NationStates Jolt Archive

If you were only watching the Gallup polls in Florida, you might think that Bush has it locked up. After all, they've got Bush up 51-43.

Now, according to the breakdown at My Due Diligence (http://mydd.com), the other polls are as follows:
ARG: Kerry 49, Bush 46 (10/25)
Insider Advantage: Kerry 46, Bush 46 (10/24)
Miami Herald: Kerry 46, Bush 46 (10/21)
Rasmussen: Kerry 48, Bush 48 (10/24)
Research 2000: Kerry 48, Bush 47 (10/21)
Survey USA: Kerry 50, Bush 48 (10/24)
Zogby: Bush 48, Kerry 47 (10/24)
Now, in case you missed it, every other poll has Kerry ahead or tied in Florida, and certainly not down by 8 with Bush over 50%. So why is Gallup so far off the mark? Here's their voter sample from MyDD.com.
Likely Voter Sample
TOTAL: 768
Rep: 341 (44%)
Dem: 273 (36%)
Ind: 146 (19%)

Registered Voter Sample
TOTAL: 909
Rep: 387 (43%)
Dem: 321 (35%)
Ind: 187 (21%)

In 2000, the electorate in Florida was 40% Democrat, 38% Republican, and 22% Independent. Gallup swung a +2% advantage for Democrats to a +8% advantage for Republicans, for an overall 10% pro-Republican swing.
The only way Gallup is even close to correct is if Florida Republicans are somehow more motivated to vote (and have magically increased their numbers) for a candidate who has done at best a marginal job than Democrats are to vote against a person they feel (with some justification) stole the election in 2000. Gee--even if the Republicans are more motivated in that scenario (not likely), how likely is it that they've increased their numbers to that degree in the last four years? Answer: not bloody likely.

So I'm making the following official announcement: for the next week, until the election, anyone who quotes Gallup as anything but a partisan poll without providing evidence of their internals that shows they're sampling the electorate at anything close to historical modesl, will be laughed at as a partisan hack, and will be dismissed.

Fair enough, so are we to use that groovy website that counts multiple polls from a variety of sources? Because that sounds pretty reliable.

Fair enough, so are we to use that groovy website that counts multiple polls from a variety of sources? Because that sounds pretty reliable.

Sure, but because it contains the unreliable Gallup poll, I would add 1% to Kerry's numbers and take 1% away from Bush (unless evidence is produced that it also contains a poll that can be proven to be unfairly oversampling Democrats.)

Sure, but because it contains the unreliable Gallup poll, I would add 1% to Kerry's numbers and take 1% away from Bush (unless evidence is produced that it also contains a poll that can be proven to be unfairly oversampling Democrats.)

Then again, upon looking at the site, it seems pretty obviously liberally slanted. That's why I watch Fox News and read Gallup...
So I know the worst case scenario and the worst possible half-truths that can come from the media.
A liberal site will parrot basically what the rest of the world, and half of the U.S. thinks... which (unfortunately) (apparently) isn't important in American politics.

Okay, I have a theory-and it doesn't have much empirical evidence to back it up so it doesn't deserve its own thread, so I'll put it here:

I think that the closeness of this race is manufactured. It may be close, but not as close as we are lead to believe, and the news corporations have a vested intrest in this being as close as humanly possible. It's more watchable. Inconsistancies with the polls, variation between them. The inaccuracies of the polls in the last few years. (Someone put up a thread a day or so ago that laid out how innacurrate the polls where last time 'round.) I think the actual break down is a complete unknown. All we'll hear is that it is close and that we should prepare for an extra week or two of bonus campaign coverage now that we discovered 'post-election controversy(sp).'

Of course if Bush wins again it's four years of constant attempts to point out how shitty he is through any artistic channel possible. And if Kerry wins we get at least four years of Karl Rove going through everything Kerry has ever done (unless there is a huge loss on the Republican side and then we get a 'collapse of Gingrich' style restructuring...)

So, yeah-not a lot of imperical evidence. Just pattern theory.

Fair enough, so are we to use that groovy website that counts multiple polls from a variety of sources? Because that sounds pretty reliable.
It wasn't last presidential election. Just saying. I'll try and find the thread.

Here it is. (http://forums2.jolt.co.uk/showthread.php?t=368357)

Then again, upon looking at the site, it seems pretty obviously liberally slanted. That's why I watch Fox News and read Gallup...
So I know the worst case scenario and the worst possible half-truths that can come from the media.
A liberal site will parrot basically what the rest of the world, and half of the U.S. thinks... which (unfortunately) (apparently) isn't important in American politics.
mydd.com is a liberal site. But Gallup has routinely found outrageous figures that no other poll (even Strategic Vision and Mason-Dixon :p) has come close to. Remember that supposed 15% lead in September? Other polls that came out in that time period averaged a 3% lead for Bush. And just last week, Bush ahead by 8% in new Gallup poll...

The fact remains, Gallup overpolls Republicans.

I made a thread yesterday pointing out just how wrong realclearpolitics.com was in 2000, I'll dig it up.

EDIT: Oh, thats ^ it. :)

Okay, I have a theory-and it doesn't have much empirical evidence to back it up so it doesn't deserve its own thread, so I'll put it here:

I think that the closeness of this race is manufactured. It may be close, but not as close as we are lead to believe, and the news corporations have a vested intrest in this being as close as humanly possible. It's more watchable. Inconsistancies with the polls, variation between them. The inaccuracies of the polls in the last few years. (Someone put up a thread a day or so ago that laid out how innacurrate the polls where last time 'round.) I think the actual break down is a complete unknown. All we'll hear is that it is close and that we should prepare for an extra week or two of bonus campaign coverage now that we discovered 'post-election controversy(sp).'

Of course if Bush wins again it's four years of constant attempts to point out how shitty he is through any artistic channel possible. And if Kerry wins we get at least four years of Karl Rove going through everything Kerry has ever done (unless there is a huge loss on the Republican side and then we get a 'collapse of Gingrich' style restructuring...)

So, yeah-not a lot of imperical evidence. Just pattern theory.

This actually makes a lot more sense than a lot of other stuff I've heard.

Okay, I have a theory-and it doesn't have much empirical evidence to back it up so it doesn't deserve its own thread, so I'll put it here:

I think that the closeness of this race is manufactured. It may be close, but not as close as we are lead to believe, and the news corporations have a vested intrest in this being as close as humanly possible. It's more watchable. Inconsistancies with the polls, variation between them. The inaccuracies of the polls in the last few years. (Someone put up a thread a day or so ago that laid out how innacurrate the polls where last time 'round.) I think the actual break down is a complete unknown. All we'll hear is that it is close and that we should prepare for an extra week or two of bonus campaign coverage now that we discovered 'post-election controversy(sp).'

Of course if Bush wins again it's four years of constant attempts to point out how shitty he is through any artistic channel possible. And if Kerry wins we get at least four years of Karl Rove going through everything Kerry has ever done (unless there is a huge loss on the Republican side and then we get a 'collapse of Gingrich' style restructuring...)

So, yeah-not a lot of imperical evidence. Just pattern theory.
I suspect that some of it is manufactured--the Gallup numbers are too weird to be anything else, although I wouldn't be surprised to see Gallup come out with a poll the day before the election, released with very little aplomb, that comes out pretty close to being right, simply so they can point to their final poll in 4 years and say they were close.

But I suspect that a larger part of it is that the pollsters are just missing a larger and larger part of the electorate, and they haven't figured out how to compensate yet. It's the cell phone/caller id problem, and until they figure it out, telephone polls are going to be inconsistent. My guess is that it's hurting Kerry more than Bush right now.

I suspect that some of it is manufactured--the Gallup numbers are too weird to be anything else, although I wouldn't be surprised to see Gallup come out with a poll the day before the election, released with very little aplomb, that comes out pretty close to being right, simply so they can point to their final poll in 4 years and say they were close.

But I suspect that a larger part of it is that the pollsters are just missing a larger and larger part of the electorate, and they haven't figured out how to compensate yet. It's the cell phone/caller id problem, and until they figure it out, telephone polls are going to be inconsistent. My guess is that it's hurting Kerry more than Bush right now.

I dunno. I think the thought that we are a little bit behind Bush is electrifying the Democratic voters, not discouraging them.

I dunno. I think the thought that we are a little bit behind Bush is electrifying the Democratic voters, not discouraging them.

My sentiments exarktly

I dunno. I think the thought that we are a little bit behind Bush is electrifying the Democratic voters, not discouraging them.
I have a co-worker who just went to Arizona to work for the Kerry campaign because it's close (but it is not close here in Santa Cruz, so we have to travel to not preach to the choir. This is the town that is still running Going Upriver, as well as having run Uncovered:The Truth About the Build Up to Iraq, Outfoxed, and Hunting the President. Farenhiet 9/11 closed like a week ago.) I don't remember anyone doing that last time 'round.

I dunno. I think the thought that we are a little bit behind Bush is electrifying the Democratic voters, not discouraging them.It's a type of bandwagon propaganda. The implied message to those leaning toward Kerry is, "A vast majority supports Bush. Why do you insist on being so far left of the mainstream?" The implied message to those leaning toward Bush is, "A vast majority supports Bush. Don't you like it when everyone agrees with you?"

Screw Gallup. Zogby is way more accurate.

I suspect that some of it is manufactured--the Gallup numbers are too weird to be anything else, although I wouldn't be surprised to see Gallup come out with a poll the day before the election, released with very little aplomb, that comes out pretty close to being right, simply so they can point to their final poll in 4 years and say they were close.

But I suspect that a larger part of it is that the pollsters are just missing a larger and larger part of the electorate, and they haven't figured out how to compensate yet. It's the cell phone/caller id problem, and until they figure it out, telephone polls are going to be inconsistent. My guess is that it's hurting Kerry more than Bush right now.
Well, yeah-if you want to be all rational about it.

No, you're right. I believe that that is really a pretty big, if not dominant part of it. To hold on to my crackpot theory, the lack of investigation in journalism and the tendancy towards 'yellow journalism' these days means that they are not pressing the numbers to notice.

It's a type of bandwagon propaganda. The implied message to those leaning toward Kerry is, "A vast majority supports Bush. Why do you insist on being so far left of the mainstream?" The implied message to those leaning toward Bush is, "A vast majority supports Bush. Don't you like it when everyone agrees with you?"
Those are the voters that most disturb me, the ones who are more interested in having voted for the person that won than the one they wanted to win. Same bastards who buy their DVDs by the number of facings or their CDs based on where they are on the top 25 (quess how I worked my way through college...)

Those are the voters that most disturb me, the ones who are more interested in having voted for the person that won than the one they wanted to win.
I remember a quote by MunkeBrain a while back along the lines of "Bush has this race won you might as well just vote for him already". :p