Another gentle reminder about polls....

[Playinghardball]

TheObamaDiary.com @TheObamaDiary
Follow
Another gentle reminder about polls

Gallup's 2012 forecast

Think about who funds these people

Then smile

And vote

6:27 PM - 1 Nov 2014

[FSogol]

1. Here's another: Eric Cantor.

[rock]

2. Yep, generally polls these days are only good

after the election, to show you how biased (and wrong) the pollsters were!

[MineralMan]

3. There is only one poll that is guaranteed to be accurate.

That's the one with the results that come in late Tuesday night or Wednesday morning next week. That poll will be accurate.

We all get to participate in it, too. That's a good thing.

GOTV 2014 and Beyond! Make that poll count.

[Peacetrain]

6. Bingo!!

[HERVEPA]

4. Still irrelevant.

This was an outlier and no one paying any attention believed it.
It's the aggregation of polls taht matters.

[tritsofme]

5. Final Gallup poll had the race Romney 50% to Obama 49%, not these numbers.

http://www.gallup.com/poll/158519/romney-obama-gallup-final-election-survey.aspx

It was well within the margin of error, like most of the close races now, which I think is a better illustration of the point you are trying to make.

[DemocratSinceBirth]

12. Better but they have become a on the whole a crap poll.

That's why aggregation is where it's at...

[rocktivity]

7. And between cell phones and social media, they're less credible than ever.

"Get thee beyond me, polls! Unclean! UNCLEAN!!!"


rocktivity

[former9thward]

10. They are very credible this close to an election.

No cell phones are an urban legend.

Gallup includes cellphones in each national Gallup poll. Gallup has been including cellphone-only households in all national telephone Gallup polls since January 2008. Further, cellphone-only households are now as likely to fall into national Gallup polls samples as those living in traditional landline households.

Gallup and the survey research industry as a whole are exceedingly aware that increasing numbers of Americans use cellphones in addition to traditional landline phones and that nearly 40% of households have only cellphones. Gallup's methodologists and survey scientists have studied, researched, conducted test polling; analyzed existing data; and examined the implications of the inclusion or exclusion of cellphones and determined that it is necessary to include cellphone numbers in survey samples despite the increased costs of doing so to ensure that the obtained samples continue to be representative of all Americans. This is particularly important given recent studies that show that people in cellphone-only households tend to be younger, are more likely to be racial and ethnic minorities, and more likely to be transient.

http://www.gallup.com/poll/110383/does-gallup-call-cell-phones.aspx

[davidn3600]

8. The difference is Democratic insiders were very confident in 2012 of an Obama victory

Those same people seem very uneasy this time around which means the internals must not be very good.

[jeff47]

9. And Republican insiders were very confident in 2012 of a Romney victory.

The guy didn't even write a concession speech until he lost Ohio.

[davidn3600]

15. I never said GOP insiders were smart

Im talking about our attitude, not theirs.

[jeff47]

17. Ours aren't smart.

If they were, they'd be pointing out that Obamacare's working, saving lots of lives and most people save money.

They'd be pointing out the deficit is tiny compared to when Obama started (that's supposed to be a good thing, according to morons...I mean pundits)

They'd be pointing out unemployment is way down and the stock market is way up.

They'd be pointing out that St. Reagan had a lower approval rating than Obama.

They'd wouldn't be desperately trying to get people to vote for Democrats who would rip their own arm off before pulling the lever for a Democrat. Instead, they'd be looking at Warren's massive crowds and say "hmm...maybe there's a different way we could go than what the pundits say we should do..."

But they aren't. Instead they're trying to convince people that Kay Hagan and Mark Udall aren't really Democrats.

So I think we shouldn't assume ours are smart.

[DemocratSinceBirth]

11. It seems those that relied on the law of large numbers ...

It seems those that relied on the law of large numbers did better than those who relied on their favorite poll... I don't believe one objective forecast model predicted a Romney win.




There are more models but I haven't looked at them in a while.


http://fivethirtyeight.blogs.nytimes.com/fivethirtyeights-2012-forecast/?_r=0

http://election.princeton.edu/2012/11/06/presidential-prediction-2012-final/

http://votamatic.org/evaluating-the-forecasting-model/

[DesMoinesDem]

13. A single poll from 6 months before an election is supposed to be a reminder of what?

That TheObamaDiary posts worthless bullshit?

[demwing]

14. I think a better example is the 2010 Nevada Senate Race

Last RCP average: ANGLE up by 2.7
Gen Elect Total - REID up by 5.6!



that a bias for Angle of over 8 points!

[Arugula Latte]

16. Are pollsters still basically talking to old people with landlines?

Or have they finally changed their ways?

[HuckleB]

18. No one believed that poll, and for good reason.

So, what's your point?