Rasmussen Polls Were Biased And Inaccurate

[IBMer]

Still pretty funny that you honestly thought Democrats would hold the House. Much funnier than my being wrong on one senate race. Did you honestly believe that, or were you just lying in an attempt to save face?

Wrong on one senate race?

Lets see em

Senate - R +10
House - R +70

 

[nick1985]

Seriously? You think you "got me" because I started a simple prediction thread an offered my prediction?

I was off on thee Senate and pretty close on the House. I wasnt saying any of that as fact, just a guess. Hence, the "prediction" thread.

http://forums.anandtech.com/showthread.php?t=2116675&highlight=predictions


I did a little better than your fellow Democrat cheerleader:

Republican gains..

Senate 5
House 37

 

[IBMer]

Thanks for calling me a democrat cheerleader, obviously you don't actually read any posts.

 

[nick1985]

Thanks for calling me a democrat cheerleader, obviously you don't actually read any posts.

You have 300 posts in the last decade, there are not many posts to read and from the ones I have seen you stump pretty hard for the D's.

 

[FerrelGeek]

Just like you kept us posted and told us that health care reform was a slam dunk and would easily pass in the summer/fall of 2009 and there was nothing the Republicans could do about it? How did THAT work out for you? Sure it passed -- in 2010, after it had been COMPLETELY gutted.

And the Dem thugs had to bully and bribe enough people to get it passed.

 

[OrByte]

NExt time someone sources Rasmussen polls it should be clear that Rasmussen has a reliability issue

 

[nick1985]

NExt time someone sources Rasmussen polls it should be clear that Rasmussen has a reliability issue

I'm still trying to figure out, other than the Angle race (which most pollsters had wrong) and the 1 Hawaii poll that was off, where were there other discrepencies?

Can you help with that one?

 

[IBMer]

I'm still trying to figure out, other than the Angle race (which most pollsters had wrong) and the 1 Hawaii poll that was off, where were there other discrepencies?

Can you help with that one?

Moreover, Rasmussen’s polls were quite biased, overestimating the standing of the Republican candidate by almost 4 points on average. In just 12 cases, Rasmussen’s polls underestimated the standing of the Democrat by 3 or more points. But it did so for the Republican candidate in 55 cases — that is, in more than half of the polls that it issued.

 

[nick1985]

Moreover, Rasmussen’s polls were quite biased, overestimating the standing of the Republican candidate by almost 4 points on average. In just 12 cases, Rasmussen’s polls underestimated the standing of the Democrat by 3 or more points. But it did so for the Republican candidate in 55 cases — that is, in more than half of the polls that it issued.

I was asking for specifc races that they had off. Based on what I saw on election night, they were pretty accurate other than the Nevada race.

Are you going to keep throwing out numbers from the NYtimes, or provide actual races where they were so off?

 

[Thump553]

nick1985 do you have any evidence at all that the conclusions of that study are wrong? Iif you want to dispute the mathematical conclusions the best thing to do is to dig up the raw data yourself and prove it. You act as if because the study is published in the NY Times that it is automatically untrue. Since we are talking basic, irrefutable mathematic calculations it should not be hard to disprove them if they published false conclusions.

If you read the study you will note that they deliberately did not rely just upon the final pre-election polls ad states the specific reason why they did not. So it is quite possible that by election night Rasmussen't projections were more in line with reality than previously.

One interesting point to me was that every polling entity studied but one displayed a Republic bias in thier polls. Seems a bit contrary to much vaunted mainstream media liberal bias theory so often propounded.

 

[nick1985]

I gave my evidence for how they were disengenious, especially with the Hawaii race. They made a claim that many polls were off by 10 points or more, yet did not provide any races for that to be the case other than the 1 Hawaii poll. I am just curious to know which those were, because besides Nevada(which wasn't even close to 10 off) they pretty much got it all within a few points.

 

[Infohawk]

Best case scenario for jpeyton: he was right about Rasmussen and wrong about much more important things. Still an overall loss for jpeyton.

 

[IBMer]

You checked all 105 surveys they were talking about to know that they weren't off on 13 of them by 10? Also remember this is Rasmussen and Pulse Opinion Research.

Also understand their number is a difference spread. Meaning if what was predicted was 4 in favor of Republicans and it was 6 in favor of Democrats, this is a 10 point difference.

 

[JeffreyLebowski]

NExt time someone sources Rasmussen polls it should be clear that Rasmussen has a reliability issue

Pretty much all the polls and sources now are biased one way or another. It's just a matter of how much work certain posters here will do to stick their fingers in their ears and yell, YOU'RE WRONG LALALALALALA!!!!

 

[her209]

Stephen Colbert explains polls
http://www.colbertnation.com/the-co...-11-2010/the-colbert-repoll---scott-rasmussen

 

[werepossum]

nick1985 do you have any evidence at all that the conclusions of that study are wrong? Iif you want to dispute the mathematical conclusions the best thing to do is to dig up the raw data yourself and prove it. You act as if because the study is published in the NY Times that it is automatically untrue. Since we are talking basic, irrefutable mathematic calculations it should not be hard to disprove them if they published false conclusions.

If you read the study you will note that they deliberately did not rely just upon the final pre-election polls ad states the specific reason why they did not. So it is quite possible that by election night Rasmussen't projections were more in line with reality than previously.

One interesting point to me was that every polling entity studied but one displayed a Republic bias in thier polls. Seems a bit contrary to much vaunted mainstream media liberal bias theory so often propounded.

In fairness to the pollsters, it's pretty difficult to poll the effect of having the SEIU in charge of the voting machines.

Seriously, Rasmussen blew it big time. In previous years they have been the most accurate, but no one pollster will always be the best.