Polls: President Obama Continues to Widen Lead Over Mitt Romney in FL, PA and OH
The latest polls (which the Romney campaign disputes out of desperation) show President Obama widening his lead. I guess “the stench” from the Romney campaign is spreading to the general public. We will update as needed through-out the day:
Florida: Obama 53%, Romney 44% (NYT/CBS/Quinnipiac)
Ohio: Obama 53%, Romney 43% (NYT/CBS/Quinnipiac)
Pennsylvania: Obama 54%, Romney 42% (NYT/CBS/Quinnipiac)
A new Associated Press-GfK poll finds 72% of Americans think President Obama’s signature health care legislation, the Affordable Care Act, aka Obamacare, is here to stay and will fully go into effect with some changes. Just 12% say they expect the law will be repealed entirely.
This was cross-posted from The Hinterland Gazette.
Share This
Hmm, above 50% in three of the competitive states this year. I wonder how the voter ID laws in FL, PA and other states will play an effect on the outcomes.
I am surprised that Romney is still as close as he is after he was caught insulting our troops as being those that do not take personal responsibility. Romney is not a patriot. He is an arrogant, greedy steaming pile of Mitt. He also insulted my mother and many other peoples mother. This is not an acceptable candidate. He should have zero percent of the vote. In addition Mitt Romney is anti-Christ and anti-American.
So the only reason to object to the CBS/NYT poll is desperation? That’s an interesting point of view.
Let’s look at the polling by party ID relative to the 2008 election. We’ll just take Florida as an example. First, we have to find the poll internals. Neither the NYT nor CBS makes them terribly easy to find for some reason, but Quinnipiac does provide the links at the bottom of this page
http://www.quinnipiac.edu/institutes-and-centers/polling-institute/presidential-swing-states-%28fl-oh-and-pa%29/release-detail?ReleaseID=1800
Now, when you go to the demographics tab, you’ll find the following information. There’s a column labeled weighted percentage and a column labeled unweighted frequency. The process for the poll is that the choice is determined by the results in the second column but the poll is “weighted” by the first. That is the frequency distribution of voters is determined by the first column. I won’t go into a long debate about why polling is increasingly conducted this way but it is. Indeed, pretty much all the polls except the tracking polls are now done this way.
So what assumption is used. From a D/R/I perspective, it’s 36/27/33 or as pollsters like to say D+9 (9 percent more democratic voters than republican ones. So the question we have to ask is whether this is a reasonable assumption. There are two data points we could bring. First, we could use the exit polls. You can find those pretty easily here.
http://www.cnn.com/ELECTION/2008/results/polls/#FLP00p1
In 2008, a pretty landmark year for Democrats, the exit polls say the voting percentages were 37/34/29 or D+3. In 2004, it was 37/41/23 or R+4. So the assumption used in the poll is that the electorate voting in 2012 will be 6 percentage points more favorable to the Democrats than the electorate was in 2008 and 13 points more favorable than in 2004.
Is there any good data to justify this assumption. Not obviously.
Gallup here reports Florida as D+2 in 2011 versus D+9 in 2009. Now it’s hard to use those data to argue that Florida’s electorate is going to be a lot more Democratic than it was in 2008.
http://www.gallup.com/poll/125066/State-States.aspx
One last thing, the same polling from Quinnipiac reports that Romney is winning 49/46 with independents in Florida. So the cause of the President’s lead is the sample, nothing more and nothing less.
Now, I’m sure all of this data is just an indication of desperation. Either that or it’s just a really bad sampling assumption. It seems to me that questioning the sampling assumption is actually pretty reasonable.