Why is the Newsweek poll showing Barack Obama with a 15 point lead getting so much attention? And why is it being accepted so uncritically? Might it be what people want to hear?
According to Backcountry Notes something looks funny about Newsweek’s numbers, especially when they are compared to the other numbers. He has a point; take a look.
Looking at the polls tracked on RealClearPolitics, I note the following:
DATE POLL OBAMA MCCAIN OBAMA LEAD
June 18 Reuters/Zogby 47 42 +5
June 19 IPSOS 50 43 +7
June 19 FOX News 45 41 +4
June 21 USA Today/Gallup 50 44 +6
June 21 Newsweek 51 36 +15
June 22 Rasmussen Tracking 49 42 +7
June 22 Gallup Tracking 46 44 +2
I don’t know how Newsweek managed to get results so far from the rest of the spread (“with six other polling organizations reporting an Obama lead in the range of +2 to +7, averaging +5, Newsweek comes out with a fifteen-point Obama lead”), but it appears even if the Newsweek number is is legitimate, it’s an outlier.
Based on the post author’s experiencing in polling, he’s smelling a rat:
I majored in political science, studied polling techniques, and took part in taking opinion-poll surveys, so I know something about this; and I am puzzled, to say the least, that the Newsweek numbers are so far out of the range of random error that something seems to be wrong.
If anyone has a better explanation, I’d be glad to hear it; but it looks to me like this is bad methodology or home cooking.
Remember this?
Perhaps the poll belongs where Newsweek put the flag.
UPDATE: Commenter Physics Geek asks a good question:
Wasn’t it Newsweek that reported back in the fall of 2000 “GORE BY 10%” on its cover?
Yes. Here’s what Newsweek said:
Just before the Labor Day kick-off of the fall presidential campaign, Al Gore has surged to his largest lead yet–ten points–over Texas governor George W. Bush. In in a new Newsweek poll, conducted Aug. 30 and 31, Gore lead Bush 49 percent to 39 percent in a four-way race. Ralph Nader of the Green Party draws three percent and Reform Party candidate Pat Buchanan, one percent. In a two-way race, Gore leads Bush by 12 points, 53 to 41 percent.
And of course it was the same Newsweek that reported the phony Koran desecration story in 2005.
I’d say the former is at least bad methodology, but the latter is definitely home cooking.
MORE: Via Glenn Reynolds, Newsweek says that Clarence Thomas is white.
Does that mean he’s an outlier too?
Comments
5 responses to “bad methodology or home cooking?”
While I’ve been hearing about this “15 point lead” issue for a few days now, this is the first time I’ve seen the numbers. What I find curious is that it’s getting played as “Newsweek polls Obama suspiciously high”, whereas the numbers indicate more something along the lines of “Newsweek polls McCain suspiciously low”. Among the other numbers, 51% for Obama doesn’t look anywhere near as strange as the 36% for McCain.
Politics has irredeemably corrupted science. I no longer consider objectivity a feature of the modern intellectual–though he expects me to continue to do so.
Yeah, that 36% looks awfully suspicious.
If you Google the phrase “15 point lead” you’ll see it’s the most exciting development since sliced bread.
Wasn’t it Newsweek that reported back in the fall of 2000 “GORE BY 10%” on its cover? I’m pretty sure that they’ve got the same pollsters working for them now.
Wasn’t it Newsweek that reported back in the fall of 2000 “GORE BY 10%” on its cover?
Yes, and it was the same Newsweek that reported the phony Koran desecration story:
http://www.cnn.com/2005/WORLD/asiapcf/05/16/newsweek.quran/