Why did Gallup’s presidential polls miss so badly?
Democratic pollster Mark Mellman believes this kind of likely-voter screen is counterproductive. Speaking at the event at the Gallup building, Mellman said that pollsters and the media have “overfetishized this whole notion of likely voters.”
“We should not be concerned about finding likely voters,” said Mellman. “We should be concerned about simulating the likely electorate.” Implicit in Mellman’s comments is the idea that the composition of that likely electorate would not have been 78 percent white.
But others say that issues with Gallup’s initial sampling may be to blame. Andrew Kohut, the retiring director of the Pew Research Center who served as president of the Gallup Organization from 1979-1989, pointed to the difference in Gallup’s results for all registered voters versus likely voters in its final preelection poll: Romney led by 1 point among likely voters, but Obama held a 3-point lead among all voters. Kohut compared that to Pew’s final results, which also showed a discrepancy of 4 net points between registered and likely voters.
“I think there’s something very basic going on with their sample. It’s not with their likely-voter scale,” Kohut told National Journal in a telephone interview, adding, “I helped create that likely-voter scale.”
Blowback
Note from Hot Air management: This section is for comments from Hot Air's community of registered readers. Please don't assume that Hot Air management agrees with or otherwise endorses any particular comment just because we let it stand. A reminder: Anyone who fails to comply with our terms of use may lose their posting privilege.
Trackbacks/Pings
Trackback URL
Comments
Most polls were off because they didn’t take cheating into account. When they polled Democrats +11, they started getting closer to the results.
The Rogue Tomato on November 19, 2012 at 8:28 AM
Because hardly any respectable pollster expected a D+6 result on Election Day. I still don’t see how that happened unless there was A)massive voter fraud and/or B)a sizable chunk of the Republican base was stupid enough to stay home.
Doughboy on November 19, 2012 at 8:34 AM
+6 on that.
ProfShadow on November 19, 2012 at 8:35 AM
As an aside from the OP, a friend in New Hampshire stated the day after the election that she had noticed crowds much higher than normal and a lot of Massachusetts license plates. Only yesterday did she read this article where someone else had noticed the same thing.
Apparently in NH a voter can register the day of the election and right there at the polling booth.
I don’t think it has to be any great conspiratorial scheme where Obama himself gives marching orders for there to be massive fraud. It could just be individuals or groups working independently; just “doing their part”.
eforhan on November 19, 2012 at 8:45 AM
D’oh. Meant to quote Rogue Tomato‘s post. Mea culpa.
eforhan on November 19, 2012 at 8:46 AM
I’m awfully interested in those places where Bammy got 108% of the vote, or where Romney got 1% & Bammy 99% etc.
If it were Hitler we’re talking about or Stalin I’d understand bcs then we’d be in a fascist or communist country.
But in America?
WTF does not the news media find this sort of thing suspicious?
Badger40 on November 19, 2012 at 9:07 AM
Because they’re racist.
forest on November 19, 2012 at 9:13 AM
They wanted to fire up the Democrats get out the vote effort and fabricated the sort of polling numbers that would best maximize the effort.
Buddahpundit on November 19, 2012 at 9:18 AM
Gallup’s polls were horrible in 2008. This is nothing new.
AngusMc on November 19, 2012 at 9:23 AM
I still think there was massive fraud. The only thing that was explained was something about double sided ballots or something.
watertown on November 19, 2012 at 9:25 AM
What if I told you… there are more democrats than republicans?
triple on November 19, 2012 at 9:25 AM
Maybe because of fraud? No!
900 votes for the Democrat in a precinct with just 7 registered voters.
http://www.breitbart.com/Big-Government/2012/11/18/Allen-West-Seeks-Emergency-Hearing-After-Opponent-Declares-VIctory-When-FL-County-Misses-Sunday-Reporting-Deadline
If we don’t confront Democrat voter fraud it’s not going to matter anymore who votes or how great our candidate is. It is so obvious that real fraud was committed in several key states and yet we ignore it. Like it really didn’t.. although we all know we’ve read and heard about the stories before during and after.. we still pretend it didn’t actually matter or make a difference.
If voter fraud doesn’t matter then why are Democrats doing it? why did they throw Republican poll watchers out of the 59 precincts? Just for the fun of it?
We are losing because we are not fighting.
JellyToast on November 19, 2012 at 9:32 AM
Fraud.
ahem on November 19, 2012 at 9:44 AM
Because the poll didn’t figure in Democrat voting FRAUD.
Axion on November 19, 2012 at 10:55 AM
Not just Gallup, but Ras’ partisan self-ID surveys. Those are among the most solid, methodologically, of any survey. Largest sample of any (15,000+), robo-calls which exclude any “human factors” in response dynamics, and a “questionnaire” that cannot introduce distortion (you just push a number to self-identify partisan leanings or registration).
This survey produced consistently historical highs for R+ self-ID. It was one of the first flashing lights that caught Barone’s eye back in the summer.
Overall, it was the consistency of all the collateral survey benchmarks – presidential job approval, right track/wrong track number, unemployment rate, etc., ALONG with the objective partial data (voter registration, likely early voter partisan break-out, etc.), ALONG with all of the objective non-quantitative “data” (a DNC convention that was broke and bizarre, an incumbent who couldn’t fill a large venue for going on 18 months, fund-raising totals, and a stunning contrast in apparent enthusiasm between the two campaigns) – that make the election results a stunner for anyone really watching closely.
Barone is no wish-casting homer – in 2008 he was the first to identify key and novel factors that made him foresee a victory by the non-entity (youth vote finally mattering for the first time ever, for example). Barone’s call this time was aggressive, and based down to the smallest details on solid evidence and historical patterns that had never been defied.
It’s the catastrophe of the country choosing decline and dishonor that matter, of course, not polling geekery, but this latest (final?) national disaster has a still completely unexplained basis. I really want to see the hardest, best estimates on likely-GOP “did not vote” numbers, esp. in the Midwest. This phenomenon is jaw-dropping in itself – if it was material to the outcome, the last tiny bit of light way down at the end of the tunnel is going to be extinguished.
IceCold on November 19, 2012 at 11:22 AM
I keep seeing Rogue tomato claim that the entire election was rigged because of 141% voter participation in st. lucie Florida. But:
1. No one has ever linked to any proof of this
2. How does this explain the victories Obama earned in every other swing state.
3. How did GOP secretary of state offices allow for Obama cheating? Is the idea that they are somehow in on it too?
4. How did voter fraud happen in states that passed anti-voter fraud measures?
So far no one as answered this question.
libfreeordie on November 19, 2012 at 12:38 PM