Are we too sensitive to partisan skew in polls?

posted at 9:21 am on September 25, 2012 by Ed Morrissey

Critics of media polling have long complained about the sample skew that tends to favor Democrats in their surveys.  We do a lot of poll analysis here at Hot Air and we routinely compare the modeling to exit polling in past elections.  National Journal reports today that pollsters, especially those who work with media outlets, have begun to fight back against the criticism leveled by conservatives of institutional sample skew, claiming that they’re seeing trends on the ground that previous electoral models won’t capture:

Critics allege that pollsters are interviewing too many Democrats — and too few Republicans or independents — and artificially inflating the Democratic candidates’ performance. Pollsters counter that the results they are finding reflect slight changes in public sentiment — and, moreover, adjusting their polls to match arbitrary party-identification targets would be unscientific.

Unlike race, gender or age, all demographic traits for which pollsters weight their samples, party identification is considered an attitude that pollsters say they should be measuring. When party identification numbers change, it’s an indication of deeper political change that a poll can spot.

“If a pollster weights by party ID, they are substituting their own judgment as to what the electorate is going to look like. It’s not scientific,” said Doug Schwartz, the director of the Quinnipiac University Polling Institute, which doesn’t weight its surveys by party identification. …

On Monday, the news website Buzzfeed interviewed a Virginia-based blogger who re-weights public polls to reflect the partisan trends reported by automated pollster Rasmussen Reports. Dean Chambers, the blogger, then presents the adjusted data in charts on his website, unskewedpolls.com.

As of late Monday, Chambers’ website claimed that an average of polls conducted since Labor Day show Mitt Romney leading Obama, 52 percent to 44 percent. The website and its findings were trumpeted on the Drudge Report, the conservative-leaning news-aggregation site that has tended to highlight polls more favorable to Romney and less favorable to the president.

The pollsters claim that they’re seeing a big shift towards identification with the Democratic Party.  If so, then Gallup and Rasmussen have both missed it.  Both organizations routinely do general-population polling for partisan identification.  In fact, the latest state-by-state polling from Gallup (August 2012) shows that the shift has gone the other way:

Thus far in 2012, the two major parties have been closely matched nationally in terms of the absolute number of states each can claim as politically favorable, representing a dramatic change from 2008 and 2009 when the Democratic Party had an overwhelming advantage on this score. This doesn’t translate directly into likely election outcomes, given differences that can exist between the party leanings of adults versus registered voters, as well as differing turnout patterns and voting behavior of Republicans vs. Democrats in some states.

Polls are intended to be predictive.  In order to be predictive, the sample has to hew closely to the turnout model of the actual election. The best way to calculate that is to check the trends from the most recent election cycles.  One can get surprised by this when turnout shifts dramatically, as it did in 2008 — but that was in favor of the Democrats for a D+7 result, and it’s unlikely to happen a second time, especially after the all-even turnout model from the 2010 midterms.  That means that D+11 on national samples aren’t going to be terribly predictive of the outcome in November, nor would R+11 samples, and so it’s difficult to take those results seriously.  Furthermore, with just a few weeks before the election, pollsters need to start finding likely voters rather than just registered voters or general-population samples if they expect consumers to rely on them for predictions of voter behavior — again, the entire point of polling.

If there is one valid criticism of conservative poll analysis, it’s that we tend to focus on just party ID rather than a broader range of demographic categories — gender, age, income, and geography.  Most of that data exists in exit polling, too, so it isn’t terribly difficult to check, but it is time consuming.  However, if a national poll features a turnout model of D+11 or R+11, that’s enough to make the results unreliable without checking the rest of the demos, just as one with 60% men would be.

The pollsters complaining in this piece sound as though they resent the idea of having their models put into question at all.  They want consumers to simply swallow what they deliver without asking any questions.  If so, they’re relying on an outdated media-consumer model.


Related Posts:

Breaking on Hot Air

Blowback

Note from Hot Air management: This section is for comments from Hot Air's community of registered readers. Please don't assume that Hot Air management agrees with or otherwise endorses any particular comment just because we let it stand. A reminder: Anyone who fails to comply with our terms of use may lose their posting privilege.

Trackbacks/Pings

Trackback URL

Comments

Comment pages: 1 2 3

Not me. I just add 10 points.

BigWyo on September 25, 2012 at 9:24 AM

Yep!

KOOLAID2 on September 25, 2012 at 9:25 AM

Partisan bias built into polls isn’t a accident and has a partisan purpose. For pollsters to insult our intelligence with such excuses only further illustrates their contempt.

Case in point… Zogby just released a poll showing Obama winning Indys by double digits to go to a commanding 9 point lead. Really? Again Zogby?

DANEgerus on September 25, 2012 at 9:26 AM

My question is what happened to all the presidential polls from all those organizations we’ve never heard of before? It seemed that when Obama was experiencing his posst convention bounce, they were produced every two hours.

Now that the bounce is history, other than Gallup and Rasmussen, we see nothing. Makes one wonder, don’t it.

dirtseller on September 25, 2012 at 9:26 AM

Polls are intended to be predictive.

Not always true. Some polls are intended to be manipulate.

dirtseller on September 25, 2012 at 9:27 AM

Polls are intended to be predictive.

Polls may be intended to be predictive. Or they may be intended to influence opinion and behavior.

petefrt on September 25, 2012 at 9:28 AM

Well, they aren’t just going to admit to being corrupt hacks.

Let’s compare their samples now to the samples they employ in the last two weeks when their reputation is on the line.

forest on September 25, 2012 at 9:28 AM

The pollsters complaining in this piece sound as though they resent the idea of having their models put into question at all. They want consumers to simply swallow what they deliver without asking any questions. If so, they’re relying on an outdated media-consumer model.

If they resent people rejecting their models, then they need to use realistic poll samples. Maybe people who still support Obama are idiots, but some of us out here in flyover country actually have a clue. And we can smell BS when it’s being served up to us.

Doughboy on September 25, 2012 at 9:29 AM

National Journal reports today that pollsters, especially those who work with

media outlets

Obama Presstitute Organs, have begun to fight back against the criticism leveled by conservatives of institutional sample skew, claiming that they’re seeing trends on the ground that previous electoral models won’t capture:</blockquote.
ReWrite™ engaged. I wonder where these pollsters were in 2010…oh wait, they were busy ignoring the trends so they could portray an ultimately-false narrative that the 'Rats wouldn't get slaughtered.

Steve Eggleston on September 25, 2012 at 9:29 AM

National Journal reports today that pollsters, especially those who work with media outlets
Obama Presstitute Organs
, have begun to fight back against the criticism leveled by conservatives of institutional sample skew, claiming that they’re seeing trends on the ground that previous electoral models won’t capture:

ReWrite™ engaged again for HTML Fail. I wonder where these pollsters were in 2010…oh wait, they were busy ignoring the trends so they could portray an ultimately-false narrative that the ‘Rats wouldn’t get slaughtered.

Steve Eggleston on September 25, 2012 at 9:30 AM

Case in point… Zogby just released a poll showing Obama winning Indys by double digits to go to a commanding 9 point lead. Really? Again Zogby?

DANEgerus on September 25, 2012 at 9:26 AM

Zogby’s such a broken outfit.

Don’t you remember that “shock poll” that had McCain down by a single point two days before he lost by seven? Yeah, that place oozes credibility. Just discard their polls without looking at them.

KingGold on September 25, 2012 at 9:30 AM

Whining about the polls is the surest sign of a doomed campaign. If Romney were doing well, we’d be talking issues and future cabinets and whether the GOP will have the guts to balance the budget, not whining about the stupid manufactured news polls.

I’ve seen this stuff before. In 2000, RCP made a big deal that the media were underplaying Bush’s pending electoral landslide. That didn’t exactly work out. I suspect that the party affiliation is part of a more complex “likely voter” model that anticipates difference between likely voters and actual voters (i.e, democrats are less likely to turn likely votes into actual votes).

Romney can beat Obama but he’s not going to do it with his supporters whining about the polls. He’ll need a good ground game, better delineation of the issues and a great debate performance. This *can* be won, but the GOP is already acting like it’s lost.

Hal_10000 on September 25, 2012 at 9:31 AM

Reminds of the Simpsons Treehouse of Terror parody of Gremlins…

Owner: Take this object, but beware it carries a terrible curse!
Homer: [worried] Ooooh, that’s bad.
Owner: But it comes with a free Frogurt!
Homer: [relieved] That’s good!
Owner: The Frogurt is also cursed.
Homer: [worried] That’s bad.
Owner: But you get your choice of topping!
Homer: [relieved] That’s good!
Owner: The toppings contains Potassium Benzoate.
[Homer stares blankly])
Owner: That’s bad

aunursa on September 25, 2012 at 9:31 AM

I am not a pollster.

But If you have a poll with 40 democrats 30 republicans 30 independents

it seems to me you cannot help but have more obama votes.

if the pollsters think were going to have the 2008 electorate they are

nuts.

it’s going to be closer to 2004.

gerry-genius pollster

gerrym51 on September 25, 2012 at 9:32 AM

Obama is going down and going down with a thud that will be heard around the world.

It’s going to be epic.

Obama is already losing big among men and middle class voters. Just wait till you see the election day exit polls for suburbanites and married women voters. Those groups will be be (at least one big part of) the story on Nov 7.

bluegill on September 25, 2012 at 9:32 AM

So we’re supposed to believe that pollsters are seeing some seismic shift toward the Dems, enough to OK samples that actually have them BEATING ’08 turnout? Come on. There was a NYT poll of VA a week or so ago that included just 23% Republicans…in what universe does anyone believe the electorate on Election day here will have just that many, when this is an R +3 state? And then there is the wildly different take on independents…some have Romney winning that group, others show Obama with a ridiculous lead among them.

I don’t know that unskewedpolls.com has it right, either…I suspect Obama is running a little ahead of Romney in OH, VA, WI, with the two more or less even in FL, IA, and perhaps CO, as well as nationally.

changer1701 on September 25, 2012 at 9:34 AM

Could someone do a posthoc statistical analysis of polls weighted by party ID vs actual electoral returns to see if there is a signficant difference? I’m thinking that you could, and that might be quite instructive.
2c

ted c on September 25, 2012 at 9:38 AM

Look… as soon as people started relying on the RCP average of polls, the gig was up. Dems had to start doing D+10 to fudge the numbers.

faraway on September 25, 2012 at 9:39 AM

Every poll is rigged including the one on election day.

bgibbs1000 on September 25, 2012 at 9:39 AM

…pollsters, especially those who work with media outlets, have begun to fight back against the criticism leveled by conservatives of institutional sample skew, claiming that they’re seeing trends on the ground that previous electoral models won’t capture:

Can previous election models capture the trend of increased bank assets of pollsters in proportion with the pro-Democrat skew of their partisan splits? If not, the pollsters are correct.

Archivarix on September 25, 2012 at 9:40 AM

******Alert **************

TRANSCRIPT:

TRANSCRIPT: Mitt Romney Delivers Remarks to the Clinton Global Initiative

9:27 am ET September 25, 2012
*******************************

http://foxnewsinsider.com/2012/09/25/transcript-mitt-romney-delivers-remarks-to-the-clinton-global-initiative/

canopfor on September 25, 2012 at 9:40 AM

Remember that in 2011, al-AP did a “poll” that sampled almost twice as many Democrats as Republicans after O’bamna strangled bin Laden.

It only resulted in a 5-6 point job approval bounce.

Del Dolemonte on September 25, 2012 at 9:40 AM

The pollsters sound like the MSM that can’t believe the sheeple don’t accepted the revealed truths the elites have always dished out to the Great Unwashed.

How dare they challenge their betters, ignore the sages on the Sunday talk shows, or change the channel to Fox News!!! How dare they go on the internet to get news from sites like Breitbart, Daily Caller, and worst of all, Drudge!!!

Don’t they know they’re supposed to read the editorial pages of the Old Media, and vote for the candidate endorsed by the NYT, WaPo, LAT, etc.???

Why, they’re thinking for themselves!!! Next thing you know they start a Tea Party or something!!

Wethal on September 25, 2012 at 9:40 AM

I suspect Obama is running a little ahead of Romney in OH, VA, WI, with the two more or less even in FL, IA, and perhaps CO, as well as nationally.

changer1701 on September 25, 2012 at 9:34 AM

Is Obama above 50% in any of those states?

faraway on September 25, 2012 at 9:43 AM

Rasmussen Daily Tracking: Obama 47-Romney 46. With leaners, tied at 47

andy85719 on September 25, 2012 at 9:43 AM

Total BS. The unrealistic oversampling of democrats has been happening for many months and not just recently… Now the media polls found a new trick to fake Obama lead which is to under sample independents who have been favoring Romney by 5 points… Folks, do not fall for it.

I created my own “Poll Adjustment Calculator” that re-calculate the biased poll based on a more realistic party ID breakdown.

You can download it form this link:

http://www.fileconvoy.com/dfl.php?id=g0f9a1c6111e9b12e14951111d0a0649192bfc3

The first section is to enter the results from a media poll, including % of democrats, % of Republicans, % of Independents, enter % Obama, % Romney, % Undecided, % Others for each of these groups ( i.e. demorats, Republicans, Independents)…

The second section is to t adjust the % democrats, % Republicans, % of Independents to match a more realistic elections turnout for each group. The worst case scenario would be democrats + 3% over Republicans. Make sure that when you do the adjustments for each of these group that they add to 100%.

The third section is the assignment of the % of undecided to each Romney and Obama. So you need to enter the % of undecided for Romney and those for Obama. Historically the undecided go 2:1 for the challenger…

mnjg on September 25, 2012 at 9:44 AM

Early voting begins next week in Ohio. There has been a huge increase in absentee ballot requests, beginning in Hamilton County, which I believe encompasses Cincinnati, which is a GOP stronghold. But the Democrats are crowing about their ground game in Ohio and the ABC-WaPO poll out this morning has Obama ahead by 7 in Ohio, which they have moved to “lean Dem.”

So, we have to contend with a lot of conflicting info and I wish we could sort some of it out.

1. Are all the polls wrong? Call me a skeptic. I looked at the Ohio 2010 gubernatorial polls, and they were all Kasich up 2 to 7 points in September. Kasich won by 4.

2. The Romney financial avalanche will win the day in the home stretch. OK, so where is it? A $50 million cash advantage is not going to defeat Obama’s MSM advantage, which is worth hundreds of millions of dollars as an in-kind gift. Evidently there is not enough money, and the Super Pacs don’t have enough money either. Too few Sheldon Adelsons.

3. The Romney camp and the Super Pacs should be expanding the field into the Philly market, Pittsburgh, Albequerque, etc. Move the needle, and force Obama to react. Why? Because Obama doesn’t have, and won’t have, the resources to fight back. DRAIN OBAMA DRY NOW. Since the GOP groups are not doing so, there must be insufficient dollars to spread around. Not a good sign.

4. Worse, if Romney-RNC husbands all of their resources for the final 3 weeks, two bad outcomes can occur: ad fatigue from the voters, further diminished by early voting; and an international event that eclipses any good the ads might do.

I don’t know how it is possible for such a horrible president to get re-elected. Our nation must be really going down the drain if people are clearly this stupid. The Romney camp and its allies should have UNLIMITED money from people who believe in freedom. Are we that far gone? I do not see how we escape a fiscal cliff and a diminished century for America if this Empty Suit is re-elected.

matthew8787 on September 25, 2012 at 9:44 AM

We know that these frauds are lying liars who lie, but does it actually work? I have a hard time believing that anyone is seriously influenced by polls.

Valkyriepundit on September 25, 2012 at 9:45 AM

I’ll give you a poll, if all the people who are upset with Obama show up and vote for Romney, Obama’s going to lose.

bflat879 on September 25, 2012 at 9:46 AM

Rasmussen Daily Tracking: Obama 47-Romney 46. With leaners, tied at 47

andy85719 on September 25, 2012 at 9:43 AM

Rasmussen is using D+3 in his poll sample because even though Romney is winning independents in the majority of his polls by an avaerage of 5 points (as well as other media polls), Rasmussen is showing that Romney is either tied with Obama, slightly ahead of Obama, or slightly behind Obama…

mnjg on September 25, 2012 at 9:46 AM

Question the polls all you want, but Romney is losing in every single poll out there. EVERY SINGLE ONE. If you want to think that’s just random chance and margin of error and bad sampling, go right ahead. But your heads will be exploading on election night when Obama wins.

BTW, Obama is opening up a decent lead in Rasmussen’s swing state poll.

gumbyandpokey on September 25, 2012 at 9:46 AM

The pollsters complaining in this piece sound as though they resent the idea of having their models put into question at all. They want consumers to simply swallow what they deliver without asking any questions.

It seems like we’ve seen this “hide the curve” attitude before.

rhombus on September 25, 2012 at 9:46 AM

Question the polls all you want, but Romney is losing in every single poll out there. EVERY SINGLE ONE. If you want to think that’s just random chance and margin of error and bad sampling, go right ahead. But your heads will be exploading on election night when Obama wins.

BTW, Obama is opening up a decent lead in Rasmussen’s swing state poll.

gumbyandpokey on September 25, 2012 at 9:46 AM

Yes liberal scum, not only we are questioning the polls but we are proving it with math and facts that they are skewed… Download my “poll adjustment calculator” and learn something…
http://www.fileconvoy.com/dfl.php?id=g0f9a1c6111e9b12e14951111d0a0649192bfc3

But may be you are too stupid to use a simple excel sheet…

mnjg on September 25, 2012 at 9:49 AM

National Journal reports today that pollsters, especially those who work with media outlets, have begun to fight back against the criticism leveled by conservatives….

Now we see the violence inherent in the system….

Hell hath no fury like Liberal push-pollsters scorned!

viking01 on September 25, 2012 at 9:49 AM

gumbyandpokey on September 25, 2012 at 9:46 AM

I tend to agree. The polls can’t all be wrong. That would require quite a conspiracy. Romney will lose, but not by as much as many of the polls show. 1 or 2 points probably. Americans are dumb and like failure.

andy85719 on September 25, 2012 at 9:50 AM

Are we too sensitive to partisan skew in polls?

How Carter Beat Reagan

whatcat on September 25, 2012 at 9:50 AM

I don’t know how it is possible for such a horrible president to get re-elected. Our nation must be really going down the drain if people are clearly this stupid. The Romney camp and its allies should have UNLIMITED money from people who believe in freedom. Are we that far gone? I do not see how we escape a fiscal cliff and a diminished century for America if this Empty Suit is re-elected.

matthew8787 on September 25, 2012 at 9:44 AM

The simple fact is they don’t think he’s been so horrible, because they don’t KNOW any better. I know people get sick of the media excuse, but they have covered for this guy from Day 1, and it shows. Not only do they not focus on what he’s done or not done, they spend most of their time attacking or mocking the GOP. Not only that, but he’s cool, he’s hip, and that’s what a lot of the low-info crowd goes for.

Is Obama above 50% in any of those states?

faraway on September 25, 2012 at 9:43 AM

I don’t think so. I was just giving my gut feeling about the race, based on what Romney’s own campaign team has said. They know OH, for example, is tough terrain.

changer1701 on September 25, 2012 at 9:52 AM

Early voting begins next week in Ohio. There has been a huge increase in absentee ballot requests, beginning in Hamilton County, which I believe encompasses Cincinnati, which is a GOP stronghold. But the Democrats are crowing about their ground game in Ohio and the ABC-WaPO poll out this morning has Obama ahead by 7 in Ohio, which they have moved to “lean Dem.”

So, we have to contend with a lot of conflicting info and I wish we could sort some of it out.

1. Are all the polls wrong? Call me a skeptic. I looked at the Ohio 2010 gubernatorial polls, and they were all Kasich up 2 to 7 points in September. Kasich won by 4.

2. The Romney financial avalanche will win the day in the home stretch. OK, so where is it? A $50 million cash advantage is not going to defeat Obama’s MSM advantage, which is worth hundreds of millions of dollars as an in-kind gift. Evidently there is not enough money, and the Super Pacs don’t have enough money either. Too few Sheldon Adelsons.

3. The Romney camp and the Super Pacs should be expanding the field into the Philly market, Pittsburgh, Albequerque, etc. Move the needle, and force Obama to react. Why? Because Obama doesn’t have, and won’t have, the resources to fight back. DRAIN OBAMA DRY NOW. Since the GOP groups are not doing so, there must be insufficient dollars to spread around. Not a good sign.

4. Worse, if Romney-RNC husbands all of their resources for the final 3 weeks, two bad outcomes can occur: ad fatigue from the voters, further diminished by early voting; and an international event that eclipses any good the ads might do.

I don’t know how it is possible for such a horrible president to get re-elected. Our nation must be really going down the drain if people are clearly this stupid. The Romney camp and its allies should have UNLIMITED money from people who believe in freedom. Are we that far gone? I do not see how we escape a fiscal cliff and a diminished century for America if this Empty Suit is re-elected.

matthew8787 on September 25, 2012 at 9:44 AM

Nice post.

As to why Romney cannot pull ahead going away well it’s either Romney is a terrible candidate(he is by the way) or his campaign is stuck on stupid(and it is) or both. Or maybe Romney doesn’t really want to win. Maybe the GOP establishment pushed him to insure that the marxist wins re-election. Yeah I know it sounds far fetched, but then so does a marxist being president of the USA.

bgibbs1000 on September 25, 2012 at 9:52 AM

First, many of the polls published today are not in fact meant to be predictive; they are meant to discourage conservatives and Republicans. And if the gnashing of teeth and rending of garments by Republican elites is any indication, their succeeding.

Second, there is one aspect of this election the pollsters are not able to capture and that is the Bradley Effect (or some call it the Wilder Effect). Most “experts have declared the Bradley Effect dead, partly as a result of society becoming more sophisticated about race and most recently, as a result of Barack Obama’s election to the presidency.

Baloney! Democrats and the media have so over-pushed the notion that anyone who doesn’t agree with Obama has to be a racist that a great many regular, every-day Americans are fearful to even casually mention to fellow workers that they are thinking of not voting for Obama.

This is going to result in a six-plus point win for Romney, and the Dems are not going to see it coming. I wrote about this a few days ago at http://alpipkin.com/blog/romneys-losing-winning-no-losing/

svar42 on September 25, 2012 at 9:52 AM

This is an interesting tabulation, although I think their results may be skewed a couple of points the other way…

http://www.unskewedpolls.com/

oddball on September 25, 2012 at 9:52 AM

If you want to think that’s just random chance and margin of error and bad sampling, go right ahead.
gumbyandpokey on September 25, 2012 at 9:46 AM

Gumbo’s right! I remember reading a study issued during the Dewey Administration that explained it all.

whatcat on September 25, 2012 at 9:55 AM

As to why Romney cannot pull ahead going away well it’s either Romney is a terrible candidate(he is by the way) or his campaign is stuck on stupid(and it is) or both. Or maybe Romney doesn’t really want to win. Maybe the GOP establishment pushed him to insure that the marxist wins re-election. Yeah I know it sounds far fetched, but then so does a marxist being president of the USA.

bgibbs1000 on September 25, 2012 at 9:52 AM

If Romney loses the monday morning quarterbacking will begin in earnest, and I’m sure there will be plenty of reasons as to why. But the idea that he doesn’t really want to win, after seeking the office for the past six years, is beyond stupid.

changer1701 on September 25, 2012 at 9:56 AM

Question the polls all you want, but Romney is losing in every single poll out there. EVERY SINGLE ONE. If you want to think that’s just random chance and margin of error and bad sampling, go right ahead. But your heads will be exploading on election night when Obama wins.

BTW, Obama is opening up a decent lead in Rasmussen’s swing state poll.

gumbyandpokey on September 25, 2012 at 9:46 AM

Awww… isn’t that cute.

Well, you know, you’re right! In fact, Obama is winning by SO MUCH, you really don’t even have to bother going to vote on November 2. Just stay home, relax, and watch the returns. You might want to have some hard liquor on hand, though. (No, not champagne.)

makattak on September 25, 2012 at 9:56 AM

“This is going to result in a six-plus point win for Romney, and the Dems are not going to see it coming. I wrote about this a few days ago at http://alpipkin.com/blog/romneys-losing-winning-no-losing/

And this is why I keep coming back. Romney will win by 6???????

This is almost like a mental illness, imo. Nobody can really believe that, can they?

gumbyandpokey on September 25, 2012 at 9:57 AM

As usual great insights Ed, I find it amusing these pollsters are seeing “things on the ground” that support their sample skews but they cannot quantify these “things” to justify those skews. We are supposed to accept as fact their data interpretations but take on faith their assumptions.

Sorry we stopped believing in astrologers, court magician and diviners a long time ago and I personally don’t plan on believing in such charlatans for the foreseeable future.

Skwor on September 25, 2012 at 9:57 AM

Now the media polls found a new trick to fake Obama lead which is to under sample independents who have been favoring Romney by 5 points… Folks, do not fall for it.

Yep. Anytime you see indies under 20 it is totally bogus. Even under 25 is suspect. Several major polls are doing this right now.

I think they are also trying to game the RCP average by throwing in junk outliers with Obama up 10 points, etc. Then they can say Obama’s leading by 3 when the average of real polls has them tied.

They will get serious in late October and start adjusting their models to reality because those are the polls on which their reputation will rest.

Although, increasingly, I think pollsters are figuring out that they can make much more money by not being accurate, because 1) polling is expensive, and good polling is even more expensive, and 2) they’re being paid by one side or the other (and it’s usually Dems) to shape opinion rather than accurately measure it. This is a disturbing trend.

Missy on September 25, 2012 at 9:58 AM

“Gumbo’s right! I remember reading a study issued during the Dewey Administration that explained it all.”

Yeah, go back to Dewey/Truman…

What will happen to all the “poll whiners” on election night if they were right?

gumbyandpokey on September 25, 2012 at 9:59 AM

I’m not gettin’ all worked up by the pollercoaster this year…I’m more concerned with voter fraud.

workingclass artist on September 25, 2012 at 9:59 AM

The pollsters complaining in this piece sound as though they resent the idea of having their models put into question at all. They want consumers to simply swallow what they deliver without asking any questions. If so, they’re relying on an outdated media-consumer model.

THIS. They are stamping their little feet that their “black box” has been opened and their biases revealed.

What really burns me is national polls that use an 800-person sample. I mean, come on. I know enough about survey research methodology to know that a national poll of 800 people is going to have a margin of error so huge that it is meaningless.

rockmom on September 25, 2012 at 9:59 AM

I tend to agree. The polls can’t all be wrong. That would require quite a conspiracy. Romney will lose, but not by as much as many of the polls show. 1 or 2 points probably. Americans are dumb and like failure.

andy85719 on September 25, 2012 at 9:50 AM

Really? On what historical basis do you make this judgement?

In fact, when was the last time the polls showed a Republican (presidential candidate) that eventually did win with a lead anywhere before the last of October?

I’ll wait. It was in my lifetime, even. And the results were SOOOOOO much in favor of the Republican that not even the polls could ignore it.

makattak on September 25, 2012 at 10:01 AM

*pollsters

makattak on September 25, 2012 at 10:02 AM

At the end of the day it’s about turnout, period. You can let the Left demoralize you, throw in the towel and declare he’s lost (as some apparently have here), or you can go out and work your a$$ off to help flip these numbers and embarrass all these pollsters. How many of the doomsayers are volunteering for the campaign? Are you making calls? Registering friends, neighbors, family…anyone that might be open to voting Romney/Ryan? Are you talking to fence sitters? Complaining about the campaign on a blog doesn’t move the numbers

changer1701 on September 25, 2012 at 10:03 AM

U.S. consumer confidence surges in September, with consumers more positive about jobs, Conference Board says.

Dumber than a box of rocks.

andy85719 on September 25, 2012 at 10:04 AM

BTW, Jim Geraghty did some outstanding reporting yesterday at NRO on how many more Republicans than Democrats have been registered in all the swing states since 2008. Pollsters are ALL missing this. Anyone that registers to vote in an election year is almost 100% guaranteed to vote. And we have a ton more of these new voters than they do.

Also BTW, this is where a lot of the RNC money has gone. Not into TV ads, but into registration drives, massive voter ID efforts and list development. Reince Priebus said yesterday in Ohio that the RNC ground game will “clobber” the Democrats when voting starts.

I started in politics in an era when it was all about identifying your voter and getting them to the polls. I worked in a county that was 3-2 Democrat and we won election after election for the GOP by just doing the basics of voter ID and turnout. The polls never reflected any of it.

rockmom on September 25, 2012 at 10:05 AM

“Gumbo’s right! I remember reading a study issued during the Dewey Administration that explained it all.”

Yeah, go back to Dewey/Truman…

What will happen to all the “poll whiners” on election night if they were right?

gumbyandpokey on September 25, 2012 at 9:59 AM

Fine then how about the Carter win over Reagan predicted by the polls, oh wait that’s right Carter got pummeled in a landslide.

Well maybe the Kerry win over Bush predicted by the most accurate of polling method, exit polls, darn again that’s right Kerry lost also.

Since most polls outside of the last one before election has been shown to be completely off the mark there is every reason to discount them, especially when the methodology this time around is demonstratively questionable.

Skwor on September 25, 2012 at 10:05 AM

Essentially all media polls are mathematical garbage. The left has an extensive effort to manipulate and skew polls as a means of influencing the voting public. A constant bombardment of positive polling data in favor of Obama will influence voter attitude and drive them to vote for the ” most popular” candidate who is of course the one leading in the polls. This must be countered by a flood of positive polling data in favor of the opposing candidate (Romney). “Fair and balanced” polling data is a no-win proposition unless you are determined to lose.

rplat on September 25, 2012 at 10:05 AM

gumby thinks undecided voters are going to wake up and vote for Obama

Obama doesnt have 50% anywhere.

Landslide.

faraway on September 25, 2012 at 10:06 AM

There is an interesting trend that nobody seems to catch. A much higher percentage of Obama voters are single or divorced – that much has been confirmed by lots of polls and isn’t questioned by anyone. Therefore, if the number of Obama voters is assumed approximately equal to that of those supporting Romney (a blood chilling thought, by the way), the number of pro-Obama poll responses will be significantly higher.

Archivarix on September 25, 2012 at 10:06 AM

US consumer confidence jumps to highest level since February on brighter job outlook

-What jobs? Aargh. People are so stupid. There aren’t any jobs yet people are excited about job prospects.

andy85719 on September 25, 2012 at 10:06 AM

“Gumbo’s right! I remember reading a study issued during the Dewey Administration that explained it all.”

Yeah, go back to Dewey/Truman…

What will happen to all the “poll whiners” on election night if they were right?

gumbyandpokey on September 25, 2012 at 9:59 AM

Rather then try to come here for a futil effort to demoralize us you need to spend more time and energy to energize your welfare queens and your young people… It takes much more energy to bring the welfare queens and airhead students (your side) to the polls than bringing the producers (the Republican side) to the polls…

mnjg on September 25, 2012 at 10:06 AM

If Romney loses the monday morning quarterbacking will begin in earnest, and I’m sure there will be plenty of reasons as to why. But the idea that he doesn’t really want to win, after seeking the office for the past six years, is beyond stupid.

changer1701 on September 25, 2012 at 9:56 AM

I don’t think Romney will lose, but if he does, there will be one reason and one reason only: our turnout sucked. And the only explanation for that is a depressed or unmotivated base. The reason I think Romney will win is because that scenario is unfathomable to me, but if we play devil’s advocate and assume a loss, the lesson the GOP better learn is nominate a bonafide conservative who isn’t afraid to take on the Democrat/media complex. In other words, someone like Newt but younger and with far less personal baggage.

Doughboy on September 25, 2012 at 10:07 AM

“Really? On what historical basis do you make this judgement?

In fact, when was the last time the polls showed a Republican (presidential candidate) that eventually did win with a lead anywhere before the last of October?”

Really? Is this post serious?

Bush lead Gore in 2000 and Bush lead Kerry by a decent margin in 04, which narrowed after his terrible first debate performance. But he always held the lead in the RCP average.

This is insanity.

gumbyandpokey on September 25, 2012 at 10:09 AM

Polls,polls,and more polls.Polls are like chicken.You can season, spice it,and cook it 100 ways.The only pole I’m interested in right now is the one the girls use at the Come and Get It Club.

docflash on September 25, 2012 at 10:09 AM

Look… as soon as people started relying on the RCP average of polls, the gig was up. Dems had to start doing D+10 to fudge the numbers.
faraway on September 25, 2012 at 9:39 AM

Yep. ONE carefully-performed scientific poll is infinitely more valuable than a thousand hack jobs.

Frankly a lot of people don’t understand how statistics work, or science in general.

logis on September 25, 2012 at 10:09 AM

Swing State Daily Tracking: #Obama 47%, #Romney 44% Ugh.

andy85719 on September 25, 2012 at 10:10 AM

Someone is wrong, Gallup and Rassmussen have consistently showed a tied race while battleground states are showing Obama recreating 2008, the numbers do not add up. By the way, any poll that shows Romney leading with independents but losing to Obama in the general is total goat manure.

rob verdi on September 25, 2012 at 10:10 AM

Yeah, go back to Dewey/Truman…
gumbyandpokey on September 25, 2012 at 9:59 AM

If you need a second opinion, ask Gov. Scott Walker:
MSM Reports Wildly Flawed WI Exit Polls To Spin For Obama

whatcat on September 25, 2012 at 10:11 AM

Bush lead Gore in 2000 and Bush lead Kerry by a decent margin in 04, which narrowed after his terrible first debate performance. But he always held the lead in the RCP average.

This is insanity.

gumbyandpokey on September 25, 2012 at 10:09 AM

Kerry beat bush via the exit polling, considered the most accurate polling method of all. Want to try again?

Skwor on September 25, 2012 at 10:12 AM

“Swing State Daily Tracking: #Obama 47%, #Romney 44% Ugh.”

Even Rasmussen is showing strong movement towards Obama in his swing state sample.

But keep denying the polls people…

gumbyandpokey on September 25, 2012 at 10:13 AM

US consumer confidence jumps to highest level since February on brighter job outlook

-What jobs? Aargh. People are so stupid. There aren’t any jobs yet people are excited about job prospects.

andy85719 on September 25, 2012 at 10:06 AM

The jobs report sucked in August(and was revised downward for June and July). And BoA just announced they’re laying over 10,000 workers by year’s end. Which consumers are they talking to?

Doughboy on September 25, 2012 at 10:13 AM

“Kerry beat bush via the exit polling, considered the most accurate polling method of all. Want to try again?”

LOL, now it’s only “exit polling” you want to look at! Have fun cherry-picking the numbers, my friend. BTW, the RCP average nailed the Bush/Kerry final result.

Gonna be fun to rub your nose into the stink on election night.

gumbyandpokey on September 25, 2012 at 10:15 AM

Despite what the doom and gloomers and Leftist plants want us to believe 47-44 with a margin of error of 3 pts may as well be even. Then consider the consistently heavy oversampling of Dimocrats and the incumbent Chicago Messiah tends to droop even further.

Remember, 2010 came up out of nowhere and never happened!

viking01 on September 25, 2012 at 10:15 AM

http://spectator.org/archives/2012/09/25/how-carter-beat-reagan/

This article explains how Carter beat Reagan in 1980. At least according to NYT polling. Don’t buy into any of the currently published polls. Not worth the paper they are written on.

COgirl on September 25, 2012 at 10:16 AM

Of course the polls reported in the MSM are lies. That’s what they do.

My family lives in a college town in New Jersey, as blue a place as exists on this earth. Believe it or not, when I walked and drove there over the last week I saw as many Mitt Romney bumper stickers right now as Obumbles/Biteme stickers. That’ll obviously change as the days tick off and the election nears, but it says quite a bit about enthusiasm (or lack of it) on the left.

Ignore the pollsters.

MTF on September 25, 2012 at 10:17 AM

What will happen to all the “poll whiners” on election night if they were right?

gumbyandpokey on September 25, 2012 at 9:59 AM

Then it will be time to blame the “TrueCons” and “purists” and “Socons” and Palinistas and Ronulans and anyone else the squish element blames instead of themselves.

I wish outlets like HA had been a weeeeeee bit more skeptical of polls when they were showing Romney as “the most electable blahblahblah”. Of course, polls were sacred…then.

ddrintn on September 25, 2012 at 10:18 AM

As Dick Morris put it on some tv show the other night [rough paraphrase]: Show me the demographic and I’ll tell you how they’re going to vote. 2 out 3 times I’ll get it right. That’s what the polls can tell you. But the polls can only tell you how they will vote, not who will vote. No one can predict that.

The pollsters and the media don’t like having their voter-turnout models (i.e., their predictions), exposed.

Candidates’ internal polls are much more focused on who is likely to turn out, using focus groups and other means to see what campaign messages are reaching which people. A much more accurate indicator of how the race is going– or certainly no worse than a random poll– is to watch what the candidates are doing, what they are saying, and where they are going or running ads to say it.

de rigueur on September 25, 2012 at 10:18 AM

We were doomed in 2004, here is a Fox news poll showing Kerry winning by 2 points the weekend before the election.

http://www.foxnews.com/projects/pdf/poll_012606.pdf

rob verdi on September 25, 2012 at 10:18 AM

Great article about the skewing of the polls for Carter.

http://spectator.org/archives/2012/09/25/how-carter-beat-reagan/

And yes, the “polls” had Carter close or ahead in October.

itsspideyman on September 25, 2012 at 10:18 AM

“Swing State Daily Tracking: #Obama 47%, #Romney 44% Ugh.”

Even Rasmussen is showing strong movement towards Obama in his swing state sample.

But keep denying the polls people…

gumbyandpokey on September 25, 2012 at 10:13 AM

Rasmussen has shown his daily national tracking poll and swing states tracking poll to go from Romney to Obama, then back to Obama, and back to Romney again… It is cycle… All with 2 to 3 points difference…

Again as I told you before instead of wasting your time here trying to demoralize us because there would nothing that would stop us from going to the polls in number never seen before to defeat Obama, you should focus your energy on getting the welfare queens and students to vote because if you do not get your base, all evidence points that you base is not energized, then Obama is certain to lose…

mnjg on September 25, 2012 at 10:18 AM

I don’t think Romney will lose, but if he does, there will be one reason and one reason only: our turnout sucked. And the only explanation for that is a depressed or unmotivated base. The reason I think Romney will win is because that scenario is unfathomable to me, but if we play devil’s advocate and assume a loss, the lesson the GOP better learn is nominate a bonafide conservative who isn’t afraid to take on the Democrat/media complex. In other words, someone like Newt but younger and with far less personal baggage.

Doughboy on September 25, 2012 at 10:07 AM

I’m sorry, but that loss will then be on US. If after 4 years of all the garbage from Obama and his pals in the media, conservatives can’t be motivated to at least vote against all that…wtf? Here’s our chance to kick this guy to the curb.

changer1701 on September 25, 2012 at 10:18 AM

“The jobs report sucked in August(and was revised downward for June and July). And BoA just announced they’re laying over 10,000 workers by year’s end. Which consumers are they talking to?

Doughboy on September 25, 2012 at 10:13 AM”

Now the Consumer Confidence Index is rigged, too?!?!

There are warning signs all over, folks. The Romney campaign here in WI is hemorrhaging volunteers and put the word out to talk radio to make announcements that they need people. That does not happen if the people think a candidate can win. We actually had many more people helping in 08 because of Sarah Palin.

Obama wins all of his 08 states minus IN and probably NC.

gumbyandpokey on September 25, 2012 at 10:19 AM

Of course the polls reported in the MSM are lies. That’s what they do.

MTF on September 25, 2012 at 10:17 AM

Right. But those “lies” are why we have Romney.

ddrintn on September 25, 2012 at 10:19 AM

“Swing State Daily Tracking: #Obama 47%, #Romney 44% Ugh.”

Even Rasmussen is showing strong movement towards Obama in his swing state sample.

But keep denying the polls people…

gumbyandpokey on September 25, 2012 at 10:13 AM

You trot out an Obama 47% poll to say that he will win?????

You will need lots of whiskey for the next 4 years.

faraway on September 25, 2012 at 10:20 AM

Then it will be time to blame the “TrueCons” and “purists” and “Socons” and Palinistas and Ronulans and anyone else the squish element blames instead of themselves.

I wish outlets like HA had been a weeeeeee bit more skeptical of polls when they were showing Romney as “the most electable blahblahblah”. Of course, polls were sacred…then.

ddrintn on September 25, 2012 at 10:18 AM

Yeah, anybody that doesn’t vote for Romney will share the blame.

changer1701 on September 25, 2012 at 10:20 AM

Really? Is this post serious?

Bush lead Gore in 2000 and Bush lead Kerry by a decent margin in 04, which narrowed after his terrible first debate performance. But he always held the lead in the RCP average.

This is insanity.

gumbyandpokey on September 25, 2012 at 10:09 AM

Why, you’re right. It was September when the polls finally said Bush had the lead, not October. (And most definitely not “always”)

makattak on September 25, 2012 at 10:20 AM

ddrintn on September 25, 2012 at 10:18 AM

of the choices we had he is the most electable.

was Santorum more electable. Was Newt more electable.Was Perry more electable.

Most of the most electable did not run.

of the one’s we had he was and is the most electable.

gerrym51 on September 25, 2012 at 10:20 AM

Gonna be fun to rub your nose into the stink on election night.

gumbyandpokey on September 25, 2012 at 10:15 AM

The important quesiton is why HotAir has not banned you yet. It is not that you are just a stupid liberal, you are a very annoying one…

How do we ask for banning on HotAir? Let me know please.

mnjg on September 25, 2012 at 10:21 AM

“Kerry beat bush via the exit polling, considered the most accurate polling method of all. Want to try again?”

LOL, now it’s only “exit polling” you want to look at! Have fun cherry-picking the numbers, my friend. BTW, the RCP average nailed the Bush/Kerry final result.

Gonna be fun to rub your nose into the stink on election night.

gumbyandpokey on September 25, 2012 at 10:15 AM

It says a great deal about you that gloating on election night is what you are aiming for rather then the state of the country. It must suck to be such sniveling person that this is your focus as the world is burning down.

rob verdi on September 25, 2012 at 10:21 AM

itsspideyman on September 25, 2012 at 10:18 AM

That’s what I was going to post! Great article!

tbrickert on September 25, 2012 at 10:21 AM

It’s not scientific,” said Doug Schwartz, the director of the Quinnipiac University Polling Institute, which doesn’t weight its surveys by party identification. …

Anyone who uses the word scientific in this way immediately discredits his sales speech.

burt on September 25, 2012 at 10:21 AM

Also BTW, this is where a lot of the RNC money has gone. Not into TV ads, but into registration drives, massive voter ID efforts and list development. Reince Priebus said yesterday in Ohio that the RNC ground game will “clobber” the Democrats when voting starts.

I started in politics in an era when it was all about identifying your voter and getting them to the polls. I worked in a county that was 3-2 Democrat and we won election after election for the GOP by just doing the basics of voter ID and turnout. The polls never reflected any of it.

rockmom on September 25, 2012 at 10:05 AM

I hope you are correct. I needed some good news.

matthew8787 on September 25, 2012 at 10:22 AM

I’m sorry, but that loss will then be on US. If after 4 years of all the garbage from Obama and his pals in the media, conservatives can’t be motivated to at least vote against all that…wtf? Here’s our chance to kick this guy to the curb.

changer1701 on September 25, 2012 at 10:18 AM

Sorry, but many conservatives are tired to death of only voting AGAINST the other guy. If Romney loses, the fault is on THOSE who nominated the guy swallowing the media-driven narrative that only Romney can win.

ddrintn on September 25, 2012 at 10:22 AM

“Great article about the skewing of the polls for Carter.

http://spectator.org/archives/2012/09/25/how-carter-beat-reagan/

And yes, the “polls” had Carter close or ahead in October.

itsspideyman on September 25, 2012 at 10:18 AM”

Again, factually wrong.

http://www.outsidethebeltway.com/debunking-a-myth-reagan-was-leading-carter-long-before-that-final-october-debate/

gumbyandpokey on September 25, 2012 at 10:23 AM

itsspideyman on September 25, 2012 at 10:18 AM

COgirl on September 25, 2012 at 10:16 AM

“How Carter Beat Reagan”
Beat ya both to it at 9:50 AM
:D

whatcat on September 25, 2012 at 10:24 AM

of the choices we had he is the most electable.

gerrym51 on September 25, 2012 at 10:20 AM

How do you know?

ddrintn on September 25, 2012 at 10:24 AM

Sorry, but many conservatives are tired to death of only voting AGAINST the other guy. If Romney loses, the fault is on THOSE who nominated the guy swallowing the media-driven narrative that only Romney can win.

ddrintn on September 25, 2012 at 10:22 AM

Which other candidate that ran in the primaries would be winning right now?

Anybody that stays home or votes for Obama will share blame, period. You’re choosing four more years of this and that’s on YOU, no one else.

changer1701 on September 25, 2012 at 10:24 AM

Gumbot’s Bidenesque bravado is sounding increasingly insecure.

I’m predicting his bottle of cold duck stays in the fridge on election night.

viking01 on September 25, 2012 at 10:25 AM

ddrintn on September 25, 2012 at 10:22 AM,
who should be the nominee?

rob verdi on September 25, 2012 at 10:25 AM

“How Carter Beat Reagan”
Beat ya both to it at 9:50 AM
:D

whatcat on September 25, 2012 at 10:24 AM

This isn’t 1980, and Romney most certainly is NOT Reagan.

ddrintn on September 25, 2012 at 10:26 AM

“We were doomed in 2004, here is a Fox news poll showing Kerry winning by 2 points the weekend before the election.

http://www.foxnews.com/projects/pdf/poll_012606.pdf

rob verdi on September 25, 2012 at 10:18 AM”

Here are the final polls in the Bush/Kerry race. Notice that Bush lead in the vast majority of the polls. And notice that the RCP average was within a point of the final margin…

Final Results – – 50.7 48.3 Bush +2.4
RCP Average 10/27 – 11/1 – 48.9 47.4 Bush +1.5
Marist 11/1 – 11/1 – 49 50 Kerry +1
GW/Battleground 10/31 – 11/1 – 50 46 Bush +4
IBD/TIPP 10/30 – 11/1 – 50.1 48 Bush +2.1
CBS News 10/29 – 11/1 – 49 47 Bush +2
Harris 10/29 – 11/1 – 49 48 Bush +1
FOX News 10/30 – 10/31 – 46 48 Kerry +2
Reuters/Zogby 10/29 – 10/31 – 48 47 Bush +1
CNN/USA Today/Gallup 10/29 – 10/31 – 49 49 Tie
NBC/WSJ 10/29 – 10/31 – 48 47 Bush +1
ABC/Wash Post 10/28 – 10/31 – 49 48 Bush +1
ARG 10/28 – 10/30 – 48 48 Tie
CBS News/NY Times 10/28 – 10/30 – 49 46 Bush +3
Pew Research 10/27 – 10/30 – 51 48 Bush +3
Newsweek 10/27 – 10/29 – 50 44 Bush +6

gumbyandpokey on September 25, 2012 at 10:26 AM

Comment pages: 1 2 3