I've read Nate Silver's Five Thirty-Eight Blog religiously for several years, even before he became a columnist for the New York Times. Silver focuses on the scientific side of polling and is a great resource for understanding the meaning of polls and how to sift through the bad ones and the good ones. In this piece, he explains that many pollsters undercount Hispanic votes due to their methodology and the undercount explains why at least two Senate seats were one by Democrats in 2010 despite pre-election polling that suggested they would go to the GOP candidates.
The subject came up because a new poll shows President Obama actually leading in Arizona, a state that most analysts have assumed was safe Romney territory. It's easy to dismiss one poll, but Silver looked at the pollster and was not ready to say their results are unreliable. From Silver:
However, there is a potentially troubling aspect of the poll for Mr. Romney. The Behavior Research Center conducted interviews in Spanish along with English. That may account for some of why it had better results for Mr. Obama than other surveys of the state.
In the past couple of elections, polls have underestimated Democrats’ standing in states with heavy Hispanic populations. (The two senate races that the FiveThirtyEight forecast called incorrectly in 2010 — Nevada and Colorado — are both states with a healthy number of Hispanic voters.)
This may be because many polling firms that conduct interviews only in English miss some Hispanic voters who are more comfortable speaking Spanish. According to Matt Barreto of the polling firm Latino Decisions, which conducts bilingual interviews, primarily Spanish-speaking Hispanic voters are more likely to vote Democratic than those who have more English fluency.
Polling firms such as Latino Decisions that have conducted interviews in Spanish have shown Mr. Obama with a larger advantage among Hispanic voters than those which interview in English only. The most recent Latino Decisions poll, for example, had Mr. Obama ahead 72-20 among Hispanic voters. This poll is not an outlier; other polling firms that have conducted Spanish-language interviews have found similar results.
The most interesting part of the article is Silver's re-running his forecasts based on two assumptions - that Romney loses by 50 points with Hispanics and that Hispanics constitute the same percentage of the vote as they did in 2008. Both are very reasonable to assume given the data we've seen this year and the fact that the number of Hispanic citizens has increased as a proportion of the entire electorate since 2008.
With revisions made to account for these two assumptions, the President's chances of winning the electoral college increase by a whopping 6%.
If you want to get an interesting read on the election, the New Yorker just released a survey of 74 DC insiders - 37 from each party - to get their views on the state of the race. Most chose to remain anonymous, though a few allowed themselves to be quoted in the story. One interesting one came from John Weaver, chief advisor to John Huntsman in his failed GOP primary bid. In response to the question of what has been Mitt Romney's biggest mistake to date, he had this to say:
The bottom line is that Hispanic voters are absolutely critical this year. I'm of the opinion that Romney's latest attempts to soften his image with Hispanic voters aren't enough. He needs to publicly break with the anti-immigrants with whom he has associated - Joe Arpaio, Kris Kobach, Steve King, for example - and make clear both what he would do regarding the President's current policies and also how he would manage to deliver on immigration reform when President Bush could not. It's not enough to offer vague promises or criticize President Obama for not achieving enough.
"Going so far right on immigration, thus trapping himself demographically."