AP Misrepresenting Dishonest Poll Results

September 21st, 2005 8:18 AM

The AP is constructing bad news for the President (Katrina Adds to Public Doubts About Bush). Again. And to do it, they're using a skewed sample poll, and then misrepresenting what it says. The latest AP-Ipsos poll is what they're reporting on. Once again, they've got a sample of adults, and it is signficantly skewed, with 49 percent Democrats and 41 percent Republicans. And they use that skewed sample as a hammer to hit the President, even if it isn't justified by the actual results.

  • An AP-Ipsos poll shows a sharp increase since the storm in the percentage of people who are most worried about the economy.
    Of course it does. Let's stop, for just a moment, and consider what was happening in the media in this country before Katrina hit. Every day, in every outlet, it was all Iraq, all the time. As soon as the storm went through, and the refineries and supply lines went down, the story changed to Katrina and the rising gas prices, and the devastating economic impact. The fact that there's been a sharp increase in people being most concerned about the economy says nothing about George W. Bush.
  • Fewer than half approve of Bush's handling of Katrina,
    It is true that the skewed sample reports overall slight disapproval of Bush's handling of Katrina. It is also true that the media coverage has been overwhelmingly negative.
    and almost two-thirds question the amount of money spent by the administration on the Iraq war.
    And this is where they just get dishonest. They've tied that response in to a statement about the President, but when they asked the question the poll, they talked about Congress. "As you may know, the U.S. Congress has appropriated 260 billion dollars to fight the war and help rebuild Iraq. What best describes how you feel about federal spending on the rebuilding of Iraq?"
    • We are spending too much We are spending the right amount We should spend more
    And they mentioned the total appropriation for the war and the rebuilding, then asked about spending on the rebuilding. The default position for any sentient being when asked about ANY government spending is going to be "we are spending too much." It's a question that is absolutely designed to get a negative response, which it did, and then is misrepresented in the article!
  • Given several choices to raise government money for Katrina recovery, people most often chose reduced spending on Iraq — named by 42 percent. About three in 10, 29 percent, wanted to delay or cancel Republican tax cuts. That's seven in 10 backing options that Bush doesn't even have on the table.
    Once again, they've constructed a question to get a specific answer. Let's look at the question and the options. If you had to choose, which one of the following options do you think is the best way for the government to pay for the relief effort for Hurricane Katrina?
    • Cut spending on Iraq
    • Delay or cancel additional tax cuts
    • Add to the federal debt and gradually pay it back
    • Cut spending for other domestic programs like education, welfare, transportation, and health care
    Look at that list. "Add to the Federal Debt?" Anyone think that's a good idea? "Cut ... education, transportation, health care?" It is specifically constructed, phrased, and limited to get exactly the response that they got.
    Budget cuts could be risky — only one in 10 favored that approach to finding money for storm recovery, the poll found.
    If they'd actually asked about "budget cuts," there's an excellent chance that it would have been much higher than 10%. But they didn't - they asked about cutting "spending on ... education...and health care." It's not the same question, and it's not going to get the same response. If there had been an option to the effect of "cutting wasteful spending in the federal budget," is it debatable that it would have gotten a higher response?

And there's more. They've got quotes from Democrats and Republicans criticizing the President. They've got an analyst who believe that there's a significant part of the Republican base that "want[s] him to select a woman or minority and perhaps avoid a bitter fight by selecting a relative moderate."

Lyflines - Lyford's other blog…