2016 Postmortem
Related: About this forumDashing Republican Hopes
The right wing's hope in this election is that nearly all the state polls are wrong and they are all wrong in one direction.
As part of their argument they argue that polls were wrong during the heated Wisconsin recall. Actually they were accurate to 1/10 of a percent:
They also argue that the polls missed the Republican wave of 2010. Actually they caught it. What did happen is that right leaning pollsters like Rasmussen and Gallup grossly overestimated the size of the wave:
And, finally they suggest that polls are skewed. That pollsters have a model that they use which pre-assigns what percentage different groups will comprise of the electorate. Most of them don't. Some of the cheap Republican robopollsters do but that's a different story. Most pollsters give the percentages they do because that's what the respondents to the poll are telling them:
http://www.theatlantic.com/politics/archive/2012/09/are-polls-skewed-too-heavily-against-republicans/262834/
Nika
(546 posts)... irrationality. They need rose colored sunglasses to face life. Because it grows bleaker every day for them.
DemocratSinceBirth
(99,710 posts)Oh, their fourth argument is that only Silver is confidently is predicting an O win. Actually, there's probably a dozen social scientists/statisticians doing the same thing and getting roughly similar results.
VirginiaTarheel
(823 posts)DemocratSinceBirth
(99,710 posts)meadowlark5
(2,795 posts)My MIL is a Fox junkie and really has no mind or reasoning of her own. A perfect minion that can be lead and thoughts influenced so incredibly easily. She told my husband on a phone call this morning that Scumney is going to win in a landslide and if Obama wins it will be just barely a win.
My husband works with a tea bagger Rush junkie. He is very happy going into the last days of the campaign, certain scumney will be winning.
I guess if they are only spoon fed propaganda and facts are to be discarded unless they favor them, then I suppose they truly believe this and are happy and optimistic
DemocratSinceBirth
(99,710 posts)Oh, their fifth argument is that because Obama isn't doing well among independents he can't win. Well, he's winning in the polls where they got that information from and that is already factored in. Also, a lot of those independents are Tea Baggers who don't consider themselves Republicans. I also see some movement in the polls among independents toward Obama. I suspect that's a function of more latent Republicans masquerading as independents now declaring themselves Republicans. That's why there's not an even bigger shift to Obama in some of the polls. The same people voting against him are still voting against him except a lot of them are calling themselves Republicans and not independents.
ffr
(22,670 posts)It's so difficult to reach the same people that polling did 20 - 30 years ago, the ones nowadays find people I don't believe accurately represent the electorate.
I'm not swayed by them at all, just as I'm not swayed to buy products advertised on television or radio. I buy what I need and I vote regardless.
DemocratSinceBirth
(99,710 posts)Res ipsa loquitar:
http://fivethirtyeight.blogs.nytimes.com/2012/11/04/nov-3-romneys-reason-to-play-for-pennsylvania/#more-37144
ffr
(22,670 posts)I stand by my position. I've know several people who haven't been polled even once this election cycle, myself included. If I had, I would have said Obama/Dems, as would these other people. Who are the pollsters polling. Not us! And we vote.
Polls are indicators and they're junk.
DemocratSinceBirth
(99,710 posts)Or how did they nail the 08 presidential results to 3/10 of a percent:
http://www.realclearpolitics.com/epolls/2008/president/national.html
Of course you didn't get polled , polling is based on inferential statistics. One can make an inference from a smaller sample to a larger sample.
Flip a con...If you flip a coin ten times you might get nine heads and one tail. Flip that coin one thousand times and I assure you it will come very close to five hundred heads and five hundred tails. Flip it a million times and I assure you it will be dead even. That's how inferential statistics work.
ffr
(22,670 posts)DemocratSinceBirth
(99,710 posts)http://www.realclearpolitics.com/epolls/2008/president/nc/north_carolina_mccain_vs_obama-334.html
That's a lot more accurate than the 5% -10% error rate you attributed to polls.
I don't think anybody expects polls to be accurate to the decimal point.
ffr
(22,670 posts)on a Web page of an unspecified date. When was this done? Is this a job to show how accurate polling is to corroborate your assertion? Why doesn't the poll averages include polls like WaPo, ABC, CBS, etc. Anybody with any deductive thought will see right through this as a straw man.
Polls are junk. You're just proving my point. Just agree to disagree already. I appreciate your taking the time to prove me wrong, but I have better things to do with my time than wasting it on a pointless subject.
You've proved you are right. I'm undeterred.
DemocratSinceBirth
(99,710 posts)"Why doesn't the poll averages include polls like WaPo, ABC, CBS, etc. Anybody with any deductive thought will see right through this as a straw man. "
Because WaPo, ABC, CBS, etc didn't poll the state of North Carolina
If you check the link it contains every published poll for the state from 2/26/08 -11/3/08
You accused me of making a logical fallacy, a straw man without an iota of evidence,
If you care to look you will find myriad peer reviewed studies that demonstrate the efficacy of aggregating polls to predict electoral outcomes.
Start here:
http://userwww.service.emory.edu/~dlinzer/Linzer-prespoll-May12.pdf
And follow the footnotes,
ffr
(22,670 posts)"Because WaPo, ABC, CBS, etc didn't poll the state of North Carolina"
And you provide me with a link that shows that they did.
http://www.realclearpolitics.com/epolls/2008/president/national.html
Dude. I've had enough. I'm undeterred. This is a straw man. This page link proves it. +7 +8. +9. Just because RCP throws together a bunch of them and you base your opinion on that, only proves they are crafty at what they do and that you think you can follow their lead. Heck, I'm fairly crafty at spreadsheets and graphs too. I could do this same bit of work without much effort.
Election 2008 National Head-to-Head Polls
Marist 11/03 - 11/03 804 LV 4.0 52 43 Obama +9
Battleground (Lake)* 11/02 - 11/03 800 LV 3.5 52 47 Obama +5
Battleground (Tarrance)* 11/02 - 11/03 800 LV 3.5 50 48 Obama +2
Rasmussen Reports 11/01 - 11/03 3000 LV 2.0 52 46 Obama +6
Reuters/C-SPAN/Zogby 11/01 - 11/03 1201 LV 2.9 54 43 Obama +11
IBD/TIPP 11/01 - 11/03 981 LV 3.2 52 44 Obama +8
FOX News 11/01 - 11/02 971 LV 3.0 50 43 Obama +7
NBC News/Wall St. Jrnl 11/01 - 11/02 1011 LV 3.1 51 43 Obama +8
Gallup 10/31 - 11/02 2472 LV 2.0 55 44 Obama +11
Diageo/Hotline 10/31 - 11/02 887 LV 3.3 50 45 Obama +5
CBS News 10/31 - 11/02 714 LV -- 51 42 Obama +9
ABC News/Wash Post 10/30 - 11/02 2470 LV 2.5 53 44 Obama +9
Ipsos/McClatchy 10/30 - 11/02 760 LV 3.6 53 46 Obama +7
CNN/Opinion Research 10/30 - 11/01 714 LV 3.5 53 46 Obama +7
Pew Research 10/29 - 11/01 2587 LV 2.0 52 46 Obama +6
You win. I give up. I'm undeterred.
DemocratSinceBirth
(99,710 posts)Marist 11/03 - 11/03 804 LV 4.0 52 43 Obama +9
Battleground (Lake)* 11/02 - 11/03 800 LV 3.5 52 47 Obama +5
Battleground (Tarrance)* 11/02 - 11/03 800 LV 3.5 50 48 Obama +2
Rasmussen Reports 11/01 - 11/03 3000 LV 2.0 52 46 Obama +6
Reuters/C-SPAN/Zogby 11/01 - 11/03 1201 LV 2.9 54 43 Obama +11
IBD/TIPP 11/01 - 11/03 981 LV 3.2 52 44 Obama +8
FOX News 11/01 - 11/02 971 LV 3.0 50 43 Obama +7
NBC News/Wall St. Jrnl 11/01 - 11/02 1011 LV 3.1 51 43 Obama +8
Gallup 10/31 - 11/02 2472 LV 2.0 55 44 Obama +11
Diageo/Hotline 10/31 - 11/02 887 LV 3.3 50 45 Obama +5
CBS News 10/31 - 11/02 714 LV -- 51 42 Obama +9
ABC News/Wash Post 10/30 - 11/02 2470 LV 2.5 53 44 Obama +9
Ipsos/McClatchy 10/30 - 11/02 760 LV 3.6 53 46 Obama +7
CNN/Opinion Research 10/30 - 11/01 714 LV 3.5 53 46 Obama +7
Pew Research 10/29 - 11/01 2587 LV 2.0 52 46 Obama +6
I thought we were discussing North Carolina polls.
I'm not basing my opinion on Real Clear Politics. I'm basing it on the Law Of Large Numbers. The law of large numbers suggest that as the size of your sample increases your accuracy increases and your findings become more robust. That's why by averaging polls the aggregate of nationals polls came within 3/10 of a percent of the national popular vote in 08:
http://www.realclearpolitics.com/epolls/2008/president/national.html
And within 7/10 of a percent in North Carolina:
Logical
(22,457 posts)DemocratSinceBirth
(99,710 posts)Who am I going to believe?
Some random man or woman on the internet or social scientists and statisticians who have submitted their methodology to peer review?
"If I can see further than other men it is because I stand on the shoulders of giants.", ergo:
ffr
(22,670 posts)why can't you just quit? I'm not after you, but you certainly are taking this personal. Relax already.
You will not convince me. I'm quite confident in my ability to do math and I know how easy it is to manipulate data to show something that looks convincing, when in fact, when you dig into it, it's BS.
Can you just let it go? Please, I beg you, let it go already. You win. I'm undeterred.
DemocratSinceBirth
(99,710 posts)If it was BS why didn't his peers call BS.
johnlucas
(1,250 posts)Show your debater this article & see if that debater says the same.
History says be skeptical of presidential polls
Excerpts
As for elections shifting in the late stages, in 2000, an Oct. 6 poll had George W. Bush at 48 percent and Al Gore at 41 percent, and as late as Oct. 26, Bush was up among likely voters in the Gallup poll by 52 percent to 39 percent. Again, the final vote was Gore 48.4 percent and Bush 47.9 percent.
Watch what I put in bold here.
And in 1976, the Gallup poll on Oct. 30 had President Gerald Ford at
47 percent and Jimmy Carter at 46 percent. Carter won 50.1 percent to 48.0 percent.
People put WAAAAY too much faith in polls as if they're all-knowing.
Check out the last line from article author Raymond J. Keating for extra marshmallowy goodness.
John Lucas
ffr
(22,670 posts)I already said you are right twice now. That's a compliment. Let me say that again. That's a compliment. And I'll say my title again. I don't take challenges personal.
However, the RCP compilation is a straw man and it proves the point that taken individually, the polling was mostly garbage and I'll bet there were a ton more taken in 2008 by less reputable pollsters that showed NC even farther from the mean than this neatly packaged list.
You win. I give up. I'm undeterred.
It's so difficult to reach the same people that polling did 20 - 30 years ago, the ones nowadays find people I don't believe accurately represent the electorate.
DemocratSinceBirth
(99,710 posts)But the Law Of Large Numbers suggests that by averaging all the polls you can come up with a rough approximation of the universe you are attempting to measure. And by rough I mean a reasonable degree of accuracy...Something much lower than five to ten percent which render a poll useless...
http://en.wikipedia.org/wiki/Law_of_large_numbers
We are discussing math. Nothing more...Nothing less...
And there might come a point where traditional polling is overwhelmed by response rate problems and polling is demonstrated to be inaccurate. We aren't at that point yet. Since you made that point at the beginning of our tet a tet it's incumbent upon you to prove it.
I'll wait for the proof.