About that CNN poll …

posted at 9:51 am on April 17, 2012 by Ed Morrissey

Allahpundit wrote a lengthy piece about last night’s CNN poll, complete with some insightful commentary about the cross-tabs.  If you missed it — and by the number of comments on the post, it doesn’t appear than many of you have — be sure to read it now.  However, the sample data, as AP pointed out, lacked a few details, most prominent of which was the partisan split of the sampling.  Given the inclination of media polls to use wildly unrepresentative D/R/I splits in their samples, the lack of transparency on that point is telling.

That might not be the biggest problem with the poll, though.  Its biggest problem is … math.  Reader Raymond O did some math and asked a rather interesting set of questions in an e-mail last night about how CNN did theirs.  First, let’s start with the topline results, as reported by CNN: Obama 52%, Romney 43% among registered voters, 53/41 among all respondents.  If that’s the case, then the number of respondents in the latter case voting for Obama should be 538, and the number supporting Romney 416.

However, when reading the questions on page 3 of the poll report, that’s not at all what we see:

BASED ON 484 RESPONDENTS WHO PLAN TO VOTE FOR OBAMA — SAMPLING ERROR: +/- 4.5 PERCENTAGE PTS.
3. (Asked of Obama voters) Is that more a vote FOR Barack Obama or more a vote AGAINST Mitt
Romney? …

BASED ON 476 RESPONDENTS WHO PLAN TO VOTE FOR ROMNEY — SAMPLING ERROR: +/- 4.5 PERCENTAGE PTS.
4. (Asked of Romney voters) Is that more a vote FOR Mitt Romney or more a vote AGAINST Barack
Obama?

Since the combined total of the two exceeds their count of registered voters in the survey (910), we have to assume this refers to the general-population response.  That’s wildly different than the 53/41 split that CNN reports from the poll.  In fact, it’s only a 48/47 split for Obama.    And given that the poll shows a slightly better result for Romney among registered voters, it’s not difficult to conclude that Romney probably led in that category before CNN’s pollster shifted the results around to this extent.

Obviously, the pollster weighted the results — dramatically.  But on what basis?  Pollsters weigh results in likely-voter models; in fact, that’s a vital part of the likely-voter modeling process.  I’m not aware of the need to weight gen-pop polls unless the demos in the survey are wildly deficient, and since CNN didn’t disclose those, we don’t know — but it hardly imbues these results with confidence.  Speaking of which, the actual 484 responses for Obama is slightly more than 10% less than what it would take to get to a 53/41 split, far outside of the margin of error announced in the data (+/- 4.5%).  In other words, CNN’s pollster produced results that we’re supposed to believe represents the general population’s views within 4.5%, and then proceeded to weight them by increasing Obama’s advantage by almost 11%.

Again, I’m not aware of a need to weight results in general-population polls, and certainly not to the extent seen here, nor did CNN disclose that they did weight the results, especially in the general-population results. Without the rest of the demographic data included in this survey and the weighting methodology, this poll should be treated as utter fantasy.


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

Like some one said yesterday, I hope they keep making up numbers till the election, then I will sit back and laugh as the REAL results come in…….

angrymike on April 17, 2012 at 10:24 AM

For the nth time: stop obsessing over every poll, especially junk CNN polls. It’s April, not late October early November.

MidniteRambler on April 17, 2012 at 10:24 AM

Look, any poll this far out should be taken with a grain of salt. That being said, the recent group of polls from Fox, Gallop, and Rasmussen has had the race either Romney ahead, or virtually tied. When a poll this out of whack comes out it’s a good idea to look into why.Thank goodness someone on our side did….

sandee on April 17, 2012 at 10:20 AM

But that’s my point. You people take Rasmussen and Foxnews at face value. Why is nobody looking at their methodology? Fox says Romney is ahead, he’s ahead. CNN says he’s losing, it must be some nefarious MSM conspiracy.

angryed on April 17, 2012 at 10:24 AM

Math is hard…

PatriotRider on April 17, 2012 at 10:18 AM

Indeed. Four polls come out at the same time, three of them show a statistical tie and the other shows +9/+12 Obama. But if you question that last one you are a paranoid nut who hates the data because you are a Hot Gas Flat Earth Romney fanboy.

forest on April 17, 2012 at 10:25 AM

You sound like a Kerry supporter in 2004 who thought every Bush voter was stupid. Try getting out of the Hannity/Limbaugh bubble some day, there’s a whole world out there where every Republican isn’t divine.

angryed on April 17, 2012 at 10:22 AM

yeah, like this coming from a stupid liberal troll has any meaning or value at all…

jimver on April 17, 2012 at 10:25 AM

Of course the pollsters from the media may want to prop up Obama’s popularity by making it seem like people will still want him for another term, but this question from the author:

Again, I’m not aware of a need to weight results in general-population polls, and certainly not to the extent seen here,

You might weight the poll in a general popularity survey to resemble demographics of the u.s. to answer the popularity contest, or general approval. But it is not useful to predict a president, because:

Our President is elected by the states, not a popularity contest.

Some of the states vote so blue or red they can be counted on to vote that way even with a large swing one way or the other, but the states that have an actual contest in them, the battleground states should be polled individually, and polled for enthusiasm to vote. That would be real info.

The media will want a popularity contest for Barach Obama. A lot of people don’t have time for that, we don’t care about how popular personally, or how much celebrity (how cool) a president is.

The economy is flat lining. Obama doesn’t seem to care, he is thinking about where he is taking Michelle on her next vacation, for god’s sake.

Fleuries on April 17, 2012 at 10:25 AM

Unless I missed it, there’s no partisan breakdown given

This is all you need to know. All the math is a waste of time.

logis on April 17, 2012 at 10:26 AM

But that’s my point. You people take Rasmussen and Foxnews at face value. Why is nobody looking at their methodology? Fox says Romney is ahead, he’s ahead. CNN says he’s losing, it must be some nefarious MSM conspiracy.

angryed on April 17, 2012 at 10:24 AM

So why don’t you look into Fox’s and Ras’s methodology, if you’re so concerned about it?

Maybe you can do that after you’ve finished up your latest entry on Daily Kos.

changer1701 on April 17, 2012 at 10:26 AM

angryed on April 17, 2012 at 10:15 AM

Well that certainly explains the Dem. Machine’s obsessive attacks now doesn’t it.

FineasFinn on April 17, 2012 at 10:26 AM

Looks Michael “hide the decline” Mann has a new job…as a pollster for CNN.

HotAirian on April 17, 2012 at 10:26 AM

The poll is accurate because it takes into account the fact that Obama voters vote more than once. It adds in a reasonable Democrat fraud factor of +10%.

zmdavid on April 17, 2012 at 10:27 AM

If you don’t get an answer you like, cheat.

GeorgieGirl9 on April 17, 2012 at 10:27 AM

You sound like a Kerry supporter in 2004 who thought every Bush voter was stupid. Try getting out of the Hannity/Limbaugh bubble some day, there’s a whole world out there where every Republican isn’t divine.
angryed on April 17, 2012 at 10:22 AM

Oh yes Mr. Ed…it is only Obama who is divine. How much more gubmint cheese will you be receiving if your Messiah wins?

BeachBum on April 17, 2012 at 10:27 AM

Translation: I don’t care what reality says, I’ll keep closing my eyes to polls I don’t like. Cool dude. I just read a poll taken by Sean You’re a Great American Hannity. It says 198% of the public will vote for Romney.

angryed on April 17, 2012 at 10:18 AM

You didn’t answer my question. The die is cast. It is Obama vs. Romney. It’s not what I wanted, but that’s the way it is. So, why do you care if the polls are right or wrong? You’re voting for Obama and most everyone else here is voting for the GOP candidate.

You might be right. If you are you can come back here a cheer if Romney loses. You can tell everyone, “I told you so.” I’m sure that’ll make you feel better about yourself.

ReaganWasRight on April 17, 2012 at 10:27 AM

I could have told you that the CNN poll, like every other CNN poll, is junk without spending half an hour typing up a summary of why the CNN poll is useless.

.

milcus on April 17, 2012 at 10:28 AM

For the nth time: stop obsessing over every poll, especially junk CNN polls. It’s April, not late October early November.

MidniteRambler on April 17, 2012 at 10:24 AM

yeah, but they are trying to shape opinions (or well, make news actually) with these garbage polls, right when the Rep candidate is actually getting in the game…this is no coincidence….

jimver on April 17, 2012 at 10:28 AM

all this is gonna do is make the congnitive dissonance the lib nutjobs experience worse. those poor dumb idiots. they have no idea whats coming. and this kind of crap just makes it worse for them. in a way, i feel sorry for them. kinda.

t8stlikchkn on April 17, 2012 at 10:29 AM

The moral of the story is the only poll to believe is the one that happens on the first Tuesday of November.

itsspideyman on April 17, 2012 at 10:18 AM

This.

It’s like Sarah Palin said, polls are for strippers and skiers.

Bitter Clinger on April 17, 2012 at 10:29 AM

So why don’t you look into Fox’s and Ras’s methodology, if you’re so concerned about it?

Maybe you can do that after you’ve finished up your latest entry on Daily Kos.

changer1701 on April 17, 2012 at 10:26 AM

like he has any idea of methodology and sampling…he’s spitting out whatever talking points du jour was given to him by his demtard bosses he takes orders from…

jimver on April 17, 2012 at 10:31 AM

Like some one said yesterday, I hope they keep making up numbers till the election, then I will sit back and laugh as the REAL results come in…….

angrymike on April 17, 2012 at 10:24 AM

I think the danger is that poll results like this will make Republican voters stay home, because they’re let to believe they can’t win. I hope I’m wrong, but that’s why it’s important to point out that the results are skewed in the first place.

rhit87 on April 17, 2012 at 10:31 AM

I love how RomneyAir dismisses any poll from CNN but anything out of Fox News is treated as gospel never to be questioned. Where was the in depth analysis of that poll? Oh that poll showed Romney winning so nobody took a second look or questioned the methodology.

angryed on April 17, 2012 at 10:10 AM

Oddly though when you blindly accept provably false polls that support your political wishes and dismiss those easily shown as having more rigor in the process that is enlightened and reality based.

Ergo when you dismiss/accept polls it is obviously correct to do, when others you disagree with do so they are obviously wrong. Truly you are a liberal with the closed mind to prove it.

I also noticed that you have not once argued the facts but instead resorted to attacking the messengers character/intentions. Classic case of misdirection when a person knows they are in a losing argument.

Why don’t you instead argue against the math and the supposition Ed put forth? Lack the intellect for it perhaps?

Skwor on April 17, 2012 at 10:32 AM

Del Dolemonte on April 17, 2012 at 10:14 AM

I’ll be laughing at you imbeciles when Romney turns out to be everything I predicted and you’re scratching your heads mumbling he was severely conservative, he was severely conservative.

angryed on April 17, 2012 at 10:16 AM

I’ve never said Romney’s a “conservative”, all I have done is correct false information put out by misinformed people, including you, that’s based on inaccurate data. One of you Hot Gas ABR people, for example, has repeatedly claimed in the past that as MA Governor 80% of his judicial nominations were Democrats, which is an absolute lie. The real number was 40%. But remember, he was dealing with an 8 member judicial board who had to approve all of his nominees; all 8 of them were Democrats. As President he won’t have quite the same hurdle, especially if the Republicans win the Senate.

And I have also seen false financial data posted here on what his spending record was as MA Governor.

A clue…he’s nowhere near as “Liberal” as you make him out to be.

Del Dolemonte on April 17, 2012 at 10:32 AM

Every poll you don’t like is biased/skewed in some way. Every poll you like is perfectly run.

angryed on April 17, 2012 at 10:04 AM

Why is it that when a poll is performed unethically and has likely been skewed intentionally, you dismiss it as sour grapes from Romney supporters not getting their way.

I asked you a question on another thread about whether or not you want the GOP to win the WH. To my knowledge you never responded. Will you allow your relentless contempt to inform your voting decision for Obami in November or will your better judgment prevail?

Slainte on April 17, 2012 at 10:32 AM

The picture is of Duke’s bench during their tourney loss methinks.

Bensonofben on April 17, 2012 at 10:23 AM

You are right. That is Coach K behind that hand.

magicbeans on April 17, 2012 at 10:32 AM

You’ve certainly raised some valid questions, Ed, and the poll might indeed be shady. Personally, I place the current margin at probably +6 in favor of Obama.

But as I said on AP’s thread, it is a testament to Romney’s weakness that he isn’t currently soundly thrashing Obama in the polls.

Why should anyone be optimistic that Romney will survive the Obama/media war that is about to be unleashed on him. He will be lucky to come out of the general election with a 52-43 loss as the CNN poll portrays.

TheRightMan on April 17, 2012 at 10:32 AM

For the nth time: stop obsessing over every poll, especially junk CNN polls. It’s April, not late October early November.

MidniteRambler on April 17, 2012 at 10:24 AM

You’re right, it’s too early to even make a difference. I think time will help Romney because of all the wacky birth control and other social agenda issues applied to the GOP brand by Santorum. Give voters a chance to realize Romney has other priorities.
On the other hand, Romney himself has admitted that his standing with the Hispanic vote is possibly his single biggest handicap (think Florida). Let’s see what good old etch-a-sketch can do to fix that.

bayam on April 17, 2012 at 10:33 AM

But that’s my point. You people take Rasmussen and Foxnews at face value.
angryed on April 17, 2012 at 10:24 AM

Could be that there is a good reason to take Rasmussen seriously.

VietVet_Dave on April 17, 2012 at 10:34 AM

Why should anyone be optimistic that Romney will survive the Obama/media war that is about to be unleashed on him. He will be lucky to come out of the general election with a 52-43 loss as the CNN poll portrays.

TheRightMan on April 17, 2012 at 10:32 AM

It’s already started and this “poll” is part of it.

forest on April 17, 2012 at 10:34 AM

like this coming from a stupid liberal troll has any meaning or value at all…

jimver on April 17, 2012 at 10:25 AM

It does seem as though s/he is a Liberal troll.

Slainte on April 17, 2012 at 10:37 AM

4/16/2012 CNN :
Obama leads Romney among women, 55-49%
Romney leads Obama among men, 46-39%
W Gap of 6 for Obama (swing of -7 from Obama 2008)
M Gap of 5 for Romney (swing of +6 from R 2008)
Gap Swing advantage from 2008 to 2012 = Romney +13

Gender Gap: http://perceptionasreality.blogspot.com/2012/04/gender-gap-president-election-exit.html

mjbrooks3 on April 17, 2012 at 10:37 AM

The funniest thing about angryed isn’t that he/she is letting the whole “I’m an angry Tea Partier who won’t vote for Romney because he is too liberal and neither should anyone else who calls themself a conservative” Moby mask slip; it is that angryed doesn’t even bother trying to hide it.

In one post ed is all about “I’m a Tea Partier who hates Obama and I’m glad the RINOs will suffer under four more years of Obama”, and then ed forgets his/her persona and starts blasting Hannity, Limbaugh, and the “conservative bubble”.

That sort of inconsistency will not help with the Mobyite efforts to suppress conservative turnout for Romney.

MidniteRambler on April 17, 2012 at 10:37 AM

I hope I’m wrong, but that’s why it’s important to point out that the results are skewed in the first place.

rhit87 on April 17, 2012 at 10:31 AM

precisely, foxnews, drudge should be on it, at least they can force them in a corner and make them reveal and release their methodology and sampling…

jimver on April 17, 2012 at 10:37 AM

Ed Morrissey on April 17, 2012 at 9:59 AM

…I get depressed when I read and hear about these polls…until I read Hot Air…then the blood pressure comes back down…THANK YOU!
(that picure! Love that picture!)

KOOLAID2 on April 17, 2012 at 10:37 AM

VietVet_Dave on April 17, 2012 at 10:34 AM

That twitter exchange was pretty funny. Annie-Rose Strasser is a genuine kook.

forest on April 17, 2012 at 10:39 AM

One of you Hot Gas ABR people, for example, has repeatedly claimed in the past that as MA Governor 80% of his judicial nominations were Democrats, which is an absolute lie. The real number was 40%. But remember, he was dealing with an 8 member judicial board who had to approve all of his nominees; all 8 of them were Democrats. As President he won’t have quite the same hurdle, especially if the Republicans win the Senate.

Del Dolemonte on April 17, 2012 at 10:32 AM

Yeah right… right until we hear excuses that Republicans do not have 60 votes in the Senate to overcome filibusters from the Dems and therefore we have to compromise.

Why is it that RINOs and their backers constantly give excuses for their failures while conservatives… simply govern.

Contrast Reagan’s achievements despite a Democrat House and later on a Democrat House and Senate to the wimpy GOP leaders we have today.

TheRightMan on April 17, 2012 at 10:40 AM

Love that picture!)

KOOLAID2 on April 17, 2012 at 10:37 AM

Mark it! It’s the official “triple-single face palm” pic for HG.

VietVet_Dave on April 17, 2012 at 10:41 AM

I think all of the polls are utter crap. Not because of skewed methodology, but because it is April. General election voters are not paying attention this early. Wait until September or October and do a poll. Then it might matter.

Shump on April 17, 2012 at 10:42 AM

Although the overall numbers polls are interesting=if they are fairly done-the only polls that really matter are swing state polls that reflect on electoral college math.

I think Romney has a decent shot in Michigan and Pennsylvania.

these combined with the normally red states give him a chance.

gerrym51 on April 17, 2012 at 10:42 AM

But that’s my point. You people take Rasmussen and Foxnews at face value. Why is nobody looking at their methodology? Fox says Romney is ahead, he’s ahead. CNN says he’s losing, it must be some nefarious MSM conspiracy.

It’s not all that hard. Most of us only want to see two things out of a poll:

1) A reasonable split between D/R/I. I don’t have numbers near at hand, but 2010 exits (a great R election) were even. 2008 exits (a great D year) were D+7 or so. There is absolutely no rational basis for ever putting D more than 7 over R. None. The Dems are not remotely in as good a position as they were in 2008. Now they may not be in as bad a position as in 2010; we don’t know. But the polls that we see (and there have been more than a few, usually CNN/WaPo/ABC) with Dem advantages of +9 to +11 are just silly. A party ID of D +0 to 4 is more in line with the truth.

2) Some attmept to screen for people who are at least somewhat likely to actually vote. A the bare minimum, an RV screen should be applied. People can argue over LV models all they want, but the RV screen should be bare minimum to take a poll seriously.

Guess what Rasmussen does? And Gallup? And Fox? They have generally reasonable D/R/I and restrict responses to RV or LV.

A pollster can come up with any damned result they want. If I tried hard enough, I could get a poll proclaiming broccoli as the nation’s favorite food. It still wouldn’t be true.

FuzzyLogic on April 17, 2012 at 10:44 AM

TRIPLE FACEPALMS all right

gerrym51 on April 17, 2012 at 10:45 AM

angryed on April 17, 2012 at 10:24 AM

I look at the sample of all the polls we talk about, including the recent FNC poll. That one appeared reasonable when I looked at the sample.

dogsoldier on April 17, 2012 at 10:46 AM

Just to clarify angrymike, no relation to Angryed ………..
No matter what will not, cannot vote for Zero……….

angrymike on April 17, 2012 at 10:46 AM

It does seem as though s/he is a Liberal troll.

Slainte on April 17, 2012 at 10:37 AM

at forst he tried to mimic the tea partier disgruntled with the Rep party, and he hid behind the ABR bunch successfully, after a while his cognitive dissonances caught up with him and got the best of him….

jimver on April 17, 2012 at 10:49 AM

Again, I’m not aware of a need to weight results in general-population polls, and certainly not to the extent seen here, nor did CNN disclose that they did weight the results, especially in the general-population results.

FIFY, Ed.

unclesmrgol on April 17, 2012 at 10:51 AM

Just want to jump in here – even if it is already well down to page 2 on the comments – and say that you do a great job of tearing apart these polls here at Hot Gas.

Gingotts on April 17, 2012 at 10:52 AM

Math. It’s what separates us from complete tyranny.

Somebody should send Raymond O. a free iPAD for his skillz.

Marcus Traianus on April 17, 2012 at 10:52 AM

I

think the danger is that poll results like this will make Republican voters stay home, because they’re let to believe they can’t win. I hope I’m wrong, but that’s why it’s important to point out that the results are skewed in the first place.

rhit87 on April 17, 2012 at 10:31 AM

well its the INTENT of these things, yes. but dont worry. 48% of this country now will crawl over broken glass to vote against obama. as with everything the MSM does, there is an inherent assumoption that the majority are stupid and lazy, and it doesnt work and really never has. because its not true. look, all this is going to become evident in august/sept. and BHO is going to be underwater 7+ points. thats when they will throw in the towel. and the truth telling will begin. so relax. thats when the fun begins. when the rest of them (dem electeds) start to throw BHO overboard, in a last ditch attempt to save themselves. and their failed ideology. its gonna be fun. relax. enjoy this.

t8stlikchkn on April 17, 2012 at 10:53 AM

Like some one said yesterday, I hope they keep making up numbers till the election, then I will sit back and laugh as the REAL results come in…….

angrymike on April 17, 2012 at 10:24 AM

The problem is that they’re setting the table for the planned massive fraud. So when it comes in at 52/43 Obama/Romney, it looks legit.

Count on it.

CycloneCDB on April 17, 2012 at 10:53 AM

Haven’t read the comment yet but did Allah pay you to use that picture, Ed? Very funny.

Cindy Munford on April 17, 2012 at 10:54 AM

I remember the Carter/Reagan “too close to call polls”. The Media carried water for Carter until the bitter end.

I will also never forget the faces of my neighbors lined up at dawn to teach that moral lecturing, sweater wearing, fear mongering, malaise pushing, peanut farmer a lesson. No one gave a damn what the polls said, we were going vote his sorry @ss out of office. And we did.

No matter how close the MFM wants us to believe this race is, today’s poor economy,high gas prices,and high unemployment would sink any incumbent. And the pResident is going to be shown to the door by 11:59 pm EST on November 6, 2011. Barky is toast.

jrgdds on April 17, 2012 at 10:54 AM

It’s not all that hard. Most of us only want to see two things out of a poll:

1) A reasonable split between D/R/I. I don’t have numbers near at hand, but 2010 exits (a great R election) were even. 2008 exits (a great D year) were D+7 or so. There is absolutely no rational basis for ever putting D more than 7 over R. None. The Dems are not remotely in as good a position as they were in 2008. Now they may not be in as bad a position as in 2010; we don’t know. But the polls that we see (and there have been more than a few, usually CNN/WaPo/ABC) with Dem advantages of +9 to +11 are just silly. A party ID of D +0 to 4 is more in line with the truth.

2) Some attmept to screen for people who are at least somewhat likely to actually vote. A the bare minimum, an RV screen should be applied. People can argue over LV models all they want, but the RV screen should be bare minimum to take a poll seriously.

Guess what Rasmussen does? And Gallup? And Fox? They have generally reasonable D/R/I and restrict responses to RV or LV.

A pollster can come up with any damned result they want. If I tried hard enough, I could get a poll proclaiming broccoli as the nation’s favorite food. It still wouldn’t be true.

FuzzyLogic on April 17, 2012 at 10:44 AM

The fact that Obama gets a quarter of the Conservative vote is another problem that isn’t supported by any other poll. Gallup has them both stealing 10% from each other for example.

Zybalto on April 17, 2012 at 10:54 AM

CNN, the most likely news station to post Kevin Bacon in front of the building, in vintage ROTC regalia, nervously shouting, “All is well! All is well!”
Its the analytical equivalent of realizing the riot crowds you’ve created are about to turn on you.
Popcorn anyone?

onomo on April 17, 2012 at 10:56 AM

Gotta love Angryed. He’s like a failed Yosemite Sam.

and then proceeded to weight them by increasing Obama’s advantage by almost 11%.

This is the same range ABC/WaPo poll used two weeks ago to reach it’s “Barry Runnin’ Away” poll.

ABC/WaPo – 51/44

CNN/ORC – 52/43

The two outliers. The same advantage range needed for Barry to reach 50+.

For April, with ABC and CNN, 47/45 Barry.

Without, 45/45.

So, at best, it’s +2 Barry, before margin of error.

IMO, CNN/ORC saw their poll would make it 45 even or Barry down in the RCP average. They wanted to avoid that storyline.

budfox on April 17, 2012 at 11:00 AM

Thanks for doing the math. I took a quick look at the poll and could tell that the results were very questionable but didn’t have time to drill down.

I’ll be laughing at you imbeciles when Romney turns out to be everything I predicted and you’re scratching your heads mumbling he was severely conservative, he was severely conservative.

angryed on April 17, 2012 at 10:16 AM

In November, barring a bizarre turn of events, one of two things wwill happen. Obama will be re-elected or Romney will be elected. Romney will govern more conservatively than Obama. Romney will be more fiscally responsible than Obama. Romney will be more socially conservative than Obama. I prefer Romney to Obama.

I’m sorry Romney isn’t conservative enough for you. I don’t know who your preferred candidate was-I assume either s/he didn’t run, or lost to Romney. Not voting or voting for Obama won’t change that. It will only increase the chance of Obama being re-elected, which most of us can agree would have horrible consequences for the future of the United States. I can’t see any circumstances where not voting for the Republican nominee will be better for the US, short term or long term.

talkingpoints on April 17, 2012 at 11:00 AM

Indeed. Four polls come out at the same time, three of them show a statistical tie and the other shows +9/+12 Obama. But if you question that last one you are a paranoid nut who hates the data because you are a Hot Gas Flat Earth Romney fanboy.

forest on April 17, 2012 at 10:25 AM

Actually math is simple. Apparently what’s hard is the lack of honesty & consistency in polling…

PatriotRider on April 17, 2012 at 11:00 AM

It’s not all that hard. Most of us only want to see two things out of a poll:

1) A reasonable split between D/R/I. I don’t have numbers near at hand, but 2010 exits (a great R election) were even. 2008 exits (a great D year) were D+7 or so. There is absolutely no rational basis for ever putting D more than 7 over R. None. The Dems are not remotely in as good a position as they were in 2008. Now they may not be in as bad a position as in 2010; we don’t know. But the polls that we see (and there have been more than a few, usually CNN/WaPo/ABC) with Dem advantages of +9 to +11 are just silly. A party ID of D +0 to 4 is more in line with the truth.

2) Some attmept to screen for people who are at least somewhat likely to actually vote. A the bare minimum, an RV screen should be applied. People can argue over LV models all they want, but the RV screen should be bare minimum to take a poll seriously.

Guess what Rasmussen does? And Gallup? And Fox? They have generally reasonable D/R/I and restrict responses to RV or LV.

A pollster can come up with any damned result they want. If I tried hard enough, I could get a poll proclaiming broccoli as the nation’s favorite food. It still wouldn’t be true.

FuzzyLogic on April 17, 2012 at 10:44 AM

Very well said. Any pollster who is doing that, gets Romney ahead. Any pollster that screws with the formula, gets outlier polls.

milcus on April 17, 2012 at 11:02 AM

Wow a CNN poll that is skewed towards the democrats. I am SHOCKED! Since 99% of the people here probably already know whatever poll you are discussing is skewed towards the left, could you guys stop putting the word “Gulp” under every poll that favors Obama?

rjcylon on April 17, 2012 at 11:02 AM

angrymike on April 17, 2012 at 10:24 AM

That’s the whole point, to skew the numbers for the desired effect to get the “popular” vote. Besides the people who vote a certain way because their family always has there has to be a fair amount of the public that wants to do what the cool kids do. How will they know unless they read the polls?

Cindy Munford on April 17, 2012 at 11:04 AM

Fantasy with a purpose. The MSM is doing everything in its power, and it has a lot of power, to get Obama re-elected. These people are not journalists, they’re Obama campaign workers.

AZCoyote on April 17, 2012 at 10:00 AM

What always cracks me up is their utter cluelessness that there will be someone on our side doing the math, finding out that the numbers don’t add up, then forwarding the results to Hot Air, where it will get forwarded to other conservative outlets. Obviously, even so, CNN and their ilk don’t care that the numbers aren’t right. They only care to ring the bell, so to speak. Clearly this needs to happen more often.

Do not believe the mainstream media. Do not believe their polls. Keep fighting. Never give up.

xrayiiis on April 17, 2012 at 10:10 AM

This should always be our motto. I know an editor at a local paper who has one of those Media Research Center bumper stickers on her file cabinet that says “Don’t Believe the Liberal Media.”

PatriotGal2257 on April 17, 2012 at 11:08 AM

I am not sure what influences CNN’s reporting more, their liberal bias or just plain incompetence.

Sometimes it is hard to tell…

William Eaton on April 17, 2012 at 11:10 AM

When we get to early November, and their credibility is on the line, you’ll see more accurate results.

Wait…did I just imply that CNN might have credibility?

moo on April 17, 2012 at 11:11 AM

This is same-old, same-old for the MSM. The news industry knows that voters tend not to turn out for a candidate they’ve been lead to believe will lose the election. By reporting low numbers for Romney, they’re influencing Romney supporters to stay home while Obama zombies come out in optimistic droves. It’s a a typical Journolister-type technique to undermine the democratic process.

Crusty on April 17, 2012 at 11:14 AM

this poll should be treated as utter fantasy.

CNN – Cartoon News Network. Where even a child will understand its not real.

nicknack60 on April 17, 2012 at 11:16 AM

The. Obama. Media.

Axeman on April 17, 2012 at 11:17 AM

I remember the Carter/Reagan “too close to call polls”. The Media carried water for Carter until the bitter end.

jrgdds on April 17, 2012 at 10:54 AM

Yep, and speaking amongst friends and co-workers about what would happen during the second Carter administration. Shortly before the election our mood started to shift and that was the first time I suggested that the networks weren’t being honest. By election day, many of us were ecstatic that Reagan could pull it off. And the day after the election we were laughing at the media for trying to sway the election. Since then I think of the (old) media the same way actors were thought of during Shakespeare’s time.

cozmo on April 17, 2012 at 11:19 AM

In other words, CNN’s pollster produced results that we’re supposed to believe represents the general population’s views within 4.5%, and then proceeded to weight them by increasing Obama’s advantage by almost 11%.

Ok, that would also explain why a poll done back in March matching up Santorum against Obama was ALSO off. If you look at the RCP poll page for Obama against Republican candidates:

USA Today/Gallup 3/25 – 3/26 901 RV 51 43 Obama +8

CNN/Opinion Research 3/24 – 3/25 925 RV 55 42 Obama +13

McClatchy/Marist 3/20 – 3/22 846 RV 48 43 Obama +5

Notice the CNN poll is way out of line with the other two polls that bracket it.

crosspatch on April 17, 2012 at 11:19 AM

RCP should remove the CNN polls from their averages until this is cleared up.

crosspatch on April 17, 2012 at 11:20 AM

Again, I’m not aware of a need to weight results in general-population polls

If they oversampled any category, say ethnic minorities, they have to weight all observations to correct for it.

DrSteve on April 17, 2012 at 11:23 AM

RCP should remove the CNN polls from their averages until this is cleared up.

crosspatch on April 17, 2012 at 11:20 AM

or at least until they reveal their methodology and sampling…

jimver on April 17, 2012 at 11:24 AM

This is same-old, same-old for the MSM. The news industry knows that voters tend not to turn out for a candidate they’ve been lead to believe will lose the election. By reporting low numbers for Romney, they’re influencing Romney supporters to stay home while Obama zombies come out in optimistic droves. It’s a a typical Journolister-type technique to undermine the democratic process.

Crusty on April 17, 2012 at 11:14 AM

Ditto

SauerKraut537 on April 17, 2012 at 11:26 AM

RCP should remove the CNN polls from their averages until this is cleared up.

crosspatch on April 17, 2012 at 11:20 AM

Again, RCP does not question polls. It is RCP’s biggest flaw in my opinion. They treat all polls as being valid, and calculate their RCP average using all polls.

milcus on April 17, 2012 at 11:28 AM

Why should anyone be optimistic that Romney will survive the Obama/media war that is about to be unleashed on him. He will be lucky to come out of the general election with a 52-43 loss as the CNN poll portrays.

TheRightMan on April 17, 2012 at 10:32 AM

Uh, because Romney ISN’T the most devisive, poorest performing, optically disasterous President in the history of the United States? Just spitballing here…

Abiss on April 17, 2012 at 11:29 AM

This poll was conducted by Opinion Research Corp.
I went looking to find any source of bias, and was surprised.

ORC was owned from 2006-2011 by Clinton-friendly InfoUSA.
In July 2011 ORC went independent with a majority stake from Lake Capital, located in Chicago.
A search of political donations from ORC revealed nothing of interest.
A search of political donations from Lake Capital revealed overwhelmingly strong Republican leanings.

Where is this bias coming from?
I doubt that the bias is coming from CNN. ORC does $500 million in revenue internationally, and isn’t going to risk their reputation on CNN’s say-so.
Is there still a big Clinton influence there?

Where is this bias coming from?

topdog on April 17, 2012 at 11:30 AM

Tell math to go solve it’s own problems.

philoise65 on April 17, 2012 at 11:30 AM

Why should anyone be optimistic that Romney will survive the Obama/media war that is about to be unleashed on him. He will be lucky to come out of the general election with a 52-43 loss as the CNN poll portrays.

TheRightMan on April 17, 2012 at 10:32 AM

Option B: Because Romney ISN’T an open Marxist? I don’t know…could be something favorable there. Maybe.

Abiss on April 17, 2012 at 11:30 AM

I’m not sure what all the concern is over. Rmoney™ will coast to a win. He is after all the only “true con” in the race. No worries.

Bmore on April 17, 2012 at 11:31 AM

Why should anyone be optimistic that Romney will survive the Obama/media war that is about to be unleashed on him. He will be lucky to come out of the general election with a 52-43 loss as the CNN poll portrays.

TheRightMan on April 17, 2012 at 10:32 AM

Because Romney didn’t steal $3 TRIL of your dollars for his rich friends, while Timmy and Ben inflated away 50% of your buying power? …To brainy? Probably. Doesn’t fit on a bumpersticker yet.

Abiss on April 17, 2012 at 11:33 AM

Pollgate has been exposed, by none other than Ed Morrissey.

Again – 1980 saw the exact same manipulation and flat out lying in polls by every media outlet. And yes I said every media outlet, sans HA.

See when Fox or the “right” channels publish these “polls” – they are complicit. Ed publishes them to expose the frauds they are. And the average HA reader both adds content – or gets the narrative right from the start “these lying POS are trying to lie to our faces”

In a Freakenomics kind of way – take a look at “common” websites with high volumes – and posts/forum answers. When Yahoo post an article and has 5,000 comments about Obama – with 70-80% in the negative – thats all you need to know.

The 18-22 bracket of “energized dopes who voted for Obama” are silent – and pissed their first forey into politics turned into an expensive dud. The new 18-22 years olds just saw tuition go up 50%, mommy and daddy stuggling to buy that Xbox game or no job at all.

OH – and speaking of optics – Hillary having a beer, in the midst of a scandal in a foreign country… supports the notion “these morons are doing nothing, no matter where they are”

It was a desperate attempt to “humanize that normal lady”… which always fails miserably.

Odie1941 on April 17, 2012 at 11:33 AM

Why should anyone be optimistic that Romney will survive the Obama/media war that is about to be unleashed on him. He will be lucky to come out of the general election with a 52-43 loss as the CNN poll portrays.

TheRightMan on April 17, 2012 at 10:32 AM

Because Romney doesn’t inspire more and more random civilians to start talking about 2nd Amendment solutions, apropos of nothing really, in the Whole Foods checkout line? Anecdotes are effective, and we’ve got a boatload of them. Come on, could work

Abiss on April 17, 2012 at 11:36 AM

Why should anyone be optimistic that Romney will survive the Obama/media war that is about to be unleashed on him.

Because Americans with a brain have learned to ignore everything coming out of the Obama “administration”?

Those without a brain will be voting Obama anyway.

NoDonkey on April 17, 2012 at 11:36 AM

I love how RomneyAir dismisses any poll from CNN but anything out of Fox News is treated as gospel never to be questioned. Where was the in depth analysis of that poll? Oh that poll showed Romney winning so nobody took a second look or questioned the methodology.

angryed on April 17, 2012 at 10:10 AM

If you follow the link to the PDF file of a Fox News poll, it usually lists the partisan breakdown of the polling sample. Fox News polls have been known to use samples favoring Democrats, but at least the partisan breakdown is given, so that analysts can adjust the results for what they believe is the true partisan split of the entire electorate (which nobody really knows before the election).

Even if a poll does not give the partisan breakdown of the sample, if the poll lists the percentages of Republicans, Democrats, and Independents that favor each candidate, it’s possible to back-calculate the D/R/I split of the sample.

If
Xr = fraction of Republicans voting for candidate X
Xd = fraction of Democrats voting for candidate X,
Xi = fraction of Independents voting for candidate X
Xt = fraction of the total sample voting for candidate X, then

XrR + XdD + XiI = Xt

The same equation can be written for the other candidate Y:

YrR + YdD + YiI = Yt

The sum of the fractions of Republicans, Democrats, and Independents in the sample must equal the total sample, or

R + D + I = 1

This results in 3 linear equations for the 3 unknowns R, D, and I, representing the fraction of Republicans, Democrats, and Independents in the sample. These can be solved by standard methods of matrix algebra or determinants, for any poll between two candidates.

Applying this method to the Fox News poll resulted in 41% Democrats, 39% Republicans, and 20% Independents in their sample. While it could be argued that Fox News under-sampled Independents, their poll sample did not favor Republicans.

Steve Z on April 17, 2012 at 11:37 AM


In a Freakenomics kind of way – take a look at “common” websites with high volumes – and posts/forum answers. When Yahoo post an article and has 5,000 comments about Obama – with 70-80% in the negative – thats all you need to know.

The 18-22 bracket of “energized dopes who voted for Obama” are silent – and pissed their first forey into politics turned into an expensive dud. The new 18-22 years olds just saw tuition go up 50%, mommy and daddy stuggling to buy that Xbox game or no job at all.

OH – and speaking of optics – Hillary having a beer, in the midst of a scandal in a foreign country… supports the notion “these morons are doing nothing, no matter where they are”

It was a desperate attempt to “humanize that normal lady”… which always fails miserably.

Odie1941 on April 17, 2012 at 11:33 AM

Yah, well…there’s that too. See TRM? Reason for an ever so faint ray of optimistic hope, do in fact, – theoretically anyway – exist. Cheer up sunshine.

Abiss on April 17, 2012 at 11:39 AM

This is same-old, same-old for the MSM. The news industry knows that voters tend not to turn out for a candidate they’ve been lead to believe will lose the election. By reporting low numbers for Romney, they’re influencing Romney supporters to stay home while Obama zombies come out in optimistic droves. It’s a a typical Journolister-type technique to undermine the democratic process.

Crusty on April 17, 2012 at 11:14 AM

This is exactly it. Someone upthread mentioned the MSM is the Fifth Column: according to Britannica.com, a fifth column refers to any clandestine group or faction of subversive agents who attempt to undermine a nation’s solidarity. This needs to be kept front and center, and the MSM needs to be called out on this over and over and over again until they are destroyed.

PatriotGal2257 on April 17, 2012 at 11:39 AM

I love the picture associated with the story (especially that woman in the back row).

teejk on April 17, 2012 at 11:41 AM

Jay Leno: “President Obama released his tax returns. It turns out he made $900,000 less in 2011 than he did in 2010. You know what that means? Even Obama is doing worse under President Obama.”

Four more years!

RobertMN on April 17, 2012 at 11:43 AM

For those that remember the media had Carter winning all the way up to election day and was surprised and heart broken when Reagan won in a landslide. Do not believe any of the MSM spin just keep working your community getting conservatives involved and informed registered and to the polls on voting day. The Senate is vital as well in saving our country.

Alborn on April 17, 2012 at 11:43 AM

Since then I think of the (old) media the same way actors were thought of during Shakespeare’s time.

cozmo on April 17, 2012 at 11:19 AM

i think of them like a dcotor that intentionally makes you sick. or an engineer that drafts up a flawed bridge by design. or a CPA that adds to your tax bill. yeah, their mothers must be proud.

t8stlikchkn on April 17, 2012 at 11:43 AM

So help out this math-deficient person. Based on the many numbers in this article, what are the actual percentages? Is it Obama 48, Romney 47?

TheLoudTalker on April 17, 2012 at 11:48 AM

If they oversampled any category, say ethnic minorities, they have to weight all observations to correct for it.

DrSteve on April 17, 2012 at 11:23 AM

Right. So say the average voter split on election day is something like 35/35/30 D/R/I and your sample of people is 50/25/25, you would weight the results of the sample to attempt to mirror what would have happened with a more accurate group.

So you would take the responses by Democrats and weight them down and take the responses of Republicans and Independents and weight those up.

In this case, it looks like CNN has their weighting completely out of whack. That is evidenced by CNN showing Obama winning the vote among men by a comfortable margin. Obama hasn’t been leading among men in any other poll for quite some time. This exposes the CNN poll as flawed, in my opinion.

crosspatch on April 17, 2012 at 11:49 AM

It served it’s purpose.

motionview on April 17, 2012 at 11:50 AM

So help out this math-deficient person. Based on the many numbers in this article, what are the actual percentages? Is it Obama 48, Romney 47?

TheLoudTalker on April 17, 2012 at 11:48 AM

That’s the way I read it. Basically a statistical tie, as are several of the other polls. The difference is within the margin of error for the poll so it is a tie.

crosspatch on April 17, 2012 at 11:50 AM

Thank you crosspatch.

TheLoudTalker on April 17, 2012 at 11:51 AM

What Bob Beckel said.

Roy Rogers on April 17, 2012 at 11:53 AM

It does the left no good to put out polls showing Obama with a good lead. All that does is pacify the liberal voters, and activate the conservative ones.

keep the change on April 17, 2012 at 11:55 AM

I love the picture associated with the story (especially that woman in the back row).

teejk on April 17, 2012 at 11:41 AM

Same here. Likes like Mike Krzyzewski of Duke. Interested to know when it was taken. The woman in the pic has a severe case of bug-eye. :)

itsspideyman on April 17, 2012 at 11:56 AM

So help out this math-deficient person. Based on the many numbers in this article, what are the actual percentages? Is it Obama 48, Romney 47?

TheLoudTalker on April 17, 2012 at 11:48 AM

yes, somewhere in that area…

this: ‘Mitt Romney is supported by 47% of national registered voters and Barack Obama by 45% in the inaugural Gallup Daily tracking results from April 11-15. Both Obama and Romney are supported by 90% of their respective partisans. The Republican results show that despite the rancor and divisiveness of the Republican campaign, the vast majority of Republicans are backing Romney in the head-to-head battle with Obama, as they have in ballot tests earlier this year.

The crucial voting bloc of independents breaks toward Romney by 45% to 39%, giving the GOP challenger his slight overall edge.

jimver on April 17, 2012 at 12:01 PM

It does the left no good to put out polls showing Obama with a good lead. All that does is pacify the liberal voters, and activate the conservative ones.

keep the change on April 17, 2012 at 11:55 AM

It indeed does those two things, but it also makes the spineless handwringers in the middle want to jump on the bandwagon of whoever appears more popular. That’s why they do it.

Kataklysmic on April 17, 2012 at 12:05 PM

CNN, Fox, ABC, CBS, NBC, etcetera, should poll their own employees. My bet is that Fox would be the only poll with less 90% voting for the ONE.
Does the CNN poll have the breakdown for the hoody-wearing demographic?

racquetballer on April 17, 2012 at 12:08 PM

Where is that photo from? Too funny.

itsnotaboutme on April 17, 2012 at 9:54 AM

Yeah, I’ve never seen that one before…

I’m a Duke fan, and even I thought it was funny. Triple facepalm, and that girl is staring down either Wojo or someone else beyond him. Her wild eyes are too funny.

ITguy on April 17, 2012 at 12:10 PM

Translation: I don’t care what reality says, I’ll keep closing my eyes to polls I don’t like. Cool dude. I just read a poll taken by Sean You’re a Great American Hannity. It says 198% of the public will vote for Romney.

angryed on April 17, 2012 at 10:18 AM

Dissecting and pointing out the deep flaws in a poll does not equal “closing one’s eyes.”

bmmg39 on April 17, 2012 at 12:11 PM

Comment pages: 1 2 3