Editore"s Note
Tilting at Windmills

Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

September 26, 2007
By: Kevin Drum

POLL LITERALISM....Everybody is beating up on David Brooks for his column yesterday, and rightly so. As he does so often, Brooks based his Tuesday column on the conceit that he has some special insight into what "ordinary" voters think, an insight that (a) usually turns out to be remarkably close to what David Brooks himself thinks and (b) bears little relation to what actual polls say.

This is one of Brooks's worst habits, and it's a tiresome one. But there's an opposite habit that can be equally dangerous: "poll literalism." This is a common malady among liberals — I'm guilty of it myself periodically — who frequently claim that "everyone agrees with us" based on poll results that show support for liberal positions on a wide variety of issues. And it's true: polls really do show this with some regularity. Unfortunately, answers to poll questions come in a vacuum. They don't show what people think once the other side has a chance to get a few licks in.

Here are two examples. First, withdrawal from Iraq. A recent New York Times poll showed that 65% of respondents want to withdraw either some or all of our troops from Iraq. Hooray! The country is with us! But then the Times asked a followup question: "What if removing troops meant Iraq would become more of a base of operations for terrorists, then would you still favor removing U.S. troops from Iraq, or not?"

Guess what? Of that 65%, only 30% still favored removal. That's a huge drop based on a single hypothetical, and in a real campaign that hypothetical would practically blanket the airwarves. It wouldn't convince everyone, of course, but it would probably convince a sizable chunk. The odds are that in real life — i.e., during a campaign in which voters were responding to actual arguments instead of casually answering poll questions over a telephone — there's something like 30% who want to stay in Iraq, 30% who want to get out, and 40% somewhere in the middle who aren't really sure what to do.

Example #2 comes from a much derided recent poll conducted by Celinda Lake for Joe Biden. The reason it was derided (aside from the fact that Washington Post reporter Chris Cillizza failed to inform his readers that Biden was behind the poll) was because of the wording of one of the questions: "Some people say [your Democratic incumbent] is a strong supporter of Hillary Clinton and will support her liberal agenda of big government and higher taxes if she becomes president," the poll stated, before asking respondents whether they would still vote for their incumbent or choose a Republican candidate.

Outrageous! And it is. On the other hand, that's exactly what Republican House candidates are going to say, isn't it? Which means that this poll, showing a 6-point lead for Democratic incumbents, is probably more useful than generic polls showing a 10 or 15 point lead.

Now, obviously this works in both direction: liberals get to make arguments during campaigns too. But conservative arguments appeal to fear pretty effectively, which means that on difficult, highly charged issues, like withdrawing from Iraq, a lot of people tend to be schizophrenic. One day they want to get out, but then they see a scary TV ad and the next day they don't.

Brooks has a bad habit of ignoring poll results that he doesn't want to acknowledge, but it can be nearly as debilitating to go in the other direction and take poll results too seriously. Middle America may not be as hawkish as Brooks imagines them to be, but they probably aren't as noninterventionist as the blogspheric left imagines either. The truth isn't always in the middle, but in this case it probably is.

Kevin Drum 12:47 AM Permalink | Trackbacks | Comments (70)

Bookmark and Share
 
Comments

Brooks based his Tuesday column on the conceit that he has some special insight into what "ordinary" voters think, an insight that (a) usually turns out to be remarkably close to what David Brooks himself thinks and (b) bears little relation to what actual polls say.

Actual polls don't always reflect what "ordinary voters" think either. They reflect what voters who don't slam the phone down want the polltaker to think about them.

Pollster: "Would you vote for a woman for president?"

Respondents: "Why SURE I would! What troglodyte would say anything else?"

"Ever eaten a whole quart of ice cream at a sitting?"

"Never!"

"Do you actually follow politics all day long, or do you, like the vast majority of Americans, actually have a life?"

"If I had a life, would I be sitting on the phone talking to you for twenty minutes?"

Posted by: harry on September 26, 2007 at 12:55 AM | PERMALINK

P.S.

...liberals get to make arguments during campaigns too. But conservative arguments tend to tug more strongly at emotions...

Oh, no, liberal political campaigns and arguments never play on emotions, do they?

Are you kidding?

Posted by: harry on September 26, 2007 at 12:57 AM | PERMALINK

Meet harry, the new Al.

Posted by: Disputo on September 26, 2007 at 1:09 AM | PERMALINK

told ya, voters are generally naive and stupid. they belive in misleading ads and smears and bogus attacks on people's character. it happens all the time and will happen again this year.

Posted by: bob on September 26, 2007 at 1:12 AM | PERMALINK

Wow- great post. Re: poll literalism, I think what happens is when we start to interpret a poll, we don't think "Well what were the respondents thinking? How would they come around to supplying that answer?" Because if we did, we'd find out that typically it looks like something a lot different than the simple "poll literalism" explanation of what happened- the voters voting one way might have two or three different explanations, none of whch is necessarily shared by everyone who gave the same answer, for answering the way they did.

When you wrote that David Brooks has "the conceit that he has some special insight into what 'ordinary' voters think" it reminded me a little bit of myself, althiugh I'm sure I probably interpret what the man in the street is thinking a lot differently that David Brooks most of the time. However, this brings me to think of another problem I think I see in how liberals view polls: I think in different social classes, because of the disparity in education and other social factors, people in different social classes see the world around them and events in it very differently. So I think a lot of relatively-upper class analysts of polls don't necessarily have a good understandig of what they're looking at when they look at how a bunch of regular, relative lower-class people responded to a question.

http://www.swanpoliticsblog.blogspot.com

Posted by: Swan on September 26, 2007 at 1:12 AM | PERMALINK

Outrageous! And it is. On the other hand, that's exactly what Republican House candidates are going to say, isn't it? Which means that this poll, showing a 6-point lead for Democratic incumbents, is probably more useful than generic polls showing a 10 or 15 point lead.

Absolutely. I think with a poll like that you're simulating somewhat the narrowing effect that occurs in the polls in the run-up to an election.

Posted by: Swan on September 26, 2007 at 1:15 AM | PERMALINK

I am glad to see someone picked up on the flip side of the Biden poll. While it certainly was a textbook push poll, I think some folks missed the real meat of the story.

A shot at Hilary, sure. But if you want to know how voters are going to respond to Republican rhetoric, it was informative.

Controlled, Pew-style polling is great for studying trends and preferences, but sometimes you need to run one up the flagpole and see who salutes.

Posted by: Matthew on September 26, 2007 at 1:15 AM | PERMALINK

Harry: I was still editing the post when you wrote your comment. I don't know that you'll like the current version any better than the original, but it's now more narrowly targeted on fear than on emotion generally.

(Though I do think that conservatives do a better job of appealing to emotions than liberals. They both do it, but conservatives do it more effectively.)

Matthew: Absolutely. This post isn't meant as a shot against polling, just a shot against taking answers to complex poll questions too literally. To determine long-term trends, you need to have straightforward questions that are the same time after time. For example, we may not know for sure how many people really want to withdraw from Iraq, but we do know that the number is a lot higher today than it was two years ago.

Posted by: Kevin Drum on September 26, 2007 at 1:24 AM | PERMALINK

The real problem with polls is they are useless to predict what the American population think. How many people do polls poll anyway? Several hundred or thousand people? Yet they use that to conclude that's what the whole population of hundreds of millions of Americans think.
I could go outside my office right now and ask several hundred people a couple of questions too, but would you accept that the same proportion of Americans would give the same answer as the people I polled? Of course not. Yet pollster think they do. This is why your criticism of David Brooks is misleading. There's no reason to think the pollsters have any greater insight than David Brooks so why shouldn't we listen to what Brooks says instead of what the pollster says his polls say?

Posted by: Al on September 26, 2007 at 1:28 AM | PERMALINK

Well I think Harry's first comment makes a good point about polls.

Poll: Are you racist? Do you have some disgusting belief?

Respondents: No, no, of course not.

Republicans: See? [holding up poll as proof that Southerners, or whoever, are not racist anymore, or whatever- implicit offering of proof that it's safe, for example, for blacks or hispanics or women or whoever to vote GOP]

Posted by: Swan on September 26, 2007 at 1:35 AM | PERMALINK

Al sez: "There's no reason to think the pollsters have any greater insight than David Brooks so why shouldn't we listen to what Brooks says...?"

SHORTER Al: The only reliable and authoritative source is one that just happens to echo my opinions and worldview, exactly. Now do you get it?

Posted by: shystr on September 26, 2007 at 1:51 AM | PERMALINK

Harry is on the right track in his first response -- something to keep in mind Kevin, every year when EVERY BLOGGER IN THE WORLD asks me to take a 20 minute survey to help determine which ads I will be blocking.

Posted by: jerry on September 26, 2007 at 1:51 AM | PERMALINK

This is an excellent post. This is something the nutroots dont understand. They read these polls and then think the Dem nominee can run way to the left and expect to win. But that doesnt work. Polls have always shown these kind of results, back when Dukakis got beat, when Mondale got clobbered,... People said the same stuff. I think alot of it has to do with people telling pollsters what they think they want to hear.

Posted by: DigDug on September 26, 2007 at 2:02 AM | PERMALINK

Ten comments - and three servings of pie.

Posted by: Blue Girl, Red State (aka G.C.) on September 26, 2007 at 2:04 AM | PERMALINK

So Al is statistically illiterate. What a surprise.

Getting a good, representative sample is hard (and for the dishonest pollster gives more ways to cook poll results than a chicken), but the math behind polling is very well-tested. (Industrial quality control is done on similar principles.)

Posted by: idlemind on September 26, 2007 at 2:06 AM | PERMALINK

Yeah, this is a really excellent post and worth keeping in mind. I wonder how we could broaden its perspective a bit so as to look at arguments besides lefty blogs vs. fatuous Beltway types. Are there some polling sources that are prone to "reverse Lakes," posing questions in ways that conservatives might make similar complaints about? I think Kevin's right about fear as a major component of Republican advertisements, but surely Democrats aren't immune to sliming and scaring. Maybe the average of two partisan polls is likely to be similar to the result found by an ostensibly non-partisan poller.

Fundamentally, people will take whatever empirical or pseudo-empirical evidence they can get, especially when the alternative is Brooks-ish bloviating.

Posted by: Zack on September 26, 2007 at 2:07 AM | PERMALINK

idlemind, a couple of weeks ago, Al tried to tell us that statisticians throw out data all the time that they don't want to consider. It's called an outlier - but I prefer to call Al an out-and-out liar.

Posted by: Blue Girl, Red State (aka G.C.) on September 26, 2007 at 2:08 AM | PERMALINK

Why do you assume the number who prefer a pull-out is less than the poll-estimated 65%? Sure, when you frame the question by implying that a pull-out will create a terrorist base, fewer favor a pull-out. But I'd bet if you framed the question emphasizing a possible regional war and thousands of more US troops killed if we stay, then the number would go up from 65%.

If half the 65% can be swayed by framing, then presumably half of the original 35% can also be swayed by the opposite framing, suggesting roughly 30% firmly against, 20% firmly for, and 50% who are wafflers. But in any case, do you really always rule out of discussion the 50% of respondents to any question who are not adamant? It still seems pretty relevant that 30% are dead set against the war, another 35% are against it under most circumstances, and possibly another 20% may be against if it is framed the right way.

Posted by: JD on September 26, 2007 at 3:09 AM | PERMALINK

Midwest High school educated women???? WHAT?

What a fruitloop, but the message is clear.

Netroots is not THAT scary. Right, but Gonzales just went bye, bye and Rudy's only BIG campaign issue is that MoveOn.org ad in the NYT. That issue and that he's official crowned himself the king of 9/11, 9/11, 9/11.

If it were not for liberal bashing, what other policy would Repugs have?? Why is Brooks writing about some itsy bitsy blog by the name of Daily Kos? I mean, Brooks did waste a whole column on it.

Hey maybe the Repugs have hear of it, you think?

AND Broder already had to the make the issue twice on MTP - you know - about those foul mouth liberal bloggers - that nobody should pay any attention too. Hey, why not just say the word "SEX" and see if anyone tunes in. Those liberal's that are 3.2 million strong over at MoveOn.org, so that before you know it, they're coming out of woodwork and lurking around every corner. But the Rove insecticide spray doesn't seem to work anymore.

Posted by: Me_again on September 26, 2007 at 3:36 AM | PERMALINK

on difficult, highly charged issues, like withdrawing from Iraq, a lot of people tend to be schizophrenic

Not so much schizophrenic as ignorant and confused. Expecting consistency from the "ordinary" voter is too idealistic. You can get anything you want by asking thoughtful questions in a poll, but come election time all it takes is a swiftboat group with a wealthy backer and poof -- there goes that imaginary majority you had in the polls.

Posted by: JS on September 26, 2007 at 3:57 AM | PERMALINK

What this (along with the comment from Bill Sammon that Kevin reports below) makes clear is that the Dems had better make clear why staying stuck in Iraq is actually BAD for our security. Of course, in order to do that, they themselves have to think much more seriously about military and security issues than the Democratic Party has wanted to do for some time now (as Matt Yglesias pointed out in an excellent American Prospect piece a few years ago).

As for that "wealthy swiftboat backer": may I suggest that Kerry would have been a lot less vulnerable to that campaign if he had actually tried to, you know, RESPOND to it? (One encouraging feature of the new campaign is that none of the Democratic frontrunners seems willing to wear a "Kick Me" sign the way those two dolts Gore and Kerry did.)

Posted by: Bruce Moomaw on September 26, 2007 at 5:19 AM | PERMALINK

Al: "I could go outside my office right now and ask several hundred people a couple of questions too, but would you accept that the same proportion of Americans would give the same answer as the people I polled? Of course not. Yet pollster think they do."

You're right, Al. I wouldn't accept the result because like most everything else you spout off about, you obviously don't know the first thing about polling.

Good polling is a fine art, and the masters always take pains to survey a truly representative sampling of the population, and to ensure that their series of questions (a good politics poll usually takes 10 minutes minimum to run by a participating individual) in no way manipultes or leads their targeted sample to a desired or predetermined outcome (a.k.a. push-polling).

Such reputable pollsters will take into account a myriad of factors, such as age, sex, race and ethnicity, level of education, location of domicile, occupation, level of income, political persuasion -- and, last but not least, consistency of both the questions and the answers they elicit.

I worked with one highly regarded pollster here in Honolulu who was also a statistics professor at the University of Hawaii. He would literally ask the same key questions four or five completely different ways during the course of an interview, and do it in such a manner that his subject would almost never catch on. From this, he could determine the degree of open-mindedness on the part of the interviewee, and the level of commitment to his or her avowed position on a given candidate or issue.

Not only was he uncannily accurate at predicting election outcomes, but he could also call voter turnout down to the district and precinct level within one or two percentage points. He correctly predicted Hawaii's 1998 gubernatorial race would go to the incumbent by a narrow 1% or 2% margin, in large part because he had determined that the challenger's public overconfidence was causing 12-15% of her staunchest supporters to believe her election a foregone conclusion, and thus they would fail to vote on Election Day. He was the only pollster in the islands who called that upset.

Individual polls merely provide an interesting snapshot of public opinion at given moments of time -- as long as one keeps in mind that public opinion can be quite fickle, and can change on a dime depending upon cirscumstances or events. If done right, it can be an invaluable tool to a campaign. If done wrong, it can hamstring you.

Posted by: Donald from Hawaii on September 26, 2007 at 6:28 AM | PERMALINK

I think the average Dem voter (like my husband and I) are more ambivalent about Iraq than the netroots. Part of it stems from fear, but also in part from the "Pottery Barn" rule: we owe these people something after coming in and destroying their country on a whim. Especially those who have risked a lot to work with us like translators, etc...

Posted by: Teresa on September 26, 2007 at 7:11 AM | PERMALINK

Make a pundit/monster out of Limbaugh, Hannity, Boortz, O'Rilley and three or four other right-wing shills; David Brooks will be worse than that.

Posted by: Bathrobespierre on September 26, 2007 at 7:36 AM | PERMALINK

drum: Brooks has a bad habit of ignoring poll results that he doesn't want to acknowledge


you can't have a republican apologist without...

denial...

its not a weapon...

its a disease..

and how's that working out?

Posted by: mr. irony on September 26, 2007 at 7:52 AM | PERMALINK

I think the average Dem voter (like my husband and I) are more ambivalent about Iraq than the netroots. Part of it stems from fear, but also in part from the "Pottery Barn" rule: we owe these people something after coming in and destroying their country on a whim.

Well, what if your Pottery Barn customer kept breaking more vases and told you not to worry, six months from now the vases will magically fix themselves? Maybe the best thing for everybody is to get this customer out of the store as fast as possible.

And yes, we owe the Iraqis refugee status. Hundreds of thousands of visas. Forget the security check: if any U.S. soldier or civilian can vouch for them, let them in. I can't see how you can do a security check on anybody over there, considering all the records are suspect or missing anyway. That's the "you bought it" part of the Pottery Barn rule.

Posted by: ericblair on September 26, 2007 at 8:31 AM | PERMALINK

David Brooks lives in a bubble of his own design...when I see him on TWWGS and actually can AGREE with something Will says in dispute of Brooks it's a happy day for me! This pompous fool is to the right what spittle boy Matthews is to the left (if in fact anyone can figure out where his behind actually comes down)...But the real truth is than NONE of these paid for TV pundits have any sense at all of what the "little people" really think/feel/deal with...and when they begin to talk about taxes and other economic issues they make me want to puke! Of course, as Sean Hannity says, there is NO HEALTH CARE CRISIS...because there isn't for him and others that make MILLIONS yearly...my GOD, people, wake up!

Posted by: Dancer on September 26, 2007 at 8:47 AM | PERMALINK

Brooks has a bad habit of ignoring poll results that he doesn't want to acknowledge

Brooks also has a bad habit of lying about or misrepresenting the polls he does reference. He once used this poll to tell us about what "poor republicans" think vs. "poor democrats."

Here's Brooks: 76 percent of poor Republicans believe most people can get ahead with hard work. Only 14 percent of poor Democrats believe that.

This study is a typology, not a demographic study, i.e., people are placed in categories based on their opinions, not their circumstances. There is no category called "poor Democrats" or "poor Republicans" in the study.

Here's the income breakdown of what Brooks calls "poor Republicans" (actually a political type in the study called "Pro-Government Conservatives")
51% make over $30,000
21% make over $50,000
10% make over $75,000

Here are the incomes of what Brooks calls "poor Democrats":
49% make over $30,000
22% make over $50,000
8% make over $75,000

He and his fellow bloviators do this all the time. We should never, ever expect honesty from these people.

Posted by: R. Porrofatto on September 26, 2007 at 9:02 AM | PERMALINK

I agree with JD.

How come they didn't ask the question:

"What if leaving troops in Iraq made it more of a base for terrorists, would you still favor leaving U.S troops in Iraq?

Of the 30% who want to stay in Iraq, I can't believe more than half of them would answer yes to that question.

Posted by: DR on September 26, 2007 at 9:12 AM | PERMALINK

And BTW, Kevin, the Republicans are ALREADY making that argument. And they have been for over 5 years. Most people don't believe them any more.

Posted by: DR on September 26, 2007 at 9:14 AM | PERMALINK

How come they didn't ask the question: "What if leaving troops in Iraq made it more of a base for terrorists, would you still favor leaving U.S troops in Iraq? Of the 30% who want to stay in Iraq, I can't believe more than half of them would answer yes to that question.

Of the 30% who want to stay in Iraq, 29% would simply put their fingers in their ears, hum loudly and refuse to answer the question.

Posted by: shortstop on September 26, 2007 at 9:31 AM | PERMALINK

This is an excellent post. This is something the nutroots dont understand. They read these polls and then think the Dem nominee can run way to the left and expect to win. But that doesnt work.

The position of the "nutroots" is not that politicians should run way to the left but that they have no reason to run away from fundamental progressive principles. The polls may be more complicated then they appear but, at a very minimum, they demonstrate that people are quite accepting of most liberal ideas. The kind of repudiation of those ideas that Democrat politicians typically engage in is, at best, unnecessary.

Liberals/progressives who raise polling data are supporting a position with evidence. Yes, Kevin, that evidence can be more complex than it appears at first glance but it is not so complex that it can mean the precise opposite of what it appears to show.

People do not like this war, for instance. There may be gradations and all sorts of qualifications to what any individual believes but the basic truth, that people do not like this war, remains. I don't agree, Kevin, that making that argument on the basis of that sort of poll is being overly literal. It is a reasonable interpretation of the evidence, even if that interpretation could be a little bit more nuanced. The same cannot be said for the arguments of Brooks and the Conventional Wisdom Brigade (CWB). Thus, I don't think its a fair comparison.

Posted by: brent on September 26, 2007 at 9:31 AM | PERMALINK

Note how Brooks had to get a dig in against Edwards in the last line.

Slimy.

Posted by: Chrissy on September 26, 2007 at 9:42 AM | PERMALINK

Excellent post re analytical scepticism re polls. Learned
from it. Thank you.

Posted by: Nietz's Epigone on September 26, 2007 at 10:00 AM | PERMALINK

I think Kevin's right about fear as a major component of Republican advertisements, but surely Democrats aren't immune to sliming and scaring. Maybe the average of two partisan polls is likely to be similar to the result found by an ostensibly non-partisan poller.
Posted by: Zack on September 26, 2007 at 2:07 AM
---

Exactly. But, it doesn't have to be "sliming". It can be scaring that is downright honest. How about a campaign ad showing the government reinstituting the draft because we don't have enough soldiers-we're here to take YOUR KID! See how that plays with the middle-class soccer moms. The problem is the lack of creativity and nerve on the part of the Democratic establishment. The 77% of Americans that want us to get out of Iraq within a year is not a mirage. All you need is the truth to activate that amygdala to keep those numbers.

Posted by: Doc at the Radar Station on September 26, 2007 at 10:11 AM | PERMALINK

Of the 30% who want to stay in Iraq, 29% would simply put their fingers in their ears, hum loudly and refuse to answer the question.

And 0% would volunteer to sacrifice their own lives, or their own childrens' lives, on the altar of the Bush's legacy and the neocons' political ambitions. 'Nuff said.

Posted by: Gregory on September 26, 2007 at 10:14 AM | PERMALINK

Kevin,

I don't think you understand why people were deriding the Lake/Biden poll. The poll found that "Whether the question named Clinton or Obama, the Democratic incumbent's lead shrank to an average of six points". Both candidates they asked about caused the exact same drop. It's likely that had they also tested the names Edwards, Dodd, Kucinich, and yes, Biden, you would have also seen similar drops.

If the poll was merely designed to show that Republican frames can hurt Democratic downticket candidates, that would be one thing. But this poll was presented as thought this were a particular problem with Obama and Clinton, when there's no evidence and no reason to believe that this very generic frame would be more potent against them, and worse, it was presented as an Obama/Clinton problem by a Biden pollster who was never identified as such.

That's the problem with the Lake/Biden poll. It doesn't show exactly what Lake was alleging it shows (that Obama/Clinton are more susceptible to "tax and liberal agenda" attacks than other Democratic candidates), and knowing that Lake works for Biden is a key piece of information to understanding why she was misrepresenting the poll.

Posted by: Royko on September 26, 2007 at 10:26 AM | PERMALINK

The real dishonesty of Biden's pollster wasn't just push-polling a probable Republican message, but doing so only for Clinton and Obama and claiming the results showed that they were worse for downticket Democrats than other unnamed, unpolled Democrats. The fact that there are legitimate answers that can be derived from that poll doesn't make that interpretation any less dishonest, or the reporter any less suspect for running with it.

Posted by: Redshift on September 26, 2007 at 10:39 AM | PERMALINK

ericbair:And yes, we owe the Iraqis refugee status. Hundreds of thousands of visas. Forget the security check: if any U.S. soldier or civilian can vouch for them, let them in.

Yep, we're fighting them over there so we can bring them back here. Thank you, GWB, for your vision and leadership in the Global War against Islamic extremists.

Posted by: PTate (back in MN, sigh) on September 26, 2007 at 10:42 AM | PERMALINK

Kevin, it is sooo easy for you to besmirch Lord David - When was the last time, you checked the pulse of the hoi polloi? Why, the good Lord took the time just last year, to put on his newly arrived mail-order Pendleton and chinos and drove into the hinterlands, to visit the gun section of a Wal-Mart - Even stopped at an IHOP, or an Appleby's, pre-IHOP, I believe to allow the masses to stroke that tartan Pendleton sleeve, or to kiss his ring, or whatever. Must mingle, Kevin, mingle.

Posted by: thethirdPaul on September 26, 2007 at 11:06 AM | PERMALINK

Sorry about the timing PTate.

You missed the Prairie Home Companion Street Dance, but on the bright side, you are home in time for winter!

Posted by: Blue Girl, Red State (aka G.C.) on September 26, 2007 at 11:07 AM | PERMALINK

One trouble with polls is that they record the opinions people verbalize, but do not necessarily record the opinions people will act on.

We are conscious of the decisions we make with our verbal consciousness. But many decisions we make have been pushed down out of our conscious mind and have become routine.These are decisions we make and are not aware of making.

Then there are those decisions we make with our emotional minds. The emotions have made decisions for mammals since mammals have existed, and They are a very powerful decision-making system. Thomas LeDoux has studied emotions for decades and has concluded that the conscious mind and the emotional mind operate using such different methods that they do not directly speak to each other.

The result is that voters make most decisions using their emotions since they generally have little direct knowledge of the individuals or situations, but they respond to polls using their conscious minds. Is it any surprise that the conscious minds tend to be used primarily to justify the decisions already make by the emotional minds?

To me the big surprise is that polls tell us much of anything at all. Ask a poll-taker about the false opinions created by respondents when they know nothing about the subject and just want to provide an answer to the poll-taker so as to not look stupid.

Posted by: Rick B on September 26, 2007 at 11:22 AM | PERMALINK

Blue Girl,

Does it really matter whether in France or in Minnesota, her writing is still as erudite and astute?

Funny thing about Prarie Home Companion - a close friend of Mrs ttP, is a strong liberal Democrat and very much an agnostic - she loved the radio program, but didn't like the movie because of the religious music - Tried to explain to her the origins of that program traversing the small towns of the upper Mid-West, where that music was a large part of local life - But, I loved the harmonizing of Streep and Tomlin, with Garrison joining on occasion.

Posted by: thethirdPaul on September 26, 2007 at 11:27 AM | PERMALINK

"I was still editing the post when you wrote your comment. I don't know that you'll like the current version any better than the original, but it's now more narrowly targeted on fear than on emotion generally.

(Though I do think that conservatives do a better job of appealing to emotions than liberals. They both do it, but conservatives do it more effectively.)"

Posted by: Kevin Drum on September 26, 2007 at 1:24 AM

Kevin,

Were you high or just sleep deprived when you wrote that? The entirety of the Democratic/Liberal election strategy consists of scaring people of what those evil Republicans are going to do if they get elected. Or have you missed the last dozen or so elections where Democrats cry how the Republicans are going to kill women, children and old people by enacting their cruel cold blooded policies of tax breaks for the rich or whatever. Just go over to today's Daily Kos to see how Republican opposition to the extention of the SCHIP program to cover children of people making $100K is characterized as letting children die instead of why the hell should government being paying for the health insurance for clearly middle class and even "rich" kids. I mean I remember the line about the America having the "worst economy since Hoover" being used in every Democratic Presidential Campaign since Bush I. I can't remember if Mondale was stupid enough to use it against Reagan in 1984, but it wouldn't surprise me if he did.

Rush was right when he said that because openly liberals or progressive candidates cannot get elected in this country, they can't run a positive campaign based on what they will do if elected. Instead, they have to scare the electorate about how much worse the other guy would be.

Kudos for acknowledging your "poll literalism" problem. As they say, admitting you have a problem is the first step on the road to recovery.

I've long believed that the media's obsessive use of polls is based on laziness and their superficial knowledge. It's much easier to talk about how Clinton's health care proposal is polling with the public instead of talking about the merits of the proposals. Anyone who uses on poll results as an argument that supports their position should instantly lose the debate because poll results are worth less than the paper they're printed on.

Posted by: Chicounsel on September 26, 2007 at 11:40 AM | PERMALINK

It's certainly fair to say that answers to poll questions have to be taken with a grain of salt because they don't reflect what people think and decide upon after all the arguments have been offered up.

That's why sensible people pay special attention to actual votes, and what those seem to indicate -- for votes are made after all the arguments have been brought out, and are also what we all actually care about.

With regard to the issue of pulling out of Iraq, it's hard NOT to see the 2006 election as a dramatic statement that the American people consider our continued presence in Iraq in numbers like those of the past several years as a major mistake. What's clear as a bell is that politicians who resist this message, and argue for that presence being continued almost exactly as before, will be punished severely in November 2008. Nothing has changed in the dynamic of the underlying facts and arguments here; Bush and friends made the same "you'll be supporting terrorism if you try to backtrack out of Iraq" argument in 2006 as they are now and will be in November 2008. It's unimaginable that that argument is going to start to win the day absent some very dramatic developments.

Democrats will be elected probably by landslides in 2008. Their marching orders will be absolutely unmistakable: they must greatly reduce our presence in Iraq. If they fail to do so, it is the Democrats who will punished in elections to follow.

Posted by: frankly0 on September 26, 2007 at 11:40 AM | PERMALINK

Kevin made the point above that he thinks the Republicans do a better job of appealing to voter's emotions than Democrats do. I think the evidence shows he is right.

Maybe is it because Democrats use poll results that are developed out of the academic tradition in while the final conclusions are evaluated on how well they explain trends and patterns. In contrast, the Republicans use poll results developed out of the Marketing tradition where results are measured by how well they influence the market to actually buy the product advertised.

Politicians don't want to explain anything to voters. They just want voters to buy the product and vote for them. Academics want to explain what his going on, while marketers want to get someone to buy the product. It would seem reasonable to expect marketers to be more successful at selling their product than academics are.

So when do we get babes-in-bikinis or good looking bad-boys included in political ads? Like Beer commercials. I'm sure Sen. David Vitter and Sen. Craig both have resources that could be used that way.

Posted by: Rick B on September 26, 2007 at 11:44 AM | PERMALINK

Chicounsel,

Some of the Democrats many have tried to frighten the voters about the Republicans. The real problem is that the Democrats just haven't been good at it.

In fact, the Democrats have been so bad at it that they probably couldn't frighten people of snakes or sharks.

Posted by: Rick B on September 26, 2007 at 11:51 AM | PERMALINK

Al is a parody troll. The "real" Al has repudiated him.

Posted by: Jeffrey Davis on September 26, 2007 at 11:54 AM | PERMALINK

Instead, they have to scare the electorate about how much worse the other guy would be.

Apparently you been haven't been paying attention.
Bush campaigned on smaller government, humble foreign policy. Uniter not divider. Compassionate conservatism.

That worked out as promised.

Posted by: Jeffrey Davis on September 26, 2007 at 11:59 AM | PERMALINK

Kevin, while I agree that David Brooks is a self-congratulatory sphincter who has about as much insight into the middle class as a cannibal does into being a vegan, he may have a point. Average people gravitate towards hawkish foreign policy because of the brainwashing done by fascists like Bush, who have convinced people that a few hundred ragtag Islamic extremists represent as great a threat to Western civilization as the Soviet Union, which is a patently absurd assertion. Similarly, average people gravitate toward conservative social policy, due to the massive right-wing propaganda machine that has convinced them that having a private insurance company act as an intermediary between you and your doctor is better than the government, even though the facts show the government can do it more efficiently and not anally rape you in the process, the way private health insurers do.

What Brooks may be saying is that the great unwashed American public is highly susceptible to propaganda and brainwashing, which fascists are extraordinarily good at.

Posted by: The Conservative Deflator on September 26, 2007 at 12:06 PM | PERMALINK

And that makes me a 40 percenter, more or less (not sure what the hell to do about removing troops from Iraq.) If it was a choice of invading Iraq, or not, my answer would be leave it alone for now. But that option is no longer available, thank you very f'ing much, Mr. President. I also thank you (NOT) for leaving us in the position of not being able to believe a word you say, true or not.

It is clear to me we must at least draw down the troops there to give the Army a chance to regroup & rebuild. We are out of replacements for those rotating out of the war zone, or the Army altogether. God help us if we face a new emergency, such as a meltdown in Pakistan, right now. And if Bush told us we were in such an emergency situation, we can neither afford to believe him nor ignore him. Again, thanks a lot.

Posted by: bob in fl on September 26, 2007 at 12:20 PM | PERMALINK

Chicounsel,

I've long believed that the media's obsessive use of polls is based on laziness and their superficial knowledge. It's much easier to talk about how Clinton's health care proposal is polling with the public instead of talking about the merits of the proposals. Anyone who uses on poll results as an argument that supports their position should instantly lose the debate because poll results are worth less than the paper they're printed on.
You represent an interesting idea here. Let me present an alternative explanation to "they are too lazy."

Journalism is primarily a daily news business. With TV it has become hourly. The journalists focus on what moves and changes rapidly, not on what does not change much. So instead of reporting on the relatively static proposals each candidate runs on, they report the fast-moving changes in the race itself. The polls give a snapshot of that movement. That's not to say that is the way they should report, but it is the way they are structured to report.

Tell me. Do you think the free market for journalism would support a news organization that did nothing but report on the differing policy proposals of different candidates? I don't. I just don't see a paying market for that kind of news reporting. You may object that they are reporting the horse-race and not on the horse statistics, history, and form, but they are paid for the former and would have to find other careers if required to report on the latter.

That's the nature of the free market.

Posted by: Rick B on September 26, 2007 at 12:26 PM | PERMALINK
Unfortunately, answers to poll questions come in a vacuum. They don't show what people think once the other side has a chance to get a few licks in.

Um, yeah, they do, since polls are usually conducted when both sides have had a chance to get quite a few "licks" in on issues.

Sure, if they were "out of the blue" polls on issues that haven't been the subject of recent and intense public dialogue, that point might be valid, but for most real polls that most people actually cite to support claims about public perception, the idea that the poll doesn't show what people think once one side gets a few licks in is pretty much a non-issue. There are plenty of ways in which polls are misread and distorted, but the "literalism" you suggest here is just not one of them.

Guess what? Of that 65%, only 30% still favored removal. That's a huge drop based on a single hypothetical, and in a real campaign that hypothetical would practically blanket the airwarves.

Sure, claims of that hypothetical might blanket the airwaves. The support only goes away if people believe it. That doesn't mean the poll does not show overwhelming support for the idea, because it does. It also shows exactly (one area, at least) where Democrats need to be proactive in heading off Republican propaganda to maintain the lead they already have.

It wouldn't convince everyone, of course, but it would probably convince a sizable chunk.

Whether it would convince anyone, much less a sizable chunk, depends on what argument is presented on the other side, and how.

Now, obviously this works in both direction: liberals get to make arguments during campaigns too. But conservative arguments appeal to fear pretty effectively, which means that on difficult, highly charged issues, like withdrawing from Iraq, a lot of people tend to be schizophrenic.

Why can't liberals appeal to fear, too, only to fear of things that there is actually a justifiable, fact-based reason to expect happening? The real consequences of conservative policies have often been things that ordinary people would have been afraid of if they had been highlighted as likely consequences when the policies were being debated (or the conservatives who embraced those policies were running for office.)

There is no fundamental propaganda advantage to conservatism; the problem is that liberal politicians and pundits are too willing to cede ground to the Right rather than pushing back hard, and attacking powerfully, as if they believe that these unilateral concessions will buy similar future concessions from the Right, rather than being exploited by the Right for more and more gain without reciprocation.

Posted by: cmdicely on September 26, 2007 at 12:33 PM | PERMALINK

Shorter Kevin: The general public is stupid and easy prey for emotional and irrational arguments based on fear, uncertainty, and bias and I am not.

Just kidding.

;-)

Posted by: anonymous on September 26, 2007 at 12:54 PM | PERMALINK

The worst thing, though, is giving any kind of a poll more than a passing nod. Unless you are able to poll large numbers of people (100K +) from a broad spectrum of the public, which is pretty near impossible, polls are worthless.

Posted by: JeffII on September 26, 2007 at 1:09 PM | PERMALINK

ah,

david broder-brooks,

the bastard son of dean david broder,

out of univ. of chicago.

better educated that the old man, but to no better avail.

brooks is afraid of the netroots.

why?

for the same reason republicans are afraid of black and hispanic voters - they might end republican power in the federal government for decades.

how to deal with his fear?

trash their reputation - in the new york times.

string together a collection of negative mini-stories, all to the point of illustrating how really, really bad these folks are.

how the "netroots" "bullY" some democrats.

how they appeal only to liberal "elites".

(broder-brooks, of course, is not, and has never been, a member of an elite. he is forced by circumstance to wear those $1000 suits.

how unmemorable and trivial netroots' educational efforts are - why, john edwards attended their convention (in sin city no less) and does not even remember it.

and how the netroots are not the "center" of american political opinion.

do not include those "real" lower-middle-class americans,

whom broder-brooks admires so much and with whom he always has stood should-to-shoulder in his writing.

"the center holds" he tells us -

like father, like son.

Posted by: orionATL on September 26, 2007 at 1:10 PM | PERMALINK

*

Posted by: mhr on September 26, 2007 at 1:16 PM | PERMALINK

David Brooks contributed an essay to a current book that recounted the reasons why the several contributors became disillusioned with liberalism and became conservatives. All very smart people and he is one of the smartest.

THAT is some prime nuttery right there. It needs to be preserved for posterity. If you think Brooks is one of the smartest, you should not even get out of your bed (sureley it has siderails) without protective headgear.

Posted by: Blue Girl, Red State (aka G.C.) on September 26, 2007 at 1:21 PM | PERMALINK

THAT is some prime nuttery right there. It needs to be preserved for posterity.

Too late--the mod deleted it. :)

Posted by: shortstop on September 26, 2007 at 1:56 PM | PERMALINK

The bane of every Poli Sci 101 professor -- poll literalism.

A poll is nothing more than a snapshot of a sample of opinion at a particular moment.

And, as we all know, they are perfectly sensitive to wording. Do you support a woman's right to choose which medical procedures she will or will not have? Sure do. Do you support a woman's right to slaughter an innocent baby in her womb who might grow up one day to cure cancer? Umm, yes?

Just don't understand why we pay any attention to them at all.

Posted by: Hemlock for Gadflies on September 26, 2007 at 2:32 PM | PERMALINK

Brooks was correct. That the netroots find it unappealing will actually prove Brooks correct.

Obama may not be the Howard Dean of 2008 (his support is a bit broader than Dean's), but he is definitely drawing a lot of his support from the same cast of characters, and Clinton clearly sees the opening that is big enough to drive a semi through- the same one Kerry used to get the last Democrat nomination.

I have admire Clinton's clear-headed political pragmatism, even if I don't think she really believes much of what she says publicly.

Posted by: Yancey Ward on September 26, 2007 at 2:51 PM | PERMALINK

Why have the pollsters not fixed this obvious problem in the obvious way? In psychology getting someone in a specific frame of mind is done with priming. Its done all the time in really cool experiments .

A poll could just do the same, ask one half:
1. Did you see the positive results presented by respected uniformed General Petraeus showing 50% progress in safety, security and boldly dealing with terrorists?

2. Iraq, stay or go?

And then the other half:
1. Did you hear how Shia Badr brigades have been fighting the Shia Medhi army of sadr, besides their usual dragging fathers and sons out of homes in the middle of the night and leave their bodies to be found the next morning on a street corner with electric drill holes and acid burns?

2. Iraq, stay or go?

You then substract the one set of results from the other, and add the difference to the "I am to stupid to know" results, which somehow, very few people seem willing to choose.

Why bother with the few percent of statistical error when the biggest margin of fuzziness isn`t in math but in man?

Posted by: rt on September 26, 2007 at 4:55 PM | PERMALINK

"Brooks was correct."

Kevin was not necessarily claiming that Brooks was incorrect. What Kevin said was that Brooks did not support his assertions with any evidence and that the available evidence directly contradicts Brooks.

"That the netroots find it unappealing will actually prove Brooks correct."

That's just silly.

Posted by: PaulB on September 26, 2007 at 5:23 PM | PERMALINK

"Were you high or just sleep deprived when you wrote that?"

I do so love this guy. Starts with an ad hominem attack and goes downhill from there.

"The entirety of the Democratic/Liberal election strategy consists of scaring people of what those evil Republicans are going to do if they get elected."

No, dear, it isn't, nor has it ever been, but don't let a little thing like reality get in your way; it never has before.

"Or have you missed the last dozen or so elections where Democrats cry how the Republicans are going to kill women, children and old people by enacting their cruel cold blooded policies of tax breaks for the rich or whatever."

The problem is, dear heart, that we witnessed the last dozen or so elections out here in the real world, not in that little fantasy world of yours. You should join us out here some time.

"Just go over to today's Daily Kos to see how Republican opposition to the extention of the SCHIP program to cover children of people making $100K is characterized as letting children die"

ROFL.... Dear heart, not only is this a silly argument (DailyKos is now part of the Democratic Party and is running commercials???) but you're wrong on the facts, since there's a distinct shortage of people on DK charactering SCHIP as "letting children die."

"instead of why the hell should government being paying for the health insurance for clearly middle class and even 'rich' kids."

Since that isn't what this bill does, forgive me if I don't take this argument any more seriously than I do the rest of your arguments.

"I mean I remember the line about the America having the 'worst economy since Hoover' being used in every Democratic Presidential Campaign since Bush I."

ROFL.... You should seek help, since someone is obviously implanting memories.

"I can't remember if Mondale was stupid enough to use it against Reagan in 1984, but it wouldn't surprise me if he did."

ROFLMAO.... And this is even sillier. Is there no end to your stupidity?

"Rush was right when he said that because openly liberals or progressive candidates cannot get elected in this country, they can't run a positive campaign based on what they will do if elected."

No, dear, he wasn't, but again, don't let reality get in your way.

"Instead, they have to scare the electorate about how much worse the other guy would be."

You mean like the Republicans did in 2002, 2004, and tried to do in 2006? You really should keep up, dear heart.

"Kudos for acknowledging your 'poll literalism' problem. As they say, admitting you have a problem is the first step on the road to recovery."

Needless to say, dear, we don't expect that first step from you any time soon, as this post clearly shows.

Posted by: PaulB on September 26, 2007 at 5:34 PM | PERMALINK

do you get a dollar every time you say "dear heart," paul? it's getting really really old. diminishes your otherwise good arguments.

Posted by: bored masses on September 26, 2007 at 6:16 PM | PERMALINK

I could buy his column if only he began with "I have gotten nearly everything wrong in the last 6 years, but I hope you don't mind my taking another oafish, blind stab at things..."

Posted by: Mark on September 26, 2007 at 9:45 PM | PERMALINK

Harry: I was still editing the post when you wrote your comment. I don't know that you'll like the current version any better than the original, but it's now more narrowly targeted on fear than on emotion generally.

Okay, we'll talk fear. And then we can get into melting icecaps and flooded cities, being one paycheck away from disaster, losing Social Security, losing all our civil rights, and every other disaster promised by the Democrats if they aren't put in charge. Everybody has their own scare stories.

Keep in mind that so far most of the Democratic scare stories are hypothetical, but there is an actual hole in Manhattan right now.

Posted by: harry on September 27, 2007 at 1:43 AM | PERMALINK

HARRY: Everybody has their own scare stories.

thank god we have turned the corner....

and terrorists are in the 2nd year of their last throes...

talk about hypotheticals..

Posted by: mr. irony on September 27, 2007 at 6:57 AM | PERMALINK


scary..."stories" ?


According to Rep. Jane Harman (D-CA), Chairwoman of the House Homeland Security Subcommittee on Terrorism Risk Assessment, the Bush administration knowingly used bogus intelligence to make lawmakers believe there was the chance of an imminent attack on the U.S. Capitol, thus frightening them into passing the temporary expansion of his powers to spy on Americans under the FISA act.


2 Patriot Act Provisions ruled unlawful

Federal Judge in Oregon Rules 2 Provisions of Patriot Act Unconstitutional

- AP 9/26/07

Posted by: mr. irony on September 27, 2007 at 8:20 AM | PERMALINK

harry: Keep in mind that so far most of the Democratic scare stories are hypothetical.

Since Bush's (and the GOP's) publically stated goal is to do away with Social Security, this is in no way a "hypothetical."

Neither was the Great Depression, brought on by conservative fiscal policy.

Neither was Katrina.

Neither has been the denial or attempted denial of civil rights throught the years by conservatives - the Jim Crow laws are only one such reality that is hypothetical only in your mendacious attacks on liberalism.

BTW, all claims of future terrorist attacks are hypothetical, no matter how many holes are in Manhattan.

There is no actuality to future events that have yet to unfold.

Only a dimwit believes otherwise.

Posted by: anonymous on September 27, 2007 at 2:27 PM | PERMALINK




 

 

Read Jonathan Rowe remembrance and articles
Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

Advertise in WM



buy from Amazon and
support the Monthly