- The Washington Times - Thursday, October 12, 2000

Presidential pollsters yesterday defended their craft in an election season of wildly fluctuating polls that have seen Vice President Al Gore's fortunes rise with a kiss and plummet with impatient sighs.

The panel of polling analysts told an audience at the National Press Club that shifting presidential polls accurately reflect voters' lack of enthusiasm for Mr. Gore and Republican presidential nominee George W. Bush.

"Public opinion is unsettled and therefore the polls are less settled," said Andrew Kohut, director of the Pew Research Center. "There's less voter conviction about these candidates."

Mr. Kohut and Humprey Taylor, chairman of the Harris Poll, also refuted the frequent complaint by conservatives that political polls reflect a liberal bias and fail to survey a representative sampling of conservatives. They point to the 1996 presidential election, in which Republican Bob Dole finished closer to President Clinton than some polls predicted.

"Most of us don't have an ideological stream of money," Mr. Kohut said. "There is no evidence of a clear liberal … bias in these national polls."

Mr. Taylor said independent pollsters would "put ourselves out of business" if they favored one party over another in their surveys.

Political polls are more numerous than ever this year, and collectively the snapshots they present can be baffling. Some surveys had Mr. Bush with a double-digit lead during the Republican National Convention in early August, only to show him falling behind Mr. Gore by a similarly large margin three weeks later, after the vice president planted a passionate kiss on his wife at the Democratic National Convention.

Likewise, most surveys had Mr. Bush trailing Mr. Gore by a few percentage points as they headed into the first presidential debate on Oct. 3. Overnight polls then showed that most viewers believed Mr. Gore had won that debate. But only three days later, most national polls showed Mr. Bush recapturing the lead by as much as six percentage points as Mr. Gore received negative media coverage for his gaffes and annoying sighs in the Boston debate.

"There appears to be less than genuine enthusiasm for either one of the candidates," said Marvin Kalb, executive director of the Shorenstein Center on the Press, Politics and Public Opinion at Harvard University. "They're simply not blown away by either one of the two candidates."

Combined with those fluctuations, studies are showing this year that media outlets are focusing on poll results more than in past presidential elections.

"Increasingly, it seems polls are something we can't live with and can't live without," said Robert Lichter, president of the Statistical Assessment Service in Washington. "There's more controversy over them than ever before."

Further, Mr. Lichter said, an analysis by his Center for Media and Public Affairs shows that reporting about the "horse race" is more favorable to Mr. Gore than to Mr. Bush. He said a study of reports on the three major television news networks has found that 80 percent of their coverage of polls is favorable to Mr. Gore.

"It's possible that voters are learning that Gore is more likely to win," Mr. Lichter said.

Mr. Taylor defended the accuracy of the surveys, saying presidential polls since 1948 have missed the actual outcome of the election by an average of 1.9 percentage points for the major party candidates. Mr. Taylor also said that publishing presidential polls before an election has not had a "bandwagon" effect voters do not, he said, switch in the final days to the candidate who is leading in the polls.

"In a two-horse race, there is no evidence the polls have an influence," Mr. Taylor said.

Mr. Kohut said primary polling and state-by-state surveys are less reliable. For example, he said, a one-night Zogby survey of South Carolina voters immediately after Arizona Sen. John McCain won the New Hampshire primary showed Mr. McCain leading Mr. Bush in South Carolina by six percentage points. The poll results were reported in 42 newspapers, he said, including 14 front-page stories.

"It created the notion … that John McCain all of a sudden was a viable candidate," Mr. Kohut said.

Mr. Bush won the state's primary less than three weeks later by 11 points. Mr. Kohut compared polls taken immediately after a political convention or a primary victory to an athlete taking his or her blood pressure immediately after running four miles.

"They confuse the heck out of people," he said.

Mr. Kalb, a former network television reporter, said there is a "direct relationship" between media coverage of a candidate and his or her fortunes in the polls, pointing to Mr. Gore's drop in popularity after he was caught embellishing stories in the first debate. But Mr. Kohut said Mr. Gore's lost lead can be attributed more to voters talking amongst themselves about the debate in the days afterward.

"I don't think you can clearly make the case that the media puts out its spin and the public follows," Mr. Kohut said.

The discussion was sponsored by the Statistical Assessment Service (STATS), a division of the nonprofit Center for Media and Public Affairs. STATS Director David Murray said the public should treat polls the same way President Reagan dealt with the Russians on arms control: "Trust, but verify."

Sign up for Daily Newsletters

Manage Newsletters

Copyright © 2020 The Washington Times, LLC. Click here for reprint permission.

Please read our comment policy before commenting.


Click to Read More and View Comments

Click to Hide