Our good old friend Nate Silver has yet another article lambasting and criticizing various pollsters... and describing which ones were the closest to reality.
Nationally at least... Quinnpac, SurveyUSA, IBD/Tipp and We Ask America were all extremely close. The biggest offenders of the reality? Gallup (horrendous) with a 7.2% republican bias, American Research Group (4.5% bias in favor of GOP) and the much touted by the political right wing Rasmussen (3.7% bias in favor of GOP).
And quoting from the article:
Rasmussen continues to get it wrong and got it wrong because of the way they conduct polling. They randomly make calls on landlines... a lot of younger voters don't have landlines with many only having cell phones. It's an outmoded way of polling and constantly had a republican bias. I believe pollsters like Rasmussen and ARG gave Romney a false sense of security and victory... along with his own campaign's internal polling. Republicans have often shot accusations that polls don't sample enough republicans... but the reality is there may not be justification to include republicans as they are a shrinking portion of the electorate.Rasmussen Reports uses an online panel along with the automated calls that it places. The firm’s poor results this year suggest that the technique will need to be refined. At least they have some game plan to deal with the new realities of polling. In contrast, polls that place random calls to landlines only, or that rely upon likely voter models that were developed decades ago, may be behind the times.
It would be interesting if Nate breaks it down regarding certain battleground states and who got it right... but for now it's interesting to see how polling companies did.