Obama leads Romney among NASCAR fans

[Raygun99]

@Simplicio

I don’t really see any reason why its worse then phone surveys, where most people also don’t respond. If nothing else, you might at least get a different skew then phone surveys, and so get some extra information out of responses.

The subsample size of NASCAR fans was 200, which gives you error bars of 7%.

Oh yeah, 200’s no good.

[Euphonious_Polemic]

@Marley23

That still sounds pretty useless unless they were able to keep track of which invitees actually responded. But even if it wasn’t a typical online poll, Zogby still sucks. Their performance in the last few elections has been terrible.

Agreed.

[Jas09]

@Raygun99

A sample size of 860 - provided it’s well selected - is more than sufficient to create an accurate poll within a few percentage points. The stuff between the dashes is the difficult part, though.

Right, but unless I’m reading this wrong they did a nation-wide sample of 860 and then released a cross-tab for NASCAR fans. I’m not sure what the size of that sub-sample is, but unless they did some over-sampling just for that demographic then I think the error margin for that group was probably extremely large.

Out of curiosity I opened up the crosstabs. Here’s from the first page:

Subsets of the data have a larger margin of error than the whole data set. As a rule we do not rely on the
validity of very small subsets of the data especially sets smaller than 50-75 respondents. At that subset we
can make estimations based on the data, but in these cases the data is more qualitative than quantitative.

For the NASCAR group they had 202 respondents - 98 for Obama, 73 for Romney, 5 Libs, 7 Greens, 4 Goodes, and 17 Not Sures.

Compare to some other small subsets in this poll: Obama is leading among military families 56-33. He’s only losing in the South 48-43.

You just can’t really draw many conclusions from that kind of data, IMO.

[moonshot925]

Zogby online poll. Pure nonsense.

[Truman_Burbank]

@Marley23

Zogby sucks and online polls are worthless. However I am loving the narrative that Romney is winning and the polls are wrong. It’s going to be even more fun hearing that one on November 7th.

On November 7th the narrative will be that Romney lost because the skewed polls depressed republican turnout.

[Marley23]

@moonshot925

Pure nonsense.

Yes, in this case it is. But so is the “the polls are all wrong!” story, and so is your theory that Romney is going to get 400 electoral votes.

[Jas09]

@moonshot925

Zogby online poll. Pure nonsense.

I agree. I can’t figure out why anybody thought it worth making a thread about…oh wait…

[Sitnam]

@Jas09

It really is awesome. The fact that somebody went to the trouble to create a website “unskewing” every poll to show a Romney landslide is amazing to me. http://www.salon.com/2012/09/25/the_website_where_mitt_romneys_winning_in_a_landslide/

Unskewed, like ‘Fair and Balanced’.

[BrainGlutton]

@moonshot925

Yahoo News: Latest and Breaking News, Headlines, Live Updates, and More

Another skewed poll.

I can’t imagine anyone whom the NASCAR could hate more than Obozo.

Actually, it’s easy to imagine the NASCAR Nation hating some uberrich hyperprivileged jobs-destroying aristo a lot worse.

[MsWhatsit]

@moonshot925

Obozo.

@Ca3799

Rmoney

Stop it.

[GrandWino]

@MsWhatsit

Stop it.

You stahp, M$Watzit.

[Euphonious_Polemic]

@moonshot925

Zogby online poll. Pure nonsense.

1. What leads you to the conclusion that JZ Analytics has erred in it’s methodology for this particular poll? Specifics please, as this is not the pit.

2. Do you believe that using an amusing variation of the president’s name gives your posting additional veracity? If not, why do you use it in this particular forum?

[Nancarrow]

@Happy_Lendervedder

Rmoneyis more of a Freudian slip than a typo.

Surely that’s a photoshop? Because if it isn’t, why the hell hasn’t it been plastered over every last square inch of the planet yet?

[Marley23]

@Nancarrow

Surely that’s a photoshop? Because if it isn’t, why the hell hasn’t it been plastered over every last square inch of the planet yet?

It’s a Photoshop and it’s been plastered over every square inch of the Internet.

[Jas09]

Yes, it’s a 'shop.

[Raygun99]

@Jas09

Right, but unless I’m reading this wrong they did a nation-wide sample of 860 and then released a cross-tab for NASCAR fans. I’m not sure what the size of that sub-sample is, but unless they did some over-sampling just for that demographic then I think the error margin for that group was probably extremely large.

Out of curiosity I opened up the crosstabs. Here’s from the first page: For the NASCAR group they had 202 respondents - 98 for Obama, 73 for Romney, 5 Libs, 7 Greens, 4 Goodes, and 17 Not Sures.

Compare to some other small subsets in this poll: Obama is leading among military families 56-33. He’s only losing in the South 48-43.

You just can’t really draw many conclusions from that kind of data, IMO.

Yes, I agree with that. I had assumed the survey was just of NASCAR fans, not that they were a crosstab pulled out of a greater sample.

[Zakalwe]

He’s certainly leading with this (white male mid-40s) NASCAR fan! I leave for Talladega in 9 days! Party time!

With that said, I know (from prior bitter experience) that I will have to walk past at least two well-trafficked Nobama tables to get to the race. Based on my personal experience, there’s no way that poll is right.