And it’s not just about politics. Pick any topic these days, and there’s sure to be a poll on it somewhere. Whether you choose to believe the results, these barometers of public opinion have become the best way to study an increasingly complex society. But how much power do polls really have? Can they sway public opinion? Newsweek’s Sarah Childress surveyed the Gallup Poll’s editor in chief, Frank Newport on polling in the upcoming White House race, the bandwagon effect–and the secret of ‘Who Wants to Be a Millionaire?’ Excerpts:
NEWSWEEK: What got you interested in polling?
FRANK NEWPORT: I got interested in human behavior, particularly the social aspects, way back in high school. I’ve always been fascinated by the power of social norms, social divisions, social stratifications on human behavior. Nobody wears a hat that says, “I’m a jock,” “I’m a nerd,” “I’m a dweeb,” but everybody knew exactly where somebody stood in those groups, and their whole life was ruled by it. One of the primary mechanisms by which sociologists understand human social behavior is polling. It’s a great scientific tool for literally generalizing for millions of people.
How do polls work at Gallup?
In all instances we have a tremendous reservoir of polling history that we draw on. Now that we’re in an election season, we don’t have to start from scratch and ask, “Who are you going to vote for,” which isn’t as simple as it seems. For example, do you say you’re going to vote for “Bush, the Republican,” or “Bush”? There’s a variety of different, seemingly simple things that you have to decide.
What do you say to people who dismiss polling?
Some people have concerns about polling because they don’t understand how it works: They say, “How can 1,000 people represent millions?” or “I’ve never been called.” Or they say, “You can make up a question to get to anything you want.” More philosophically, should leaders even be paying attention to polls? Some even go so far as to say, people are stupid, who cares what they think? Polling is scientific and accurate. If done right, it’s a miracle. One thousand people can represent the responses you would get if you interviewed 100 million. Elections show that. I think there’s great wisdom in the collective opinions of the people–tremendous wisdom–and we need to pay more attention to it.
How can a mob be smarter than a single person?
It’s a remarkable process. There’s been research that shows the averaging out of different opinions typically comes out to be more accurate than any one person. That’s why, in the old Regis Philbin show, ‘Who Wants to Be a Millionaire,’ the ‘Ask the Audience’ lifeline was often more accurate than calling an expert. There’s something about averaging across the audience. If I were on ‘Millionaire,’ I’d always have chosen the audience response over calling a genius at home, because on average they’re more likely to answer the right response.
Do we need other people’s opinions to tell us what we think?
It’s unclear to the degree that people do that, but people certainly have an interest in what other people around them think and feel, and I think that’s legitimate. We talk to families and friends and say, “What did you think about Kerry’s speech at the convention?” and form a consensus. We should be able to take into account what people think and feel.
Can’t ‘group think’ also be harmful?
Often, group think is more wise than it isn’t. There’s no evidence that shows we don’t have any new ideas. We need out-of-the-box thinkers, and I think we have that. There’s no evidence that people won’t think differently because they’re constantly trying to modulate themselves to what others think. We have plenty of people with new ideas and new thoughts. Then the public, monitors them like a jury: Did these ideas percolate through, and did the public pick them up or not?
Can polls sway public opinion?
The research literature is unclear on that, especially when it comes to elections. But as George Gallup pointed out a long time ago, obviously not, because if there were a real bandwagon effect, every candidate would win with 100 percent of the vote. Once polls showed a candidate ahead by 10 points, everybody would say, well, I want to vote for the guy who’s ahead, and he’d be ahead by 20. Obviously there’s no real empirical evidence that there’s these massive bandwagon effects. But it would be all right if people did take into account what other people think and feel. It’s legitimate input.
Did the bandwagon effect come into play with John Kerry’s nomination?
He was behind, so obviously if people had just been paying attention to polls this year, Dean would be our nominee. [Dean] was leading in all the polls. When people came to actually vote, they voted for Kerry instead. It was more the actual vote than polling that really affected Kerry. Our own polling showed him down in single digits, so he should’ve given up, just based on polling. But Democrats, although few said they would vote for Kerry, changed their minds. That’s good evidence that polling doesn’t change elections.
In your new book, Polling Matters [Warner Books, July 2004] you note that politicians often view the public as “ballot cattle”–the group they appeal to only during election season. That doesn’t seem like much incentive to get people involved in politics.
That may be part of the problem. Voting participation is low. Part of that is the public is cynical because politicians seem to ignore them. There is scientific research that shows they misjudge their votes and are significantly off-kilter from the people they represent. Part of it is because they don’t use polling until they want to get re-elected. In between elections, they say, “I’m going to do what I think is best,” or listen to lobbyists or special interests. The most recent example was the Clinton impeachment. The representatives in the Senate had to make a big vote. A few of them courageously said, “I’m going to take a poll and see what the people think.” Others said, “I’m going to go off on a mountaintop by myself and meditate about what to do.” I don’t want you meditating, I want you trying to divine wisdom from the public.
Why don’t politicians feel they need polls?
There is a practical problem. Some have said, “I would pay more attention to polls, but I don’t have a good sense about what the polling says.” To the representatives’ credit, they do get confused. Lobbyists will come in saying, here’s a poll showing people favor this. In an hour, another lobbyist comes in with another poll showing people oppose it. After awhile, the politicians throw up their hands. That’s why politicians should have a scientific pollster so they don’t have to rely on lobbyists. I believe there are some Reps who philosophically believe they shouldn’t pay attention to the people. That’s what I call the trustee model-I’m sent off to do what I think is best. And some think the people are too stupid.
So the best candidate is really the “Pollster Candidate,” who does only the will of the people?
Certainly that’s the extreme example–the blank-slate candidate, who actually ran with no policies, no positions, no nothing, except to pledge that he or she would pay attention to the people. If a lobbyist comes to them, they say, hey, convince the people, and I’ll vote that way. We do need representatives–society’s too complex. But it’s gotten out-of-kilter. I think they slight the fabulously important input of the people.
Which presidential candidate pays more attention to polls?
It has become unfashionable for a candidate to say they listen to the polls. I don’t have any inside info on what degree either Kerry’s or Bush’s staff pays attention to polls. Publicly, Bush has said, “I will not and do not pay attention to polls in making decisions.” We do know his re-election campaign pays attention to polling, so certainly his campaign is not adrift without polls. That’s just his own philosophical position. Bush says he does “what he thinks is right.” I don’t know whether Kerry would publicly adopt a similar position or not.
Polls show that Bush and Kerry are neck-and-neck. Can you predict who’s going to win?
Only a fool at this point would say who’s going to win. We do know, however, that Bush is vulnerable. Based on all of our intensive Gallup poll analysis in the past, those incumbents who have breezed on to re-election since 1950 all were considerably ahead of where Bush is at this point. On the other hand, Bush is not as bad off as the last two incumbents who lost: Bush the elder and [Jimmy] Carter. Bush’s father had a 29 percent job approval rating [in] July of 1992. Bush’s is 49, so he’s clearly not in as dire straits as his father was. But it’s unclear whether Kerry can take advantage of that vulnerability and win. Kerry will get a bounce out of the Democratic convention. He’ll be ahead, unless history really is not followed, but then Bush will whittle away at that lead and get his own bounce in the Republican convention. So it’ll really be after the dust settles after Labor Day that we’ll get a better feel for where we stand.
