Wednesday, February 28, 2007

On Biased Polls



I wrote about this particular poll for TAPPED last week. It is a biased poll, intended to produce biased answers. Glenn Beck has used it and now Fox News is using it, too:

On the February 27 edition of Fox News' America's Newsroom, anchor Bill Hemmer cited the results of a "new poll" by Public Opinion Strategies that 53 percent of voters say "victory in Iraq is still possible." Hemmer failed to acknowledge that, as noted by Media Matters for America (here and here), POS has described itself as a "Republican polling firm," and a Republican pollster has reportedly stated that the question Hemmer cited was worded in a "completely unprofessional" manner.

The POS poll, conducted February 5-7, asked voters to rate their level of agreement/disagreement with the statement "Victory in Iraq, that is creating a young but stable democracy and reducing the threat of terrorism at home, is no longer possible for the US." Media Matters has noted that, according to blogger Greg Sargent, Republican pollster David E. Johnson, CEO of the Strategic Vision polling firm, criticized the poll as, in Sargent's words, "leading and designed to elicit the answers they got." Johnson also asserted, according to Sargent, that the wording of the poll's statement was "completely unprofessional" because "[i]t's designed to confuse the respondent. People are being asked whether two different things can be accomplished -- establishing democracy in Iraq and reducing the threat of terrorism at home -- and doesn't clarify which one people are talking about."

The danger I see in all this is a total corruption of polls as at least partly useful strategies for learning about voters' preferences. If the right-wing is going to start praising biased polls on purpose it will not be very long when the left-wing must do the same or lose the game. And the overall result is that nobody will trust any polls at all. Part of the return to the faith-based times?

Actually, this trend has existed for a while. The conservative think tanks have been spewing out biased research (research without the proper anonymous refereeing of academic papers or the tournament of seminars where it is ripped apart if it is bad) for a long time, and this research has been given the same respect as research that came out of the peer-review system. I'm sure similar examples can be given from the left if one digs deep enough.

This is dangerous, because it removes another leg from the stool of reasoned arguments on which we all try to sit. (And yes, I did have the mental image of all people jostling to try to sit on the same stool. My sense of humor is quite sick.)