Skip to comments.Poll Accuracy in the 2008 Presidential Election
Posted on 11/05/2008 9:55:06 PM PST by Para-Ord.45
On average, preelection polls from 23 public polling organizations projected a Democratic advantage of 7.52 percentage points on Election Day, which is only about 1.37 percentage points away from the current estimate of a 6.15-point Obama margin in the national popular vote.
1. Rasmussen (11/1-3)** 1. Pew (10/29-11/1)** 2. YouGov/Polimetrix (10/18-11/1) 3. Harris Interactive (10/20-27) 4. GWU (Lake/Tarrance) (11/2-3)* 5. Diageo/Hotline (10/31-11/2)* 5. ARG (10/25-27)* 6. CNN (10/30-11/1) 6. Ipsos/McClatchy (10/30-11/1) 7. DailyKos.com (D)/Research 2000 (11/1-3) 8. AP/Yahoo/KN (10/17-27) 9. Democracy Corps (D) (10/30-11/2) 10. FOX (11/1-2) 11. Economist/YouGov (10/25-27) 12. IBD/TIPP (11/1-3) 13. NBC/WSJ (11/1-2) 14. ABC/Post (10/30-11/2) 15. Marist College (11/3) 16. CBS (10/31-11/2) 17. Gallup (10/31-11/2) 18. Reuters/ C-SPAN/ Zogby (10/31-11/3) 19. CBS/Times (10/25-29) 20. Newsweek (10/22-23)
What were they predicting 1 month, 3 weeks, 2 weeks, 1 week out?
I think we learned that the polling in the Internet age is very sophisticated.
The deal is that it was pretty darn accurate.
And, if there was voting fraud....chase it, and enforce it. Kinda late now.........
OK, eggheads, what is the ACORN factor? +6, +8, +18?
Nice comeback for Rasmussen after some busts in the primaries and 2004 miss.
No surprise to see Mr. Special Sauce, John Zogby on the bottom along with the always laughable Newsweek.
It`s the ACORN factor PLUS Democrat SoS Brunner in Ohio !
I’d say it is the McCain factor.
“And, if there was voting fraud....chase it, and enforce it. Kinda late now......... “
There is another election in 739 days. WE MUST ELIMINATE THE VOTING FRAUD OR DEMOCRATS WILL RUN THE TABLE!!!
oh yeah, forgot the McCain suppress the base factor
“What were they predicting 1 month, 3 weeks, 2 weeks, 1 week out?”
Very similar amounts when you average the polls together. A lot of folks on here were very quick to somehow spin an Obama+6 average into something that meant little and part some giant conspiracy. Many posters even predicted McCain landslides up until election day. If you questioned the idea of him getting well over 300 they’d attack you.
Some of the mental gymnastics were impressive. Some posters could take an Obama +6 poll and use fancy math to make it look like McCain was up. “If we change the percentage of young voters to this, change party ID numbers to this, change how we weight blacks, etc, etc we have McCain by 19 in California!, etc” There was a lot of wishful thinking and funny math going on.
Nate Silver at fivethirtyeight.com was pretty accurate, even under estimated the result.
no words needed
aside from the outliers at 12 and 15 points
the RCP averages were pretty close
and that crazy 20% they kept telling us where the internals showed 15-20% of respondents calling themselves conservatives and still supporting Obama
i thought that was nuts just like everyone
well, shucks, that is exactly how it showed in the exit polls, apparently 15-20% of those polled really did think they were conservatibe and did support and vote for Obama
the polls from 2000 on have been unsettling fairly close even though we all hoped they weren’t
I admitted worrying about it and got some big doses of freeper love for my efforts
people were so emotional they would claim anything and deny any precedence staring them in the face..
but don’t worry, next election the same folks will be back to tell us all how the polls are all lies again
FR at election time is a zoo of near psychotic delusion...all meant well and for a good cause of course
“but dont worry, next election the same folks will be back to tell us all how the polls are all lies again”
What I never understood was the question of motive. Why would the pollsters be involved in some grand conspiracy? Even if they are lib (and with some of them we have no way of knowing) they have to strive for the mark. Miss it by too much too often and you go out of business and somebody else takes your place. Its how the market works. If the pollsters had been predicting Obama +7 and he lost the industry would implode and they’d all go under. Even if they were rooting for Obama I can assure you that they don’t want to put themselves out of business.
I’m pretty sure that in my eight years here that I had more invectives hurlked my way hourly by some freepers when I dared question the notions you just described.
don’t get me wrong, I thought the polls were probably off a bit too...maybe 3-5 points given the high Dem sampling but I never thought they were irrelevant or a grand conspiracy
and it turns out the Dem sampling was for a reason on likely voter polls, the friggin Dems did turnout much more in comparison to low GOP turnout
and folks would take one of the two polls in 2004 out of 17 that actually showed Kerry leading at election day and tout that as proof that “in 2004 they were wrong to...see? see?”
I was about ready to ask who took the damned strawberries around here..lol
but it’s all over now....and the nasty gnats will go back to being just a little snarly instead of nuts
i had one girl call me pissy when i said I was worried about the election....I have never been called pissy..lol...not in 50 years...it made me feel gay...lmao
in all seriousness though I saw some newbies come on and speak with trepedation about the election and get scalded so bad I bet they won’t ever be back
one hapless freeper claimed Mccain was in trouble about 7PM CST election nite and got slammed so hard you woulda thought he said Magnus really did dye his hair after all
yep....FR is pretty touchy feely around elections when it looks bad for us and the customary denial sets in
This thread is perfect evidence of why... because they are held accountable for only one poll - the last one.
So even though Pew had Obama +16 polls for the final weeks, they are labeled “acurrate” because they issued a +6 poll at the last minute. Same with CBS and many others.
The lesson is to watch Rasmussen. These other polls are all over the place, and I think a lot of it is for manipulation purposes. Rasmussen polling is very much steady as she shows, which is more reflective of reality than Pew/Zogby/Gallup/CBS polling with 10 point shifts in just days.
“and folks would take one of the two polls in 2004 out of 17 that actually showed Kerry leading at election day and tout that as proof that in 2004 they were wrong to...see? see?”
The cherry picking was impressive. We also saw that with the one day Zogby sample from a three day poll. Never mind we’ve lost a hundred and fifty or so straight and most by 5+. Those don’t matter. This subset of a poll from one of the less accurate guys proves we must be winning!
Its started in fact already after this one. I saw some people using the fact that a couple polls were bad outliers as proof that the polls don’t matter. They ignored the fact that the averages were spot on. It was pretty mind-boggling to see it already starting a day after the disaster.
I was told the same in 2008 but was skeptical of my fellow Freepers.
In 2010 I'll believe the polls!
“This thread is perfect evidence of why... because they are held accountable for only one poll - the last one.”
You are cherrypicking though. I don’t pay attention to one poll. I pay attention to the average. The industry average stayed pretty constant. There was no gigantic move at the last second.
Look at the RCP averages. No big swing period.
I pay attention to polls, not averages.
A better method than averaging may be a median so outliers on both ends get tossed.
I hope no one listens to Zogby/Newsweek/CBS again after this election.