New Iowa Caucuses GOP Poll: Both Accurate and Meaningless
August 16, 2010 · 3:17 PM EDT
Can a poll be both correct and totally, utterly meaningless? Absolutely, and the new TheIowaRepublican.com/Concordia Group LLC survey conducted by Voter Consumer Research is a great example.
GOP pollster Jan van Lohuizen’s firm is one of those survey research firms that is widely trusted by political handicappers and veteran journalists, and I have no doubt that the survey of likely voters accurately reflects current opinion about 2012.
But, as my friend Chris Cillizza of the Washington Post points out, the new results are nothing more than a test of name recognition.
Pollster Andrew Kohut put his finger on the issue in a June 21, 2007 New York Times piece on early national polling in the 2008 election cycle by the late Robin Toner. “People aren’t paying a lot of attention,” said Kohut “Their attitudes are soft, so tests of their preferences aren’t reliable.”
“I’ve never seen so many pay attention to so much that means so little,” said Democratic consultant James Carville in the same Toner article.
Early polls in Iowa have also been misleading.
A May 2007 Des Moines Register poll of likely Iowa caucuses attendees showed Mitt Romney leading the GOP field with 30 percent, followed by John McCain at 18 percent and Rudy Giuliani and 17 percent. On the Democratic side, John Edwards was leading the field with 29 percent, while Hillary Clinton was second and Barack Obama was third.
Of course, Obama won the Democratic caucuses by eight points over Edwards and nine over Clinton, while Mike Huckabee came in first in the GOP contest, nine points ahead of Romney. Giuliani, by the way, drew 4 percent.
In the past, many Iowans haven’t made up their minds up until late – often the last couple of weeks – exactly what they’ll do at the caucuses.