GOP takes aim at ‘skewed’ polls

GOP takes aim at ‘skewed’ polls

The Romney campaign and other Republicans say polls showing President Obama with a significant lead over their candidate are inaccurate. 

They argue many mainstream polls skew in Obama’s favor because of sample sizes that base 2012 turnout projections on 2008, when Democrats — and Hispanics, blacks and young voters in particular — turned out in record numbers. 

“I don’t think [the polls] reflect the composition of what 2012 is going to look like,” Romney pollster Neil Newhouse said in an interview.


Frustration that polls are skewed in favor of Obama has escalated among some on the right in recent weeks. One website,, recently began re-weighting the mainstream polls to closer track the demographic assumptions of conservative polling outlet Rasmussen Reports. The re-weighted polls all show Romney ahead in the race, with leads of between 3 and 11 percentage points.

Democrats counter that the sample sizes used in polls are accurate because there is no reason to think the makeup of the 2012 electorate will be proportionately different than in 2008. They also point to census data that shows minorities making up a greater share of the population, something driven by the surging Hispanic population. 

 “With African-Americans, there’s no question they will turn out at the same rate, for reasons that are obvious,” said Anna Greenberg, senior vice president at liberal research group Greenberg Quinlan Rosner. 

“Moreover, given demographic changes in this country, four years later now there are more minority and younger voters, just by natural demographic changes alone. We’ve seen a more and more diverse electorate over time,” she said.

Whether polls are accurately reflecting the race is critical to how it is reported, and to decisions on how millions of dollars in campaign funds will be spent in the six weeks before Election Day. 

According to the RealClearPolitics average of polls, Obama enjoys a 3.7-percentage-point advantage over Romney in the race. 

But Newhouse argues the RCP average, which is designed to neutralize outlying polls, is just an average of skewed polls that have a built-in bias for Obama. 

“Averaging bad polls that include skewed samples will get you a skewed result,” Newhouse told The Hill.

Nobody disputes that Obama maintains big leads over Romney among blacks, Latinos and younger voters, but Republicans argue the economic downturn that has hit these groups hard is likely to lower their turnout in 2012, especially since 2008 turnout was boosted by the historic implications of electing the nation’s first black president. 

As a result, they say, polls projecting similar turnout among these voters are inaccurate. 

Dick Morris, the political consultant and pollster (who writes a column for The Hill), wrote last week that pollsters using 2008 turnout models in weighting their samples are likely to be incorrect given the turnout results that year. 

Blacks made up 14 percent of the electorate in 2008, when they had traditionally cast about 11 percent of the vote, while the Latino share of the electorate rose by 1.5 percent and college-aged voters doubled their share, Morris wrote.

The problem with the polls, say conservatives, is the assumption most pollsters make on the party identification of the 2012 electorate. 

In 2008, Democrats had a 7-percentage-point advantage in party identification over Republicans, which was close to the final margin of victory Obama had over Sen. John McCainJohn Sidney McCainGOP senators appalled by 'ridiculous' House infighting MSNBC's Nicolle Wallace, Chris Christie battle over Fox News Trump's attacks on McConnell seen as prelude to 2024 White House bid MORE (R-Ariz.).

Newhouse and other Republicans say it’s foolish to expect a similar proportion of Democratic voters in 2012. They argue a smaller proportion of Democrats are likely to come to the polls in November, while a larger proportion of Republicans eager to deny Obama a second term can be expected to vote. 

Polls that assume the makeup will be the same as 2008 don’t take this into account, they say. 

“You had an extraordinary 2008 turnout among rural evangelicals, but Republicans stayed home in larger numbers,” Newhouse said. “This time you have a more enthused and energetic Republican electorate, and because of that, you’re not going to see the margins go up, you’ll see it narrow. So instead of a 7 [percentage-point advantage for Democrats], I anticipate something smaller than that.”

Those defending the accuracy of polls showing Obama with a lead argue Team Romney is overestimating the proportion of Republicans who will make up the 2012 electorate, and underestimating the proportion of minorities. 

Among Hispanics in particular, Greenberg said Romney’s tack to the right during the GOP primaries had recharged the group “after a fair amount of softness the last couple of years,” and that Democrats had seen a sustained increase in enthusiasm across the board since the convention earlier this month.

Greenberg acknowledged voter identification has narrowed, but she said Democrats still have a 5-point advantage. 

Veteran GOP consultant Roger Stone, who in recent years has worked on libertarian causes, argues the polls are correct because any benefit Romney gets from lower turnout by Democratic voters is likely to be evened out by softer conservative turnout. 

“Romney doesn’t have a lock on conservatives,” Stone said. “And he’s losing some, just a couple of points, to [Libertarian candidate] Gary Johnson. That’s a problem.”

Some Democrats argue that polls that show Romney and Obama running neck and neck are skewed because not enough of those polled rely on cellphones. 

“If there’s less intensity among Democrats, that’s balanced out by lack of cellphones in samples done by [conservative polling outlet] Rasmussen,” Stone said. “They’re under-sampling cellphone users, who we know are usually young, black, Latino or lower-middle-class working people.”

“The race is always going to be perceived as competitive, which it still is,” Stone said. “But the president has a slight edge.”