Shortcomings Of Exit Polls

In future elections, political analysts may put a bit less stock in the inside track data from exit polls.

That's because some of the early numbers that got into circulation Tuesday turned out to be misleading. CBS News Correspondent Richard Schlesinger reports on the erroneous exit polls.

Natalia Busey and about 1,500 exit pollsters like her spent Tuesday gathering some of the most closely watched numbers in the political world.

"I'm taking a national election poll,'' says Busey, preparing to ask someone how they voted.

Yesterday a few of those early numbers leaked to some bloggers ... and today there are a few red faces in cyberspace.

It's not supposed to be leaked to the campaigns or to characterize who's ahead or who's behind. That's not the purpose of the first wave of exit polls,'' said Joe Lenski, with Edison Media Research.

It wasn't long before reports surfaced on the Internet showing a close race…but one that Kerry was on the way to winning. According to one post ... Kerry would win not just ohio, but Florida as well.

The networks, including CBS, struggled with the early numbers but resisted the temptation to call races based on exit poll.

CBS did report the optimism in the Kerry camp early in the night. Of course that optimism turned out to be misplaced -- based largely on early exit polls that are notoriously unreliable.

So why are they so unreliable?

"The first wave of data has a smaller sample size and is only interviewing people that voted that morning,'' says Lenski. "So they're not necessarily going to accurately reflect the entire day."

Election experts have warned for years that poll numbers are given too much credibility by a nation eager for quick information. This year it was Internet bloggers who learned the hard way not to consider those numbers a snapshot of voter attitude when they are really just a small part of a much larger picture.