Hi folks. I missed you all in the election season, but has having wayyyy to much fun on some college football boards giving the mouth breathers there a hard time. Okay, so let's look at how the pollsters did: http://themonkeycage.org/
So there you have it. Lesson learned: Gallup, Rassumussen, Democracy Corps - all bad; PPP, YouGov and Ipsos/Reuters - good. One other thing: Polling mode (whether its RDD land line, cell phone, robo call, internet) does NOT matter. Understanding the nature of the underlying population - in this case, the voting public - is key.
Good stuff. Reveals once again how damaging Fox is to the U.S. Even as the polls overall were slightly biased in favor of Republicans, Fox convinced its audience that the polls were a vast conspiracy against Republicans. So viewers who watched Fox were steered away from the truth, and into bitterness and disappointment and cynicism. How is that good for them or this country? Now they feel cheated, and going forward they are likely to be angrier and more partisan. Great. Just great.
Well, the Republican War on Science came back and bite their own asses right off. So much for the new Republican Polling Statistics.
It's worth pointing out that PPP, by their admission, got lucky. They kept the likely voter model the same as 2008, and that was based mostly on faith that the Obama ground game would get the votes out. There's no way to reliably model a ground game until the votes are cast.
And for a bellyful of laughs, Unskwedpolls.com These jokers had Romney winning with around 330 EVs, but I see they changed their final forecast to 275 for Romney.
Did PPP really say they were lucky? I think the smarter pollsters ended up assuming that the share of white voters would decline, while the Republicans assumed that minorities would not show up to the polls. Some of the better analyses I've seen have been on independents and Republicans, with an understanding that many Tea Party lunatics switched from Republican to Independent. thus, the polls with large (6-8) point Dem advantages were correct.
Okay, so "lucky" wasn't the word: But Jensen conceded that the secret to PPP’s success was what boiled down to a well informed but still not entirely empirical hunch. “We just projected that African-American, Hispanic, and young voter turnout would be as high in 2012 as it was in 2008, and we weighted our polls accordingly,” he explained. “When you look at polls that succeeded and those that failed that was the difference.” Given the methodological challenges currently confronting pollsters, those hunches are only going to prove more important. “The art part of polling, as opposed to the science part,” Jensen said, “is becoming a bigger and bigger part of the equation in having accurate polls.”So you're right about how they got to the likely model, but it was still a hunch, based partly on the empirical but also partly on the immeasurable - enthusiasm and effectiveness of the operation.
Thanks for finding that specific quote - and great point about the "hunch" - which is a good way to put the pollsters' estimates on turnout. I think that under normal election years, most of the electorate give the pollsters the benefit of the doubt on turnout models; after all, these guys are the professionals. This year was off because the Wing Nut War on Science turned its blunderbusses on polling, a lot of people really started to question the turnout models. Otherwise, I don't believe folks would have paid so much attention.
A very brief distraction... https://gawker.com/5958847/lets-play-drunk-nate-silver-the-hilarious-new-twitter-game Hail Silver!
I'm still unsure of Twitter's long-term uselessness, but I like how these instamemes seem to bring smart, funny people together for fleeting moments. See also: #philipriversexperience
The reason Gallup was so wrong is that they projected the % of white voters this year to be the exact same as it was in 2010. Just right off, you KNOW that was a stupid assumption. They didn't average 2008 and 2010 or anything, just straight up used 2010. Why would you compare a presidential election t an off year election rather than the previous presidential? Makes no sense. Doesn't that make Gallup an organization run by incompetents?
Gallup didn't think that perhaps nonwhite voters might show up more if Barack Obama is on the ballot?
No, not incompetent. They've been doing this a long time. However, they did seem rather naive. Bells should have gone off at Gallop when they were seeing their poll was a consistent outlier. Now, Rasmussen - that's incompetence. Or just using Republican War on Science Math.
But Rasmussen's audience is prospective clients on the right. Feeding bullshit through public polls is marketing.
Until the internet, I had no idea how many funny people there are in this world but I'm always blown away by it.
Anyone have any idea if this is the most the polls have been off in any of the recent elections? Seems like it. If so was it mostly down to the media trying to portray a race as close that wasnt really close just for ratings or some other reason?