Yesterday afternoon Campaign Desk chided the National Review Online and other blogs for prematurely releasing early exit polls from yesterdays Democratic primaries and caucuses.

It’s not rocket science to figure out what’s wrong with that policy, but some readers have asked. We’ll have more to say about this later, but here’s the Cliff Notes version of our answer:

Early exit polls have the potential to distort the actual vote to come. It’s easy to imagine a voter on his way to the polls hearing that his candidate is getting creamed, and throwing in the towel … or to envision an undecided voter who is swayed at the last minute by reports of how his neighbors have voted.

That danger increases exponentially when news wires capitulate to pressures to chase blog reports, and exit poll results begin to appear on the web sites of traditional newspapers.

Exhibit A: Yesterday, at around 6 p.m. EST, a United Press International wire story showed up on The Washington Times web site specifically citing the preliminary exit poll numbers from the Arizona primary as reported by the National Review Online. The UPI story ran a full three hours before the polls closed in Arizona, leaving 180 minutes for a Dean voter in Arizona to conclude - “If Kerry has such a commanding lead, then why bother voting?”

Francis Coombs, the managing editor of the Washington Times, told Campaign Desk the paper has a policy against prematurely releasing exit poll information. So what happened?

Well, Coombs explained, the paper doesn’t bother to monitor the UPI stories, which are uploaded automatically by the wire service.

Coombs said he would be seeking an explanation from UPI itself. Like him, we have a call into UPI. We’ll let you know how they explain themselves.

Thomas Lang

Thomas Lang was a writer at CJR Daily.