If the Polls are "Skewed"

[Zander]

I think we are seeing the exact same pattern we see every cycle- crazy polls that mean nothing. Political polling is not an exact science- it's a SWAG.

I do remember Carter being way, WAY WAY!! ahead of Reagan according to Gallup in October of 1980.....we all know how that worked out!

Bottom line - This is a very close race and it's all going to come down to voter turnout in the swing states like Ohio and Florida. Hopefully the media continues to advance the narrative that "Obama's got this"....that way more of his supporters stay home on election day..why bother to vote? He's got this....he can't lose.

A republican hoping for a low voter turnout. Shocking.

that's not the "Gallup" poll that I cited.... but nice try. Gallup showed Carter leading by a few points right up to the election....He lost in a landslide

and yes, I want Obama to lose. I don't hide that fact. A low turnout by Democrats is highly probable - Obama is not blank canvas that he was in 2008- he has a record now...it stinks.

 

[AceRothstein]

Fact remains that democrats are being oversampled by most pollsters, and they're even admitting as much in the disclosure of their methodologies.

But don't believe your lying eyes.

Why do you think this is? What do private polling companies have to gain from being wrong as all the cons are claiming? I think this is the fifth time I've asked this question, maybe this time I will get an answer.

I don't delve into speculation of motivations of people I don't know.

Fact remains that when the methodologies are being revealed, democrats are being oversampled....Why the thumb is on the scale is irrelevant to the fact that it is.

Maybe they aren't wrong, we'll see.

I don't understand why people think the polling agencies would purposely put out incorrect numbers. They aren't in the business of being wrong, they do polling/surveys for more than just elections. Who is going to hire someone that is consistently wrong?

 

[Article 15]

I think we are seeing the exact same pattern we see every cycle- crazy polls that mean nothing. Political polling is not an exact science- it's a SWAG.

I do remember Carter being way, WAY WAY!! ahead of Reagan according to Gallup in October of 1980.....we all know how that worked out!

Bottom line - This is a very close race and it's all going to come down to voter turnout in the swing states like Ohio and Florida. Hopefully the media continues to advance the narrative that "Obama's got this"....that way more of his supporters stay home on election day..why bother to vote? He's got this....he can't lose.

A republican hoping for a low voter turnout. Shocking.

that's not the "Gallup" poll that I cited.... but nice try. Gallup showed Carter leading by a few points right up to the election....He lost in a landslide

and yes, I want Obama to lose. I don't hide that fact. A low turnout by Democrats is highly probable - Obama is not blank canvas that he was in 2008- he has a record now...it stinks.

You can bitterly cling to whatever one individual poll from back in 1980 that you want to, the fact remains that the idea that Carter was somehow beating Reagan is a myth.

 

[NYcarbineer]

Fact remains that democrats are being oversampled by most pollsters, and they're even admitting as much in the disclosure of their methodologies.

But don't believe your lying eyes.

Why do you think this is? What do private polling companies have to gain from being wrong as all the cons are claiming? I think this is the fifth time I've asked this question, maybe this time I will get an answer.

I don't delve into speculation of motivations of people I don't know.

Fact remains that when the methodologies are being revealed, democrats are being oversampled....Why the thumb is on the scale is irrelevant to the fact that it is.

So the Foxnews poll has 41% Democrats and 38% Republicans.

Why is that number wrong, and what is the right number, and how do you know what the right number is?

 

[nodoginnafight]

I think we are seeing the exact same pattern we see every cycle- crazy polls that mean nothing. Political polling is not an exact science- it's a SWAG.

I do remember Carter being way, WAY WAY!! ahead of Reagan according to Gallup in October of 1980.....we all know how that worked out!

Bottom line - This is a very close race and it's all going to come down to voter turnout in the swing states like Ohio and Florida. Hopefully the media continues to advance the narrative that "Obama's got this"....that way more of his supporters stay home on election day..why bother to vote? He's got this....he can't lose.

A lot of very good points I have to agree with here. Yes, Reagan overcame a significant lead - most credit his debate performace for that. And yes, we still have the debates to go. No reason Romney can't pick up a lot of votes in the debates. And I also agree that turnout is crucial. I don't think many polls are lying or skewing anthing, but yes - SOME do. I think that's why I always trust the RCP poll average more than any individual poll. They tend to average out the outliers.

 

[Oddball]

Why do you think this is? What do private polling companies have to gain from being wrong as all the cons are claiming? I think this is the fifth time I've asked this question, maybe this time I will get an answer.

I don't delve into speculation of motivations of people I don't know.

Fact remains that when the methodologies are being revealed, democrats are being oversampled....Why the thumb is on the scale is irrelevant to the fact that it is.

Maybe they aren't wrong, we'll see.

I don't understand why people think the polling agencies would purposely put out incorrect numbers. They aren't in the business of being wrong, they do polling/surveys for more than just elections. Who is going to hire someone that is consistently wrong?

Maybe they are...Maybe they aren't...But the oversampling is right there in black and white.

Personally, I think they're using turnout numbers from '08 as a demographic template...That was a record year for democrat turnout, unlikely to repeat itself.

Meanwhile, a lot of polls cant get Obiedoodle over the 50% threshold, despite the oversampling of democrats...That's not good news.

 

[nodoginnafight]

I don't delve into speculation of motivations of people I don't know.

Fact remains that when the methodologies are being revealed, democrats are being oversampled....Why the thumb is on the scale is irrelevant to the fact that it is.

Maybe they aren't wrong, we'll see.

I don't understand why people think the polling agencies would purposely put out incorrect numbers. They aren't in the business of being wrong, they do polling/surveys for more than just elections. Who is going to hire someone that is consistently wrong?

Maybe they are...Maybe they aren't...But the oversampling is right there in black and white.

Personally, I think they're using turnout numbers from '08 as a demographic template...That was a record year for democrat turnout, unlikely to repeat itself.

Meanwhile, a lot of polls cant get Obiedoodle over the 50% threshold, despite the oversampling of democrats...That's not good news.

by "right there in black and white" are you reffering to your post or a blog?
let's see your evidence

 

[nodoginnafight]

So these "skewed" polls... have they changed their formula for this election? Have they departed from the formulas they used in past elections to more accurately predict outcomes than the "unskewed" poll so often cited?

 

[nodoginnafight]

Unfounded declarations - no matter how forcefully presented - just don't outweigh the numbers and the facts. Is it over - not by a long shot. Is Obama ahead right now - sure looks like it, Is the margin of Obama's lead inflated? In some polls - yes, I believe it is inflated. In some polls it is underestimated. Are the vast majority of polls being "skewed" in Obama's favor - nope, no one has presented a shred of evidence to support that dubious claim. We will see.

 

[NYcarbineer]

ALL the polls currently have Obama winning, if the election were held today.

Rasmussen is the outlier, at +1 today.

Take note, Rasmussen was close in 2008, but Rasmussen was in line with everyone else throughout the 2008 race.

Rasmussen gets it right when most others get it right.

 

[nodoginnafight]

about this time in 2008 Rassmussen was calling the race a tie in a poll conducted between 9/17 and 9/20.
less than a week later 9/24 - 9/26, he had Obama +6.

I smell a face-saving correction. Or does he expect us to believe that 6% of likely voters changed their minds in four days?

He is all over the place this year too.
9/7 - 9/9 he's got Obama +5
less than a week later (9/14-9/16) he's got Romney at +2

 

[mamooth]

Rasmussen always tightens up their polls in the last 2 weeks, after sucking hard for the previous months. That way, they can point to the final result and say "Look how accurate we were!". It fools the low-IQ crowd, the conservative true believers.

The conservative argument about skewed polls is "More people are now self-identifying as Democrats, so the polls must be rigged!". Alas, the conclusion does not follow from the premise. More people self-identifying as Democrats simply means ... more people are self-identifying as Democrats. No conspiracy is necessary to explain it. The polls are doing what polls are supposed to do, track and report opinions. It's only Rasmussen that twists the data by applying some bizarre artificial turnout model.

 

[healthmyths]

Even the latest FOXNews poll has Obama up 5 points nationally.

The people who keeping banging the skewed polls drum are just trying to cope.

Simple test for you..
OK if 100 democrats and 50 GOP And 25 Independents that lean Democrat are asked:
Who do you think will be elected Obama or Romney?

What do you think the chances are that there will be MORE people saying Obama?

 

[nodoginnafight]

Even the latest FOXNews poll has Obama up 5 points nationally.

The people who keeping banging the skewed polls drum are just trying to cope.

Simple test for you..
OK if 100 democrats and 50 GOP And 25 Independents that lean Democrat are asked:
Who do you think will be elected Obama or Romney?

What do you think the chances are that there will be MORE people saying Obama?

or perhaps a better question:
If 37 people who identify themselves as democrats, 34 people who identify themselves as republicans, and 29 people who identify themselves as independents are asked their preference, who do you think will come out on top?

(I don't think there is any reason to exagerate the splits - unless you are trying to create an unrealistic picture of what is happening)

 

[nodoginnafight]

Rasmussen always tightens up their polls in the last 2 weeks, after sucking hard for the previous months. That way, they can point to the final result and say "Look how accurate we were!". It fools the low-IQ crowd, the conservative true believers.

The conservative argument about skewed polls is "More people are now self-identifying as Democrats, so the polls must be rigged!". Alas, the conclusion does not follow from the premise. More people self-identifying as Democrats simply means ... more people are self-identifying as Democrats. No conspiracy is necessary to explain it. The polls are doing what polls are supposed to do, track and report opinions. It's only Rasmussen that twists the data by applying some bizarre artificial turnout model.

It certainly happened in 2008 if you look at the Rassmussen polls in September and October of 2008. In one week during September, Rassmussen went from dead heat to Obama +6.

But they didn't correct the state polls and those continued to show the 4 point republican bias.

And in 2010 they didn't even appear to make any pretence. In Hawaii they overestimated the republican sentorial candidate's support by 40 points.

 

[Zander]

I think we are seeing the exact same pattern we see every cycle- crazy polls that mean nothing. Political polling is not an exact science- it's a SWAG.

I do remember Carter being way, WAY WAY!! ahead of Reagan according to Gallup in October of 1980.....we all know how that worked out!

Bottom line - This is a very close race and it's all going to come down to voter turnout in the swing states like Ohio and Florida. Hopefully the media continues to advance the narrative that "Obama's got this"....that way more of his supporters stay home on election day..why bother to vote? He's got this....he can't lose.

A lot of very good points I have to agree with here. Yes, Reagan overcame a significant lead - most credit his debate performace for that. And yes, we still have the debates to go. No reason Romney can't pick up a lot of votes in the debates. And I also agree that turnout is crucial. I don't think many polls are lying or skewing anthing, but yes - SOME do. I think that's why I always trust the RCP poll average more than any individual poll. They tend to average out the outliers.

garbage in = garbage out

 

[nodoginnafight]

I think we are seeing the exact same pattern we see every cycle- crazy polls that mean nothing. Political polling is not an exact science- it's a SWAG.

I do remember Carter being way, WAY WAY!! ahead of Reagan according to Gallup in October of 1980.....we all know how that worked out!

Bottom line - This is a very close race and it's all going to come down to voter turnout in the swing states like Ohio and Florida. Hopefully the media continues to advance the narrative that "Obama's got this"....that way more of his supporters stay home on election day..why bother to vote? He's got this....he can't lose.

A lot of very good points I have to agree with here. Yes, Reagan overcame a significant lead - most credit his debate performace for that. And yes, we still have the debates to go. No reason Romney can't pick up a lot of votes in the debates. And I also agree that turnout is crucial. I don't think many polls are lying or skewing anthing, but yes - SOME do. I think that's why I always trust the RCP poll average more than any individual poll. They tend to average out the outliers.

garbage in = garbage out

I don't understand the point you are making here. Could you clarify for me?