Herald pulls manipulated poll
By Brent Curtis Rutland Herald - Published: September 30, 2004
RUTLAND — Editors at the Rutland Herald pulled an online poll Monday night after a local radio talk show host urged his listeners to skew the results.
On Saturday, the Herald posted the question "Who is your likely choice for governor in the Nov. 2 vote?" The poll included the top four candidates running for that office — Republican Gov. James Douglas, Democrat Peter Clavelle, Liberty Union candidate Peter Diamondstone and Libertarian Hardy Machia. The informal poll was intended to remain posted at
www.rutlandherald.com until Friday.
Editors at the Herald decided to remove the poll Monday night after Tim Philbin, host of "On the Air with Tim Philbin," told his listeners that the poll was biased and described how they could skew the results by voting more than once.
Philbin, who often discusses politics on his morning show broadcast by WSYB in Rutland, said he was appalled to check the poll Monday to find that Clavelle had the lead after about 450 people had voted.
"The point is, please don't tell me that represents reality," Philbin said after his show Tuesday. "That's called manufactured news."
Philbin said the unscientific poll was a sham because it flew in the face of other polls he has seen this election year that show Douglas in a commanding lead.
He said he suspected the newspaper's reason for posting the poll was to shape public opinion, not reflect it.
"If they have a poll that can be manipulated, the results of which should represent reality, you can manipulate reality to represent what the press wants," Philbin said.
To prove his point, he told his listeners how to effectively stuff the ballot box.
Hours after his radio show Monday, the number of votes cast on the Web site climbed from about 450 to 1,003.
After the editorial staff at the newspaper heard about Philbin's broadcast, the decision was made to pull the plug on the poll Monday night.
"We ended the poll when we realized it had been rigged," city editor John Dolan said Tuesday. "We assumed something like this could happen in small amounts, but not this kind of organized rigging."
Dolan said the decision to withdraw the poll was not a partisan issue. He said the results would have been removed no matter which candidate's results were rigged.
While the online poll doesn't follow standard polling methods, Dolan said the results — available on the Web site and in the newspaper's Street Talk section — provided readers with a picture of public opinion and an opportunity to participate in the process.
"It's valuable because it lets people participate willingly instead of waiting to be called and asked," he said. "And while the results are not scientifically accurate, they give some indication of what the community is thinking."
Dolan also pointed out that Philbin conducts his own informal polls during his two-hour radio show.
"It's interesting to note that today, Mr. Philbin boasted about his prank and then spent the remaining hour of his show conducting a poll on the Leahy-McMullen race," he said. "One person called. We had 450 people vote in the two days before his show. Which poll was more useful? People can decide for themselves."
However, the number of participants in a poll might not be an accurate gauge of their usefulness or accuracy, according to professors at the Poynter Institute, a journalism school in St. Petersburg, Fla.
Aly Colòn, an ethics group leader at Poynter, said Tuesday that online polls and polling in general could ruin a newspaper's credibility if the publication wasn't upfront about its methodology and accuracy.
"If people think it's definitive information that's pure and unbiased, then they see the results and aren't sure how the findings could happen. They will think the newspaper tried to skew the results in a particular way even if it wasn't," he said.
Al Tompkins, who teaches broadcasting and online issues at Poynter, was even more skeptical of online polling.
"The big issue with online polling is that no matter how many responses you get, it's not 100 percent accurate," he said. "What you end up with is a very lopsided demographic. It gives you a wild idea at best of what a community is thinking."
Tompkins said online polling was once a popular way for newspapers to get their readership involved. But, he said, many newspapers have evolved to using online forums and community chat rooms to have interactive discussions with their readers.
"Most of the time, I've found the online polls to be a complete waste of time because they're so wildly unscientific," he said. "They're called polls, but they're not really polls at all. One is borderline voodoo, the other is scientific polling."
Contact Brent Curtis at
brent.curtis@rutlandherald.com.
http://www.timesargus.com/apps/pbcs....23/1003/NEWS02