Examining the pull of the poll The practice of polling proliferates, even as politicos and the public deny the surveys matter

Ideas: Politics.

October 11, 1998|By Carl Schoettler | Carl Schoettler,Sun Staff

In the midst of White House scandal and fall elections, an impromptu poll of pollsters and political scientists on polling produces the usual mixed results, with at least a 4 percent margin of error.

Despite doubts, dismissals and denials, polls of every sort continue to proliferate like rabbits on Viagra and spout out information as fast as a well-oiled Gatling gun.

Politicians from President Clinton to the local dogcatcher seem to take polls daily, if not hourly.

Indeed, Clinton uses polls more than any other president has, says Iva Deutcher, a political scientist at Hobart and William Smith Colleges, in Geneva, N.Y.

"If there are not 10,000 polls done this month, I'll be surprised," says Guido Stempel, an Ohio State University professor who polls for the Scripps Howard News Service.

"Easily," agrees Ted Arrington, chairman of the department of political science at the University of North Carolina. He notes every member of the House of Representatives is up for election in November, along with senators, governors and a wide assortment of state and local officials.

They all take polls, though many of them deny it.

"Most politicians believe if they follow the polls, it shows they are feeble-minded," says Tom W. Smith, director of the General Social Survey at the University of Chicago's National Opinion Research Center. "A lackey, not a leader."

He recalls a 19th-century British politician who, when asked if he led or followed public opinion, said: "I met it along the way."

Deutcher, the professor from Hobart, says she doesn't believe politicians who say they don't believe in polls.

"When a poll shows a candidate losing by 20 percent and he says, 'I pay no attention to polls,' you know that's a lie," she says. "Their staff is biting their fingers to the bone."

Richard E. O'Connor, associate professor of political science at Pennsylvania State University, says that when people reject the findings of polls, "It often means they don't like the results.

"For instance," he says, "Clinton-haters would gainsay polls reflecting strong public support for Clinton."

As a matter of fact, pollsters point to the consistency of polls showing approval ratings for Clinton in the 55 percent to 65 percent range as a measure of the validity of even "quick and dirty" overnight polls.

"A number of polls in the last six weeks have been designed very quickly and carried out very quickly," says Tom Smith, the University of Chicago pollster. "If you want to know [what the public thinks] three hours after Clinton makes a national confession, you're going to have to sacrifice some quality."

Nevertheless, Smith says, the very consistency of polls showing support for the president shows the quickness of the turnarounds "hasn't led to any highly aberrant results."

Lee Miringoff, a professor who directs the Institute for Public Opinion at Marist College, in Poughkeepsie, N.Y., says he can start polling at 6 p.m. and have the results ready for the 11 o'clock news.

4 "We're shaping coverage very strongly," he says.

But Deutcher, the Hobart professor, says, "This is the first time I have seen the public as resistant as they are to the elite media."

Only Thursday, a poll taken for the media think-tank Freedom Forum showed 47 percent of Americans believe media coverage is unfair to Clinton.

The media usually act as an agenda-setter, the political scientists say. But not this time. The message, Deutcher says, is, "We're not interested. We're turning to the crossword puzzle."

Still, the Freedom Forum survey found 81 percent of their respondents follow the story closely.

Miringoff says that when the Monica Lewinsky scandal hit, pundits felt "the guy would be ridden out of town in a matter of minutes."

Then the polls came in and showed the public splitting 50-50 on Clinton.

"His ability to come off the floor was confirmed by the polls," says Miringoff.

If the public is skeptical of the pundits, they are sometimes equally skeptical of polls. The idea that polling 1,500 people can reveal the opinions of the nation goes against the American grain, Miringoff says.

"Millions of dollars are spent to tell people you're an individual, you're unique. And pollsters tell you, 'You're just part of an interest group.' "

But you don't have to taste the whole pot to know the soup needs salt, he says. "And if the doctor takes a blood sample, you only want him to take a little."

Present-day polling is ruled by the mathematics of probability and the law of large numbers. Arrington, from North Carolina, uses the example of flipping a coin.

If you flip a penny five times, you might get four heads and one tail. If you flip it 1,500 times, you're pretty sure to get 750 heads and 750 tails.

"We know how to do polls that are remarkably accurate," says Stempel, the Ohio State pollster. "If we interview 1,000 people, we get an answer within 4 percent of what we'd get if we interview everybody in the country."

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.