice, a chance to react to the candidates' issues and provide information to other citizens,'' Traugott said.
Polls seem a modern invention, firmly intertwined with the telephone in American society. But they are much older and seem to reflect a need among citizens to find out what the rest of the country thinks.
As far back as the 1820s, and with increasing popularity in the latter 1800s, newspapers reported ``straw polls,'' completely unscientific counts of any sort of gathering - on trains and steamers or in factories and drugstores. They were often wrong, but it did not diminish their popularity.
The straw poll peaked with the Literary Digest, which used mail-in ballots to accurately predict the next president from 1920 to 1932. But in 1936, an unknown named George Gallup rose to challenge the Digest, claiming that his scientific survey of a few thousand people would be more accurate than the Digest's millions of ballots.
The Digest's ballots picked Alf Landon; Gallup said Franklin D. Roosevelt would win. The fact that you've likely never heard of either Alf Landon or the Literary Digest, which closed the following year, shows that Gallup's risk paid off.
He had successfully introduced the notion that you can find out from a thousand people what one million think, or 10 million, or 100 million. The size of the population is irrelevant. However, this is not widely understood even today.
After 1936, ``straw polls'' nearly disappeared and the rise of the professional pollster began.
A few straw polls survive. In 1968, General Cinema Theatres began setting out drink straws with each of the parties on them, and has a perfect record of picking the winner in presidential elections (although liberal and independent candidates tend to do better in the ``straw vote'' than in the election).
General Cinema even repeats, perhaps unknowingly, the Literary Digest's old claims: ``The 1996 Strawvote poll . . . will survey over 3 million moviegoers. Professional polling services typically only survey approximately one thousand people.''
This arises from the misunderstanding Gallup demolished in 1936: For a good poll, there is no relationship between the number of people polled and the size of the larger population.
Straws notwithstanding, today most polling comes from major news organizations, many smaller ones, and the professional pollsters.
``Polls give people a sense of connection to the rest of the electorate,'' Traugott said. ``Although it's interesting that as they get closer to Election Day, people prefer to make the final choice on their own unencumbered by other information.
``Americans think of this decision as a private one, made with the sanctity of the secret ballot.'' ILLUSTRATION: Graphic
HOW THEY STAND
SOURCE: THe poll of 1,027 Americans from Sept. 4-8 was conducted by
Chilton Research Services of Radnor, Pa. for ABC News Nightline.
The margin of error was +/- 3.5 percentage points for the entire
sample
[For complete graphic, please see microfilm]
KEYWORDS: PUBLIC JOURNALISM POLLS by CNB