Out-of-Control Numbers

What to do when public polling becomes the story…
Public poll results make journalists happy. Candidates are winning, losing or in a dead heat. Approval numbers are up or down. There are statistics and charts to back it all up. Plus, the frequency of public polls is increasing as new polling companies pop up—just as the frequency of story deadlines is increasing (although, unfortunately, not because new journalism jobs pop up).
 
Public poll results do not make happy consultants, however. What makes a good story for a news outlet tends to send campaigns into overdrive. “Candidates will quote polls if they think it will help their candidate and a week later they will criticize the same poll if they think that it is hurting their candidate,” says Frank Newport, editor-in-chief of the Gallup Poll, one of the most respected public pollsters out there. “They will attack the mother of the pollster, the grandmother of the pollster, the ethnicity of the pollster—politics is vicious.”
 
Public polls usually ask different types of questions from campaign polling and often yield signi%uFB01cantly different results. On the state and local level, it seems as though new public pollsters are a dime a dozen and many employ smaller sample sizes and methodologies that many consultants question. The biggest complaint from frustrated campaign consultants: too often reporters don’t go beyond the top-line results, which means a bad poll can get front-page treatment.   
Brad Todd, of On Message Media, says pollsters often used by media organizations aren’t invested in particular races in the same way that campaigns are. “Public pollsters don’t have the pressure to get the race right,” he says. “You see a lot more uneven polling in public polls than you do in the private campaign polls because they don’t have a stake in the campaign—it matters to them to get a story.”
The advice from most pollsters is to keep an eye on the results, but don’t let them run your campaign. Campaign polls, which are usually kept out of public view, are most important in determining strategy. “The results you get in public polling, by and large, is pretty feeble stuff,” says Vic Fingerhut, a longtime Democratic pollster. “It’s kind of useful, but they don’t do anything near what we do in terms of very speci%uFB01c use of language and sophisticated message testing.” Consider the numbers “freebie data,” Fingerhut says, “I look at it, and it’s useful stuff to have, but I don’t normally react to it.”
But, in the new media world, campaigns can’t let news stories based on poll results slide by without comment. “You can’t let a bogus public poll sit there without much counter in the press,” says Todd. “In any sort of consumer activity we always look for the validity of an independent authority, and a lot of folks treat an independently conducted survey as that kind of ‘Good Housekeeping Seal of Approval.’”
Yet diverting resources away from substantive campaigning to rebut results could turn a public poll into
a self-ful%uFB01lling prophecy. “I advise my campaigns to keep their focus on things they can control,” says Todd. “Attempt to knock it down with your donors if you believe it’s methodologically %uFB02awed, but don’t spend too much time on it.”
There are times, however, when public poll results drive a storyline that really damages a candidate and the campaign has to mount an all-out offensive. Advisers for Sen. Richard Burr (R-N.C.) have lately put a lot of energy toward knocking down stories about his apparently low approval ratings based on numbers from the %uFB01rm Public Policy Polling. With the monthly polls driving stories about Burr’s vulnerability in 2010, his campaign strategist Paul Shumaker is actively questioning the impartiality of the numbers. “The name Public Policy Polling implies an independent, objective organization, but it is everything but independent,” Shumaker says. “It is clearly an extension of Democratic campaign operations.” The Raleigh-based polling agency asked respondents if they thought the North Carolina senator deserved a second term in of%uFB01ce or ought to “give someone else a chance.”
 
Of those polled, 49 percent thought someone else should govern while only 29 percent wanted for a second term for Burr. Shumaker faults the media for inaccurately conveying these results to potential voters. “It is hard for reporters to be pollsters,” says Shumaker, who believes journalists rely too heavily on top-line numbers rather than “internal information” like cross-tabulations.
Such criticism is nothing new to public pollsters and members of the news media. PPP, which serves Democratic clients on the federal, state and local levels, dismisses the charges that the numbers are inaccurate. “We regularly get attacked from the right and the left for telling the truth,” says PPP President Dean Debnam. “We are often telling Democrats that we work for that they are going to lose, not because we wanted them to lose but because we know that if they don’t change their strategy they will lose.”