Posted on 09/14/2012 9:03:41 PM PDT by Leto
Polls have caused a lot of consternation among my friends here. Some say to ignore polls, but this election season I have found them to be very consistent even thought the results vary quite a bit.
The key is understanding the turnout model the poll is using. In 2008 the turnout was +7.6% dem, in 2010 it was +1.3% R. In the last Rasmussen survey 8/2012 the party affiliation is +4.3% R.
Most polls are using turnout models that vary from +6 to +10% D. This is why Obama is leading in most polls.
The latest NYT/CBS poll has Obama +3 among likely voters 49-46%. Their turnout model is +6% D close to the 2008 turnout model. 35D/39R/32I for their likely voters in this poll.
I have a spreadsheet I use to see the results if the 2010 turnout model is used or any other model.
Below is the CBS NYT poll 'normalized' to reflect both the 2010 turnout model and the current Rasmussen party id.
Two points:
1) Although I don’t expect the 2012 turnout proportions to be as good for Democrats as it was in 2008, I don’t expect it to be as bad for Democrats as it was in 2010: Obama wasn’t running in 2010, and there are still plenty of kool-aid-drinkers who worship the ground he walks on. They’ll vote in 2012 even if they couldn’t drag their worthless butts out of bed in 2010 to vote for some lame-o donkey Senator or Representative.
2) The MSM polls will have the effect of energizing Democrats (another chance to dance with the winner) and discourage Republicans (can’t beat this dirtball no matter how bad he is). That further moves the turnout ratios in the wrong direction away from the 2010 numbers - unless we keep from falling for their machinations, and do all we can do keep our acquaintances from getting taken in as well.
CNN/ORC poll
50.4 percent Democrats and
45.4 percent Republicans
4.2 percent independents.
Check out the independents. Independents under-sampled and Democrats are over-sampled 12.1 percent.
“Unskewing this data to make up for the likely 25 percent under-sampling of independent voters shows the results are quite different. With the weightings of independents, Democrats and Republicans conducted by the Rasmussen numbers, this poll’s data would indicate a Romney lead over Obama of 53 percent to 45 percent. That is almost exactly the reverse of the 52 percent to 46 percent lead it reports in favor of Obama. The sampling skew of this poll actually reverses the result that should be shown by the data.”
http://www.examiner.com/article/mitt-romney-would-lead-eight-unskewed-data-from-newest-cnn-orc-poll
WSJ? Nah,this is a very simple spreadsheet, the pollsters understand the game, my goal here is for my friends here to understand the process and assumptions made when polls are put out.
They are guesses based on assumptions. If your assumptions are correct the results will be accurate, if not........
The MSM knows the enthusiasm for Obozo isn’t what it was, their game is to discourage conservatives and encourage the dem base by creating the impression that the election is Obozo’s to lose.
This of course is pure crap.
I want my fellow freezers to understand this.
Rasmussen says Republican affiliation is at a historic high.
Also, trying to discourage us did not work in 2010 or with the Scott Walker recall and I seriously doubt that anger has subsided. I guess we’ll have to wait and see.
My hope is that it will backfire— Republicans will realize they MUST get out and vote. Kind of like when I was reading, off subject but interesting, the other day where the show “Family Ties” was written by liberals who wanted to have the hippie left parents be the driving force behind the show and have the conservative kid shown as an idiot these parents were trying to raise. But it completely backfired as Alex P Keaton became the reason for the shows success and the public identified with him.
“but this election season I have found them to be very consistent even thought the results vary quite a bit.”
Arrrrrrrrrrrrrggggggggggghhhhhhhhhh.....
The Ras party ID numbers and the voter enthusiasm numbers indicate that this election could be more anti dem than 2010, we shall see.
Of course the events of this week in the ME is a wild card that could affect thing in an unpredictable manner,
Looking at the comments on the Examiner and saw overwhelmingly positive response for Romney, even from moderate Dems. Good sign!
Rasmussen threw out a big hint this morning in response to allegations about his poll being so far off from the others — that his was the only poll to include data on days after the embassy.
I think the embassy thing pissed off a LOT of conservatives.
Paul Ryan was so fiery today. Too bad he isn’t on the top of the ticket. Romney needs to man up.
v.g.
Although the subsamples are small, by combining the polls you get a reasonably sized tracking poll of independents. Romney’s winning with them. To off-set this, the media polls over-sample Democrats.
Regarding the argument that you can’t compare 2010 to 2008, because one is a midterm election and the other a presidential, compare 2010 to 2006. A swing of 6 points toward the Republicans. Therefore, a swing of 6 points from 2008 to today is not unreasonable. (Rasmussen thinks the swing will be larger.)
By their own numbers:
Total Respondents 1,301 Non-Weighted/ Weighted
Total Republicans 343 / 285 (22%)
Total Democrats 451 / 461 (35%)
Total Independents 507 / 555 (43%)
Total Registered Voters 1,170 1,083
Total Likely Voters 1,162* 999
Republican likely voters 289 (29%)
Democratic likely voters 351 (35%)
Independent likely voters 359 (36%)
*Not all likely voters are assigned the same probability of voting.
I see a D + 13 overpoll
What do these folks think we’re smoking??
No . . 35, 29, 36 . . . watch your typing! OK, OK, it’s late and we’re all tired.
Your points are well-taken - and I should have added in my original reply, a really nice job with the original numbers.
Hopefully, the ever-developing fiascos of this regime (although consistently covered for by the MSM) will continue to erode BO’s chances, but short of that, it looks like this election comes down to four (maybe five if it’s really close) things:
1) base turnout (can we get the GOP/donkey ratio close to the 2010 numbers?)
2) the extent of a preference for Romney among the independents (how much should R&R suck up to the squishy moderates)
?
3) vote fraud (the donkeys always play with a deck with more than four aces, the extras being up their sleeves)
4) the nature of this year’s “October surprise” (I expect Obama has all of Romney’s tax returns, and just like Jack Ryan’s divorce records, they’ll magically appear late next month).
and then, if it’s really, really close:
5) how many votes will Virgil Goode pull in Virginia?
You know, I think it is even worse, because they not only routinely oversample Democrats, but they also oversample women. THAT is more subtle, but I have seen it done.
LOL CNN. Dems haven’t made 50% in party affiliation since probably 1964.
Since President is in reality a group of 51 elections, there should be a different turnout model applied to each state, especially the battlegrounds. And it could prove difficult to decide on predictive models because each state’s voter turnout can be heavily influenced by down-ballot items, like propositions.
Yes it’s ridiculous. Since they know Romney is winning independents by 10 to 14 points they under-sampled them more than 25% to get the number they wanted.
Thanks you Leto. If true this will be as big as the Reagan win over Carter.
Something else reminds me of that election but I can’t quite put my finger on it.
Well done.
See my tagline.
I don’t think anyone but you has considered the 2010 results and that the anger towards leftists has grown.
Ping.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.