I have been regularly writing about the pro-Republican slant of the Gallup/CNN and CBS News/New York Times polls. On Tuesday, MoveOn.Org took out full-page ads in major newspapers to further publicize what it described as “Gallup-ing to the Right.” The ad highlighted Gallup’s assumption that the Republican turnout on election day will exceed the Democrats by 6-8%, while in 1996 and 2000 Democratic turnout exceeded Republican by 4-5%. This discrepancy turns Bush’s 14 point Gallup lead into a one point advantage. Yesterday, the Times wrote a story about the MoveOn ad, but blamed “partisanship” rather than defective polling for the uproar.
The theme of Jim Rutenberg’s September 29, 2004 Times story, “A Request to Partisans: Don’t Shoot the Pollster,” is summed up in this excerpt:
Many pollsters interviewed Tuesday, while not claiming to be perfect, said
their more vociferous critics were often trying to shout down messengers
delivering news that runs counter to the version of reality they want to see
In other words, MoveOn is mad because Gallup shows Kerry far behind, and is attacking the pollster for simply reporting political reality.
But as the ad clearly demonstrated, Gallup’s polling is based on faulty methodology. In fact, Ruy Teixiera and others at emergingdemocraticmajority.com have shown that Gallup’s results are so inconsistent with other polls on a state-by- state basis that the skewed Republican sample is but one many serious irregularities.
Since Gallup’s polls are widely publicized by their partners at CNN and USA Today, their methodology is important. The Times would thus be expected to uncover the scientific basis for the company’s polling a far higher percentage of Republicans than will actually vote on November 2.
But the Times story simply quotes the Gallup Organization’s view that people who say they are supporting Bush are likely to identify themselves as Republican. In truth, it is more the case that Republicans identify themselves as Bush supporters.
As Teixera and others have pointed out, Gallup’s premise is that since early September America has witnessed the greatest shift between Republicans and Democrats since the Civil War. No other polling organization has witnessed such a shift.
So why would the Times bend over backwards to defend junk science? The best answer may be that its own Times/CBS News poll also fails to reflect the party composition of the likely electorate.
The Times poll is almost as skewed with excess Republicans as Gallup, so MoveOn’s full-page ad hit a bit too close to home. As John Nichols pointed out in our interview last week, media organizations have paid too much too polling organizations to not use the results.
No matter how flawed.
What’s also interesting about the Times’ “Don’t Shoot the Messenger” theme is that the article makes no reference to the paper’s Sunday, September 26 front-page story on the record increase in new voter registrants in many swing states. These new registrants could clearly impact the outcome of the race Nor does the piece mention the exclusion of cell phones from polls, a fact that has led many experts to doubt the certainty of any short-term telephone polling.
All this talk about polls would be meaningless if it were not creating a false bandwagon affect for the President. People like to go with the winner, and if they feel Bush is clearly ahead they will go with him.
Polls have clearly shaped the spin for tonight’s debate. The prevailing frame is that Kerry needs a knockout to win, when the usual test for a challenger is whether they can hold their own with the incumbent.
I wrote yesterday about the need for activists to get involved in making sure that the post-debate spin does not unjustly favor Bush. Toward this end, America Coming Together has put together a Truth Squad to hit the ground Saturday to make sure that voters rely on the facts, not the pundits.
We will have our own analysis of the debate in Friday’s edition. For those who are tired of the usual post-event punditry, the Daily Show will have a special show following the debate that will probably provide far greater insights than would be found elsewhere on television.