Tweet
Pt I
Eight Reasons Polls Showing Obama Winning Are Dead Wrong
by John Nolte 1 Nov 2012, 9:41 AM PDT
The National Journal's Reid Wilson has written a superb piece that goes a long way towards explaining the one issue bringing right and left together this campaign season: What is going on with these polls? There's no question that there's one consensus of pollsters that show Mitt Romney leading (Rasmussen, Gallup) and another, larger group showing Obama with an edge -- especially in the all-important swing states.
In a nutshell, what we have are two camps of pollsters predicting two very different types of electorates will show up at the polls or have already voted. This, by the way, includes the very detailed internal polling the Obama and Romney campaigns are spending millions on. Boiled down: Will the 2012 electorate look like 2008, 2004, or somewhere in-between?
2008 was a wave election, where demoralized Republicans stayed home and Democrats, energized by the hope and the change, came out in record numbers. The result was an electorate in which Democrats enjoyed a record turnout advantage of 7% over Republicans: 39% Democrat, 32% Republican, and 29% Independent -- or D+7.
In 2004, the turnout was 37% Democrat, 37% Republican, and 26% Independent -- or D+0.
With this in mind, what we're seeing in many of these state polls showing Obama leading are pollsters who believe 2012 is going to look a lot more like 2008 than 2004. And it has nothing to do with pollsters putting a thumb on the scale and weighting Party ID in a way that benefits Obama. It also has nothing to do with response rates, cell phones, or people lying to pollsters.
There's no magical thinking behind what we're seeing in these polls. It all comes down to choices pollsters are making with respect to demographics and screening for the all-important Likely Voter.
DEMOGRPAHICS:
National Journal:
This explains more than why the campaigns' internal polling looks so different; it also explains what we're seeing in this daily flood of public polls. Marist, Quinnipiac, etc. all apparently agree with Team Obama's vision of what the 2012 electorate will look like. Because of this, they're including larger samples of groups inclined to vote for Obama. This is why Obama leads in most swing state polls and explains the D+5 to D+8 samples everyone's talking about.
Likely Voter Screens
Something else is happening, though, that the National Journal missed. Pollsters aren't just making demographic choices favorable to Obama, some are also loosening the Likely Voter screening process, sometimes to absurd levels.
Pollsters generally work with three pools of voters: All Adults and Registered Voters (RVs) -- which is self-explanatory; and Likely Voters (LVs) -- those whom pollsters believe will actually turn out to vote. It's just a fact that, as you tighten your screen from All Adults to LVs to RVs, more and more Democrats are filtered out. But the screen for LVs is a subjective one; it's left up to the pollster to decide who qualifies as an LV based on whatever set of questions are asked.
And in the end, published poll results and the ensuing headlines are based on LVs.
Once again, pollsters can claim they aren't weighting Party ID to benefit Obama, but if their LV screen is loosened, it has the exact same effect, because more Democrats make it into the LV pool of voters. The best example of this arrived yesterday with three Quinnipiac polls that showed Obama winning in Ohio, Virginia, and Florida.
In the wave election of 2008, Democrats enjoyed turnout advantages of D+5 in OH; D+6 in VA; and D+3 in Florida. Yesterday's much publicized Quinnipiac polls predicted Obama's turnout advantage in those states would actually increase this year: VA: D+8 -- FL: D+7 -- OH: D+8.
But if Quinnipiac doesn't weight for Party ID, where did that result come from?
Those three Quinnipiac polls are really RV polls. Normally, when looking for the more accurate LV pool of voters, most pollsters tighten the screen to qualify as a LV with the kinds of questions that screen out somewhere between 20 to 30% of RVs. Quite incredibly, Quinnipiac only screened out 4% of RVs. If you release a poll where 96% of RVs make it through the LV screen, you are releasing a RV poll -- which of course includes more Democrats and favors Obama.
Quinnipiac is just one example of another reason we're seeing most state and some national polls claiming Obama has a small but persistent lead thanks to a D+5 to D+8 turn out advantage.
Where this matters most, naturally, is with a CorruptMedia that reports these skewed polls as gospel, which in turn serves the Obama campaign's Inevitability Narrative.
The real question, though, is who's right? Will Tuesday's electorate look like the wave election of 2008 or something closer to 2004.?
Obviously, no one can answer that question with any certainty, but there is some data and no small amount of common sense that says pollsters using the pro-Obama 2008 model are the ones climbing further out on a limb than the rest.
by John Nolte 1 Nov 2012, 9:41 AM PDT
The National Journal's Reid Wilson has written a superb piece that goes a long way towards explaining the one issue bringing right and left together this campaign season: What is going on with these polls? There's no question that there's one consensus of pollsters that show Mitt Romney leading (Rasmussen, Gallup) and another, larger group showing Obama with an edge -- especially in the all-important swing states.
In a nutshell, what we have are two camps of pollsters predicting two very different types of electorates will show up at the polls or have already voted. This, by the way, includes the very detailed internal polling the Obama and Romney campaigns are spending millions on. Boiled down: Will the 2012 electorate look like 2008, 2004, or somewhere in-between?
2008 was a wave election, where demoralized Republicans stayed home and Democrats, energized by the hope and the change, came out in record numbers. The result was an electorate in which Democrats enjoyed a record turnout advantage of 7% over Republicans: 39% Democrat, 32% Republican, and 29% Independent -- or D+7.
In 2004, the turnout was 37% Democrat, 37% Republican, and 26% Independent -- or D+0.
With this in mind, what we're seeing in many of these state polls showing Obama leading are pollsters who believe 2012 is going to look a lot more like 2008 than 2004. And it has nothing to do with pollsters putting a thumb on the scale and weighting Party ID in a way that benefits Obama. It also has nothing to do with response rates, cell phones, or people lying to pollsters.
There's no magical thinking behind what we're seeing in these polls. It all comes down to choices pollsters are making with respect to demographics and screening for the all-important Likely Voter.
DEMOGRPAHICS:
National Journal:
Republicans and Democrats alike believe the African American vote is unlikely to change between 2008 and 2012. But they differ dramatically on the number of Hispanic voters who will show up at the polls -- a key factor in critical battleground states like Colorado and Nevada. Republicans believe turnout will be down, depressed by Obama's failure to pursue immigration reform during his first term. Democrats think the booming number of Hispanic residents means their share of the electorate will only increase.
The same argument happens over younger voters. In 2008, 18 percent of the electorate was made up of voters between 18 and 29 years old. That's higher than the percentage has been in recent presidential years, when the youth vote has made up around 15 or 16 percent. Republicans believe the younger share of the electorate will slide slightly, and that Obama will win fewer of those voters anyway.
The manifestation of these disagreements is evident in polling weights. Most Republican pollsters are using something close to a 2008 turnout model, with the same percentage of white, black and Hispanic voters as the electorate that first elected Obama. Most Democratic pollsters are a little more bullish on minority turnout, which helps explain some of the difference between the two sides.
The same argument happens over younger voters. In 2008, 18 percent of the electorate was made up of voters between 18 and 29 years old. That's higher than the percentage has been in recent presidential years, when the youth vote has made up around 15 or 16 percent. Republicans believe the younger share of the electorate will slide slightly, and that Obama will win fewer of those voters anyway.
The manifestation of these disagreements is evident in polling weights. Most Republican pollsters are using something close to a 2008 turnout model, with the same percentage of white, black and Hispanic voters as the electorate that first elected Obama. Most Democratic pollsters are a little more bullish on minority turnout, which helps explain some of the difference between the two sides.
Likely Voter Screens
Something else is happening, though, that the National Journal missed. Pollsters aren't just making demographic choices favorable to Obama, some are also loosening the Likely Voter screening process, sometimes to absurd levels.
Pollsters generally work with three pools of voters: All Adults and Registered Voters (RVs) -- which is self-explanatory; and Likely Voters (LVs) -- those whom pollsters believe will actually turn out to vote. It's just a fact that, as you tighten your screen from All Adults to LVs to RVs, more and more Democrats are filtered out. But the screen for LVs is a subjective one; it's left up to the pollster to decide who qualifies as an LV based on whatever set of questions are asked.
And in the end, published poll results and the ensuing headlines are based on LVs.
Once again, pollsters can claim they aren't weighting Party ID to benefit Obama, but if their LV screen is loosened, it has the exact same effect, because more Democrats make it into the LV pool of voters. The best example of this arrived yesterday with three Quinnipiac polls that showed Obama winning in Ohio, Virginia, and Florida.
In the wave election of 2008, Democrats enjoyed turnout advantages of D+5 in OH; D+6 in VA; and D+3 in Florida. Yesterday's much publicized Quinnipiac polls predicted Obama's turnout advantage in those states would actually increase this year: VA: D+8 -- FL: D+7 -- OH: D+8.
But if Quinnipiac doesn't weight for Party ID, where did that result come from?
Those three Quinnipiac polls are really RV polls. Normally, when looking for the more accurate LV pool of voters, most pollsters tighten the screen to qualify as a LV with the kinds of questions that screen out somewhere between 20 to 30% of RVs. Quite incredibly, Quinnipiac only screened out 4% of RVs. If you release a poll where 96% of RVs make it through the LV screen, you are releasing a RV poll -- which of course includes more Democrats and favors Obama.
Quinnipiac is just one example of another reason we're seeing most state and some national polls claiming Obama has a small but persistent lead thanks to a D+5 to D+8 turn out advantage.
Where this matters most, naturally, is with a CorruptMedia that reports these skewed polls as gospel, which in turn serves the Obama campaign's Inevitability Narrative.
The real question, though, is who's right? Will Tuesday's electorate look like the wave election of 2008 or something closer to 2004.?
Obviously, no one can answer that question with any certainty, but there is some data and no small amount of common sense that says pollsters using the pro-Obama 2008 model are the ones climbing further out on a limb than the rest.
(seriously, can we lose that 10000 character limit on posts?)
Comment