Balkinization  

Monday, November 03, 2008

Why the big surprise?

Sandy Levinson

The New York Times headlines one of its articles "Level of White Support for Obama a Surprise." The obvious question is, to whom? Much has been written of the "Bradley effect," which has been criticized as inapplicable even to the original 1982 election that give the ostensible "effect" its name. But even if, for sake of argument, the polls in that election of 26 years ago were distorted by the unwillingness of people to admit that they simply wouldn't vote for the African-Ameican candidate, isn't it, shall we say, unforgivably stupid to believe that nothing has changed since then in America and that what was true in 1982 is, of course, true today?

As Ronald Reagan might have put it, it is morning in America, and one of the things we're all learning is that the country really and truly has changed. It's not that there are no racists left, but surely there are far fewer. And, to give credit where credit is due, one should acknowledge that George W. Bush apparently had no hesitation at all in naming African-Americans as Secretary of State and National Security Advisor. He may be many things, but a racist is not one of them.

One should be exhilarated that Obama is pulling more white support than any other Democratic candidate, including Bill Clinton, but shouldn't we stop being surprised and take real pleasure, whatever our politics, that this is the case? (Perhaps even McCain supporters will realize that there are silver linings even in what they will view as a very dark cloud.) Perhaps the media next time will write fewer articles on the probable unwillingness of white working class Americans--i.e., the "real Americans"--to cast their lot with an African-American candidate who just happens to speak to their interests and, dare one say it, aspirations and hopes for a better country. That analysis is turning out to be no more sophisticated than the belief that Democratic supporters of Hillary Clinton would actually prefer a 72-year-old Republican just because he picked a hard-line anti-reproductive choice, anti-welfare state, know-nothing woman.



Comments:

"One should be exhilarated that Obama is pulling more white support than any other Democratic candidate, including Bill Clinton,"

Only people who think Obama would make a good President should be "exhilarated" about him polling well in this group or that. Multi-racial bad judgment is still bad judgment.
 

Sandy:

The NY Times finding was based upon their own skewed NYT/CBS News poll, which until today claimed that Obama had a 13 point lead among voter - the epitome of an outlier joke derived from over counting Dem votes.

Counting coastal white Dems as 1.5 voters in your likely voter model hardly means that Mr. Obama is more popular with flyover country "bitter clinger" conservative Reagan Dems than was Bill Clinton. Indeed, the polls could not manage to find these voters during the primary season as they routinely underestimated Hillary Clinton and overestimated Barack Obama's actual votes in blue collar state primaries across the country.

As to the so called Bradley Effect where pollsters overestimate the votes case for African American candidates because respondents do not want to be though of a racist if they oppose the African American, there is plenty of recent evidence of the phenomenon.

In the five 2006 biracial statewide contests, there were unusually large groups of "undecided" respondents in the polls and the undecided on overage broke 2:1 for the white candidate and in four of the five races broke 4:1 for the white candidate.

In the 2008 Dem primaries, the undecided routinely broke for Hillary Clinton and the polls underestimated her votes by between 3 to 10 points. The only exception were southern states with very heavy African American Dem voting blocks who went nearly completely for Obama.

The Bradley Effect is real and will show up tomorrow. The only question is the extent of the effect - a 10 point over estimate of Obama strength like NH or a close 3 point overestimate as in PA.
 

To give credit where credit is due, one should acknowledge that Sauron apparently had no hesitation at all in naming Southrons as Mouth of Sauron and as the First Lord of the Nazgul. The Lidless Eye may be many things, but a racist is not one of them.
 

Jeez, Bart. Don't have a cow, man.
 

It's a surprise because a Democrat hasn't polled this high among whites in a long time. He may even be beating Carter numbers. Of course, Obama's going to be the most successful Democratic candidate, electorally, since LBJ. So I don't quite know what's so surprising about his success among whites; I think what they really mean is that it's surprising that of all the Democrats who have run since 1968, Obama's going to get a bigger share of the popular vote than any of them.
 

this is for you prof levinson... uncommon sense.

i know you've read it, but it's worth repeated reading.


Mr. President

I confess that there are several parts of this constitution which I do not at present approve, but I am not sure I shall never approve them: For having lived long, I have experienced many instances of being obliged by better information, or fuller consideration, to change opinions even on important subjects, which I once thought right, but found to be otherwise. It is therefore that the older I grow, the more apt I am to doubt my own judgment, and to pay more respect to the judgment of others. Most men indeed as well as most sects in Religion, think themselves in possession of all truth, and that wherever others differ from them it is so far error. Steele a Protestant in a Dedication tells the Pope, that the only difference between our Churches in their opinions of the certainty of their doctrines is, the Church of Rome is infallible and the Church of England is never in the wrong. But though many private persons think almost as highly of their own infallibility as of that of their sect, few express it so naturally as a certain french lady, who in a dispute with her sister, said "I don't know how it happens, Sister but I meet with no body but myself, that's always in the right — Il n'y a que moi qui a toujours raison."

In these sentiments, Sir, I agree to this Constitution with all its faults, if they are such; because I think a general Government necessary for us, and there is no form of Government but what may be a blessing to the people if well administered, and believe farther that this is likely to be well administered for a course of years, and can only end in Despotism, as other forms have done before it, when the people shall become so corrupted as to need despotic Government, being incapable of any other. I doubt too whether any other Convention we can obtain, may be able to make a better Constitution. For when you assemble a number of men to have the advantage of their joint wisdom, you inevitably assemble with those men, all their prejudices, their passions, their errors of opinion, their local interests, and their selfish views. From such an assembly can a perfect production be expected? It therefore astonishes me, Sir, to find this system approaching so near to perfection as it does; and I think it will astonish our enemies, who are waiting with confidence to hear that our councils are confounded like those of the Builders of Babel; and that our States are on the point of separation, only to meet hereafter for the purpose of cutting one another's throats. Thus I consent, Sir, to this Constitution because I expect no better, and because I am not sure, that it is not the best. The opinions I have had of its errors, I sacrifice to the public good. I have never whispered a syllable of them abroad. Within these walls they were born, and here they shall die. If every one of us in returning to our Constituents were to report the objections he has had to it, and endeavor to gain partizans in support of them, we might prevent its being generally received, and thereby lose all the salutary effects & great advantages resulting naturally in our favor among foreign Nations as well as among ourselves, from our real or apparent unanimity. Much of the strength & efficiency of any Government in procuring and securing happiness to the people, depends, on opinion, on the general opinion of the goodness of the Government, as well as of the wisdom and integrity of its Governors. I hope therefore that for our own sakes as a part of the people, and for the sake of posterity, we shall act heartily and unanimously in recommending this Constitution (if approved by Congress & confirmed by the Conventions) wherever our influence may extend, and turn our future thoughts & endeavors to the means of having it well administred.

On the whole, Sir, I can not help expressing a wish that every member of the Convention who may still have objections to it, would with me, on this occasion doubt a little of his own infallibility, and to make manifest our unanimity, put his name to this instrument.
 

here is a wager for you bart declueless.

if obama wins you don't post here for a year.

if mccain wins you have blogging privileges for a year.

deal?

it's time to put your money where your mouth is...
 

This comment has been removed by the author.
 

Mr. DePalma,


As to the so called Bradley Effect where pollsters overestimate the votes case for African American candidates because respondents do not want to be though of a racist if they oppose the African American, there is plenty of recent evidence of the phenomenon.

In the five 2006 biracial statewide contests, there were unusually large groups of "undecided" respondents in the polls and the undecided on overage broke 2:1 for the white candidate and in four of the five races broke 4:1 for the white candidate.

In the 2008 Dem primaries, the undecided routinely broke for Hillary Clinton and the polls underestimated her votes by between 3 to 10 points. The only exception were southern states with very heavy African American Dem voting blocks who went nearly completely for Obama.



Just read http://www.fivethirtyeight.com/2008/10/bradley-effect-or-elephant-effect.html

This explains how your numbers come from cherry picking one poll. In actuality, in 2006, African Americans outperformed the polling averages in in their respective races. The African Americans who did the worst (compared to their polling) were republicans, and the causation for their larger than predicted losses makes more sense to be due to party affiliation than the color of their skin.

In fact, most models would predict the undecideds to break slightly in favor of McCain, not because of racism, but the fact that most undecideds this year are disaffected republicans and Perot-types, who will come home to McCain when push comes to shove.

The Bradley Effect, to the very little bit it may have actually existed, no longer exists. (Most evidence shows there was no Bradley effect in the actual Bradley election, his internals had gone down to a mere one point lead before the election). There is no evidence to support it, and actually most African American candidates over perform (by a very small .3%) in the final election. I wouldn't stake your hopes on this. In fact, if I were you, I wouldn't want to use 2006 as a benchmark, since it had little if anything to do with race, and more to do with party. 2006 was a wave year, as 2008 looks to be, in that case, the undecideds may actually break Obama.

Obama is over 50% in lots of states, including PA, which means you are no longer simply talking about persuading undecideds, but actually taking votes away - requiring the existence not only of the non-existent Bradley Effect, but a "Strong" Bradley Effect (where not only are people lying about their preferences by claiming they are undecided, but actually affirmatively claiming support for Obama with intentions to vote for McCain) - to win those states. (the only potential for PA to be in play is Murtha's jackass comments, but you never know, by accusing West PA'ians of being racists, maybe they'll show their displeasure by voting for the black guy and kicking Murtha's ass out). Factor in the Cell Phone Onlys and potential reverse Bradley Effect, etc, and I think you are gonna see almost 400 EV (Kerry + OH, FL, VA, NM, NV, CO, IA, MO(close), MT, ND(close), IN(close), NC, and maybe NE Dist 2, Obama pulling off GA, and WV will be really close), and probably about a seven to nine point victory for Obama, with double digits not outside of the realm of possibility.
 

Count me as being one person that will be jubilant that the racist Harold Arlan "Bart" DePalma will get handed a big sh*t sandwich to munch on tomorrow.

We've had enough of the party of racists running the U.S. (and just to be clear, that includes Reagan with his "Cadillac queen" race-baiting). Time -- nay, way past time -- to join the developed world.

Put a stake in the heart of the Republican party, shoot it with a silver bullet, and pour 100 tons of concrete on the tomb just to be sure. "Never again...."

Cheers,
 

I look at the (Bill) Bradley Effect as a positive for Obama, a slam-dunk. While Bill Bradley was not successful in his presidential campaign, who can forget how his African American peers in the NBA accepted and admired him during his playing days and his political days. It's time for role reversal, for white Americans to accept and admire Obama as a savvy political leader. Words matter, my friends. Obama has chosen them well. Here comes Obama, set for his shot, it's up, "Good, like Needicks!" (I can just hear Marty Glickman.)
 

nerpzillicus said...

BD: In the five 2006 biracial statewide contests, there were unusually large groups of "undecided" respondents in the polls and the undecided on overage broke 2:1 for the white candidate and in four of the five races broke 4:1 for the white candidate.

Just read http://www.fivethirtyeight.com/2008/10/bradley-effect-or-elephant-effect.html


My figures are derived from the Silver data in that article. Indeed, in the midst of critiquing Greener for cherry picking stats, Silver inadvertently proved Greener's point with more comprehensive data. Silver's claim that the obvious Bradley effect was a consequence of a good Dem year does not explain how the the Dem Duval in MA suffered the same effect.

As for the the rest of my post, Pew found both a Bradley and a Reverse Bradley Effect in the Dem primaries as the polls overestimated Obama votes in 28 states.
 

arne langsetmo said...

We've had enough of the party of racists running the U.S. (and just to be clear, that includes Reagan with his "Cadillac queen" race-baiting). Time -- nay, way past time -- to join the developed world.

The Bradley Effect involves white Dems and the Reverse Bradley Effect involves black Dems.

Stones and glass houses, hero.
 

Bart, enough on the Bradley effect. Are you hanging your hopes for Old Man and the Twit to become president via racism? That's your grand scheme? Hoping that we're as racist and backwards and embedded in culture wars as the moral-capital wasting Baby Boomers? Your story is finished. Get outta Dodge. You're passe like a keyboard necktie. Evolve. Or not. Either way, we've got it from here. Smell you later.
 

"Bart" DeRanged:

The Bradley Effect involves white Dems and the Reverse Bradley Effect involves black Dems.

Stones and glass houses, hero.


WTF are you talking about?!?!?

The Rethuglicans have been race-baiting ever since Nixon's "Southern Strategy" (and they wonder why blacks don't vote Rethuglican ... GOOPers are stoopid too, I guess).

They're proud of their racism, and won't deny it. They weren't voting Democratic anyway.

But the existence of a "Bradley effect" is disputed (even for Bradley's campaign). And there is no "reverse Bradley effect" ... except in your fevered imagination.

But, returning to the point, that was not the gravamen of my post.

Cheers,
 

Ol Mucky:

Evolve. Or not.

Yeah. One of my newest T-shirts (for Palin, and the other "creationist" nut-jobs on the rights, as well as for "Bart" here) is: "OK. I evolved. You didn't."

Cheers,
 

BTW, initial reports are that turnout will be high. Very high.

That ought to scare the pants off of McCain and Palin. And the resident LSRs.

Cheers,
 

not the gravamen of my post

Extra points for gravamen. If you can work in philippic or jeremiad, major props.

Just saying.
 

arne:

I want an very large turnout because the Dem strategy has been to depress GOP and Reagan Dem turnout.

I assumed that the Dems would show for The One. My question is whether the GOP would show for the Maverick who has dumped on them in the past.
 

Meanwhile, Dem elections officials in Philly are kicking out GOP poll watchers in violation of a court order. I guess they figure they can stuff a few ballot boxes by the time the court gets around to doing something.
 

"GOP Poll Watchers" that's about as innocuous sounding as "Patriot Act" - must be innocent!
 

If Mr. Obama wins, I will congratulate the Obama supporters here for their win, say a prayer that Mr. Obama is wise enough to adjust his policies to benefit the country and work to get the GOP back to conservative first principles to earn the trust of the voters again.

If Mr. McCain wins, I do not expect any sort of reciprocal courtesy, Rather, I expect the Obama supporters to accuse the voters of being racist and/or stupid and the GOP of again "stealing" the election.

Same ol, same ol.

# posted by Bart DePalma : 11:27 PM


Less than 12 hours later...

Meanwhile, Dem elections officials in Philly are kicking out GOP poll watchers in violation of a court order. I guess they figure they can stuff a few ballot boxes by the time the court gets around to doing something.

# posted by Bart DePalma : 10:18 AM


You are the best.
 

Mr. DePalma,


My figures are derived from the Silver data in that article. Indeed, in the midst of critiquing Greener for cherry picking stats, Silver inadvertently proved Greener's point with more comprehensive data. Silver's claim that the obvious Bradley effect was a consequence of a good Dem year does not explain how the the Dem Duval in MA suffered the same effect.


No, no, no. With the black candidate overperforming, even though the undecideds broke for the republican, the more likely explanation is a shy tory effect, or disaffected republicans coming home. This is simply no evidence for the Bradley Effect.

More fundamentally, assuming the Bradley Effect exists, this proves it ain’t gonna take votes away from Obama, it only may result in a disproportionate undecided break (which, as I said, is vastly more reasonably explained in other ways – in fact, I would be surprised to see the undecideds break towards Obama). But since he is over 50%, it doesn’t matter – a “Bradley Effect,” as “proven” by your analysis, will simply result in a smaller margin of victory for Obama, say 5 points. I’ll take it.

As for the rest of my post, Pew found both a Bradley and a Reverse Bradley Effect in the Dem primaries as the polls overestimated Obama votes in 28 states.
# posted by Blogger Bart DePalma : 9:24 AM



No, no, no. First, Pew found three states – CA, RI and NH that had potential Bradley Effects. CA and NH are right at the margin of error. CA is more reasonably explained by lack of enthusiasm since Obama was gonna lose, and NH due to angry female voters after a lot of dismissive reporting and the Rush Limbaugh insults directed at HRC. The RI number is based from a poll with less than 1000 people, i.e. small sample. If you allow for the small sample, I guess this may show a Bradley Effect in RI. If you’re counting on a Bradley Effect in RI to save McCain, its gonna be a long night for you.

Second, are you ever honest? “the polls” did not overestimate Obama in 28 states - the exit polls overestimated Obama. Exit polls are absolute garbage – you might as well read tarot or the stars. The only thing they prove is that you will accept bad data to back up flimsy theories.



The Bradley Effect involves white Dems and the Reverse Bradley Effect involves black Dems.


No, no, no. The reverse Bradley effect, if such a thing exists, is explained by both African Americans hiding true preferences, and Republicans hiding true preferences. Since Obama is really close to his ceiling in AA voting (90%+), a reverse Bradley Effect, if any, is going to come from republicans. I expect a small amount of this.
 

nerpzillicus said...

No, no, no. With the black candidate overperforming, even though the undecideds broke for the republican, the more likely explanation is a shy tory effect, or disaffected republicans coming home. This is simply no evidence for the Bradley Effect.

I suspect that there is a substantial overlap between the Bradley Effect as I interpret it (conservative blue collar voters avoiding pollsters because they are unwilling to be perceived as racists for supporting the white candidate) and the Shy Tory Effect (conservative voters who avoid pollsters for indeterminate reasons - in my case, because I do not want to waste my time with them).

I doubt there is an effective difference between the two effects in this race. The bottom line is that polls in general and exit polling in particular undercount conservative voters.
 

jpk:

Extra points for gravamen. If you can work in philippic or jeremiad, major props.

How about "polemic"? Or "diatribe"?

You ever seeThe Shooting Party? One of my favourite passages:

(an animal rights protester has just disrupted the shooting]
Sir Randolph Nettleby{: This is a very well produced pamphlet. Where do you get a thing like this printed? Is it expensive? You don't mind my asking you?
Cornelius Cardew: Oh no, not at all. I know a very good printer in Dorking, just near where I live. An excellent man of anarchistic views. He gives me very good rates.
Sir Randolph: Ah, special terms. Hmm. He wouldn't give me such good ones, I suppose.
Cardew: Are you a pamphleteer too, sir?
Sir Randolph: Well, I was thinking of making a sort of foray in that direction.
Cardew: A polemic, would you say?
Sir Randolph: Yes, I think that's the right word.
Cardew: Would you call it a *diatribe*?
Sir Randolph: Yes, I [...chuckle...] could call it a diatribe.
Cardew: The Ruin of Rural England - A Diatribe
Sir Randolph: Precisely. I don't think we should continue our discussion here... my fellow murderers are rather hot blooded.


Cheers,
 

The bottom line is that polls in general and exit polling in particular undercount conservative voters.

Including the poll in the voting booths. Suck on it. There's a great Silent Majority out there to be won....

But I'd mention in passing that Silver's analysis includes comparison of polling data to the electoral results (as does any reputable pollster in constructing polling models; not a single pollster uses raw poll data).

Cheers,
 


I suspect that there is a substantial overlap between the Bradley Effect as I interpret it (conservative blue collar voters avoiding pollsters because they are unwilling to be perceived as racists for supporting the white candidate)


But this isn't the Bradley Effect - a conservative, to be a Bradley Effect Voter, cannot "avoid" a pollster, but has to answer incorrectly, which is a huge difference. I have ignored a lot of phone calls coming form area codes I don't know anyone in, because I suspected they were robocalls or something. But that isn't the Bradley Effect.


and the Shy Tory Effect (conservative voters who avoid pollsters for indeterminate reasons - in my case, because I do not want to waste my time with them).

But this isn't the shy tory effect, either. "avoiding" the pollster does not result in a shy tory - a shy tory is one who claims to be undecided or voting for the liberal side. But they have to affirmatively say something contrary to what they will do. Your avoision ("its a word, look it up" K. Brockman) of the pollster does not result in an undercount of conservative voters, as the measure of the number of conservative voters is generated through a party id weighting. the assumptions of that may be off, but that has nothing to do with whether conservatives don't answer their phone.


I doubt there is an effective difference between the two effects in this race. The bottom line is that polls in general and exit polling in particular undercount conservative voters.


Actually, that's the whole point of the shy tory theory. Its not that pollsters undercount conservatives, but that some conservatives lie about their preferences to pollsters. This doesn't miscount the conservatives, but instead, through the false answers, poll results show more strength for the liberal and lesser strength for the conservative. exit polls have a habit of undercounting conservatives, which conservative avoision of exit pollers probably has some effect on. however, there are a lot of other reasons while exit polls are crap, too.

the big question is why there are shy tories, but not really shy liberals.

you really should try to understand these concepts before pontificating on them.
 

"The bottom line is that polls in general and exit polling in particular undercount conservative voters."

Are you joking? You really think it's the conservative voters that will be undercounted in this election? Bart, you will see, give it a few hours, or a day. You will see that it was liberals who were undercounted; blacks, hispanics, and the youth vote.

I'd say you're clutching at straws, but they've burned up and the ashes blew away and all you're holding onto are a couple hydrogen atoms in a near-vacuum where sound, logic, and facts lack representation.
 

The thing about the Bradley Effect I can't compute is why a person who is racist (or isn’t racist but doesn’t want to be perceived as racist) would feel it necessary to lie to an anonymous pollster. I mean, come on: lots of ppl are voting for the white guy, so what’s wrong with just saying “I’m voting for ______ (the white guy).” That doesn’t imply you’re racist. You’re not required to tell the pollster “I’m voting for the white guy because I’m a raging racist.” Now I can see how people might lie instead of say THAT, but that’s not what’s going on here. It’s not like you’re talking to your neighbors who you want to maintain a nice relationship with (and might be ambivalent or even lie about your preference to). This is a random pollster. We as a society have so little respect for random ppl who call us on the phone that we created a “Do Not Call List.” So random people who don’t know us who call us are as low as dirt, basically. Who cares what they think of you. And you don’t have to tell them WHY you’re not voting for the black guy – just that you’re not (like thousands of other ppl so why feel uncomfortable?). That’s what I mean by not understanding the psychological basis for the Bradley Effect, and therefore doubting it’s existence (based on stuff I’ve read doubting it’s existence).
 

nerpzillicus said...

BD: I suspect that there is a substantial overlap between the Bradley Effect as I interpret it (conservative blue collar voters avoiding pollsters because they are unwilling to be perceived as racists for supporting the white candidate)

But this isn't the Bradley Effect - a conservative, to be a Bradley Effect Voter, cannot "avoid" a pollster, but has to answer incorrectly, which is a huge difference. I have ignored a lot of phone calls coming form area codes I don't know anyone in, because I suspected they were robocalls or something. But that isn't the Bradley Effect.


By avoiding the pollster, I mean physically avoiding them at exit polls, hanging up on them as I do and giving intentionally misleading or false answers of being undecided or supporting the African American candidate. All these things avoid providing the pollster with accurate information.

...and the Shy Tory Effect (conservative voters who avoid pollsters for indeterminate reasons - in my case, because I do not want to waste my time with them).

But this isn't the shy tory effect, either. "avoiding" the pollster does not result in a shy tory - a shy tory is one who claims to be undecided or voting for the liberal side.


Neither Red nor Blue offers a pretty good analysis of the Shy Tory Factor - which is when conservatives refuse to participate or provide misleading answers. In reality, the pollsters are guessing as to how conservatives are avoiding the polling because no one has been able to identify and study those who avoid the polls.
 

The Bradley Effect has been debunked, but it's the only hope Baghdad Bart has left, so he's going to flog that horse until it drops.
 

Mr. DePalma,

By avoiding the pollster, I mean physically avoiding them at exit polls, hanging up on them as I do and giving intentionally misleading or false answers of being undecided or supporting the African American candidate. All these things avoid providing the pollster with accurate information.


No, no, no. As I said in my last post, avoiding exit pollers may be a factor in an overestimation of liberal support at exit polls, but not answering the phone does not result in a loss of information for the pollster. But exit polls are crap anyway, so no rational person is concerned about them.

The third part, misleading answers, is the Bradley Effect or Shy Tory Effect.

The second part, hanging up the phone, has absolutely, positively, no effect on a poll. (again, excluding exit polls) I'll repeat, hanging up, or not answering, does not affect the accuracy of a poll. only misleading answers (and the pollster’s incorrect assumptions) have an inaccuracy effect.

...and the Shy Tory Effect (conservative voters who avoid pollsters for indeterminate reasons - in my case, because I do not want to waste my time with them).

But this isn't the shy tory effect, either. "avoiding" the pollster does not result in a shy tory - a shy tory is one who claims to be undecided or voting for the liberal side.

Neither Red nor Blue offers a pretty good analysis of the Shy Tory Factor - which is when conservatives refuse to participate or provide misleading answers. In reality, the pollsters are guessing as to how conservatives are avoiding the polling because no one has been able to identify and study those who avoid the polls.
# posted by Blogger Bart DePalma : 12:01 PM


Please, this is not rocket science. Even the link you provided talked about inaccuracy through avoision at exit polls. These are two very different things (i thought the previous post, illustrating your inaccurate claim that Obama underperformed in 28 states made this abundantly clear). No one cares about those, and certainly none of us do at this point, when none have been released. We are talking about the accuracy of scientifically done pre-election polls. “refusing to participate” does not have an effect. Only the misleading answers can result in either effect. The polls done prior to an election are weighted by estimated party id, which has to do with registrations and historical data, not whether, out of 1000 calls, I get a whole bunch of liberals but no conservatives to talk to me. If I do a poll of a hundred people, make 200 calls, and 60 liberals answer, 40 independents and only one lone conservative responds, I’m not going to come up with a result of Obama 90%, McCain 10%. The conservative will be weighted higher, and I, using party id weighting, end up with Obama 55, McCain 45. Of course, this is an extreme example, the real way it works is I keep calling until I find a substantial number of conservatives to form a good sample. But if it only takes me 50 calls to find enough liberals, and 1000 calls to find enough conservatives, it doesn’t matter if I weight the results.

The shy tory effect, ignoring exit polls [‘cause no one cares], only exists when conservatives lie. Not answering your phone does not result in a lower estimate of the number of conservatives. Only lying results in an inaccurate interpretation of relative strength of the parties in elections.

There is no systemic miscounting of conservatives. The polls are accurate as long as conservatives are not lying about their preferences, and the pollster’s assumptions are correct. The most significant factor in this election will be the cell phone only people.
 

The conservative blogs are reporting heavy turnout in their districts, but no youth vote.
 

nerpzillicus said...

The polls are accurate as long as conservatives are not lying about their preferences, and the pollster’s assumptions are correct. The most significant factor in this election will be the cell phone only people.

The pollsters claim that they correct for having less cooperation from conservatives by weighting conservative responses higher according to historical norms. However, in this race in particular, the pollsters are using ahistorical weights for Dems and against GOP respondents in anticipation of a swarm of new Dem voters. Thus, the pollsters are not only missing conservatives, they are underweighting those they get to cooperate.
 

The conservative blogs are reporting heavy turnout in their districts, but no youth vote.

Every time you post something like this 538 lower's McSame's chances of winning. It's now down to 1.1%
 


The pollsters claim that they correct for having less cooperation from conservatives by weighting conservative responses higher according to historical norms. However, in this race in particular, the pollsters are using ahistorical weights for Dems and against GOP respondents in anticipation of a swarm of new Dem voters. Thus, the pollsters are not only missing conservatives, they are underweighting those they get o cooperate.


This is a completely different argument, and one that, at least in theory, makes sense. However, you are ignoring the fact that there is good reason to suspect dem turnout will be ahistorically high. If you have good grounds for your claim that the weighting is inaccurate, i would love to see it. However, most pollsters are basing their weighting on registrations, for instance:

An estimated 153.1 million Americans have registered to vote, an increase of 10.1 million that sets the stage for a likely large turnout on Election Day, a non-partisan study reported Sunday.

American University's Center for the Study of the American Electorate estimated 73.5% of citizens 18 or older have registered, compared with the previous high of 72.1% set in 1964.

Curtis Gans, director of the center, said that the registration gains "could lead to a turnout of as many as 135 million" voters or about 65% of eligible citizens. That would be the highest since 67% of citizens 18 and over voted in 1960, he said.

Democrats gained an estimated 2.9 million new voters, while Republican registration declined by about 1.5 million, the report said. The study estimated those figures based on registration in 19 of the 28 states and the District of Columbia that register voters by party.


http://www.usatoday.com/news/politics/election2008/2008-11-02-vote_N.htm


As you can see there is evidence that party id has shifted in democratic favor from the historical numbers. This is actually a double whammy, because not only has the dem/repub id difference increased in the dems favor, but new registrations favor the challenging party, since typically people choose to newly register because of displeasure with the incumbent party. in the final analysis, it might not mean anything, but there is evidence.

Moreover, enthusiasm is higher among dems, early voting indicates higher turnout for dems, and voters not normally caught by polling models (new, lapsed, cell phone) overwhelmingly trend democratic. The incumbent party's president is historically unpopular, fundraising favors the dems, and organization favors the dems.

None of this is a smoking gun, we still have to wait until the end. but you claim, with no evidence or reasoning, that the estimates of party id are undercounting repubs. The pollsters have plenty of empirical, factual reasons for their assumptions. What are yours?
 

... when conservatives refuse to participate or provide misleading answers.

Why, that's Dubya (and his maladministration) to a 'T'!!! ;-)

I'm looking forward to his doing it under oath....

Cheers,
 

nerpzillicus:

Registration forms do not equal actual living and eligible human beings (as opposed to the dead, members of the Dallas Cowboys and cartoon characters), nevertheless folks that in fact turn out to the polls
 

The conservative blogs are reporting heavy turnout in their districts, but no youth vote.

The conservative blogs are also "reporting" that Obama had his grandmother offed for the sympathy vote (read at your own risk, NSFWOAODP). They're just as screw-loose as you, "Bart", if not more so....

Cheers,
 

50% of registered Colorado voters (including yours truly) have voted early and the number of Dem and GOP ballots are tied.

Colorado elections officials do not anticipate a finally tally until Wednesday morning because of the large number of mail in ballots.

I have yet to see these legions of new Dem voters that the Colorado polling assume.
 

Mr. DePalma

nerpzillicus:

Registration forms do not equal actual living and eligible human beings (as opposed to the dead, members of the Dallas Cowboys and cartoon characters), nevertheless folks that in fact turn out to the polls


if your client claims the light was green, that doesn't mean that it was, it means there is evidence that the light was green. increased registration (and more significantly, decreased republican registration) is not dispositive, but it does represent evidence. "evidence" is what the pollsters have for their assumptions - it is what i am seeking from you for your critiques of the pollster's assumptions. I never claimed that was dispositive. besides, the study was based on the finalized registration results of the registration offices, so the offices had already rejected the faulty or fraudulent registrations, so the numbers have a decent likelyhood of accuracy.

please read for comprehension.

and again, where is your "evidence"? (note: not asking for god's word proof, just a scintilla of evidence, i won't even hold hearsay against you, if you can find something reliable for why there won't be an increase in dem turnout and affiliation.)
 

FiveThirtyEight.com: Electoral Projections Done Right: Today's Polls, 3 AM Edition (11/3): Barack Obama's position has become somewhat stronger since our update this afternoon. We now have him with a 5.8 point lead in the national popular vote, and winning the election 96.3 percent of the time.... Obama's win percentage has ticked upward again for a couple of reasons. Firstly, he's gotten some relatively good numbers out of Pennsylvania since our last update, with PPP and Zogby giving him leads of 8 and 14 points, respectively, and Rasmussen showing his lead expanding to 6 points after having been at 4 before. (The Zogby poll is probably an outlier, but may serve to balance out outliers like Strategic Vision on the other side). Secondly, McCain's clock has simply run out.....

Related to this is the fact that there are now very, very few true undecideds left in this race. After accounting for a third-party vote, which looks as though it will come in at an aggregate of 2 percent or so (after doing some work on this tonight, I concluded that I had been slightly underestimating the third-party vote before), I am showing only about 2.7 percent of the electorate left to allocate between the two major-party candidates. Even if John McCain were to win 70 perecnt of the remaining undecideds (which I don't think is likely), that would only be worth a net of about a point for him. Frankly, McCain's winning scenarios mainly involve the polls having been wrong in the first place -- because of a Bradley Effect or something else. It is unlikely that the polls will "tighten" substantially further -- especially when Obama already has over 50 percent of the vote...



so who you gonna believe?
 

I have yet to see these legions of new Dem voters that the Colorado polling assume.

# posted by Bart DePalma : 1:17 PM


You didn't see the big Dem gains in 2006 coming, either. Your cluelessness is well documented.
 

Mr. DePalma,



50% of registered Colorado voters (including yours truly) have voted early and the number of Dem and GOP ballots are tied.

Colorado elections officials do not anticipate a finally tally until Wednesday morning because of the large number of mail in ballots.

I have yet to see these legions of new Dem voters.


Better. I see you are starting to understand the concept of “evidence.” You leave out a few key facts:

Dems actually have a slight lead. Also they have a lead in registrations.

Furthermore, according to the Colorado Secretary of State, in Oct. of 2004, there were:

1,125,374 registered repubs 947,866 registered dems 1,028,886 registered unaffiliated

In Oct of 2008, there were:

1,063,347 registered repubs, 1,051,643 registered dems, and 1,069,294 registered unaffiliated.

For differences of:

-61,727 repubs, +103,777 dems, and +40,408 unaffiliated.

George W. Bush won the state total of votes of 2,422,251 by 1,101,256 to Kerry’s 1,001,725, for a difference of 99,351, or 4.1%

The registered electorate grew from 3,114,566 in 2004, to 3,203,583, an increase of 2.8%. With a net increase of dem/non-repub voters of 165,504, (a 5.1% shift) dems should win CO by at least 1%, with the afore mentioned increase in turnout relative to republicans of 5.1%

Now these are estimates, and luckily we’ll see soon enough. Nevertheless, 5.1% is a fairly large number. And since we do not know how the voters actually voted, we’ll have to wait and see whether there are party defectors on either side, and how the indies break. But thanks for pointing out CO, it proves a very good point about increased dems.
 

Garth:

I have a C-Note that says Silver is wrong and Obama will not win by 5.8% or more.
 

nerpzillicus:

Which part of voted early did you not understand? I could care less how many folks are registered. The polling is predicting that Colorado Dems will cast substantially more ballots than will GOP voters. Here is the early voting results from CO:

Out of 1,704,280 "early votes" cast, over 50% of registered voters, the percentage breakdown was:

Dem 33.44% (569,875)
Rep 32.64% (556,241)
Ind. 24.71% (421,124)
 

"Bart" DeClueless:

Out of 1,704,280 "early votes" cast, over 50% of registered voters, the percentage breakdown was:

Dem 33.44% (569,875)
Rep 32.64% (556,241)
Ind. 24.71% (421,124)


Were they taking early peeks at the ballots? Because otherwise this would just be stated party affiliation, not the actual vote.

But nonetheless, that does sound like more Democrats are voting than Republicans, which seems to be a singular event for Colorado in recent years, and hardly one that McSame can take comfort in....

Cheers,
 

Mr. DePalma,

I apologize. I forgot you are incapable of rational thought. I brought up registrations because dems gained almost 10% in registrations in 4 years. you asked me where all the dems are.

In early voting in 2004, Repubs had a 42% to 36% advantage. The dems have cut that down to a slight lead. I call that a massive increase in dem turnout. Noone ever said a state like Colorado would suddenly be 60% dem, but the increase is really quite obvious.

This also should present a serious problem if you look at it. Even though registered dems, in 2004, were only registered at 30.4% of the registered voters (947,866/3,114,566), they got 46.7% of the vote(1,001,725/2,147,224). The republicans, who were registered at 36.1% (1,125,374/3,114,566) got only 51.3% (1,101,256/2,147,224), a slight breaking towards the dems (16.3% dems v 15.2% repub). If the voting is now even, and registration is now close to even, the CO break should spell doom for the repubs. Moreover, the raw total of dem vote was 1,001,725, a + 53,859 over the registration numbers. On the other hand, the repub raw number was 1,101,256, a -24,118 total. Sadly, I have no by party breakdown, so we’ll have to guess how this happened. This would seem to mean:

1) repubs don’t vote in the numbers they are registered
2) repubs defected to Kerry
3) indies disproportionately broke to Kerry
4) some combination

now, in 2008 vs 2004, we have: a six point loss in early voting; a 5.1% swing in registrations; and a sub-registration republican turnout numbers to begin with. Well, I’m no math major (actually, I am), but that means a lot more dems and dem indies.



Which part of voted early did you not understand? I could care less how many folks are registered. The polling is predicting that Colorado Dems will cast substantially more ballots than will GOP voters. Here is the early voting results from CO:



First, it is hysterical for you, of all people, to ask if I understand a term. We have already corrected your misunderstanding of Bradley, reverse Bradley, and shy tory effects.

Second, please show me the polls predicting dems will out-ballot repubs in CO. I believe the polls are showing Obama will win, probably because of an increase in dems, to balance out the repubs, and then the traditional CO disaffected repubs and dem-favoring indie swing will take care of the rest. Do you really believe those votes are straight party line? Do you think the indies will break substantially for McCain. If so, why, and where is you evidence?
 

I have a C-Note that says Silver is wrong and Obama will not win by 5.8% or more

And you had a $50 bill that said McSame IETM shares were a bargain at $0.20 at the time. They're now trading at $.06.
 

jpk said...

And you had a $50 bill that said McSame IETM shares were a bargain at $0.20 at the time. They're now trading at $.06.

The only figures that mean anything for my winner takes all contracts are $1.00 per contract if McCain wins or $0 if he loses.

Half of my money went into contracts betting in the voting margins. That is already money in the bank.
 

The only figures that mean anything for my winner takes all contracts are $1.00 per contract if McCain wins or $0 if he loses.

He's going to ride it into the ground.

Half of my money went into contracts betting in the voting margins. That is already money in the bank.

I call BS.

Cheers,
 

From the liberal Atlantic magazine:

I'm a student at Carleton College who has contributed (financially and with my time) to the Obama campaign. I just got an automated call from Jeff Blodgett, the chair of the MN campaign. The text of the call was:

"Hello, this is Jeff Blodgett from the Minnesota for Obama campaign. Our initial data shows this election is significantly closer than the polls predicted. We are putting out an urgent call for volunteers... We are organized groups to knock on doors at five P.M., or earlier if you can, for our final GOTV operation." This was followed by different numbers to call based on your residence.


McCain close in Minn?
 

Oops...

Fox News is reporting that the number of new voters in the exit polling is 10%, one point lower than the 11% for 2004.

It is possible that all the hoards of new Obama voters cast their ballots early...except that the exit polling for early voting did not show this.
 

He's going to ride it into the ground.

Of course. Fools do.

The guy who leaps off the burning Hindenburg at the point it hits the ground, screaming he's been validated, this shows that hydrogen is the future of lighter than air transportation, has, how do we say this politely, demonstrated merely the disadvantages of extreme denial, not to mention the worthlessness of his predictions. His airship analysis is not more impressive than if he'd lept sooner. The folks who impress us are the ones well clear of the crash zone, who took the other end of the bet.
 

The first exit polling has exceedingly narrow Obama margins in Pennsylvania, Virginia, North Carolina and Florida.

Given that exit polling overstated votes for the Clinton, Gore and Kerry by 2-3% and for Obama in the primaries by 7%, watch out folks!
 

They're now trading at $.06

Update: $.05
 

jpk:

Its too late to send the college more money to put on McCain or I would definitely send them another $100.

These exit polls are fantastic!
 

McCain is pulling in 16% of Hillary supporters in Ohio in the exit polling according to Fox.
 



McCain close in minn?


more likely colman/frankin is close

ps

enjoy the exit polls. the kerry suppoters did took
 

I would definitely send them another $100

Yes, after your flight on the Hindenberg crashes, the thing to do is get on the very next hydrogen dirigible. Shows how smart you are. Shows how good your predictions are.

Enjoy your ride!
 

There will be no Obama blowout as predicted by the media polls. None of the battleground states where Obama supposedly had high single or double digit leads have been called.

The second wave of exit polling has larger Obama leads. Keeping in mind that early voting and evening voting has not been counted, if these exit poll figures follow the 7 point over count Obama had in the primaries would mean a very comfortable McCain win with him taking PA, NH and MN. If the exit polling follows the 3 point over count enjoyed by Dems since Clinton, then Obama wins a Bush 2004 level victory and picks off a handful of red states.

http://citizen-pamphleteer.blogspot.com/2008/11/first-exit-polling-shows-very-tight.html
 

There will be no Obama blowout as predicted by the media polls. None of the battleground states where Obama supposedly had high single or double digit leads have been called.

The second wave of exit polling has larger Obama leads. Keeping in mind that early voting and evening voting has not been counted, if these exit poll figures follow the 7 point over count Obama had in the primaries would mean a very comfortable McCain win with him taking PA, NH and MN. If the exit polling follows the 3 point over count enjoyed by Dems since Clinton, then Obama wins a Bush 2004 level victory and picks off a handful of red states.
 

Keeping in mind that early voting and evening voting has not been counted, if these exit poll figures follow the 7 point over count Obama had in the primaries would mean a very comfortable McCain win with him taking PA, NH and MN.

"Clap! Clap harder! If you clap hard enough, Tinkerbelle won't die...."

Cheers,
 

... a very comfortable McCain win with him taking PA, NH and MN ...

Bart puts me in mind of a very well known conservative, a Mr. William Buckley:

"That's interesting, from a pathological point of view."
 

Mattski:

Bart puts me in mind of a very well known conservative, a Mr. William Buckley:

"That's interesting, from a pathological point of view."


You owe me for a new keyboard. I spewed my Guinness though my nose all over it. ROFLMAO.....

Good news for the Bartster: PA called for Obama by MSNBC. Time to sell his IEM shares ... you know, buy high, sell low (but before they're worthless....)

It's starting to snowball.... Wheeeeeeeee!!!!

Cheers,
 

You owe me for a new keyboard

I'm good for it, Arne!

Glad to know you're enjoying a Guinness. I'm favoring some quality Tequila myself.
 

mattski:

I'll have to wait on the hard stuff ... if I want to last until the CA elections (the propositions there are critical).

NP on the keyboard; just joking. But not about the Guinness.

Cheers,
 

Well, it appears that Obama drew enough of the Reagan Dems to gain the first majority since Carter.

Congrats to you Dems on a well earned victory.

Enjoy the win. I have enjoyed the last 28 years. Now it is your turn.

DC is now yours with no excuses. The next four years will determine whether this is a realignment or a repeat of 1992.
 

Congrats to you Dems on a well earned victory.

Thank you, Bart. That is gracious of you.
 

"Bart" DeLuded:

Well, it appears that Obama drew enough of the Reagan Dems to gain the first majority since Carter.

BS.

Congrats to you Dems on a well earned victory.

Indeed. We earn it. Not steal it.

Enjoy the win. I have enjoyed the last 28 years. Now it is your turn.

You also enjoy torture. You deserve worse than just this humiliation.

Cheers,
 


They're now trading at $.06

Update: $.05


Update: $.01

What famous blowhard said they were a bargain at $.20?

Who was it who said he wished he could put in another $100 when they were $.05?

Credibility: what you don't have when you make big blowhard predictions that were unlikely then and didn't happen in the event.

Here's a tip: indifference to the facts doesn't help with that prediction thing.
 

Famous last words (from up the thread):

Its too late to send the college more money to put on McCain or I would definitely send them another $100.

These exit polls are fantastic!


Yes, too bad. One almost wishes he put his kid's college fund (or the mortgage) into it.

This needs bookmarking for the unfortunately too-common occasion in the future when our reznit eedjit "Bart" will pontificate mindlessly once again. And on that prediction, I'm very confident.

Cheers,
 

And in case anyone wants to know if I'm going to be gracious in victory to our reznit eedjit, the answer is "no". He has done nothing to deserve it. Conceding after the lights have gone out doesn't cut it. It doesn't atone for his previous sins, and is hardly a concession under the circumstances.

Cheers,
 

It doesn't atone for his previous sins, and is hardly a concession under the circumstances.

Well, I'd grant Bart credit for a concession, but that's about it. He's still one of the world's great ignoramuses and his 'insights' aren't worth a wad of pigeon shit on the sole of your shoe.
 

Ah, the grace and class of the left never ceases to amaze me.
 

Ah, the grace and class of the left never ceases to amaze me.

Polish it all you want, Bart. It's still a turd.
 

"Bart":

You were wrong. Just like we pointed out. Joust like you were wrong in 2006. I know it's a lot to ask, but this time, do you think you could twig to a clue or two? If not, could you indulge us and SFTU for a year or so ... or until you manage some sentience?

Thanks in advance.

Cheers,
 

Ah, the grace and class of the left never ceases to amaze me.

# posted by Bart DePalma : 11:09 PM


What does a lying scumbag like you know about grace or class?
 

BDP to Garth:I have a C-Note that says Silver is wrong and Obama will not win by 5.8% or more.

Does Garth get to collect his C-note now?

CO:
Obama 53%
McCain 46%

Nationwide:
Obama 52%
McCain 46%

Isn't that a >5.8% victory?
 

@Pat,

You know he won't answer here (unless to spite me for saying so). More probably you'll have to spring this on him, cut-and-paste, after he makes an early post on a fresh thread. He's a slippery one.

Or you could join me in the plea for self-regulation and we could really make a go of creating a "Balkinization Comments Community" here. Can you be enticed?
 

HD kaliteli porno izle ve boşal.
Bayan porno izleme sitesi.
Bedava ve ücretsiz porno izle size gelsin.
Liseli kızların ve Türbanlı ateşli hatunların sikiş filmlerini izle.
Siyah karanlık odada porno yapan evli çift.
harika Duvar Kağıtları bunlar
tamamen ithal duvar kağıdı olanlar var
 

Post a Comment

Home