RCP Poll - Yikes!

[SFC Ollie]

OK...Your guess....Which shows itself to be anything but educated.

Glad we cleared that up.

Can you offer a more plausible explanation as to why Rasmussen intentionally skews its presidential approval poll to the right?

I'm sorry, how does using likely voter skew to the right again? I can understand how it can make the prediction of an election more accurate but not how it can lean left or right, Unless you are admitting that the left is more likely not to vote?Then I would want to know , WHY?

 

[Sinatra]

And of course the thread is designated to the RCP average - Obama's polling numbers have declined steadily, and are now below 50% for the RCP average.

The simple fact is, this president is in some serious trouble...

 

[Foxfyre]

Rasmussen didn't do as well when they first started out but have been revising and improving their methodology shooting for accuracy, not ideology. They were pretty bad prior to 2004, began getting better, and according to a Fordham University study were #1 in accuracy by the time they polled ahead of the 2008 election:

The following list ranks the 23 organizations by the accuracy of their final, national preelection
polls (as reported on pollster.com).
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)
http://www.fordham.edu/images/acade...ccuracy in the 2008 presidential election.pdf

If you poll only neighborhoods or area codes with largely blue county/state demographics or call only homes during the day where you'll find non-working folks, many dependent on the government, you likely get a much different result than doing it in the evening when more folks are likely to be home and polling likely voters only.

It was interesting to me that CNN somewhat beat Fox in polling accuracy, and the ultra leftwing Daily Kos beat ALL the mainstream media outlets except for CNN. But Rasmussen and Pew did their homework and they got it the most right.

And if you put your money on anybody to be most correct in analyzing what the American voter is thinking, I would put it on them now too.

 

[saveliberty]

And? The numbers used by the guy who wrote that are incorrect. It says in the document that he's basing his number off of an Obama win of 6.15 percent. The real margin was 7.26 percent.

Still having trouble with that margin of error concept I see. Accuracy is not only having the right numbers, but stating a degree of reliability also.

Does anyone besides you make that claim?

Pretty much all the statisticans and pollsters do, yes. It is like the bell curve, where you are showing the likelihood your answer could be off. I think it is two standard deviations out and then what is left over is your percentage off you can be. For the poll in question it was plus or minus 3% for Rasmussen.

Rasmussen tends to have lower deviations because they make everyone an agree or disagree, so it adds up to 100%. Then there sample size is a little bigger in most cases. Finally, there's the likely voters group themselves. They tend to be a little better informed and to have an opinion.

 

[Dr.House]

because I have you to do that. And, btw, good job!!

So you admit they're outliers, yet have no problem with including them?

Good job, indeed...

If you simply use the average then that problem is solved. It's the people singling out Rasmussen because it suits their agenda, when their approval polls are intentionally skewed, that are the problem.

Yet more "outliers" skewed toward the boi king kind of helps his cause, wouldn't it?...

I posted the data minus the "outliers", which puts him at +1.75... Amazingly unimpressive, wouldn't you say?...

 

[Polk]

I don't know that it "tilts to the right".

I merely recognize that they're the only polling agency to screen for likely voters. Which, judging by their track record, appears to add to the accuracy of their polling.

If you want to dream up paranoid fantasies as to their motivations for doing so, that's your strait jacket.

They're not the only polling firm that screens for likely voters.

 

[Polk]

OK...Your guess....Which shows itself to be anything but educated.

Glad we cleared that up.

Can you offer a more plausible explanation as to why Rasmussen intentionally skews its presidential approval poll to the right?

I'm sorry, how does using likely voter skew to the right again? I can understand how it can make the prediction of an election more accurate but not how it can lean left or right, Unless you are admitting that the left is more likely not to vote?Then I would want to know , WHY?

People who would be considered likely voters under most likely voter screens are people who have voted in several consecutive elections. Those people in general tend to disproportionately older, wealthy, and more white than the general population, which in turn happens to be a more conservative demographic.

 

[Polk]

It was interesting to me that CNN somewhat beat Fox in polling accuracy, and the ultra leftwing Daily Kos beat ALL the mainstream media outlets except for CNN. But Rasmussen and Pew did their homework and they got it the most right.

Because pollsters in general are looking to get the most accurate result possible.

 

[NYcarbineer]

Still having trouble with that margin of error concept I see. Accuracy is not only having the right numbers, but stating a degree of reliability also.

Does anyone besides you make that claim?

Pretty much all the statisticans and pollsters do, yes.

Good, then you should be able to easily cite me several examples of that, and perhaps even one that cites the 2008 election.

Why, btw, do you simply ignore the fact that the Fordham study, which was done the day after the election, uses the wrong margin of victory?

 

[NYcarbineer]

I don't know that it "tilts to the right".

I merely recognize that they're the only polling agency to screen for likely voters. Which, judging by their track record, appears to add to the accuracy of their polling.

If you want to dream up paranoid fantasies as to their motivations for doing so, that's your strait jacket.

Likely voters is more accurate for an approval poll only if the approval poll is measuring that specific subset. The right conveniently, and arbitrarily, takes the stand that using that subset is better because that subset gives them results they like better.

 

[NYcarbineer]

[People who would be considered likely voters under most likely voter screens are people who have voted in several consecutive elections. Those people in general tend to disproportionately older, wealthy, and more white than the general population, which in turn happens to be a more conservative demographic.

And Rasmussen knows that.

 

[NYcarbineer]

So you admit they're outliers, yet have no problem with including them?

Good job, indeed...

If you simply use the average then that problem is solved. It's the people singling out Rasmussen because it suits their agenda, when their approval polls are intentionally skewed, that are the problem.

Yet more "outliers" skewed toward the boi king kind of helps his cause, wouldn't it?...

I posted the data minus the "outliers", which puts him at +1.75... Amazingly unimpressive, wouldn't you say?...

I tossed out the high and the low and got roughly the same number as when they were in. What did you do, arbitrarily set your 'outlier' at 7 so you could toss out CNN at 13 but not Foxnews at zero?

good one

 

[Dr.House]

If you simply use the average then that problem is solved. It's the people singling out Rasmussen because it suits their agenda, when their approval polls are intentionally skewed, that are the problem.

Yet more "outliers" skewed toward the boi king kind of helps his cause, wouldn't it?...

I posted the data minus the "outliers", which puts him at +1.75... Amazingly unimpressive, wouldn't you say?...

I tossed out the high and the low and got roughly the same number as when they were in. What did you do, arbitrarily set your 'outlier' at 7 so you could toss out CNN at 13 but not Foxnews at zero?

good one

I tossed the "outliers" in the spread column, dipshit...

I thought even you, with your diminished brain capacity, could have figured that out... I guess I gave you too much credit...

 

[Polk]

[People who would be considered likely voters under most likely voter screens are people who have voted in several consecutive elections. Those people in general tend to disproportionately older, wealthy, and more white than the general population, which in turn happens to be a more conservative demographic.

And Rasmussen knows that.

He does, but I wouldn't be so quick to label it an issue of bias. Remember, he has to be able to sell polling to clients. That's only going to be effective to the extent his polling is accurate.

 

[SFC Ollie]

You have all been shown that Rasmussen is / has been the most accurate. Yet you still insist that they are not and that they are biased. Must be because he occasionally appears on FOX news. Are we that Biased ourselves?

 

[NYcarbineer]

You have all been shown that Rasmussen is / has been the most accurate. Yet you still insist that they are not and that they are biased. Must be because he occasionally appears on FOX news. Are we that Biased ourselves?

No, the accuracy claim for Rasmussen is in error. And even if it were true it means nothing.

 

[NYcarbineer]

Yet more "outliers" skewed toward the boi king kind of helps his cause, wouldn't it?...

I posted the data minus the "outliers", which puts him at +1.75... Amazingly unimpressive, wouldn't you say?...

I tossed out the high and the low and got roughly the same number as when they were in. What did you do, arbitrarily set your 'outlier' at 7 so you could toss out CNN at 13 but not Foxnews at zero?

good one

I tossed the "outliers" in the spread column, dipshit...

I thought even you, with your diminished brain capacity, could have figured that out... I guess I gave you too much credit...

From realclear? If so, you tossed out 1 on the low side and 2 on the high side. Why didn't you toss out 2 on each side?

 

[PoliticalChic]

Rasmussen is STILL an outlier. 13 points outside the average.

49.9-46=13 And liberals wonder why we think they're dipshits.

No, we wonder why homeschooling doesn't focus more on math.

Here, here, point of order:

In a study of over 5,000 home-schooled students, they outperformed public school students by 30 to 37 percentile point in all subjects (from “Strengths of Their Own-Home Schoolers Across America,” Dr. Brian D. Ray)


Recent statistics from The College Board and the American College Testing Program (ACT) indicate that home schoolers are exceeding the national average test scores on both the SAT and the ACT college entrance exams. In 1999, the 2219 students who identified themselves as home schooled students on the SAT test, scored an average of 1083 (verbal 548, math 535), 67 points above the national average of 1016. A perfect SAT score is 1600. Also in 1999, 3616 home school students taking the ACT scored an average of 22.7, compared to the national average of 21, a perfect score being 36.
College-bound Home Schoolers Make Headlines (HSLDA | National Center News)


And what school is it that is proud to point to you as a grad?

 

[Polk]

You have all been shown that Rasmussen is / has been the most accurate. Yet you still insist that they are not and that they are biased. Must be because he occasionally appears on FOX news. Are we that Biased ourselves?

A claim that was made based on the results on one race, and even in that race, it's been shown the researcher who stated Rasmussen was the most accurate used inaccurate numbers. It wasn't due to malice, he just jumped the gun on putting out his results.

 

[SFC Ollie]

You have all been shown that Rasmussen is / has been the most accurate. Yet you still insist that they are not and that they are biased. Must be because he occasionally appears on FOX news. Are we that Biased ourselves?

A claim that was made based on the results on one race, and even in that race, it's been shown the researcher who stated Rasmussen was the most accurate used inaccurate numbers. It wasn't due to malice, he just jumped the gun on putting out his results.

http://www.fordham.edu/images/acade...ccuracy in the 2008 presidential election.pdf

And what made Rasmussen not the most accurate here?