US 2012 Presidential Election Polls

dublinman1990

A poll has been compiled by The Washington Post in a survey of likely voters show that Barack Obama has widened his lead over his Republican rival Mitt Romney in the state of Ohio by 8 percentage points.

Obama = 52%

Romney= 44%

This is a state in which every Republican candidate for the US Presidency has won the race to The White House.

In Florida though, the lead for Obama is only half the margin at 4 percentage points when G.W. Bush won there in 2000.

Obama = 51%

Romney= 47%

http://www.sfgate.com/business/bloomberg/article/Obama-Leads-Romney-in-Florida-and-Ohio-in-3892887.php

In the same post poll, it showed an approval rating that was taken from registered voters. It found that Obama's rating at a margin of 55% in Florida and 56% in Ohio respectively. 52% of voters in Florida and 53% in Ohio said that they approved the way in how Obama was handling the U.S. Economy.

In both two states, Obama has been highly favoured among registered voters at a rate of 59%. Although, 40% view him as highly unfavourable in Florida and 39% view the same way in Ohio.

In Romney's case it is though more negative. In Florida, 50% of them said they highly favour Romney as POTUS while 48% view him as being unfavourable. In Ohio, 47% said they view him as favourable while 50% view him as an unfavourable candidate.


The poll was taken on the 19th to 23rd of September.

Amerika

Duck Soup wrote: »

New Quinnipac/CBS/NYT swing states poll out today [conducted 18th to 24th September]:

Florida
Obama 53
Romney 44

Florida 2008 exits (Exit Polls): 37% Democrat, 34% Republican, 29% Independent.
Florida New York Times/Quinnipiac 2012 sample: 36% Democrat, 27% Republican, 33% Independent.
Each party’s share only shifts a few percentage points, but the overall split goes from D+3 to D+9.

Ohio
Obama 53
Romney 43

Ohio 2008 exis: 39% Democrat, 31% Republican, 30% Independent.
Ohio New York Times/Quinnipiac 2012 sample: 35% Democrat, 26% Republican, 35% Independent.
In this sample, the partisan split is D+9 compared to D+8 four years ago, and the GOP is five percentage points smaller than in 2008.

Pennsylvania
Obama 54
Romney 42

Pennsylvania 2008 exits: 44% Democrat, 37% Republican, 18% Independent.
Pennsylvania New York Times/Quinnipiac 2012 sample: 39% Democrat, 28% Republican, 27% Independent.
Somehow a D+7 split has turned into D+11 split, and Republicans’ share of the electorate is nine percentage points less than they were four years ago.

Pollsters... meet grain of salt.

http://www.nationalreview.com/campaign-spot/328555/mornings-polls-project-more-heavily-democratic-electorates-2008[/SIZE]

Black Swan

Romney has not led in the polls reported by RCP for a month (since 27 August 2012 by Gallup and then only by +1). These 24 polls have been conducted by 19 different organisations, with Obama leading from +1 to +8 for this time period (with only 2 ties). Rasmussen Tracking varies daily (even within the same day), and during this time period has ranged from a tie to +5 for Obama. The last Republican leaning FOX News poll on 11 September shows Obama +5.

Amerika

Rasmussen seems to be the only pollster that uses sample sizes with any substantive nuance. And of all the polls, including the RCP average, Rasmussen is the only ones who I have any confidence in this time around. Now if all the other pollsters believe that the voting turnout will be even more heavily Democratic this year than it was with the record Democratic turnout in 2008, then they should make their case. Also, Democratic enthusiasm is down, and the president has lost considerable ground with independents. So I doubt they can make the case, and therefore think they all should be taken with a grain of salt.

joseph brand

The National Review's editor is Rich Lowry. He's quite the conservative. A few years ago he got very excited when Palin apparently winked at him through his tv set.

I'm sure I'm not the only male in America who, when Palin dropped her first wink, sat up a little straighter on the couch and said, "Hey, I think she just winked at me." And her smile. By the end, when she clearly knew she was doing well, it was so sparkling it was almost mesmerizing. It sent little starbursts through the screen and ricocheting around the living rooms of America. This is a quality that can't be learned; it's either something you have or you don't, and man, she's got it.

He needs to get out more.

Anyway, Criticism of Romney's Campaign Grows; Six in 10 Rate His Efforts Negatively

Sixty-one percent in a new ABC News/Washington Post poll hold an unfavorable view of how Romney's handling his presidential campaign, up by 12 percentage points since mid-July. Far fewer, 35 percent, rate Romney's performance positively, essentially unchanged.

Barack Obama's ratings for handling his campaign are substantially better, 54-43 percent, favorable-unfavorable. And while ratings of Romney's campaign have grown more negative, favorable ratings of Obama's campaign efforts have gained 8 points since July.

Even among Republicans, more than one in four rates Romney's efforts negatively - 27 percent. While essentially unchanged since July, that's substantially more than the share of Democrats who respond negatively to Obama's work on his campaign, 11 percent.

Black Swan

Amerika wrote: »

Rasmussen seems to be the only pollster that uses sample sizes with any substantive nuance. And of all the polls, including the RCP average, Rasmussen is the only ones who I have any confidence in this time around. Now if all the other pollsters believe that the voting turnout will be even more heavily Democratic this year than it was with the record Democratic turnout in 2008, then they should make their case. But I doubt they can, and therefore think they all should be taken with a grain of salt.

There were some questions regarding Rasmussen's survey research methodology, sampling in particular, between the 2008 and midterm 2010. It was reported that Rasmussen did not call mobiles (often called cell phones in the States), only landlines, and that the population demographics regarding mobile adoption to replace landlines has markedly changed since 2008 (some voters do not have landlines now, having totally replaced them with mobile phones).

Although anecdotal, most of my friends no longer have landlines, because they are old tech, redundant with mobiles, and a waste of money to have both.

If Rasmussen Tracking continues this practice of calling only landlines for 2012, it may make their predictions both invalid and unreliable. There was an article which suggested this inaccuracy during the 2010 midterms, and may produce different but similar "Dewey Wins" inaccurate predictions for 2012. Do they still use landline-only sampling per 2008 and 2010 for 2012?

Amerika

As the skepticism continues in earnest regarding the pollsters samplings and methodologies, it will be interesting to see how the different pollsters rate against the actual results come 6 November. I’m guessimating Rasmussen still comes out on top. But only time will really tell.

Black Swan

Amerika wrote: »

Rasmussen seems to be the only pollster that uses sample sizes with any substantive nuance.

Not sure what you mean by this when referring to sample "size" with "any substantive nuance." Sampling is a very complex method when attempting to derive valid and reliable estimates of population parameters within reasonable confidence levels and confidence intervals for the US voting population.

If you are not referring to the very complex notion of "size" from a technical/statistical standpoint as pertains to estimating population parameters, but merely noting that Rasmussen Tracking typically uses n=1500 in their recent polls, then both Gallup (n=3050) and Pew Research (n= 2268) have surpassed that number, showing Obama leading +6 and +8 in their last polls per RCP.

Duck Soup

Black Swan wrote: »

There were some questions regarding Rasmussen's survey research methodology, sampling in particular, between the 2008 and midterm 2010. It was reported that Rasmussen did not call mobiles (often called cell phones in the States), only landlines, and that the population demographics regarding mobile adoption to replace landlines has markedly changed since 2008 (some voters do not have landlines now, having totally replaced them with mobile phones).

Although anecdotal, most of my friends no longer have landlines, because they are old tech, redundant with mobiles, and a waste of money to have both.

If Rasmussen Tracking continues this practice of calling only landlines for 2012, it may make their predictions both invalid and unreliable. There was an article which suggested this inaccuracy during the 2010 midterms, and may produce different but similar "Dewey Wins" inaccurate predictions for 2012. Do they still use landline-only sampling per 2008 and 2010 for 2012?

Yep, they still only call landlines.

http://www.rasmussenreports.com/public_content/about_us/methodology

As the drift away from landlines to mobiles continues, Rasmussen is likely to get ever-worsening results, as evidenced by their precipitous drop in accuracy of forecasting from 2008 to 2010.

Note that "To reach those who have abandoned traditional landline telephones, Rasmussen Reports uses an online survey tool to interview randomly selected participants from a demographically diverse panel."

This is a key element in the broader criticism of Rasmussen - that they have cut back on costs and this is causing their poll accuracies to suffer. Automated rather than live interviews. Online surveys rather than calling mobiles. Seriously?

Amerika

I think it has a lot to do with how Rasmussen handles stratification. I think they better at keeping their samples close to the real voting population… by party, likely to vote, sex, ethnicity, etc.

And yes they don’t call cell phones. They handle those with online sampling tools. People with cell phones and no landlines typically are young and tend not to vote. Also, they have caller ID and are much less likely to take a live poll call because they have to pay for minutes. But times are changing and that demographic will also change.

http://www.rasmussenreports.com/public_content/about_us/methodology

It will be interesting to see how the various polling firms do in the final few days of the election. Will their numbers change dramatically, or keep in tune what they have been reporting to us all along. Will they parallel Rasmuessen in the final days and then claim they were accurate, when they were so different for so long? Again, the actual results will identify the pollster winner.

Black Swan

Amerika wrote: »

And yes they don’t call cell phones. They handle those with online sampling tools.

It's unfortunate that Rasmussen fails to provide any details regarding survey research design or sampling methods for this segment of the voting population, other than the below vague statement. Without more specificity, their representativeness for this segment becomes more faith based than fact based.

To reach those who have abandoned traditional landline telephones, Rasmussen Reports uses an online survey tool to interview randomly selected participants from a demographically diverse panel.

It would appear that a huge number of US households are in this segment that are mobile only, without landlines:

• More than one in four U.S. homes, or 26.6%, had only a wireless phone as of June 2010
• Low-income homes and those in poverty are more likely to be wireless-only homes
• Renters are also more likely to have wireless-only households

Overheal

Amerika wrote: »

They handle those with online sampling tools. People with cell phones and no landlines typically are young and tend not to vote.

Uhm, reality called but you weren't home to answer. Did you have a cell number it can reach you at?

TwoShedsJackson

Amerika wrote: »

As the skepticism continues in earnest regarding the pollsters samplings and methodologies, it will be interesting to see how the different pollsters rate against the actual results come 6 November. I’m guessimating Rasmussen still comes out on top. But only time will really tell.

Shockingly, Rasmussen are also the only poll that are anywhere near what you would like the result to be.

spank_inferno

TwoShedsJackson wrote: »

Shockingly, Rasmussen are also the only poll that are anywhere near what you would like the result to be.

The only stat that matters is that Rasmussen provides the outcomes Amerika wants, 100% of the time

Amerika

Overheal wrote: »

Uhm, reality called but you weren't home to answer. Did you have a cell number it can reach you at?

If you have more up-to-date information that differs significantly, please correct me, and I will be happy to relent. Otherwise. have fun in that alternate reality.

Research done this year by the National Center for Health Statistics found that 14.5 percent of households cannot be reached by a typical survey because they do not have landline phones.

The majority of those in cell-phone-only households are ages 18 to 24. Given these numbers, experts agree that it is a problem that will get worse in the future.

A Pew Research study done this year titled "Cell Phones and the 2008 Vote" found that "23 percent of cell-only young respondents say they 'always vote,' compared with 41 percent among the landline respondents."
"Younger people tend not to vote, (compared to) people who are age 35 and older," said Molly Longstreth, director of the Survey Research Center at the University of Arkansas. "If younger people actually go out and vote then it will show that polls that didn't use cell phone data will not be as accurate as polls that did."

http://www2.ljworld.com/news/2008/oct/26/cellonly_users_ring_warning_bells_polls_rely_landl/

Amerika

TwoShedsJackson wrote: »

Shockingly, Rasmussen are also the only poll that are anywhere near what you would like the result to be.

spank_inferno wrote: »

The only stat that matters is that Rasmussen provides the outcomes Amerika wants, 100% of the time

And if Rasmussen once again is determined to be the most accurate of the pollsters in the presidential race, will there be an apology or just the standard silence of denial?

Who hear really thinks Democrat turnout for this election is going to be stronger than it was in 2008 (besides many of these "reputable" polling firms, that is)?

Memnoch

Amerika - I urge you to read the following article:

http://fivethirtyeight.blogs.nytimes.com/2012/09/27/sept-26-could-2012-be-like-2008/

Yes it is hosted on the NY times, but it is probably the most informed source you could read on the mathematical issues of polling. The math is rigorous and takes all the polls into accounts. This guy has been spot on time and time again and I would say is the best meta-pollster in the US right now.

I know you won't like it and you won't want to like it, but you should give it a read nonetheless. I'll happily agree to read any article of a similar length (election related) that you chose to put up in exchange.

Amerika

Memnoch wrote: »

Amerika - I urge you to read the following article:

http://fivethirtyeight.blogs.nytimes.com/2012/09/27/sept-26-could-2012-be-like-2008/

Yes it is hosted on the NY times, but it is probably the most informed source you could read on the mathematical issues of polling. The math is rigorous and takes all the polls into accounts. This guy has been spot on time and time again and I would say is the best meta-pollster in the US right now.

I know you won't like it and you won't want to like it, but you should give it a read nonetheless. I'll happily agree to read any article of a similar length (election related) that you chose to put up in exchange.

A good read. It won't be long to see who is right. But does the NY Times feel the democratic turnout will be even stronger than it was in 2008? I haven't seen any indication that that is the case. And if there isn't viable proof of it, don't you think the NY Times really should be taking the other pollsters, and not Rasmusssen, to task for bumping democrat samplings even higher than 2008 results or republicans less than 2008 results?

And I'll take you up on your offer. Like you, I know you won't like it and you won't want to like it.
http://hotair.com/archives/2012/09/26/which-polls-are-or-arent-legitimate/

TwoShedsJackson

Amerika wrote: »

And if Rasmussen once again is determined to be the most accurate of the pollsters in the presidential race, will there be an apology or just the standard silence of denial?

Who hear really thinks Democrat turnout for this election is going to be stronger than it was in 2008 (besides many of these "reputable" polling firms, that is)?

I won't have anything to apologise for, I'm not carrying out my own poll, or even saying that Rasmussen are wrong, or will be wrong when we know the election result.

What I am saying is that it is interesting that the one poll source that you 'guesstimate' will be the accurate one, is the only one whose result comes anywhere near the result you wish to see, and all the other poll results are dismissed out of hand.

Amerika

TwoShedsJackson wrote: »

I won't have anything to apologise for, I'm not carrying out my own poll, or even saying that Rasmussen are wrong, or will be wrong when we know the election result.

What I am saying is that it is interesting that the one poll source that you 'guesstimate' will be the accurate one, is the only one whose result comes anywhere near the result you wish to see, and all the other poll results are dismissed out of hand.

Okay, fair enough. I guess what I'm saying is I trust Rasmussen more because unlike the majority of other polling companies, I don’t think democrats voters will miraculously fall from the sky like manna from heaven into the voting booths, while republicans will somehow forget which day to vote on.

joseph brand

According to Fox and Friends, the polls are rigged, even their own. Ahh, don't laugh.

http://2012.talkingpointsmemo.com/2012/09/fox-friends-polls-rigged-conspiracy.php?m=1

It’s clear that Fox & Friends has a tricky relationship with data. But on the show Thursday morning, the hosts took that relationship to a whole new level.

After a raft of new polls showed Obama opening up leads in swing states, the Friends flew in to full-blown conspiracy mode about what’s really behind the data.

Parroting the latest Republican meme that national polls oversample Democrats, host Steve Doocy threw in to the mix the possibility that pollsters are using voter turnout from 2008 to guide who they should be asking.

Better link here: http://www.youtube.com/watch?v=zkPyseb8tWo&feature=plcp


Republican minds can't process facts. "Thinkin's hard."

Overheal

I've seen enough uncanny stupid **** from Fox and Friends that I pray nobody takes them seriously

Paleface

Obama has shortened again with Paddy Power whilst Romney has lengthened accordingly.

Obama 1/6
Romney 7/2


I haven't time to check Ladbrokes or William Hill but they usually align with each other on these kinds of markets.

Memnoch

Amerika wrote: »

A good read. It won't be long to see who is right. But does the NY Times feel the democratic turnout will be even stronger than it was in 2008? I haven't seen any indication that that is the case. And if there isn't viable proof of it, don't you think the NY Times really should be taking the other pollsters, and not Rasmusssen, to task for bumping democrat samplings even higher than 2008 results or republicans less than 2008 results?

And I'll take you up on your offer. Like you, I know you won't like it and you won't want to like it.
http://hotair.com/archives/2012/09/26/which-polls-are-or-arent-legitimate/

It's just the same over-sampling stuff all over again. I switched on Fox news for a few minutes there yesterday and serendipitously they had on a couple of women talking about this.

Both agreed that the polls are not skewed. One of them made the point that they aren't talking about registered voters, where the sampling is concerned but rather people who 'identify' one way or the other. And that this is usually from independents who will identify as republican or democrat based on who they are going to vote for.

This tends to skew the sampling in favour of the side that's usually gaining more support from independents and applies to both republicans and democrats.

Like I said in my last post. Nate Silver's math and method are really top notch (he was mentioned in the fox news discussion also). The article you posted seems to be purely partisan opinion. Nate's articles just don't read that way.

joseph brand

"Mitt Romney Favorability Lower Than George W. Bush, Poll Finds." Ouch!

If Mitt Romney was hoping to distance himself from former President George W. Bush, a new poll has some news that might trouble the Republican presidential nominee: Bush posted higher favorability ratings than Romney.

The national survey, conducted by Bloomberg News and released Wednesday, found that Bush received a favorable rating of 46 percent, while Romney's favorable rating was 43 percent. Forty-nine percent of respondents gave Bush an unfavorable rating, compared to 50 percent of respondents who gave Romney an unfavorable rating.

A wave of recent polls has shown Romney struggling to keep up with President Barack Obama. A lower favorability rating than Bush, who was recently named the most unpopular living U.S. president, shows how difficult it has been for the GOP nominee to resonate with voters during his current presidential bid.

The only hope for the GOP after Obama wins his second term, is to avoid appealing to the aged, the racist, the rich and the religious fundamentalists. They need to start fresh, because the electorate have gotten wiser. They need to change with the times. In effect, be more like the Dems.

Amerika

joseph brand wrote: »

They need to start fresh, because the electorate have gotten wiser. They need to change with the times. In effect, be more like the Dems.

So we need to learn how to better lie, cheat, steal and pull the wool over the voters eyes eh?

karma_

Amerika wrote: »

So we need to learn how to better lie, cheat, steal and pull the wool over the voters eyes eh?

The GOP have nothing to learn on that front, they practically invented it.

joseph brand

Amerika wrote: »

So we need to learn how to better lie, cheat, steal and pull the wool over the voters eyes eh?

Lie, cheat, steal: (Romney Ryan)

Pull the wool over the voters eyes: Fox 'news'.

You guys got it covered.

Memnoch

Amerika wrote: »

So we need to learn how to better lie, cheat, steal and pull the wool over the voters eyes eh?

The fact that you can say that and believe it says everything anyone needs to know about the GOP and those who would vote for them. How can you be so utterly blind to such trenchant hypocrisy?

Also please quit with the smileys. They add nothing to the discussion and just get people riled up. The content of your posts is disingenuous enough as it is.

joseph brand

Romney's 47% Video Has Been Viewed 3 Times As Often As His Convention Speech






A picture is worth thousand words, they say.

The internet (information) is the GOP's biggest enemy. Just like Jay Rockefeller said.

And we all know what type of scum the Rockefellers are.