Push Me, Poll You
Push Me, Poll You
How you look at things.
Feb. 15 2000 9:30 PM

Push Me, Poll You

John McCain says supporters of George W. Bush have been "push polling" in South Carolina—calling voters on the phone and asking loaded questions designed to smear McCain. The Bush camp admits that it is asking tough questions but argues that they're part of a real poll, not a "push poll." This controversy, which has consumed the media for the past week, misses the point. Every campaign poll that asks about an opponent's flaws is a push poll. Its purpose is to figure out how to push voters away from the opponent—if not through the telephone, then through the television.


Over the weekend, the Bush campaign released the script of its survey. It asks whether you approve of McCain's "legislation that proposed the largest tax increase in United States history" or his plan to "increase taxes on charitable contributions to churches, colleges and charities by $20 billion." It also asks whether you would be more likely to vote for or against McCain after learning that his "campaign finance proposals would give labor unions and the media a bigger influence on the outcome of elections."

William Saletan William Saletan

Will Saletan writes about politics, science, technology, and other stuff for Slate. He’s the author of Bearing Right.

McCain's beef is that the poll "pushed" respondents toward a negative view of him. According to a McCain release, "Push polling is the practice, condemned by political professionals in both parties, of conducting a phony poll which actually attacks an opponent with false or misleading accusations." McCain spokesman Howard Opinsky told the Associated Press, "Push polling has nothing to do with how many people are polled. It has to do with the negative distorted information displayed as fact under the guise of a real poll." This complaint implies that unlike "phony" polls, "real polls" (conducted by "political professionals") don't distort information or attack the opponent.

The Bush camp draws the same distinction. Bush spokesman Ari Fleischer told reporters that the Bush phone calls were not a "push poll" but a genuine "survey" to "determine how much familiarity [voters] had with Sen. McCain's record." According to Fleischer, "Surveys are meant to figure out what people are thinking." The Bush campaign urged reporters to read political analyst Stuart Rothenberg's Feb. 14 column in Roll Call, which articulates this distinction. Rothenberg writes that real polls such as Bush's "seek to measure public opinion," whereas push polls are designed "to alter opinion" on behalf of a campaign that "is interested in delivering a message, not in measuring sentiment." Such "advocacy calls," unlike true surveys, "seek to change opinion by pumping up one candidate or raising questions about another."

This distinction is a sham, and the Bush survey illustrates why. Like other "real polls" designed by "political professionals," its purpose is not just to measure public opinion but to alter it, by figuring out how to "deliver a message" full of "negative distorted information" that will push voters away from McCain. Political pros aren't paid "to figure out what people are thinking." They're paid to figure out how to make people think what the campaign wants them to think. They're paid to tell the candidate what he can say that will persuade you to vote for him, or at least not to vote for his opponent. They're paid to gather information about you the way a hawk gathers information about a rabbit.

Look at Bush's survey. Do you approve of "the largest tax increase in United States history"? Do you want to "increase taxes on charitable contributions to churches, colleges and charities by $20 billion"? Are you more likely to vote for McCain because his "campaign finance proposals would give labor unions and the media a bigger influence on the outcome of elections"? Please. These questions aren't designed "to figure out what people are thinking." They're designed to figure out exactly how many South Carolina voters Bush can turn away from McCain by making each of these accusations.

Nor are the questions designed, as Bush's spokesman pretends, to measure voters' "familiarity" with McCain's record. Many of you are already familiar with McCain's efforts to make tobacco companies pay $500 billion for smoking-related medical costs. You already know that he's been trying to strip deductions from the tax code and to ban unregulated contributions to political parties. What you don't know is that these efforts, respectively, will impose "the largest tax increase in United States history," "increase taxes on charitable contributions to churches," and "give labor unions and the media a bigger influence on the outcome of elections." You don't know these things because they're not knowledge. They're spin. And the purpose of Bush's poll is to measure how many of you will swallow them, so he can decide which of them to dish out.

Is push polling worse than "real" polling? Yes, because it's more dishonest. In a push poll, Rothenberg explains, "there is little or no give-and-take between the caller and the respondent." The caller shoves his cattle prod into you and hopes you'll jump. In a "real" poll, however, the caller cares about your feedback. He wants to learn how high you'll jump when he inserts the prod. This may not strike you as much consolation. But of such things the ethics of "political professionals" are made. As Rothenberg puts it, "Part of [Bush's] poll involved a few questions testing reaction to potential Bush messages. That's standard practice in polling, and nobody ever complains about it." Well, perhaps somebody should. 

  Slate Plus
Dear Prudence
Feb. 8 2016 2:46 PM My Wife Won’t Stop Flirting on Facebook Dear Prudence answers more of your questions—only for Slate Plus members.