CBS/Quinnipiac have released a poll of likely voters in Florida, Pennsylvania and Ohio today that purports to show that President Obama has reached 50% in all three swing states, and leads Mitt Romney in all of them by several points.
The poll finds Obama leading in Florida by 51-44. That’s a gaudy number, but take a look at the crosstabs and we find that the poll used a sample that gave Democrats a 37-26 partisan advantage. Four years ago, Florida gave Democrats a three point advantage. But two years ago, Florida elected Republican Marco Rubio to the US Senate. An eleven-point advantage for the Democrats in Florida in 2012 is extremely unlikely.
Same story in Ohio. President Obama is campaigning there today, where he is greeted by layoffs in the Brilliant, OH coal mine that are his fault. Obama’s regulations forced the layoffs. The CBS/Quinnipiac poll finds Obama with a 50-44 lead in Ohio, but the crosstabs reveal more weighting shenanigans. The poll goes in giving Democrats a 35-27 advantage. Eight points, CBS? Really?
The poll does a little better in Pennsylvania, only giving the Democrats a 38-32% advantage.
Taken all together, we have a swing state poll that appears to have been engineered to give the impression that President Obama is pulling away from Mitt Romney in three vital swing states.
A 37-26 advantage in Florida is an 11 point advantage, not 9. Which makes it even worse, of course. But then again, pollsters are using “lies, d*n lies, and statistics.”
You’re right, obviously. I blame a lack of caffeine, and fixed the post.
Stop your whining it’s only one poll and I’m sure Rasmussen is scurrying around and will shorty come out and refute those numbers – maybe you should check his crosstabs.
So I just looked at FL where you are carping about and in their very own numbers they have 42% Dems and 36% Rep with 22% ind or other. The current registration in FL is 40% Dems and 36% Rep. What that’s telling me is that 4% of registered Dems consider themselves independent and 9% of Rep consider themselves independent or as I suspect those % prefer to be called independent but are still registered in their party. The republican party on FL with Scott has been an embarrassment so that might have something to do with it was well.
This poll also determined a “likely voter” as someone who simply said they were going to vote in the election. That’s an extraordinarily weak way to assess whether someone is a likely voter, and makes this more like an LV poll, not an RV poll. That, in turn, favors President Obama.
Other polls I’ve read go into it much deeper than that by inquiring about actual voting history, etc. before designating someone a “likely voter”. I would think most people have good intentions about voting (or if they don’t, they won’t admit it). But many just won’t get there when the time comes.
I’m sorry, I meant to say that this is more like an RV poll than an LV poll.
Kick ‘em in November.
This poll only shows how likely the word “Unexpectedly!” will appear on Drudge on the afternoon of the November election.
Mark my words, this election will be “unexpectedly” over before it leaves the Eastern Time Zone.
“That’s a gaudy number, but take a look at the crosstabs and we find that the poll used a sample that gave Democrats a 37-26 partisan advantage…”
That’s criminal. Gallup’s latest partisan index puts Democrats at 30%, Republicans at 28%, and all important Independents at 41%.
That’s a +7 for Dems, -2 undersampling for GOP voters, and a massive -9% undersampling for independents. Who shorts indie voters in a swing state poll by almost a third? Qunnipiac, apparently.
All in all this poll is overestimating Obamee’s swing state numbers by 5-7%. There is no conceivable way Obamee is at 50% in any swing state right now. This poll is worthless to the point of being garbage.
Their numbers might be closer to the truth than you are willing to admit. They seem to have applied the “dead and/or illegal voters”* correction to the sample to give a truer idea of what will happen in November.
Note the current DOJ efforts to justify that correction in both Florida and Pennsylvania.
*also known as the Demofraud effect
Conservatives main character trait: whine. Whine when attacked. Blame the media. Whine when the polls give you bad news (which you’ll be getting LOTS of this fall) and blame the pollster.
Whine. Whine. Whine. Whine.
You are so victimized.
Dude at 8: No, whining is when you’ve been president 3.5 years and whine that the problem is your predecessor’s fault; that the situation was far worse than you thought; that everyone takes your words out of context; that people are so *mean* because they actually want to see your *original* long-form birth certificate–the one with the actual signatures of the attending physician and obstetric nurse, which has STILL, to this day, never been produced.
Whining is spending a trillion dollars–that’s one-thousand Billions–on vague “stimulus” projects–all supposedly “shovel-ready”, of course–because you claim that will bring unemployment down to five percent, but then when unemployment remains at 8.2 percent, you claim we didn’t do enough. Even as you quietly admit “there were very few shovel-ready projects.
Whining is saying “those mean Republicans criticized me” for having 35 (or was it 50?) unvetted, unconfirmed “czars” in charge of government, bypassing the cabinet secretaries; for ordering federal agencies not to enforce bona fide laws of the U.S.
It’s NOT whining to point out a significant methodological flaw in a poll.