FL: Rasmussen: McCain 49, Obama 44

[phk]

McCain will win by 3 to 5.

My Cuban friend in Miami says that the Spanish commentators are calling him a communist, practically toxic to the Cuban-vote.

Are you referring to Obama? A lot of the Hispanics I know have a lot of distrust of Obama. The blacks where I work, of course, are acting completely racist and get a free pass (as usual). I've heard stuff like "McCain is a honky" "Palin is a bitch/whore/honky".

Yep. But I think its primarily just Cubans that are GOP. I don't think the Puerto Ricans or Mexicans have that strong an affinity toward the GOP.

[??????????]

McCain will win by 3 to 5.

My Cuban friend in Miami says that the Spanish commentators are calling him a communist, practically toxic to the Cuban-vote.

Are you referring to Obama? A lot of the Hispanics I know have a lot of distrust of Obama. The blacks where I work, of course, are acting completely racist and get a free pass (as usual). I've heard stuff like "McCain is a honky" "Palin is a bitch/whore/honky".

Yep. But I think its primarily just Cubans that are GOP. I don't think the Puerto Ricans or Mexicans have that strong an affinity toward the GOP.

I only know one Cuban, the rest I know are Mexican and Puerto Ricans and they have no love of Obama. Very few Cubans around here, mostly Mexicans.

[J. J.]

PA and the United States:  They may be mostly independent of each other, but not purely so.  I mean, Pennsylvania is a component part of the United States.  They're not totally independent, and I do imagine they're more likely to correlate than to inversely correlate.  I'm arguing likely here, not definite.  I think it's more likely that PA moves with the nation than directly against, even if the latter is possible.

They are not totally independent, but not related.  The numbers are Bush's difference between his national margin in the popular vote:

2000
-0.51  US
-4.17  PA  
GAP:  3.66

2004
+2.46 US
-2.50 PA
GAP:  4.96

Even though Bush's national numbers improved and his margin in PA was narrower, in comparison with the national numbers, Bush ran more below his national average.

You must be logged in to read this quote.

"Unseen" is your word, not mine.  I would not use it.  I would use "unpolled."  The Bradley Effect implies people are being polled.  "Cell phones users only" would be an example of "unpolled."  A religious group that doesn't use phones would also be "unpolled."

[Alcon]

A gap between PV locally and nationally has nothing to do with local movements correlating with national trends...I didn't claim a perfect correlation, just one.  Besides, you just proved my point.  The national margin moved right, the PA margin also moved right.  That was my point...that event being more likely than the inverse.

So, your rule only applies to situations where people are being polled, but you claim they're giving inaccurate answers?  It doesn't apply to claims that people aren't polled at all?  Um, why?  That doesn't really make sense.  All things being equal, if a group has x property and isn't polled, that's going to affect the poll results more than if they have x property, are polled, and lie at <50% net rates.

[J. J.]

A gap between PV locally and nationally has nothing to do with local movements correlating with national trends...I didn't claim a perfect correlation, just one.  Besides, you just proved my point.  The national margin moved right, the PA margin also moved right.  That was my point...that event being more likely than the inverse.

Alcon, it's not strong enough, especially at the lower levels. 

Here is the Reagan numbers:

1980
+9.74 US
+7.19  PA 
GAP:  2.55

1984
+18.21US
+7.34 PA
GAP:  10.87

Even though the was a gigantic increase in the national margin, the state margin was virtually unchanged.  Though not my favorite recent elections, you had something similar in 1992 and 1996 for Clinton.  The GAP went up from 9.02 to 9.20.  From 1988 to 1992, the GAP for Bush actually did 5.41 to 3.53, though he lost the second one.  I'm seeing a particularly strong correlation.

You must be logged in to read this quote.

I don't think you really understand the Bradley Effect.  It assumes that voters are actually polled, but give inaccurate answers, i.e. they lie.  Why?  Because they don't to vote for a black candidate, but don't want to say it.  This occurs in exit polling as well.  It has nothing to do with the group not being polled; it, as a condition, requires them to be polled.

http://en.wikipedia.org/wiki/Bradley_effect

J,J,'s Second Rule applies to claims that some group isn't being polled at all.  They only have cell phones.  There will be a mass of people that have never voted that will overwhelm the polls and sweep the candidate into office.

Those two things are very different.

[Lunar]

It's easier for a state to trend in an opposite direction than the country over four years with different candidates than one month with the same candidates.  In fact, that's to be expected.

btw Alcon understands the Bradley effect.  He's saying there's no difference between missing a voter and catching the voter but missing the voter's vote.

Again, no clear Bradley effect occurred in the primaries.

[J. J.]

It's easier for a state to trend in an opposite direction than the country over four years with different candidates than one month with the same candidates.  In fact, that's to be expected.

btw Alcon understands the Bradley effect.  He's saying there's no difference between missing a voter and catching the voter but missing the voter's vote.

Again, no clear Bradley effect occurred in the primaries.

I say there is a very big difference between missing a voter and getting an inaccurate accounting of how the voter voted or intends to vote.  Except in some very rare situations, a missed voter will be taken care of in a sample size; the pollster will ask the next guy on the list.  A false responder gets recorded.

I'm not too interested in the primaries.  The last one I saw strongly documented was 2006 in MI, on a referendum.

[Lunar]

Well, the Bradley effect is a personal relationship between the voter and the candidate - thus it's highly individualized and does not always occur.  It's NEVER been documented before for national candidates.  It HAS been documented strongly in two cases: Bradley in CA and Wilder in VA.  One could argue that the nature of a presidential race and the saturation of media information results in a presidential election, voters will have plenty of reasons to cite to themselves for voting against the black candidate when they used to be Democrats.  In fact, I think we can already see some of this in bitter ex-Hillarites.  All they have to do is tell themselves they support McCain because of the generic anti-Obama Hillary stuff (Obama is inexperienced, sexist, untrustworthy), and the problem is solved.  Because the Bradley Effect probably occurs most often within party loyalists (thus the lie), there are a LOT of reasons to believe the Bradley effect might not substantially outweigh other things.

In addition, a  simple counter-argument could say that some people who will vote for Obama are claiming to be undecideds or voting McCain simply because they are bitter about the primary loss of Hillary, especially in a state like PA.  Who's to say that these people won't outweigh the Bradley effect or at least cancel some of it?  I know this effect hasn't been documented in PA before, but neither has the Bradley effect.  Again, the Bradley effect is not a substantiated claim for a presidential election - it has no history and has yet to demonstrate itself whatsoever during the 50 state primary (where Obama was often underpolled and not overpolled).

It is political bias that is causing you to give McCain an extra %1 in this scenario.

When we talk about missing voters, we don't mean random voters.  We mean very specific demographics that are getting underpolled, which CANNOT be fixed by increasing the sample size.  In fact, increasing the sample size would increase the effects of these things.  These people are just as relevant as the Bradley effect.  There is no substantive difference between miss-polling voters and missing voters with unique characteristics.  You are really using a technical detail to avoid violating your second rule but you are still 100% violating it in principle.

If you read your own wikipedia article, you can see that the Bradley effect is not unanimous.  In fact, there's evidence that Barack might not face it.  In the case of Jindal, he might have faced the Bradley effect the first time he ran in LA, but the second there was no recorded effect.  Maybe once voters are familiar with a candidate, they lie to pollsters less?  I mean, I have a lot of different points in this post as to why you are wrong to flatly assume a 1% Bradley effect, so don't jump down to this one, but it's something to keep in mind.

[Lunar]

In addition, your own wikipedia article (It's yours now for the rest of time btw) says this:

You must be logged in to read this quote.

I know you don't think the primary results matter, but the matter of the fact is that it is thousands of times more useful than California and Virginia governor elections from twenty years ago.  Plus, if the Bradley effect does exist, it almost certainly would not exist in Republicans who would have substantive policy disagreements to justify their vote and less racial guilt.  So we would almost certainly see some form of it in the primaries if it were clearly going to occur in the general.

My point is that you cannot justify an automatic 1% of hidden votes in PA under this idea without relying on faith.

[J. J.]

Well, the Bradley effect is a personal relationship between the voter and the candidate - thus it's highly individualized and does not always occur.  It's NEVER been documented before for national candidates. 

Ah, we've never had a national that was African American before, so I don't find that argument compelling.

You must be logged in to read this quote.

What makes you think it is limited to party loyalists?  In CA and VA, it seemed to have occurred within the general electorate.  Likewise, the voter can just declare himself undecided, which is possibly why the Obama campaign is saying that 60% will break for McCain.

You must be logged in to read this quote.

No, I said up to 1%, enough that where I see a tie, I'll say McCain, barely.

You must be logged in to read this quote.

You may call those "missing voters," but prefer "unpolled," or even "underpolled/"  That is different than a pollster giving a false answer.

You must be logged in to read this quote.

First, there has to be such a group.  Where is this group.

You must be logged in to read this quote.

If you would have comprehended the article, you would have notice the heading "Possible diminished effects."  Originally it was 9-10 points.  I'd be surprised if it was above 1 point.  It's enough for me to say, where there is a true tie, I'll give it to McCain.

A small effect won't show up easily, but I think it's there.  I've been surprised at some of the more open hostility I've heard regarding Obama's ethnicity, even in Phila.

[J. J.]

In addition, your own wikipedia article (It's yours now for the rest of time btw) says this:

You must be logged in to read this quote.

I know you don't think the primary results matter, but the matter of the fact is that it is thousands of times more useful than California and Virginia governor elections from twenty years ago.  Plus, if the Bradley effect does exist, it almost certainly would not exist in Republicans who would have substantive policy disagreements to justify their vote and less racial guilt.  So we would almost certainly see some form of it in the primaries if it were clearly going to occur in the general.

My point is that you cannot justify an automatic 1% of hidden votes in PA under this idea without relying on faith.

You know in both cases, Bradley and Wilder, they won their primaries.  I don't recall it showing up in a Democratic Primary.

It's not hidden; we know that so far, this has effect occurred.  I'm telling you now; I think it there.  My guess is something not exceeding 1%.  It's enough to to make me calculate it there.

[Lunar]

Well, the Bradley effect is a personal relationship between the voter and the candidate - thus it's highly individualized and does not always occur.  It's NEVER been documented before for national candidates. 

Ah, we've never had a national that was African American before, so I don't find that argument compelling.

EXACTLY.

That's why you can't assert one phenomenon into another context under the guise of fact.  Governor elections in the 80's in two states and exit polling in a ballot initiative in Michigan in 2006 is enough to prove that Obama hasn't crossed the threshold to the point where people will find real reasons to give themselves a reason to vote against him.

Unless you can explain to me, piece-by-piece, and not through anecdotes how you can cross-apply the Bradley/Wilder/MI2006ballotinitiatveexitpolling effect to Obama in 2008, I'm going to reject your argument.  I've given you far more relevant evidence that if anything, we might witness the reverse of a Bradley effect.

You must be logged in to read this quote.

What makes you think it is limited to party loyalists?  In CA and VA, it seemed to have occurred within the general electorate.  Likewise, the voter can just declare himself undecided, which is possibly why the Obama campaign is saying that 60% will break for McCain.[/quote]

I think it might be somewhat limited to party loyalists because people in the opposing, Republican party have substantive policy disagreements with the black Democratic candidate.  They might be racist but they can tell themselves their vote is for other reasons.  The Bradley effect, and I'm no psychology PhD, could be attributed to people who *should* vote Democratic and know it but when it comes down to what lever to pull, their racism wins them over.

Increased saturation, such as this presidential race, would combat this as people would get to know the two candidates better and find better reasons to like/dislike each of them to coordinate with their own racism.

You must be logged in to read this quote.

No, I said up to 1%, enough that where I see a tie, I'll say McCain, barely.[/quote]

Whatever.  The point is whether this effect, which is real and I agree with you, possibly less than 1%, outweighs the rest of the noise (people lying saying they'll vote vote McCain, non-response bias, etc).

You must be logged in to read this quote.

You may call those "missing voters," but prefer "unpolled," or even "underpolled/"  That is different than a pollster giving a false answer.

You must be logged in to read this quote.

First, there has to be such a group.  Where is this group.[/quote]

EVERYWHERE.  There are THOUSANDS of these groups.  These are people who work irregular shifts.  People who own multiple houses.  People with no landlines.  Non-responders.  And yes, people who lie are counted in this group.

You must be logged in to read this quote.

If you would have comprehended the article, you would have notice the heading "Possible diminished effects."  Originally it was 9-10 points.  I'd be surprised if it was above 1 point.  It's enough for me to say, where there is a true tie, I'll give it to McCain.

A small effect won't show up easily, but I think it's there.  I've been surprised at some of the more open hostility I've heard regarding Obama's ethnicity, even in Phila.
[/quote]

Your own article also indicates that Obama thus far has had the opposite of a Bradley effect.  This is again, more relevant than exit-polls on a ballot initiative in Michigan.

[Lunar]

In addition, your own wikipedia article (It's yours now for the rest of time btw) says this:

You must be logged in to read this quote.

I know you don't think the primary results matter, but the matter of the fact is that it is thousands of times more useful than California and Virginia governor elections from twenty years ago.  Plus, if the Bradley effect does exist, it almost certainly would not exist in Republicans who would have substantive policy disagreements to justify their vote and less racial guilt.  So we would almost certainly see some form of it in the primaries if it were clearly going to occur in the general.

My point is that you cannot justify an automatic 1% of hidden votes in PA under this idea without relying on faith.

You know in both cases, Bradley and Wilder, they won their primaries.  I don't recall it showing up in a Democratic Primary.

It's not hidden; we know that so far, this has effect occurred.  I'm telling you now; I think it there.  My guess is something not exceeding 1%.  It's enough to to make me calculate it there.

Did they win their primaries by more than they polled?  Wilder also won his election to be governor but he still faced the Bradley effect.

[Robespierre's Jaw]

Not surprising as it was known that McCain was going to defeat Obama in Florida this November. However I am surprised that according to the poll that both Bob Barr and Ralph Nader are polling at two percent. However this could be yet another servere case of overpolling as seen in the 2004 Presidential Election when many predicted, at least according to CNN that Ralph Nader was overpolling.

[Lunar]

Hey, we're trying to talk about the Bradley Effect in Pennsylvania here and you're trying to talk about FLORIDA?

[J. J.]

Well, the Bradley effect is a personal relationship between the voter and the candidate - thus it's highly individualized and does not always occur.  It's NEVER been documented before for national candidates. 

Ah, we've never had a national that was African American before, so I don't find that argument compelling.

EXACTLY.

That's why you can't assert one phenomenon into another context under the guise of fact.  Governor elections in the 80's in two states and exit polling in a ballot initiative in Michigan in 2006 is enough to prove that Obama hasn't crossed the threshold to the point where people will find real reasons to give themselves a reason to vote against him.

Last one noted was a referendum in MI in 2006.

You must be logged in to read this quote.

You are free to reject it, but I expect it to be there.  No, you have produce evidence that it is weakening, which I readily concede.  If I though it was within Wilder range, I'd be saying McCain will take NY.  There is one factor from the Primaries that might indicate it.  Caucuses.  Obama did better when people were watching, he tended to worse when secret ballots were used.

You must be logged in to read this quote.

And the of Robert Byrd is?  I really doubt it is party loyalists.

You must be logged in to read this quote.

Or, say "I'm undecided."

You must be logged in to read this quote.

No, I think it is a problem, though as great as it was 20 years ago.  I said a few hours ago that I'd like a poll before I call PA for Obama.  I thought it would be 2-3 points for Obama, so it wouldn't be a factor.  Now it's a factor, so I won't ignore it.  

You must be logged in to read this quote.

Getting close a violation J. J's First Rule.  Most of what you are referring to there tends balances itself out.  Except for the folks that lie, none go into the sample.  You'll also note that they have not appeared to skew other polling.

You must be logged in to read this quote.

Who won most the secret ballot primaries?  Who won most the public caucuses?  Now, I'd suspect MOST of that better organization, but not all of it.  It's enough for me to say, a tie goes to McCain.