McClatchy Washington Bureau | 10/28/2008 | Poll: McCain has cut into Obama’s edge on jobs, economy
WASHINGTON — “Joe the Plumber” may be paying off for John McCain.
The Arizona senator scored sharp gains on the pivotal issue of jobs and the economy in the past week, helping him gain a bit on front-runner Barack Obama and narrow the presidential race as it heads into the final week, according to an Ipsos/McClatchy Poll released Tuesday.
I cannot, in my lifetime, recall a time when the polls were as all over the place, and as contradictory, as they have been during the campaign.
It is, frankly, impossible that they can all be accurate, or even close to it. I think what we are seeing is the apotheosis of gimmicked polls as political propaganda. Which is a real shame, because it simply removes yet another tool that could help the average observer to figure out what is actually going on. You can’t trust the “analysis,” you can’t trust the “objective, unbiased reporting,” and now you can’t even trust the “accurate measuring methods.”
A free and healthy democracy cannot function when it has no way to determine reality. We have our corrupt, degenerate, dangerous mass media to thank for this situation. The solution is simple: Destroy them. Don’t read them, buy them, or support them in any way. As somebody else said, “Let them die. I’ll dance on their graves.”
Whatever replaces our current media cannot possibly be any worse, and most likely will be better. More transparent, at least.
UPDATE: Welcome, Instapundit readers!
Submit to Stumbled Upon!
Somehow I get this gut feeling that most of MSM is blowing Obama because they are going broke and Obama has promised to subsidize them. It fits the socialist plan…
I’d be more convinced, guyk, if the media hadn’t been decidedly biased for as long as I can remember. Yeah, it seems particularly bad this year but I dunno if there is quite the quid pro quo you suggest. Though a reinstatement and vigorous enforcement of the Fairness Doctrine might help (or, at least, perhaps the MSM believes it will).
But I agree with Bill’s point about the polls. Further, if McCain somehow pulls it out, I don’t see how anyone can possibly trust the media polls anymore. I for one would start affording them the same level of reliability as astrology…
While I’m not a statistician by trade, I have some training it, as my job description absolutely requires it. And there is something really, really wrong with someone’s methodology this year. Instead of a normal-type distribution, we’re getting what amounts to a bimodal distribution in which one peaks predicts an Obama landslide while the other one predicts a tight race, with McCain possibly winning. Obviously, both scenarios cannot happen, so someone’s assumptions are grossly wrong this year. I guess that we’ll find out Tuesday night.
BTW, Bill, this is a friendly reminder for a Put Up or Shut Up type post in which we all do our level best to predict how things will go. As volatile as things are, it might be wise to wait until Monday, because the “McCain performs human sacrifice” expose on 60 Minutes will have aired by then.
Also the ‘I’m richer than hell so I can afford to pre-empt the World Series for a half-hour informercial” from Hussein.
I also feel that the media may be gently lowering the lead in order to prevent people from staying home because they think Obama is going to win hands down.
However, something tells me that the 30 minute delay of baseball’s crown jewel isn’t going to endear Obama to a large segment of the crowd he’s trying to reach.
Ask, and ye shall receive. I suppose I don’t need to post this, since y’all probably read Instapundit anyways.
Robb,
I hope that you’re correct. However, I’ve never been in the habit of overestimating the viewing publics IQ and I don’t plan to start now.
Wouldn’t it be something if Barry’s infomercial backfired and actually pissed off enough voters to cost him the election?
I see the same thing. I believe it is a result of the turnout models being used. Traditional models are showing a close race. The “Obama” enhanced models show a big lead for the One. They cannot be averaged to get a fair result. One or the other is correct, or at least as correct as possible. I think we are going to find the O enhanced models to be nothing more than leftwing dreams. Even the traditional models are often off as much as 5%, IIRC. I’ll go out on a limb and predict a narrow McCain victory. He just won’t get my vote. I’ll leave the pres selection blank. Of course, in close states that will be deduced as a vote for Obama in the recount.
Most of the right will not buy or read the damn things now. And when the polls prove to be wrong (again) and McCain wins, the left will burn the buildings in which the editors and writers work to the ground.
Either way, the life of the big dailies is over. Local papers will survive if they had sense enough to keep half the crap off their pages and stick with local events, if not, there’s always the PennySaver.
I’ve always thought of Polling companies as Advertising companies that are in disguise. Basically speaking, a Polling company will attempt to steer as many people as possible in the direction their customer requests.
I think the large newspapers are cutting their throats and I can’t figure out why. Here’s my personal example. I’m 55 and last year I stopped my subscription to the regional newspaper after 17 years, The (Norfolk) Virginian Pilot. Before that, I bought it daily from the local coffee shop. And my Dad had a subscription since I was a boy. I understand editorial pages and have no problems with their biases. But they would slant national news articles, mostly AP stories, way left. I finally cancelled out of disgust. I kept the local paper and now I can get national news on the web. If the country is spilt 50/50, wouldn’t they want to just write accurate news stories? Times change I guess. But I haven’t missed the paper and I’ve saved about $13 a month since.
Are the polls really all over the place? They all show leads for Obama, at the state level and at the national tracking poll level.
Then don’t leave it blank. I’m holding my nose and voting for McCain because the alternative is too horrible to contemplate. I hope to God there are about 100 times more people like me than you, or Obama gets crowned.
Faster please!
I am a married grad degreed woman with 2 small children who is already John Galting it as a stay at home mom.
To cut costs, we stopped buying anything but $8 a month cable, which means only 5 networks, 3 cspans, and 2 weather channels. I don’t buy newspapers. I don’t watch any of the 3 networks’ national news shows or their hour long analysis shows. I don’t buy magazines.
I’m already doing what I can to bury them, but it’s not happening fast enough. I can’t opt out anymore than I already do, and we’re still gonna get slaughtered financially by a BHO win.
We need the equivalent of missile defense and Rejkyavik to bankrupt the MSM NOW, not later.
Robb Allen: I also feel that the media may be gently lowering the lead in order to prevent people from staying home because they think Obama is going to win hands down.
There’s another, simpler, more mundane explanation. The MSM may be in the tank for Obama, but at the end of the day they are still a bunch of for-profit businesses who need to squeeze every last ratings point, every last copy sold, and every last advertising dollar out of their election coverage as they can - now more than ever. If people get it in their heads that Obama’s already won this thing, they’ll stop following the campaign and all that income will dry up. So, the MSM finds itself on the horns of an exquisite dilemma, one that pits their two most vital interests - the election of BHO and their own bottom line - directly against one another.
The polls are not all over the place, but they do seem that way because one’s intuitions about how variable the polls should be under perfect conditions are invariably wrong. I explain here.
Personally, I think democracy benefits from the break down of polls–provided that breakdown is widely recognized and people stop paying attention to polls. The “horse race” form of coverage was always the worst, basest kind. People who vote based on how popular this or that candidate is should not be allowed to vote.
Let the polls fail. God forbid we have to actually look at the candidates and figure out which one we like.
McCain by 7-10%.
I believe the polls are unusually unreliable this year, and the tendency as well as the wish to understate Republican support is magnified.
Add the high hangup rate. I don’t understand why people believe polls are accurate when over half the population refuses to participate.
Stir in the PUMAs.
Shake well.
And hang on. This will be a helluva ride.
Well, I read anonymous @ *16 link and while I agree with what he said there, I still think he’s making a big mistake. (To boil it down, if the polls were unbiased and truly randomly sampled, 100 polls or so would look much like the ones we’re seeing.)
The problem is, if over the past 36 years (since the pretty decent polling of ‘72) the polls were generally unbiased (i.e. without systematic error) and even close to truly random, then they wouldn’t have so egregiously shown losing Democrats ahead so often.
There is almost certainly a systematic error in the polling practices and they certainly are nowhere near to as close to random as in previous years, in my estimation. And it’s not just me. While I merely teach (occasionally) survey methodology and statistics, my wife works in the field, and for the past 4-5 years *all* their models (mostly in non-political surveys) have gone at least partially hinky on them. Her company (and she does stats on one of the biggest and best known general surveys) and their stats and survey specialists have come to believe that we’re no longer getting samples that are random enough to generalize from.
Believe me, they’re tearing their hair over this. I’m willing to believe that the day of useful telephone surveys may be over.
I also have a friend who works for one of the really giant survey research companies, and he’s doing a longitudinal study for the US govt that involves 50,000 randomly chosen citizens for face-to-face surveys that take 4 hours once a year. To get them they had to pretty generously reward the participants. The study has been going on for about 6 years now, and, given normal attrition, it is still working very well.
I don’t think most political polling, outside of very well-funded presidential campaigns perhaps, can afford to randomly select even a 3300 or so base and reward them enough to get good results.
Basically, I don’t think we can trust any of the polls currently available for methodological reasons, let alone the more prevalent problem of MSM bias. (With the exception of the Pew polls, which I am beginning to distrust after listening to Andy Kohut the other night explain that conservative Republicans are basically uneducated and kind of stupid, the outliers are MSM based.)
I can’t for the life of me understand why someone who ignores the MSM line on Obama/McCain would nevertheless believe their polls. If they’re going to lie about the fundamental question, then they’re going to lie about everything else too. (I personally know many right-leaning McCain supporters who have given up all hope.)
The fact is, there is one and only one reason why anyone would believe that Obama has already won, and that is that the media polls (CBS, NYT, WaPo, etc.) show Obama leading by an impossibly great margin. What if they are lying? Or to put it more mildly, exaggerating?
There are three recognizable groups of pollsters who have some incentive to be right: Gallup, Rasmussen and Zogby. Personally, I disregard all the others. The reason why they have an incentive to be right is that between elections, their main business comes from the marketing side of major corporations, who would like to see proof that they are accurate. The media pollsters have no such incentive — in fact, their incentive is quite the opposite of what an ordinary citizen would like to see in a polling organization — they exist solely to serve their masters, who are large media groups that are by and large enemies of the people. (I wouldn’t mind including the Battleground poll in my (all too) short list of credible pollsters, but they are exclusively political.)
None of these wants to be the first to show McCain with a lead — they will only show McCain with a lead if he is ahead by more than the “margin of error”, which at one time used to be a meaningful measure of a poll’s accuracy.
As election day approaches, you can expect to see an increasing divergence between the credible pollsters and the rest. Nothing will stop the media at this point.
I already afford them the same level of reliability as astrology. Why? Because I don’t buy into the science. Like many people on my side of the political spectrum, I believe that people act as individuals more than they act as members of groups–or at least the kind of groups that are used in polling. Sure, all Masons or Teamsters or NRA or Code Pink members might answer poll questions the same, but does anyone really expect all members of a racial, gender, age or income group to answer the same, simply because they’re in that group? And if not, doesn’t this blow the whole idea of polling completely out of the water, no matter how “sound” their methodology may appear to be?
When someone starts giving too much credence to polls, I ask them first of all if they participated in that poll. When (as always), the answer is “no,” I then ask them who they consider to have spoken for them and which personal trait–race, gender, income, etc.–qualifies that person to do so. Nobody ever has an answer.
This may sound trite, but the only poll that matters is the one that will take place on November 4…
Actually, there’s nothing wrong with the science, as far as it goes, which isn’t nearly as far as the wielders of the science would like us to believe. The problem is that pollsters use the science purely as spooky mumbo-jumbo, hoping we won’t notice the guy behind the curtain.
I wonder if part of it is the changes in the census over the past few decades. I’ve read that more and more people are identifying themselves as something other than white, even with one white parent, whereas before they would identify themselves as white. This may be due to more government and private programs for things like college tuition, and also I thought I read the census questions themselves were changed to allow people to pick differently to try to get a better idea of our demographics.
The problem this makes is poll companies have adjusted their weightings of responses so that there are more minorities, and thus overweighting their views. I remember reading during the 2004 campaign that one company (I forget who but one of the major ones) weighted self-identified republicans as if they were only 28% of the country. They did this because that’s just how it came out after weighting for race after the 2000 census numbers came in. They defended it but it’s ridiculous on its face, Republican voters make up more than 28% of the country and no Democrat campaign manager would ever believe that.
So I wonder if those mistakes are still propagating. It’s been the past decade where polling has been really off by several percentage points. So something is really wrong with either the pollsters or the census data.
To Physics Geek,
I am looking at the same polls and thinking how could they be so different from each other. After all sampling is sampling and we would believe everyone is using the same statistical method more or less… You’d expect that the differences would be with the statistical margin.
The only way I can rationalize the difference among the polls is to think that the universe of likely voters is being interpreted differently. Therefore the sampling would be off. With the Obama campaign’s claims of millions of new registrations, samples must contend with this - how many are real? If they’re real, how many will likely vote - and what factor do I put into the sample to account for this? I would think that this factor and the resulting smoothing efforts is one reason for the inconsistencies.
why do you cowards hide you email
nwerle at yahoo
and with all the mccain lies no wonder people are confused
Nick,
Can I call you Nick? “Mr Subhuman Moron with an Unhealthy Fascination with Obama’s Penis” is so unwieldy.
If only we had an “internet driver’s license” with a minimum IQ requirement. Then we wouldn’t have to waste our time with retards who can’t find the site owner’s email address. It’s on the front page, but hidden behind a dumbass filter.
As for McCain, lies, and confusion, well, he certainly does lie—he is a politician, after all—but the lies confuse only the ignorant, the idiotic, and the inbred. Which are you?
Sincerely,
The Daily Pundit Customer Service Representative