`Blog' blunder sparks debate on US exit polls

"A hot story tends to spread exponentially, so by early evening, the early exit-poll results were all over the blogosphere."

Steve Outing, a senior editor at the Poynter Institute for Media Studies

Internet ended up with egg on their face this week after releasing early exit-poll data from Tuesday's vote suggesting Senator John Kerry was on his way to a victory against US President George W. Bush.

Blogs, short for Web logs, became a major phenomenon in this year's campaign. But the mistakes, while not of the magnitude of the 2000 election fiasco, opened a debate over the credibility of the sites and of the exit polls being used.

Although the preliminary exit poll data were not widely used by television networks and other mainstream media, the misleading news spread like wildfire and even prompted a selloff late in the day on Tuesday on Wall Street when it appeared Bush was in trouble.

One site distributing the early poll data, the Drudge Report, attracted 978,000 hits on Tuesday, and several other blogs collectively attracted more than a half a million visitors, according to comScore Networks.

"People jumped the gun and started interpreting the poll data before it had been completed," said Michael Cornfield, a senior research consultant at the Pew Internet and American Life Project.

"There's a game in Washington -- you feel like an insider if you get the early data and distribute it, and I did myself ... I've been writing e-mail notes all morning apologizing," he said.

But Cornfield said the release did not have the same impact as the mistaken call of Florida in the 2000 election.

"When it goes through television networks, it acquires the authority the bloggers don't have," he said. "It also affects people who are voting ... if a network calls a state [there is the feeling that] it's over."

Still, there was no doubt about the impact of the blogs.

"The thing about blogs, of course, is that a hot story tends to spread exponentially, so by early evening, the early exit-poll results were all over the blogosphere," said Steve Outing, a senior editor at the Poynter Institute for Media Studies.

"Bloggers, in particular, are loose cannons when it comes to information that established media seeks to control. Mostly independents with no corporate masters to abide -- and often no traditional journalistic training or standards -- bloggers cannot be expected to play by old media rules," Outing said in a Web column.

Cornfield he believes the exit-poll data -- a pool effort of the major media -- was generally accurate but only after late-day voters were included.

Still, bloggers and others posting messages on blogs were smarting and trying to find an explanation, with some suggesting manipulation of electronic voting systems.

"Analysis of the polling data versus actual data and voting systems supports the hypothesis that e-voting may be to blame in the discrepancies," said one Web poster, identified only as "SoCalDemocrat."

"Nevada has e-voting but with verified receipts. In that state the exit polling matches the actual results within 0.1 percent accuracy. However for other swing states Bush has unexplainable leads."

"Notice, if you will, that states with a narrow or wide Bush margin of victory NOT called Ohio or Florida, project perfectly," said another Web poster, London Yank. "It is a clear and blatant sign of voter fraud."

Cornfield that he did not believe the outcome of the vote had been influenced by the early release or by fraud.

"I'm willing to believe there was a level of fraud in Ohio and Florida, and so did the Kerry campaign," Cornfield said. "But they did their arithmetic, and if there was fraud it was not enough to change the results."