Author Topic: GOP pollsters missed big in Virginia  (Read 178 times)

0 Members and 1 Guest are viewing this topic.

Online mystery-ak

  • Owner
  • Administrator
  • Hero Member
  • ******
  • Posts: 250,510
GOP pollsters missed big in Virginia
« on: June 12, 2014, 11:14:03 AM »
http://dyn.politico.com/printstory.cfm?uuid=77C0B157-BA8F-44CF-997E-3733DDC4AEFB

 GOP pollsters missed big in Virginia
By: Steven Shepard
June 12, 2014 10:43 AM EDT

Eric Cantor did not see it coming — and neither did some GOP pollsters.

A late-May survey conducted for the ousted House majority leader’s campaign showed Cantor with a 34-point lead over Dave Brat. And while a later public survey showed the race closer than that, Cantor was campaigning like an incumbent who expected to receive around 60 percent of the vote, spending much of Election Day at work in D.C.

Primary elections are notoriously volatile, but the failure of Cantor’s internal polling — along with an automated Republican survey conducted for a conservative news website — are the latest black eyes for the GOP’s professional survey researchers, who are still seeking a way forward after a difficult 2012 cycle. For Cantor’s campaign pollster, it was another embarrassing misfire in one of their publicly-released surveys.



Cantor’s campaign last week released an internal poll from McLaughlin & Associates, which conducts surveys for Republican candidates across the country and has offices in Virginia’s D.C. suburbs and upstate New York. That poll was conducted just after the Memorial Day holiday, and it showed Cantor with a massive lead, 62 percent to 28 percent.

That survey was released in response to a poll from new GOP robopollster Vox Populi Polling, conducted for the Daily Caller website, which showed Cantor with a smaller lead in the low double-digits.

Both polling firms – which compiled their sampling frames from lists of previous GOP primary voters – cited an unexpected boost in turnout for the yawning discrepancies between their numbers and the eventual results.



According to results collected by Cantor pollster John McLaughlin, slightly over 46,000 voters turned out in 2012, but more than 65,000 voters showed up this time. McLaughlin said he thought Democratic voters were being urged to vote in the open primary. Voters don’t register by party in Virginia, and voters simply need to declare in which party primary they wish to participate.

“We were polling Republican primary voters,” said Cantor pollster John McLaughlin in an email. “[W]ho were the new 20,000 primary voters? They probably aren’t Republicans. Certainly the extra voter surge of non-Republican primary voters seriously hurt.”

Similarly, Vox Populi cited energized turnout for boosting Brat, the increased difficulty of polling in a state with open primaries and Brat’s late-breaking momentum.

One indication that voters who don’t typically show up in Republican primaries hurt Cantor: Both pollsters showed him running stronger with self-identified Republicans than those who said they were independents.

“In our poll two weeks out, Eric was stronger with Republicans at [70 percent to 21 percent],” but Cantor was running under 50 percent among non-Republicans, McLaughlin said.


Vox Populi also had Cantor up big among Republicans. But he was actually trailing among independents – who comprised 30 percent of the survey – 42 percent to 24 percent.

The polling failures come at a precarious time for GOP survey firms, who are looking to rebuild their credibility after a rough 2012 cycle. GOP internal polls painted a rosier picture for the party in the presidential race and key contests down-ticket than turned out to be the case — and many Republicans were left as stunned at the breadth of their defeats two years ago as Cantor was Tuesday night.

The National Republican Congressional Committee and Republican National Committee launched comprehensive reviews of the party’s polling practices – recommending the party contact more voters via cell phone, experiment with listed samples versus random-digit dialing and conduct over-samples of key, hard-to-reach demographics.

McLaughlin & Associates attended meetings during the review process. Though very few of a pollster’s private surveys eventually become public, McLaughlin had a number of notable misses in recent years:

— They had Republicans Mitt Romney and George Allen leading in Virginia in Oct. 2012.

— McLaughlin had Republican Randy Altschuler leading Democratic New York Rep. Tim Bishop in New York by five points – with Romney ahead of President Barack Obama by 12 points in the district. Bishop won reelection by four points, and Obama narrowly carried the seat.

— In the waning days of the 2012 Indiana Senate campaign, McLaughlin had Democrat Joe Donnelly and Republican Richard Mourdock tied – though Donnelly won by six points.

— McLaughlin’s final poll of the 2013 Massachusetts special election for GOP super PAC funder John Jordan showed Democrat Edward Markey and Republican Gabriel Gomez in a dead heat, despite other polls showing it a double-digit race. The other polls were right, and Markey won by ten points.

While McLaughlin’s miss in Virginia on Tuesday has little impact on GOP polling for the 2014 midterms – since the seat isn’t likely to factor into how the party spends its resources to maintain their House majority – it suggests the party hasn’t solved its polling problem.


McLaughlin does poll for a number of other members and challengers, including Carlos Curbelo in a battleground South Florida district and former New York Rep. Nan Hayworth, who is running for her old seat against Rep. Sean Patrick Maloney. They have also polled in the past for the NRCC.

Cantor spent more than $5 million in his losing bid – far more than Brat – through the pre-primary period in late May. The biggest recipients of his campaign cash included direct-mail firm Creative Direct and media buyer Mentzer Media Services, Inc.

Support the USO

Online mystery-ak

  • Owner
  • Administrator
  • Hero Member
  • ******
  • Posts: 250,510
Re: GOP pollsters missed big in Virginia
« Reply #1 on: June 12, 2014, 11:15:56 AM »
http://www.newsmax.com/Newsfront/News-Cantor-primary-loss/2014/06/12/id/576571/

Newt: Cantor Should Sue His Pollster, Campaign Consultants


Thursday, 12 Jun 2014 06:17 AM

By Elliot Jager

Eric Cantor can yet make a political comeback. In the wake of his defeat by tea party-aligned economist Dave Brat, the departing House majority leader needs to first listen to the voters' message that "power really does come from the people," writes former House Speaker Newt Gingrich.

He should also "sue his pollster and his consultants for malpractice," Gingrich advises. "They outspent the challenger 25 to 1 and lost. They spent almost as much on steakhouse dinners ($170,000) as Brat's entire campaign spent (about $200,000)."

Contrast Cantor's shocking loss in his Richmond, Virginia, district with Sen. Lindsey Graham's big win in the South Carolina Republican primary, writes Gingrich.

Both lawmakers were criticized by tea party opponents. "Graham understood the first rule of American politics, which is that the voter is King. The voter loans power to the elected official but they can always take it back. Wise elected officials always remember that power really does come from the people."

The senator regularly returned to his state. He spent time with his constituents, heard their complaints, "explained his policies and emphasized the points on which they agreed."
Voters in Cantor's district felt he'd forgotten them.

His attack ads against Brat only reminded them that they had an alternative. Where Brat signed a pledge developed by the Independent Women's Voice organization to repeal Obamacare, Cantor declined to do so.

"Clearly, repealing Obamacare will have to be a major Republican pledge this fall," Gingrich argues.

Cantor is a gifted man with an estimable political record. "Losing a race does not have to be the end of the line," writes Gingrich. He could yet become a governor, senator, or cabinet member.

People were angry not at Eric Cantor the man as much as they were signaling frustration at the failure of Republicans to mount an effective fight against President Barack Obama's policies.

Support the USO

Online mystery-ak

  • Owner
  • Administrator
  • Hero Member
  • ******
  • Posts: 250,510
Re: GOP pollsters missed big in Virginia
« Reply #2 on: June 12, 2014, 11:17:02 AM »
http://www.newsmax.com/Newsfront/Luntz-Cantor-Virginia-primary/2014/06/12/id/576575/

Frank Luntz: Cantor's Polling 'a Whopper for the Ages'



Thursday, 12 Jun 2014 06:49 AM

By Elliot Jager


Polls cannot forecast election outcomes with certainty. "They cannot and should not be used as infallible crystal balls," writes veteran Republican pollster and political consultant Frank Luntz in a New York Times op-ed.

Polling goes wrong when practitioners and clients forget that public opinion surveying is an art as well as a science. Republicans were unable to predict House Republican Majority Leader Eric Cantor's loss because the human element in the polling was missing; the sentiments that can't be captured by technology alone.

Cantor's pollster, McLaughlin & Associates, estimated that he would handily overcome tea party-aligned economist Dave Brat by 34 points. Cantor lost by 11 points. As mistakes go, "This was a whopper for the ages. McLaughlin didn't merely get it wrong; this was quantitative malpractice — a mind-blowing modern-day 'Dewey Beats Truman' moment."

Moreover, if it's true that Cantor's poll was disclosed to cower the opposition then "not only was the poll inaccurate, the tactic was inept" because it may have spurred Brat's camp to redouble its efforts, Luntz writes.

That said, polls are not infallible. Statistically, one in 20 get it wrong. Even when conducted by the book, 5 percent of all polls wind up on the wrong side of the margin of error.

What polls can do is provide insight into the thinking of a target audience. They are a snapshot of attitudes valid at the time the poll was taken. "Without qualitative insight — talking with voters face to face to judge their mood, emotion, intensity and opinion — polls can be inconsequential, and occasionally wrong," writes Luntz.

There is a limit to what can be illuminated by simply asking dispassionate "yes" or "no" questions. Today's pollsters provide not only public opinion surveying but also strategy and public relations. Yet to get a handle on voters' real aspirations and concerns pollsters need insights into, "What anxiety keeps them up at night? What will be weighing on their conscience when they pull the electoral lever?"

That can only be revealed through human contact, "asking the right questions of the right people, and demanding honest answers."

Support the USO


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf