http://www.motherjones.com/politics/2012/10/polls-undercounting-latino-voters
Shortly after the 2010 midterm elections, Jim Margolis, a longtime Democratic pollster who's now a top media consultant for Obama's reelection campaign, cowrote a memo outlining how the Democrats had managed to save his then-boss, Senate Majority Leader Harry Reid (D-Nev.). During the fall, Reid had looked like he was going to get swept away in the tea party wave that handed the GOP a majority in the House of Representatives. Polls showed Reid trailing his Republican opponent, Sharron Angle, by an average of nearly 3 points. Republicans were about to knock off the second Democratic Senate leader in a row, having ended Tom Daschle's Senate career in 2004.
Then something weird happened. Reid won—by almost 6 points. In Colorado, another state with a large Latino population, Democratic Senate candidate Michael Bennet eked out a 1-point win despite polls showing his GOP rival, Ken Buck, up by an average of about 3 points. For good reason, politicos have calloused fingers from hitting refresh on New York Times' numbers guru Nate Silver's website, but even his model predicted likely Republican wins in Nevada and Colorado in 2010.
"Nobody had Reid winning, nobody had Reid ahead," says Brad Coker, a pollster for Mason-Dixon, the firm hired by the Las Vegas Review-Journal to poll the state. "People underestimated the ability to turn out Hispanic voters the way they had turned out for Obama." In the memo written after Reid's win, Margolis and Reid pollster Mark Mellman said the same thing: Latino voters, undersampled by pollsters and written off as unlikely voters, had made a huge difference for Democrats.
How did this happen? According to Matt Barreto of Latino Decisions, a polling firm that specializes in public opinion surveys of the Latino community, pollsters were using outdated methods that simply missed the Latino voter surge. Small sample sizes of Latinos, ineffective or inadequate efforts to reach "Spanish-dominant" Latino voters, and failing to account for how many Latino voters would turn out kept most pollsters from identifying what immigration reform advocate Frank Sharry called the Democrats' Latino "firewall" in 2010. Barreto argues that the polls often miss working-class Latinos, who are more likely to use cellphones, work long hours, and prefer to speak Spanish. Instead, the polls are more likely to reach more financially comfortable, "English-dominant" Latinos who are more likely to vote Republican. Robopolls, in which an automated prompt calls voters, tend to be particularly bad at measuring Latino public opinion, even when they have a Spanish-language option, Barreto says.
"When you start polling in any state that's competitive with a big component of the electorate being Latino, you tend to see that they tend to underestimate the Latino vote," says David Damore, a political science professor at the University of Nevada. "[The result] tends to be more Republican than it actually will be."
Barreto says it's happening again this year. "Pollsters who are missing that component of the correct proportion of Spanish interviews, they are completely underestimating a growing part of the electorate, and this is the part that is most heavily Democratic," he argues. "They're are operating on models that are at best 20 years old." (Barreto expanded on his theory in a post written Monday.)
Shortly after the 2010 midterm elections, Jim Margolis, a longtime Democratic pollster who's now a top media consultant for Obama's reelection campaign, cowrote a memo outlining how the Democrats had managed to save his then-boss, Senate Majority Leader Harry Reid (D-Nev.). During the fall, Reid had looked like he was going to get swept away in the tea party wave that handed the GOP a majority in the House of Representatives. Polls showed Reid trailing his Republican opponent, Sharron Angle, by an average of nearly 3 points. Republicans were about to knock off the second Democratic Senate leader in a row, having ended Tom Daschle's Senate career in 2004.
Then something weird happened. Reid won—by almost 6 points. In Colorado, another state with a large Latino population, Democratic Senate candidate Michael Bennet eked out a 1-point win despite polls showing his GOP rival, Ken Buck, up by an average of about 3 points. For good reason, politicos have calloused fingers from hitting refresh on New York Times' numbers guru Nate Silver's website, but even his model predicted likely Republican wins in Nevada and Colorado in 2010.
"Nobody had Reid winning, nobody had Reid ahead," says Brad Coker, a pollster for Mason-Dixon, the firm hired by the Las Vegas Review-Journal to poll the state. "People underestimated the ability to turn out Hispanic voters the way they had turned out for Obama." In the memo written after Reid's win, Margolis and Reid pollster Mark Mellman said the same thing: Latino voters, undersampled by pollsters and written off as unlikely voters, had made a huge difference for Democrats.
How did this happen? According to Matt Barreto of Latino Decisions, a polling firm that specializes in public opinion surveys of the Latino community, pollsters were using outdated methods that simply missed the Latino voter surge. Small sample sizes of Latinos, ineffective or inadequate efforts to reach "Spanish-dominant" Latino voters, and failing to account for how many Latino voters would turn out kept most pollsters from identifying what immigration reform advocate Frank Sharry called the Democrats' Latino "firewall" in 2010. Barreto argues that the polls often miss working-class Latinos, who are more likely to use cellphones, work long hours, and prefer to speak Spanish. Instead, the polls are more likely to reach more financially comfortable, "English-dominant" Latinos who are more likely to vote Republican. Robopolls, in which an automated prompt calls voters, tend to be particularly bad at measuring Latino public opinion, even when they have a Spanish-language option, Barreto says.
"When you start polling in any state that's competitive with a big component of the electorate being Latino, you tend to see that they tend to underestimate the Latino vote," says David Damore, a political science professor at the University of Nevada. "[The result] tends to be more Republican than it actually will be."
Barreto says it's happening again this year. "Pollsters who are missing that component of the correct proportion of Spanish interviews, they are completely underestimating a growing part of the electorate, and this is the part that is most heavily Democratic," he argues. "They're are operating on models that are at best 20 years old." (Barreto expanded on his theory in a post written Monday.)
Last edited: