Poll Crazy in Campaign Coverage, 10/20/04
http://www.pbs.org/newshour/extra/features/july-dec04/polls_10-20.html

Related Lesson Plan
http://www.pbs.org/newshour/extra/teachers/lessonplans/media/polls_10-20.html

While the media often emphasize the latest polls to determine who's winning a presidential campaign, experts say the public should focus on the issues and character of the candidates, rather than the "horse race."

"It's a draw!" No, wait -- "the Democrat won by 13-points!" That may sound like an unusual sportscast, but it actually refers to the news coverage following the last presidential debate between Republican President George Bush and Democratic challenger Senator John Kerry.

Just 25 minutes after the debate on October 13, ABC News declared the face-off a "draw," according to its poll of 566 "registered voters." Forty-two percent said Sen. Kerry won the debate, while 41 percent said President Bush did a better job, according to the poll. The poll had a "margin of error" of 4.5 percent, meaning the results could have been 45.5 for the president and 37 percent for Sen. Kerry or vice versa, a notable difference.

The news media's love of polls

Sometimes polls -- and the competition between candidates -- can dominate news coverage entirely, at the expense of more informative stories that could actually help voters decide which candidate they prefer.

For instance, during the 2004 Democratic primary, the largest group of news stories covered the competition (68 percent), followed by character (18 percent) and lastly policy (14 percent) of each candidate, according to a recent study by the University of Missouri's Communication Department.

Most people assume that the news informs them about who will make a better president, but what the news is best at is informing them about who is the most efficient campaigner, according to William Benoit, the professor who led the study.

Favoring polls over substance

Given these apparent contradictions in polling data and their lack of substance, why do news organizations cite polls on a daily basis?

Media experts say covering elections like a competitive sporting event, like a horse race, creates a more exciting and compelling story for the audience.

"Polling is one way to create a veneer of excitement about the race," said Jay Rosen, a journalism professor at New York University.

Issue-based political coverage, such as the policy positions of each candidate, can seem boring, according to Stephen Hess of the Brookings Institution.

"Substance stories are about issues and candidates' qualifications for office. Horse race stories are more fun. Substance stories are more important," Hess wrote.

Journalists also want to report new information. "A poll just out is new; Sen. Kerry's or Mr. Bush's stand on an issue is not new...I think a reason the media over-emphasize polls is that it allows a story to tell something new," Benoit said.

This trend is nothing new. Polls have been widely used by the media since the 1950s, according to an analysis by the University of Missouri's Communications Department.

"There's a premium on information about the election that seems objective, scientific, nonpartisan, not just 'spin' -- facts that are free from all those tricky problems of human judgment. Polling seems like one of those things -- here are the numbers," Rosen said.

The problems with surveys in news coverage

But, even if polls provide "spin-free" facts during a presidential election, experts strongly caution against taking poll results at face value because the method of collecting data for the survey may be flawed.

"There is no standard methodology in the polling profession for identifying likely voters...Each [polling] firm uses a calculus of its own...This is part of the art of polling, rather than the science," Michael W. Traugott, a communications professor at the University of Michigan, recently told the Boston Globe.

Even the most reputable polling firms, such as Gallup and Zogby International, have their own guidelines for selecting a group of people, or sample of the population, to survey. Even polls on the same subject can therefore have very different results.

Substantive news or just numbers?

Most importantly, experts say the media often report polling data in a misleading way.

"Too much emphasis is on what the latest poll numbers show. Not enough emphasis is on what do these polls mean... very often the typical news report is just a recitation of a top line...and that confuses more than it informs," Andrew Kohut, president of the Pew Research Center, told the NewsHour.

News reports often do not mention the "margin of error" or provide specifics about the group of people surveyed, said Benoit. Details about the sample -- such as "likely voter" and "registered voter," or "Republican" and "Democrat" -- are important because they can dramatically affect the results of a poll.

Despite these flaws, polling is not likely to disappear from the news. Instead, the public can learn to interpret polls intelligently on their own.

Benoit advises: Know the polling sample, keep in mind the margin of error of the poll and the date of the poll since people's opinions can change after the poll. Finally, ask yourself if it is a reputatable and unbiased polling organization, or a political group, that is conducting the poll.

In other words, always read the fine print.

-- By Liz Harper, Online NewsHour

© 2004 MacNeil/Lehrer Productions