Posted on 09/26/2012 6:04:30 PM PDT by SeekAndFind
The overwhelming majority of public opinion polls show President Obama cruising to re-election. State polls of Ohio, Florida, Virginia, Wisconsin and Iowa are now consistently showing Obama with leads of five points or more.
But Romney is not behind in the polls because liberal Republicans are being turned off or independents are moving toward Obama. Most polls show Romney doing very well within his own party and holding his own with independents. Romney is dropping in the polls because pollsters are predicting a Democratic edge in turnout that will make 2008 look like a good year for the GOP.
For example, the most recent NYT/CBS/Quinnipiac poll of Virginia shows Romney carrying Republicans 95-3 and independents 54-43. In fact, if the poll were weighted the same way that the August NYT/CBS/Quinnipiac poll of Virginia was weighted, Romney would be leading Obama. But because of the increased Democratic lean of the sample, he instead trails Obama by four points.
But the oversampling of Democrats is only a small part of a larger problem. In addition, pollsters are undersampling groups that are sympathetic to Republicans, like evangelicals and people who make over $100,000 a year, and oversampling groups that are sympathetic to Democrats, like mainline Protestants and people who make less than $50,000 a year.
Here’s what Marist is predicting the 2012 electorate will look like in various swing states, relative to 2008:
Republicans Democrats
Iowa -6.1% Iowa +5.9%
Colorado +3.2% Colorado +13.3%
Virginia -21.2% Virginia -20.5%
Florida +11.7% Florida +10.8%
Ohio -9.7% Ohio -2.6%
Income >100k Income <50k
Iowa -19% Iowa +11.6 %
Colorado -25% Colorado +68%
Virginia -14.3% Virginia +20%
Florida -29.2% Florida +30.8
Ohio -14.3% Ohio +11.4%
Evangelicals
Iowa -22.5%
Colorado -19%
Virginia -25%
Florida +4%
Ohio -30%
I don’t mean to single out Marist. It’s representative of most of the media polls showing the president on track for re-election.
So why are these polls skewing so far left? Conservative conspiracy theories tying the polls to mainstream media bias fail to acknowledge the need for pollsters to retain professional credibility. They can’t do that if they are consistently making bad predictions. I think the problem is that pollsters are so focused with ensuring that Democratic-leaning groups — especially minorities — are fairly represented in their polls that they’re failing to ensure that Republican-leaning groups are also fairly represented in their polls.
Of course, it’s possible that the pollsters are right. Perhaps minority voters will turn out in even higher numbers in 2012 than they did in 2008. Maybe Republicans who are not sold on Mitt Romney will stay home. If that’s the case, President Obama really will cruise to re-election.
Brandon J. Gaylord, the editor-in-chief of HorseRacePolitics.com, is a graduate of George Washington University’s Graduate School of Political Management. Brandon got his start in politics as an intern in Vice President Richard Cheney’s Office of Political Affairs.
I’ve been hearing this word ‘oversampling’ a lot lately, which reminds me of the time I oversampled in Napa Valley, California. But isn’t it true that more citizens of this veil of tears register as Democrats than as Republicans?
The oversampling is intended to reflect the statistical skew in turnout. It’s not a bad practice providing it’s relevant.
Risking credibility with the knowledge they and their surrogates can be agents provacateur both during and after election day, crowing incessantly about poor, disenfranchised morons and the evil Tea Party voters. They will work at promoting fraud behind the scenes while covering it up in the public arena. Their credibility depends on it.
In 2010 we saw an unbelievable turnout for the GOP in what was widely considered to be a referendum on Obama. The pollsters completely ignore that. 2008 was the “hopey changey thing” and that is gone.
We will see.
I wouldn’t be too sure of the premise that pollsters’ credibility ride on the election results. Nate Silver and the MSM widely discredit Rasmussen, despite Ras nailing 2008 and 2010. The consumer is always right, and the consumer is the MSM.
Depends on the state. A reasonable sample would be Dem +4 nation wide. That assumes that the election will fall some place between 2004 and 2008 as far as turn out goes.
However here is the problem for the pollsters. Since 2010 Reps have been posting significantly higher registration numbers in most of the battle ground states. So it is possible that even a +4 Dem sample is still too high based on current party registration numbers
So Dems point to 2008 and claim they should get a bigger sample Repbs point to 2010 and say the Dems should get a smaller sample
Problem is the pollsters are using 2008 as a base line and assuming Obama will do better in 2012.
See any evidence on the ground that Obama is doing better with the voters in 2012 then he did in 2008?
OK, I think I get it now.
Dem voters take notice:
It’s over.
The election will be a landslide for O’Bozo.
No need to drop the bong, stop chooming or stop trolling for gays just to go vote against Romney.
You can stay home on election day - just watch ABC, NBC, CBS or MSNBC and CNN - your side has already won.
In a word, no. He is preforming worse in every group. Even african americans who came out in big number in 2008 will fall some. Democrats always get 90% of the votes, it will just be a bit smaller total votes this time.
Why is it so hard for Freepers to admit Romney is getting his ass handed to him?
This may not even be close folks.
Why are we losing? Hell if i know.
RE: Why is it so hard for Freepers to admit Romney is getting his ass handed to him?
TRANSLATION: You believe the polls.
Problem with your assumption is it is wholly based on pollsters who are using 2008 turn out as a base line for their polling sample an are assuming Obama will do better in 2012.
See any evidence on the ground that Obama is doing better with the voters in 2012 then he did in 2008?
OK, so we say that the bias, or sampling should reflect the most recent turnout numbers to get most accurate results. But then, turnout varies from election to election, as you point out, and is not going to be in 2012 like 2008 or 2010, which in turn tells me what I’ve always thought that is that these polls are in the end more or less worthless.
If you get into weighting poll results to reflect other polls showing a different distribution by party respondents you'll invariably screw up ~ several reasons for that we need not get into here.
However, the number of so-called independents has continued to increase such that you have pollsters telling us about 35% are Republicans, 35% are Democrats and 30% are Independents.
So, an answer to a question ~ what party do independents regularly vote for? it could be they don't actually vote ~ they just use that self-description as a way to cover up that fact. with a known 90 million adults in this country not voting at all, there are plenty of potential voters to go around to fill that bill. Besides, you don't want to accidentally or otherwise let a Democrat group find out you're a Democrat or they'll add your number to their robocall operation ~ and that can be a fate worse than death in some areas!
i tend to discount whatever it is pollsters say independents want ~ same with 'the broad middle' ~ a theoretical construct that never existed in American politics ~ not at any time in more than 2 centuries!
For the most part as long as Obamugabe is at less than 50% Romney and the Mittbots will probably win.
That’s a very good explanation.
My own is that people lie, people equivocate, people don’t commit, people say what they think others want to hear, people say whatever comes to their minds when no cost to themselves is involved.
And, I just don’t believe the very premise of polling (and the “science” of statistics), which states that a poll of a careful selection of, say, 1024 respondents can accurately reflect opinions of millions. And the 3% or 4% error margin that the pollsters cite is of course their own self serving number pulled out of a hat. (Lovers of statistics will call me an ignoramus, I expect.)
Used to be so...but not anymore :-)
In 2010 the Republican turnout was 44 million.
In 2008 the Democrat turnout was 69 million.
In 2010 the Democrat turnout was 39 million.
What that means is the stunning Republican turnout in 2010 represented a fall off of 15 million voters. The Democrat turnout in 2010 represented a fall off of 30 million voters.
(That's based on House votes since not every state had a Senator up for election)
That's a partial answer to why pollsters might well avoid trying to normalize current data on the 2010 results.
All a poll is telling you is that it's likely the case that if you continued to poll the target population you would likely continue to get the same old answer with little variation.
A good case can be made that folks played with derivatives following parrando's paradox and endedup busting the world financial system.
I"ll tell you this if I"d made a few hundred mill doing that you'd never get to see my tax returns.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.