By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Politics Discussion - Biased polls? Yes...

Maybe there's another explanation: do Republicans own more phones than Democrats?



Stwike him, Centuwion. Stwike him vewy wuffly! (Pontius Pilate, "Life of Brian")
A fart without stink is like a sky without stars.
TGS, Third Grade Shooter: brand new genre invented by Kevin Butler exclusively for Natal WiiToo Kinect. PEW! PEW-PEW-PEW! 
 


Around the Network
Alby_da_Wolf said:
Maybe there's another explanation: do Republicans own more phones than Democrats?


Not only more phones, but more likely to be in the vicinity of a phone all day as compared to, say, a labourer.



fordy said:
Alby_da_Wolf said:
Maybe there's another explanation: do Republicans own more phones than Democrats?


Not only more phones, but more likely to be in the vicinity of a phone all day as compared to, say, a labourer.


Yes, quite likely too.



Stwike him, Centuwion. Stwike him vewy wuffly! (Pontius Pilate, "Life of Brian")
A fart without stink is like a sky without stars.
TGS, Third Grade Shooter: brand new genre invented by Kevin Butler exclusively for Natal WiiToo Kinect. PEW! PEW-PEW-PEW! 
 


theprof00 said:
Kasz216 said:
theprof00 said:
I think one of the points is that people like nate had the correct data for a while 21 days and more and lots of people, including many members of this site, chose to believe differently based on their own suspicions of bias, and even down into the last days denied the truth even when their trusted polls began to slip.


This sentence is kind of confusing, though no, that really isn't a point, at least not in regards to this thread.

I also get the feeling that you don't quite get what Nate Silver does when using the term "correct data".

Since everybody had the same data... and scrutinizing data sets is only ever a good thing, at least among researchers.

If you'll note, practically all the current coverage is actually just that, analysing why things turned out why they did, confused why republicans didn't vote in higher numbers.

You always have to comment on the semantics. Sorry! He had the correct study results, the correct model.

The way things turned out as they did because they weren't the majority. Not sure what else there is to add to that.


I'd hope something... since it's totally unrelated to what you originally commented on.



Alby_da_Wolf said:
Maybe there's another explanation: do Republicans own more phones than Democrats?


Almost, you may have missed it, but it's not that Republicans own more phones... but that republicans own more landline phones.


The arguement is that the polls that swung Republican were mostly robo callers.

Robo Callers can't call cellphones.

 

Younger people in the city are more likely to not use landlines and just cellphones.



Around the Network
Kasz216 said:
Alby_da_Wolf said:
Maybe there's another explanation: do Republicans own more phones than Democrats?


Almost, you may have missed it, but it's not that Republicans own more phones... but that republicans own more landline phones.


The arguement is that the polls that swung Republican were mostly robo callers.

Robo Callers can't call cellphones.

 

Younger people in the city are more likely to not use landlines and just cellphones.

Its going to be interesting to see how they deal with this going into the future, as I only see the issue becoming more pronounced. Personally, I think its only part of the problem, as far as this election is concerned.

In terms of land-line calls, the robodial surveys seem to be far outnumbered by other methods (particularly live calls), and most of the polling houses that used robodials only conducted a few surveys (1-4) each (except Rasmussen which is a mixture & Survey USA), and most of these surveys still moved to a more pro-Obama prediction if they conducted later polls. Some of the polls are just strange as well.....I mean +15.7 bias for Romney.....I don't think anybody would have trusted that at any point in the campaign.

My opinion, and I've said this before, is I would like to see the likely voter screening for some of the bigger polling houses (Gallup and Rasmussen). People tend to overreport voting by about 10-20 percent in post-election polls (in 2008, it was about +15). I would assume people would also overreport intending to vote as well, which could throw off likely voter predictions, depending on how they determine likely voters.

Big picture though, the polls weren't really that inaccurate. If I had to guess, I would say final week polls underestimated Obama's victory by about 2 percentage points. It could be better, but still not bad given the margin of error.