Surprise! Oversampling Dems Puts Obama in Lead

[Dr.House]

How is a pollster supposed to predict what the ratio of Democrat to Republican turnout will be?

Plenty of methods for garnering current voter/party enthusiasm or apathy, then apply to sampling...

you mean like doing a poll of registered voters, asking party affiliation? That kind of method?

You can hang your hat on polls that oversample by wide margins, but if Barry doesn't get that turnout the results won't be good for your side...

Live by the 2008 sword, die by the 2008 sword...

 

[NYcarbineer]

Irrelevant to this year's election...

If 0bama is only up by small numbers when oversampling democratics by 10% or more in some cases, then if your boy doesn't pull in those turnout percentages in November, he is toast...

Clue: He's not going to pull in 2008 numbers...

How is a pollster supposed to predict what the ratio of Democrat to Republican turnout will be?

Plenty of methods for garnering current voter/party enthusiasm or apathy, then apply to sampling...

If you want to pretend that voter turnout will be like it was in 2008, nobody can stop you...

But looking at some of the recent ones, they are going way beyond the difference in 2008... Bad news for you fluffers is that Barry is not doing well with even THAT sampling...

Barry is in trouble... I know it, he knows it, but it looks like you fluffers are oblivious...

Or, they could do what Rasmussen does, make people pay to see the internals of their polls.

The president has 94 electoral votes to work with from 2008. That's 94 electoral votes he can afford to lose from 2008 before he loses in 2012.

 

[8537]

Plenty of methods for garnering current voter/party enthusiasm or apathy, then apply to sampling...

you mean like doing a poll of registered voters, asking party affiliation? That kind of method?

You can hang your hat on polls that oversample by wide margins, but if Barry doesn't get that turnout the results won't be good for your side...

Live by the 2008 sword, die by the 2008 sword...

So you claim there are "plenty of methods" but you can't actually explain what they might be?

How do you know they over-sampled Democrats?

 

[Dr.House]

you mean like doing a poll of registered voters, asking party affiliation? That kind of method?

You can hang your hat on polls that oversample by wide margins, but if Barry doesn't get that turnout the results won't be good for your side...

Live by the 2008 sword, die by the 2008 sword...

So you claim there are "plenty of methods" but you can't actually explain what they might be?

I don't work for a polling company... I'm confident they have methods, though...

 

[8537]

You can hang your hat on polls that oversample by wide margins, but if Barry doesn't get that turnout the results won't be good for your side...

Live by the 2008 sword, die by the 2008 sword...

So you claim there are "plenty of methods" but you can't actually explain what they might be?

I don't work for a polling company... I'm confident they have methods, though...

Your "confident they have methods" because...well...that just "feels" right or what?

I agree - they have methods. Namely, they do a poll of a few thousand randomly-sampled people asking party affiliation.

 

[Dr.House]

How is a pollster supposed to predict what the ratio of Democrat to Republican turnout will be?

Plenty of methods for garnering current voter/party enthusiasm or apathy, then apply to sampling...

If you want to pretend that voter turnout will be like it was in 2008, nobody can stop you...

But looking at some of the recent ones, they are going way beyond the difference in 2008... Bad news for you fluffers is that Barry is not doing well with even THAT sampling...

Barry is in trouble... I know it, he knows it, but it looks like you fluffers are oblivious...

Or, they could do what Rasmussen does, make people pay to see the internals of their polls.

The president has 94 electoral votes to work with from 2008. That's 94 electoral votes he can afford to lose from 2008 before he loses in 2012.

Which boils down to a few key swing states...

Barry ahead in polls that sample heavily toward D's by small margins is not good for the empty chair...

 

[Dr.House]

So you claim there are "plenty of methods" but you can't actually explain what they might be?

I don't work for a polling company... I'm confident they have methods, though...

I agree - they have methods.

I'm sure they do... They may not all be the same, though...

 

[8537]

I don't work for a polling company... I'm confident they have methods, though...

I agree - they have methods.

I'm sure they do... They may not all be the same, though...

They all use the same method for most purposes. Namely, they poll a whole bunch of people and ask - and that's why all of the samplings from Fox to Gallup to PPP return breakdowns within each other's margin of error.

 

[Care4all]

Obama's convention bounce seems to have evaporated. Rasmussen today has Romney up 3 points. Today's Gallup tracking poll had Obama up 5, a loss of 2 points over two days. Gallup uses a 7 day rolling average, so expect this number to continue to come down over the next several days. Undaunted, the media are trumpeting new polls with show Obama with 3 point lead nationally and significant leads in the battleground states of Florida, Ohio and Virginia. As is becoming routine, these new polls again oversample Democrats.

I realize oversampling Democrats is simply part of an effort to create an air of inevitability for Obama's reelection. Its a narrative the media is desperate to foster.

Surprise! Oversampling Dems Puts Obama in Lead

Registered Democrats still dominate the political playing field with more than 42 million voters, compared to 30 million Republicans and 24 million independents.

Voters are leaving main parties in droves

There ARE more registered democrats than registered republicans, even with democrats loosing many (more than 2.5 million) of their registered voters to independents since 2008.

The polling should reflect this....?

 

[wjmacguffin]

DOJ threaten to sue Gallap because of the poll numbers before were not in favor of The DEMS Messiah .. Department of Justice sues Gallup | The Daily Caller

Actually, the DOJ joined a lawsuit (no threat) about a month ago accusing Gallup of overcharging the federal government. There's nothing about poll numbers listed in the story you quoted. Did you not actually read the story you quoted or are you just that bad at lying?

Perhaps you're one of those people who eschew facts for conspiracy and believe the whole overcharging aspect is a smoke screen to bully Gallop into giving Obama better numbers.

 

[Dr.House]

I agree - they have methods.

I'm sure they do... They may not all be the same, though...

They all use the same method for most purposes. Namely, they poll a whole bunch of people and ask - and that's why all of the samplings from Fox to Gallup to PPP return breakdowns within each other's margin of error.

Where does voter apathy factor into that method when using Registered Voters?

 

[mamooth]

There was a record turnout in 2008 -- for _both_ sides. The Democrats didn't get any advantage from the high turnout. The Republican advantage in percentage of registered voters who voted was pretty much the same as it always was, the same as it is now. That's why it's silly to claim that a 2008 turnout model is favoring Democrats.

 

[NYcarbineer]

Plenty of methods for garnering current voter/party enthusiasm or apathy, then apply to sampling...

If you want to pretend that voter turnout will be like it was in 2008, nobody can stop you...

But looking at some of the recent ones, they are going way beyond the difference in 2008... Bad news for you fluffers is that Barry is not doing well with even THAT sampling...

Barry is in trouble... I know it, he knows it, but it looks like you fluffers are oblivious...

Or, they could do what Rasmussen does, make people pay to see the internals of their polls.

The president has 94 electoral votes to work with from 2008. That's 94 electoral votes he can afford to lose from 2008 before he loses in 2012.

Which boils down to a few key swing states...

Barry ahead in polls that sample heavily toward D's by small margins is not good for the empty chair...

For the third time, Obama leads by 5 in Ohio in the PPP poll that sampled +4 Democrats over Republicans,

when the 2008 Ohio exit polls counted +8 Democrats over Republicans actually voting.

There's your lower D turnout factored in, and Obama is still ahead.

 

[NYcarbineer]

It's interesting...

The 2 pollsters in Realclear's current average giving Obama the biggest lead?

CNN and Foxnews.

The 2 pollsters in 2008 that hit the 7 point Obama win on the nose?

CNN and Foxnews.

 

[kaz]

Guy, you act like we really CARE if you Teabagging idiots have any ideas.

If you guys are against it, it's probably because it's humane and sensible.

But I'll give you one example- YOu guys were all for individual mandates- until the black guy did them.

Romneycare was wonderful... until they started calling it Obamacare.

Then it became "Marxists" - Which is teabagger for the N-word.

Socialized medicine is a reason that fiscal conservatives would support Obama. I can't say you ever pass on an opportunity to show you're mentally retarded.

So in your mind fiscal Conservatives support saving money over saving lives? One, socialized medicine sacrifices quality and quantity to provide cheaper services so that is a problem. Second, we are not just talking socialized medicine we are also talking about subsidized medicine and in the end it will end up costing more not less.

Senior moment there? I'm against socialized medicine, I was mocking Joe the Racist. I asked how Obama was conservative and he said Obama's for socialized medicine.

 

[kaz]

A better question, what is the Democrat versus Republican ratio for the general population being sampled?

That's an irrelevant question. The oversampling argument isn't that they sampled more Democrats than Republicans, it's that they sampled more Democrats than there are. One reason for that is that Democrats are very eager to tell people they are for the Marxist in the Oval Office and the rest of us are more likely to not answer the phone. I live in North Carolina and I get endless calls from polling companies, none of which have gotten past my answering machine. Another reason is that Democrats are less likely to show up at the polls, particularly if it's raining or cold. This is referred to as "likely" voters, and determining who is "likely" to vote is pure art. There are lots of ways they do it, none of them without shortcomings.

 

[kaz]

Plenty of methods for garnering current voter/party enthusiasm or apathy, then apply to sampling...

you mean like doing a poll of registered voters, asking party affiliation? That kind of method?

You can hang your hat on polls that oversample by wide margins, but if Barry doesn't get that turnout the results won't be good for your side...

Live by the 2008 sword, die by the 2008 sword...

That is a huge issue for Democrats. Republicans had that issue with McCain. Many people who would have voted for him had they gone to the polls didn't. This election, Obama is dominating the "don't give a fuck" crowd.

One reason for that is that they frequently determine "likely" voters based on the question, "did you vote in 2008?" A lot of people who say Obama and yes will be busy not caring enough to show up this time. It's really hard to think of a sequence where anything breaks in favor of Obama unless the economy improves, and there isn't much time for that and no indication at all it's going to happen.

 

[expatriate]

you mean like doing a poll of registered voters, asking party affiliation? That kind of method?

You can hang your hat on polls that oversample by wide margins, but if Barry doesn't get that turnout the results won't be good for your side...

Live by the 2008 sword, die by the 2008 sword...

That is a huge issue for Democrats. Republicans had that issue with McCain. Many people who would have voted for him had they gone to the polls didn't. This election, Obama is dominating the "don't give a fuck" crowd.

One reason for that is that they frequently determine "likely" voters based on the question, "did you vote in 2008?" A lot of people who say Obama and yes will be busy not caring enough to show up this time. It's really hard to think of a sequence where anything breaks in favor of Obama unless the economy improves, and there isn't much time for that and no indication at all it's going to happen.

admittedly, I am physically removed from the electorate - or most of it anyway - but I have lengthy political discussions with my three adult children living stateside... I have asked them if they, themselves are "not giving a fuck" about this election, and, if their friends and colleagues are apathetic about this election. They don't see it. I think it is kind of funny - in a pathetic sort of way - how Romney supporters will grasp at anything to try and spin the polling data and convince us - but most probably, themselves - that even though Obama leads in all the key swing states and odds makers have him as a 2 to 1 favorite, Mitt is actually doing quite well and will obviously prevail on election day.

 

[JoeB131]

Socialized medicine is a reason that fiscal conservatives would support Obama. I can't say you ever pass on an opportunity to show you're mentally retarded.

YOu jokers are nominating Romney, the guy who made the Beta Version...

Except when Romney did it, it was called "A free enterprise solution to the Problem"... The Cato Institute and Heritage Foundation used those exact words.

Until the Black Guy Did It.

Then it became "Marxism" and "Socialized Medicine".

 

[8537]

I'm sure they do... They may not all be the same, though...

They all use the same method for most purposes. Namely, they poll a whole bunch of people and ask - and that's why all of the samplings from Fox to Gallup to PPP return breakdowns within each other's margin of error.

Where does voter apathy factor into that method when using Registered Voters?

The most recent polls aren't registered voters. They are almost exclusively Likely Voters.