An interesting post-election thread is the issue of why so many polls underestimated voter turnout, specifically in races where the Democratic candidate won. The starkest example comes from the state of Nevada where the Democratic candidate for Senate, Harry Reid, beat his Republican challenger, Sharron Angle, by 5 points. Polls published by the Las Vegas Review-Journal had Angle leading by 4 points just days before—a 9 point gap. The New York Times’ Nate Silver had Angle ahead by 2.3 points, with Reid eventually winning by 5.6 points—nearly an 8 point gap. Why the disparity?

The Las Vegas Review-Journal writes today that its own polls “wound up having about as much predictive power as the Old Farmer’s Almanac in forecasting the winter snowfall.” Nate Silver wrote in the New York Times that “It’s fairly unusual, however, to have the consensus of polls off by 7 or 8 points in an extremely competitive Senate or gubernatorial general election.” Silver then went on to speculate why numbers were so off in the Silver State:

“I speculated, for instance, that the fact that Mr. Reid is the sort of candidate whom one votes for unenthusiastically might have skewed the turnout models…There is another theory, however, which was proposed to me last night by Matt Barreto of the polling firm Latino Decisions…that Latino voters—somewhat against the conventional wisdom—were relatively engaged by this election and for the most part were going to vote Democratic. Mr. Barreto also found that Latino voters who prefer to speak Spanish—about 40 percent of Latino voters in California meet this description, he told me—are particularly likely to vote Democratic. Pollsters who don’t conduct bilingual interviewing at all, or who make it cumbersome for the respondent to take the poll in Spanish, may be missing these voters.”

LatinoDecisions, the polling group which focuses on states in which the Latino vote will play an important role in elections, and who brought this to Nate Silver’s attention, writes in great detail about the challenge and ongoing undercount of minority voters on their blog:

“The problem of faulty exit poll data for Latinos is not new, yet very few in the media have expertise in polling Latinos and analyzing Latino vote data, and as a result are not in a position to assess on election night the veracity of the Latino results…the National Exit Pool surveys, systematically underestimate Latino and African-American Democratic vote share by over-representing higher income, higher education, and more socially integrated minority voters than their share of the electorate warrants.”

The Las Vegas Review Journal is beginning to agree and in taking a look at its own polling wrote:

“R-J critics think they know exactly where the breakdown occurred: Its pollster, relying on old-fashioned random dialing to land lines, doesn’t account for voters who only have cell phones, and undercounts Hispanic voters who may be reluctant to participate in surveys. Both groups are heavily Democratic.”

And they note the bottom line problem:

“They’re not drawing a population that looks like the electorate,” said Dave Damore, a UNLV political scientist who studies public opinion data.

At the end of the day, pollsters and the politicians who depend on them are at great political peril if they willfully ignore huge gaps in their poll samples. Undercounting Latinos, cell phone users, or any other group that has clearly emerged as undercounted and who are continuing to grow in size and importance each election day, is clearly political and reputation suicide.

Photo by barackobamadotcom.