It’s depressingly predictable: Every time a poll comes out showing Mitt Romney trailing nationally or in key states, Romney supporters scream or tweet: “Oversample Ds!” They claim that the pools of respondents — either accidentally or even deliberately — contain too many Democrats, meaning they don’t resemble the electorate that will turn out on Election Day, skewing the results against Romney.
But Steven Shepard calls up the pollsters themselves and gets them to explain what’s really going on here. Most pollsters don’t weight for party ID, for good reasons:
Pollsters counter that the results they are finding reflect slight changes in public sentiment — and, moreover, adjusting their polls to match arbitrary party-identification targets would be unscientific.
Unlike race, gender or age, all demographic traits for which pollsters weight their samples, party identification is considered an attitude that pollsters say they should be measuring. When party identification numbers change, it’s an indication of deeper political change that a poll can spot.
“If a pollster weights by party ID, they are substituting their own judgment as to what the electorate is going to look like. It’s not scientific,” said Doug Schwartz, the director of the Quinnipiac University Polling Institute, which doesn’t weight its surveys by party identification.
This isn’t to say that a poll can’t be inaccurate or fail to perfectly represent what the electorate will, in fact, look like. Rather, the point these pollsters make is that not weighting for party I.D. is a better course of action than weighting for it. If a sample shows more Dems than seems likely, then that could mean that “there are more people who want to identify with the Democratic Party right now than the Republican Party,” as Schwartz puts it. If you try to wipe out that finding by reweighting the sample, you risk losing sight of what may be an actual shift in the electorate. What’s more, weighting for party I.D. would require pollsters to predict themselves what the electorate will look like, which only risks introducing more error.
It’s true that Dems offered complaints about polls in 2004 that are similar to the ones Republicans and conservatives are making now. But you frequently see Romney partisans attacking the polls in truly outlandish ways. It’s not unusual to see them Tweeting that the polls are deliberately being mis-weighted to skew towards Dems or even that these moves are part of a larger liberal media conspiracy to create a false narrative of inevitable victory for Obama.
Part of this is probably attributable to what Jonathan Bernstein has called the “conservative closed information feedback loop,” in which those who get their information mainly from Fox News or Rush Limbaugh have come to believe that anything that contradicts that info simply reflects the corrupt mainstream media’s efforts to keep Americans in a state of mass liberal ignorance, by shielding them from conservative truths. And so it’s not uncommon to hear that the only polls that “really count” or are “really telling the truth” are Gallup and Rasmussen, as if all other polls done by major news organizations are mere tools of that conspiracy. But in the real world, dismissing all the polls that don’t tell you what you want to hear as flawed or corrupt is hardly the way to win elections.
* Pro-Romney Super PACs operating from failing playbook: Jeremy Peters has a must-read report on how all the pro-Romney Super PACs are coordinating to unite behind a simple message: The Obama presidency is an utter, abject failure. Here’s how Tim Phillips of Americans for Prosperity puts it:
“To defeat the president, the public needs to understand the utter failures of his economic policies.”
And it’s true that variations of the word “failed” are ubiquitous in anti-Obama Super PAC ads. But again, there is a great deal of polling that suggests swing voters are not accepting the idea that Obama was a resounding failure on the economy. If it turns out that all this Super PAC cash has less of an impact than promised, their reliance on a flawed messaging strategy may be a key reason why.
* Obama grabs leads in Ohio and Florida: New Washington Post polls show Obama leading Romney in Ohio by 52-44 among likely voters; in Florida the spread is 51-47. Romney’s path to victory is exceedingly narrow if he only loses one of those states, let alone both.
* Obama holds edge on economy in Ohio: Some key Post findings:
Slightly more than half of all Ohio voters — 53 percent — give Obama positive marks for in dealing with the economy, with more — 56 percent approving of his overall performance....Matched against Romney, 50 percent of all voters say they trust the president more to deal with the economy; 43 percent say so of his Republican challenger.
Obama is now leading on the economy in this pivotal state. Eight national polls, and a raft of polls in many key swing states, show that Obama has pulled into a tie with Romney on the issue, after Romney had held the advantage on it for many months.
* Ohio now leaning Obama: The Fix crew moves Ohio into the “leans Obama” column, which is significant, because it reflects the notion that movement towards Obama in Ohio may no longer reflect any convention bump.
* The dueling ground games: With early voting underway in a number of key states, The Hill games out the advantage the Obama campaign has built up over Romney and Republicans when it comes to on the ground organization. This advantage will be key to neutralize whatever spending advantage the Romney camp enjoys when the final flood of Super PAC ads washes over the key battleground states.
* Elizabeth Warren hits back at Indian attacks: She’s up with a new ad that takes on the accusations about her Native American heritage with unusual directness; it features her talking directly to the camera, clarifying that she never received any benefits as a result of her background. Also note the pivot to her message about fighting for ordinary families.
* Dems use Romney to hit House Republicans: The DCCC is up with its first ad of the cycle that ties a vulnerable Republican member of Congress — Chris Gibson, in New York’s 19th district — to Romney. The ad shows Gibson’s face alongside Romney’s, and claims they both support the Paul Ryan plan to “end the Medicare guarantee.”
It’s a test as to whether Dems can use Romney as a drag on down-ticket candidates (though this will obviously be easier in New York).
* And we’re all makers and takers: Suzanne Mettler and John Sides have a great piece outlining a survey that shows that virtually all Americans have partaken of government services at one time or another. As they conclude: “Almost everyone is both a maker and a taker.”
The irony here is that, in their quest to paint Obama’s vision of government as radical, Romney and Ryan are the ones who are advancing a radical “makers and moochers” vision that is out of touch with the reality of how Americans really experience government.