February 7, 2008 | Read Time: 4 minutes
It wasn’t quite “Dewey Defeats Truman,” but the fact that pollsters were flatly wrong about Sen. Hillary Rodham Clinton’s victory in the New Hampshire Democratic primary will probably rank a close second in the history of polling blunders.
Pollsters are still debating exactly what went wrong in New Hampshire. The fundamental problem in this case, however, wasn’t that the polls were badly carried out. The real problem is that people are counting on survey research to do the wrong things. And perhaps more worrisome, people are becoming more skeptical of what public-opinion surveys can accomplish because of their disappointment in the type of horse-race polling that got the primary results so wrong.
Foundations and advocacy groups often make this mistake when they look to polls for guidance on issues. Like political operatives and the news media, they too are looking for winners and losers — except that the emphasis is on proposals rather than politicians.
They’re frequently looking for a yes or no answer from the public on a specific, detailed policy idea. That’s a perfectly legitimate thing to seek, but nonprofit officials cannot expect a single horse-race-style survey question to do the job.
An election comes down to choosing between “A” and “B,” but public-policy choices on, say, health care or energy are rarely so clear cut. People often haven’t been thinking much about the issue and can barely understand, based on a brief survey question, how these ideas would work in practice. Even though people are generally happy to answer the question, what you’re really getting are their best guesses. It’s not that people are lying to the polls, but their views are far from solid.
Until people have some time to absorb an idea, hear pros and cons, and consider how they think it will really work in real life, they don’t give very reliable responses. When nonprofit officials and political leaders take these results at face value, they often end up fooling themselves. More than one proposal has gone down in flames because policy makers thought they had more solid support than they really did.
So when polls are wrong, you end up with the political commentator Arianna Huffington calling for Americans to hang up on poll takers, comparing them to “horoscopes and political betting lines.” You end up with Vice President Cheney dismissively saying “polls change day by day, week by week.”
You end up with advocates and policymakers using poll results as disposable talking points, a way of “proving” the public is on their side. And that’s the worst part of all this debate.
It makes people skeptical of the truly useful information that surveys can provide, by examining the public’s values, finding out what worries them, and asking them to set priorities for the nation. The next president will be faced with an unpopular war in Iraq, a health-care system badly in need of change, a troubled economy, and a federal budget that’s headed for a long-term disaster. It’s up to the next president to lead on these questions, not just “follow the polls.”
But the next president will lead more effectively if he or she understands the insight that surveys can provide.
The public often doesn’t understand the details of public policy the way that experts and advocates do. But people’s values and priorities, unlike their support for a candidate, generally don’t change overnight.
Those values and beliefs shape how people respond when they start seriously thinking about policy alternatives. And leaders simply cannot govern in a democracy unless they have some sense of what the public wants. If a policy is going to survive, it has to be grounded in the public’s sense of what’s right and wrong.
Surveys can show that. Properly crafted, surveys can tell the difference between knee-jerk reactions and settled judgments. They can tell the difference between serious concerns and back-burner problems, and the kinds of tradeoffs people are willing to make to solve them. That has little to do with predicting elections. But that knowledge is the raw material that lets leaders build sound public policy.
The danger is that the cynicism engendered by election polls will taint all surveys. For all their flaws, surveys are still one of the best available tools for figuring out what the public wants.
Nonprofit officials, policy makers, the news media, and the public all need a better sense of what surveys can and can’t do. And in the long run, true public engagement, bringing the public into the process, is a better way of giving voice to the public’s concerns. But in the meantime, surveys can make a real contribution to democracy — if we use them wisely.
Scott Bittle is executive vice president for public-issues analysis at Public Agenda, a nonprofit, nonpartisan research organization in New York. With his colleague Jean Johnson, he is co-author of the forthcoming book Where Does the Money Go? Your Guided Tour to the Federal Budget Crisis (HarperCollins).
