-
Dismantling Networks, Preventing Footholds
May 23, 2013 | 2 comments
-
I Agree With Andy McCarthy: No Special Counsel
May 23, 2013 | 5 comments
-
Issa Recovers
May 22, 2013 | 4 comments
-
Colorado Governor Gives “Reprieve” to Murderer
May 22, 2013 | 1 comment
-
Steven Crowder: Jesus vs. Muhammad
May 22, 2013 | 2 comments














Butch| 10.12.12 @ 1:54PM
Of course it has. I have had a dual career as a Professor and professional researcher specializing in survey research (mostly commercial, but a little political). I can tell you that exactly the same techniques and principles used to create accurate results can be reversed to create biased results.
Pollsters work for money. If a client tells them to weight the results unrealistically or bias their sample (the most subtle kind of bias, which no one ever detects), they will. However, they will keep them accurate in the race's final stages to protect their reputations, because their final predictions compared to the actual outcome is the only thing people ever remember.
Rush intuited it correctly. Everything is a propaganda tool to the left, including polls. They are used in the early stages to encourage their base and discourage the opposition. But as the election grows near, the pollsters themselves must reflect the electorate's predisposition accurately.
Ross Kaminsky| 10.12.12 @ 2:00PM
Butch,
I agree with your premise but would add that for the largest polling firms, a reputation for accuracy is important.
Rasmussen is only highly respected these days because he was the most accurate pollster in the 2008 election (and one of the most in 2004, if I remember correctly).
So they do have an incentive other than just the wishes of a client for a particular outcome.
JP| 10.12.12 @ 2:32PM
Like you said, Rasmussen hit a grand slam in 2008. Until he proves otherwise, I will go with his numbers. But, I still have a few questions:
1)As far as I know Rassmussen, like the other polling firms, use 2008 turn-out models for their 2012 surveys. Does Rassmussen really believe that turn-out for Obama/Biden in 2012 in states like Indiana, Virginia, Florida, and North Carolina will be a repeat of 2008?
2)Does Rasmussen rely totally on phone surveys (ie land-lines). Not to pick on the Elderly, but most Americans under 50 have dropped land line phone service since 2008. Does he canvas voting ages in proportion to past demographics (ie is he sampling too many elderly)?
Butch| 10.12.12 @ 3:01PM
Ross, I respect Rasmussen the most because of his ethics. In fact, it's the only one I pay attention to in the early stages. You have given me an interesting idea for an academic study: testing the hypothesis that poll accuracy affects polling firm reputation more the closer the poll is taken to the actual election; I would almost bet that for most firms (maybe not Rasmussen), R-square grows the nearer you get to the actual election.
By the way, Rasmussen nearly hit the 04 election right on the nose. Guess how he did it? With a one-minute voice mail poll (!), that asked just three questions: President, Senator in your state, and Congressman in your district. The message guaranteed that it would take less than a minute. It was taken the last 48 hours before the election.
JP: Mobile phones complicate telephone survey research, but you can still make unbiased contact with random digit dialing. Most RDD programs allow you to paramaterize area codes and three-digit prefixes in the area, and the rest is random numbers generated at the speed of light. Some are hits. It can be done.
Al Adab| 10.12.12 @ 2:28PM
Still, all the polls and maps leave the GOP short of 270 even if Romney runs the table with FL, NC, VA and Ohio.
Simon Templar| 10.12.12 @ 3:34PM
The election has never been close. Do you ever have an original thought or observation other than what you hear in the MSM and the polls controlled by them? Every damn election it is the same. Shit, they do not even give up the day of election with their stupid exit polls. Every poll up till now has been widely corrupted, rigged, and biased.
Independents are going to Romney now at 20. He has locked up the male vote and is gaining on women. He is winning in all states that have been held up as absolutely necessary for election or reelection. The swing since the debates has been gigantic. Crowds are record high. This administration is now mired in a scandal that dwarfs Watergate. I do not see you writing much about that. Forty one million Tea party patriots are dying and ready to vote.
Do you want Obama reelected? Did you have a problem last night with Ryan's religious belief and are now worried about a pogrom?
I really do not understand you....who are you?
Dai Alanye | 10.12.12 @ 4:12PM
Unless it turns out that Romney had a DUI in his twenties, I say he gets between 320-360 electoral. And I don't even like the guy.
Al Adab| 10.12.12 @ 4:24PM
Please see my 2:28 above. His solid states plus those four equal 256. Add MO and even WI (a long shot even with Ryan) and it's 276. Trade WI for Iowa and it's 272. Where can he find those votes?
Simon Templar| 10.12.12 @ 3:49PM
John Zogby to Newsmax: Ohio Momentum 'Clearly With Romney'
Simon Templar| 10.12.12 @ 3:51PM
The next I will hear if he does not carry all 5o states he will lose. And do not forget that planet in the Qurzt galaxy.
JD| 10.12.12 @ 9:20PM
America should listen to Thomas Peterffy:
http://www.youtube.com/watch?v=N2QtDExs6lM