| UnSkewed Choices QStar News Quick Poll UnSkewed Average -- Romney vs. Obama UnSkewed Average -- Presidential Approval
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Posted on 09/19/2012 6:54:14 PM PDT by SeekAndFind
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
I’ve already explained stratified random sampling to you on another thread. In this case they are assuming that the August 2012 party affiliation survey by Rasmussen was representative and are doing stratified random sampling based on that weighting of party affiliations, rather than party affiliations from 2008 exit polling data.
prayers that this is what happens!
Thank you for that brief interlude for sanity...and now back to our regularly scheduled DNC propaganda.
You are clearly not a numbers guy, you never have illustrated any ability to crunch, analyze or explain data or numbers of any kind.
You can safely be ignored if the subject involves math.
Flaws of the top of my head:
1. I don't see anything concerning registered voter v. likely voter. This makes a huge difference in poll results.
2. Sample size. What is it? That makes a huge difference.
3. Method. What process did they use to determine who they picked and where they resided? How did they go about making these random? Cities v. suburbs? Northern or Southern? A southern democrat can be more conservative than a northern republican. These things all have a major effect on polling results.
Devil’s Advocate Moment:
But how do we KNOW that there aren’t lots of Dems? Face it, Obama milks the popular culture like a serious pro. Due to the TV and movies megaphone, it is widely considered incredibly cool to be a Democrat.
If you ask "are you politically liberal or conservative?", you will not get the same answer. Some folks are personally conservative, but feel like being "liberal" in politics is a way to be "big-hearted" on the cheap.
I think the real tell is the Abyss rating: Strongly Disapprove - (Strongly Approve + Somewhat Approve). IOW, all the approvers can’t overcome the those most fed up with this putz.
Strongly Disapprove 48.78%
Somewhat Disapprove 6.29%
Somewhat Approve 19.67%
Strongly Approve 25.64%
48.78 - (19.67 + 25.64)
48.78 - 45.31
Abyss = 3.57!
The trick, of course, is to get the ABO turnout to reflect this...
Boom! I think you've nailed it.
I do not trust these/this poll(s) :)
thanks!! blind pig meets acorn
The proposal here is not stratified random sampling. They are taking existing polls and simply REPLACING the results with numbers they prefer!
We all encountered this way back in 5th grade when our math teacher ~ the special one brought in from the university ~ referred to "the Egyptian variable" ~ which is the number you multiply your answer with to get the correct answer, or vice versa!
Weighting samples for better fit to the population is also a perfectly reasonable practice if one knows the demographic composition of the population, and knows that the survey or other measurement is drastically out of line.
Here, of course, the real population doesn’t exist yet — those who vote on Nov. 6 — and one can debate whether using the party affiliation rates from the August 2012 Rasmussen party affiliation survey is a good weighting, but they’re not just jiggering the numbers to get a desired result: they are reweighting the samples from recent polls that provide enough internals to fit the Rasmussen party affiliation data.
It appears to me that the goal of every poll is to show a statistical tie of around 46-47%,
so they take their poll and adjust the sample in order to get these results every time.
Their enthusiasm is way down since 2008, but...
even though they stay home, they might “vote” anyway.
After all, they’re democrats.
Within a few years of the founding of the United States under the 1790 Constitution, Thomas Jefferson and his associates created the Democratic Republican party ~ the purpose of which was to WIN ELECTIONS.
Of necessity a political party is a coalition of similarly minded people ~ but not identically minded. They feel they have interests in common with the others, although those need not be the same interests.
However, to win in a single member district you usually need 50%+1 vote. To successfully challenge a winning party you will need to sop up all the other "interests" in the same district so that you have some chance at getting that 50%+1 yourself!
A third party has little chance at overcoming the organizational inertia created by the two earlier parties so most of the time a third party will need to infiltrate an existing party and take over the top end structure.
Obviously not everybody agrees about everything all of the time, else these parties would tend to grind down to a 50/50 situation every election. On the other hand they do come close to 47% to 47% every single time.
Now here's the dilemma for the politicians. If they believe that you can identify and take the political pulse of every single individual with polling that 47/47 split looks like it has 6% available for competition.
Actually that 6% doesn't exist. It's an artifact of the sampling process itself ~ either you catch people at the wrong time, or they don't want to talk, or your questions were defective, or you didn't understand what they said ~ we have some FReepers here who've said as much ~ they get polled and go out of their nuts pushing back at the pollster ~ are they counted as part of the 6%? Or are they simply deleted from the log pages? Only the pollsters know.
In reality the additional 3.1% of the vote the politician needs is already part of his party ~ just unreachable by the polls at the sensitivity available at the price anyone is willing to pay.
This is where the hard part comes in ~ everybody doesn't vote! So, you've got 47% ~ but where is that 47%. If you figure 140 million people are going to vote, that's just under 70 million votes, and your opponent has that amount as well.
However, there are another 60 million VOTERS who usually don't turn out for these elections. If you can get an additional 10% of your own kind of voters out of that nonvoting body to turn out for you, that's an extra 6 million! You win. The other guy loses. If he can get 11% to turn out you lose.
If you can convince some of his regular voters to not bother showing up, you can edge him out. Same on his side.
If you do both ~ like Obama did in 2008 ~ get regular nonvoters to show up to vote for you, and at the same time suppress the other side, you may beat them by 10 million votes.
This phenomenon ~ to wit, that many of his own party followers don't bother to vote most of the time, is critical to electoral victory. If you imagine that there's a 6% plum of undecided out there and it's your job to battle for them with ideas and wit, you'll probably lose unless your opponent does the same thing.
On the other hand, if you recognize that your winning combination lies in the hands of your own party followers who don't usually vote, then all you have to do is get them out to vote and you will almost always win (if the other guy's get out the vote efforts are weak).
This is one of the reasons you don't want to anger your other party coalition members ~ they might not help you get out the vote! When that happens you lose.
There's very little ideology in this ~ just coalitions of related interests as perceived by similarly minded individuals ~ all of whom are locked into a single member district where there can be only one winner.
Polls that show wide departures from a fairly even split are reflecting one of two things ~ recent reapportionment after a census, or an improperly constructed poll.
Did you ever stop to think that these pollsters have a really serious public confidence problem? Here we have FReepers imagining that these guys have stratified samples and all they really have is a boiler room operation that calls random numbers around the country until they get about a few hundred respondents ~ with a minimum of more than 100 who'll say they are Republicans, another minimum who'll say they are Democrats, and maybe about 60 who'll swear that they are independents.
That gives you a sort of statistically valid floor but just enough so you have fewer accidents in your selection of numbers to call.
Unless someone pays them an awful lot of money they don't do much more than that on any of these polls. And, as you know, the more characteristics you test for, the more sampling it takes to get statistically valid results for every sample cell in your matrix.
I'd like to know ALL the characteristics, not just the R/D/I business. How about the non responses? If you don't have those in your data base you don't even know if either candidate has a popular following!
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.