Non-Gallup/Rasmussen tracking polls thread

[Alcon]

And Zogby is a Democrat, and I'd though R2K's founder was actually a GOP-leaning independent.  I think we're putting too much stock into the R2K/Kos-as-political-propaganda idea.

[Alcon]

And Zogby is a Democrat, and I'd though R2K's founder was actually a GOP-leaning independent.  I think we're putting too much stock into the R2K/Kos-as-political-propaganda idea.

The only problem I have with it is the weighting.

Yea, OK, cool, so basically it's like Hotline, but with objectionable weighting.  I actually agree with Spade that Hotline has some strange methodology quirks too, so I put them basically in the same caliber (question, maybe tinker, but don't ignore)

[Alcon]

Hotline doesn't have the track record of R2Kos, but if the weighting was fixed, I'd rank with Hotline.

Eh?  Hotline has the same track record as a national poll, and less as a state poll.

Del Ali - I don't think so.  Besides, R2000 has a history of polling for Democrats.

And btw, I'm not putting too much stock into it, Alcon.  But the fact is that it can't be ignored, which is the reason why I mention it. 

Cool with me.  And I actually think the Zogby thing, which I hadn't heard, if it's true, is worth mentioning.  Zogby's...news-making...might be influenced by his clients.

[Alcon]

Hotline doesn't have the track record of R2Kos, but if the weighting was fixed, I'd rank with Hotline.

Eh?  Hotline has the same track record as a national poll, and less as a state poll.

I don't remember Hotline from 2004?

Yeah, my point man

[Alcon]

Cool with me.  And I actually think the Zogby thing, which I hadn't heard, if it's true, is worth mentioning.  Zogby's...news-making...might be influenced by his clients.

The only thing I found supporting ChrisfromNJ's claims (on Google) is from this Daily Kos diary and it deals with a Zogby Interactive poll.  It would also make an important difference whether this was done before the horserace question or after it.

http://www.dailykos.com/story/2008/10/15/13844/116

Poisons his later samples, since they're almost all the same.   I don't understand why Zogby would have done that...he's a Democrat, IIRC.

[Alcon]

Yeah, I know, but unless the GOP is investing in him (in some sense), wtf?

[Alcon]

Lunar, the subset problem has been discussed.

Where, and to what conclusion?

[Alcon]

Lunar, the subset problem has been discussed.

Where, and to what conclusion?

This thread, IIRC, about 4-5 days ago.  It was just too small a sample size to be able to generalize.

Um, that's not true.  You can still calculate a Margin of Error for a (relatively) small sample size, and apply it.  The MoE is higher.  It isn't high enough to make that result not highly suspicious.

[Alcon]

Um, that's not true.  You can still calculate a Margin of Error for a (relatively) small sample size, and apply it.  The MoE is higher.  It isn't high enough to make that result not highly suspicious.

MOE = 150%

MoE = +/-10%

That subsample = Obviously flawed well out of MoE range

[Alcon]

Um, that's not true.  You can still calculate a Margin of Error for a (relatively) small sample size, and apply it.  The MoE is higher.  It isn't high enough to make that result not highly suspicious.

MOE = 150%

MoE = +/-10%

That subsample = Obviously flawed well out of MoE range

Actually, there were others with similar results.  I raised the question about it.  It seems to be a very small sampling.

Which makes it immune to the laws of margin of error? 

[Alcon]

FTR, anyone wanting to calculate an MoE, it's 100/[sqrt(sample size)]. Assuming IBD/TIPP has about 9-10% of respondents as 18-25, that's around 100 respondents, for an MoE of about 10%. 74-22 McCain under such as MoE means that McCain is at at least 64% among the group--definitely not true.

In fact, 538 did a calculation to Obama being ahead by twenty points among the youth and the odds of being that far off. I think 538 might be slightly overestimating Obama's actually lead among under-25s, but he comes out to the odds of a 74-22 McCain sample being 55 billion to one. (Take an Obama lead of 9 points, Kerry's lead in the demographic, and the odds are still well into the millions to one.)

Thanks I was trying for that calculation myself, and failing miserably.

It points to a methodology error.

[Alcon]

Um, that's not true.  You can still calculate a Margin of Error for a (relatively) small sample size, and apply it.  The MoE is higher.  It isn't high enough to make that result not highly suspicious.

MOE = 150%

MoE = +/-10%

That subsample = Obviously flawed well out of MoE range

Actually, there were others with similar results.  I raised the question about it.  It seems to be a very small sampling.

Which makes it immune to the laws of margin of error? 

No, just using a really bad sample for that age group.

Right...which shows that there was probably a deep methodological flaw in this poll, or its sample, or something.  It's not "too small of a sample to be able to generalize" -- the generalization is that the chance that would occur, without their being a sampling issue, is infinitesimal.  Add that into the fact that the poll seems like an outlier in the top-line, and I'm beginning to smell the strong scent of fish.

[Alcon]

Well, I'm look looking at this subsample, which is off.  As a full sample, until today, it was in line with the other major polls, even more pro-Obama than BG.

OK, so you think it's a good poll, and it's just a one-in-several-million error?  Interesting.

[Alcon]

Sorry, but I don't see the importance of a bad subsample, especially since it was repeated during samples that matched other polls.

Because it demonstrates that either:

1. There's a methodological error; or,

2. A one-in-many-million chance event has occurred.

Which seems more likely to you, especially when the poll soon thereafter shows a result that seems like an outlier?

[Alcon]

Sorry, but I don't see the importance of a bad subsample, especially since it was repeated during samples that matched other polls.

Because it demonstrates that either:

1. There's a methodological error; or,

2. A one-in-many-million chance event has occurred.

Which seems more likely to you, especially when the poll soon thereafter shows a result that seems like an outlier?

It hasn't been showing the results as being an outlier, at least until today.  It was in line with Gallup and Rasmussen. 

In seems likely  subsample is clearly an outlier, but the whole poll.

So, you're maintaining that the subsample was just a one-in-several-million event? Yes or no?

[Alcon]

I'm saying the subsample is bad, but a bad subsample doesn't invalidate the poll.

So, in other words, you are maintaining that it's a one-in-several-million event, and not a flaw in the methodology.  Why?

[Alcon]

What type of lead among this age group are other pollsters showing?

All, that I know of, are showing at least 15-20 points.

Could this skew the top line #'s?

To some (small) degree, but I think the bigger point is that it points to a likely flaw in methodology.  That kind of result doesn't just materialize out of statistical noise.

[Alcon]

I want to see the internals first.

Let's examine every internal then to see where there could be flaws to add points to your candidate.

Time elapsed:  2 minutes, 51 seconds.

[Alcon]

I'm saying the subsample is bad, but a bad subsample doesn't invalidate the poll.

So, in other words, you are maintaining that it's a one-in-several-million event, and not a flaw in the methodology.  Why?

No, I'm saying that the poll isn't primary designed to accurately measure a subsample.

It has to be either a one-in-a-million event or a flaw in the methodology.  It can't be neither, lol.

[Alcon]

The problem that I'm having with this entire argument is that, even with a bad subsample, the poll is conforming to the other polls, or at least was until today.

So it doesn't matter how invalid the methodology of the poll is, as long as it comes within close range of other polls?

hellooo arg

[Alcon]

The problem that I'm having with this entire argument is that, even with a bad subsample, the poll is conforming to the other polls, or at least was until today.

So it doesn't matter how invalid the methodology of the poll is, as long as it comes within close range of other polls?

hellooo arg

No, it means that there probably is not a problem with the methodology that affects the result.

So, you think that it was just a one-in-multi-million error?  You think that is more likely than a sampling error?

Thanks for confirming that.

[Alcon]

No, I'm saying that whatever the problem, it doesn't seem to effect the result.  Please don't put words into my mouth.

I'll let you put words in your own mouth, then.  There are three options:

1) The poll's sample is not as "off" as it seems. Obama is performing worse than Kerry among youths -- probably much worse.

2) The one-in-many-million event occurred, and the poll was just off by chance.

3) The poll had a methodological flaw.

You need to pick one.  There are no other possible options here -- either the poll isn't off, it is off by random chance, or it is off by error.  Which option are you arguing?

[Alcon]

No, I'm saying that whatever the problem, it doesn't seem to effect the result.  Please don't put words into my mouth.

I'll let you put words in your own mouth, then.  There are three options:

1) The poll's sample is not as "off" as it seems. Obama is performing worse than Kerry among youths -- probably much worse.

2) The one-in-many-million event occurred, and the poll was just off by chance.

3) The poll had a methodological flaw.

You need to pick one.  There are no other possible options here -- either the poll isn't off, it is off by random chance, or it is off by error.  Which option are you arguing?

4)  The poll is not designed to accurately measure the subsamples.  That isn't a methodological flaw because it's not designed to get that type of information.

That makes no sense at all.  How can the poll be overall sound, while its sub-samples can readily fall ridiculously outside the MoE?  That means it's not an accurate sample.  Margin of error applies at low samples, too, just more liberally.

How exactly does one design a poll to not accurately measure a subsample within MoE, without having the poll sample incorrectly?

[Alcon]

Thanks for the daily tracker...tracker, Sam.

[Alcon]

IBD/TIPP
Obama 45.8%(+1.0)
McCain 42.3%(-1.4)

Probably a bad sample dropping.

Yeah, a good Mac day fell off.

Probably not the methodology.

Is that your unsubtle substitute for responding to my last post?