Public opinion

What the people think when they’re really thinking

By James Fishkin

It’s hard to get a reading on public opinion from people whose information consists of sound bites and headlines. Immersing citizens in the detail of competing policy options can yield better results.

Most of the time, in most countries around the world, the public is not very attentive or well-informed about politics or policy. Social scientists have an explanation. They call it “rational ignorance.” If I have one opinion among millions, why should I pay a lot of attention to complicated public issues? My individual views will not have much effect in a large community or nation-state. We all have other, more pressing concerns, areas of life where we can, we think, make more of a difference.

Most public issues involve complicated trade-offs that are difficult to capture in sound bites for a public that is barely paying attention. Yet most governments purport to be democratic and open to listening to the public in some way. How are governments to listen? What are they to listen to, when most of the public is unengaged and uninformed? Many efforts to solicit public input are open to capture by organized interests, mobilized groups, or paid advocates, all of whom are happy to take the opportunity to impersonate public opinion as a whole to serve some interest—perhaps for a public purpose, perhaps for a private one.

So what are governments to do about public opinion? Some poll incessantly. But polls of an inattentive public often give little more than an impression of sound bites and headlines. Sometimes they even convey phantom opinions or “nonattitudes.” People just do not like to say they “don’t know,” even when that is the case; they will, almost randomly, select an alternative rather than admit ignorance. Philip Converse of the University of Michigan famously discovered this in the 1960s, and George Bishop of the University of Cincinnati illustrated it with polls about the fictional Public Affairs Act of 1975, about which people happily offered responses.

At least polls with random samples offer a picture of public opinion that is representative. Even this merit is lost in the current fashion of eliciting Internet commentary or holding town-hall meetings with open invitations. When the Obama administration created a Citizen’s Briefing Book and invited the public to select its top priorities online, voting showed overwhelming support for putting the legalization of marijuana at the top of the national agenda—in the midst of two wars and the Great Recession. There is a kind of paradox of openness. The most natural strategy for public input is simply to invite public comment from anyone who is interested. But such an open, unstructured process will reliably lead to unrepresentative results, open to capture by those who feel intensely. The rest of the public will be left out, and a distorted picture of public priorities will result. This happened with the briefing book, and it also happened with congressional town-hall meetings on health care reform, which were descended upon by impassioned groups in the summer of 2009.

When a random sample thinks

Deliberative Polling provides a simple solution to these conundrums. It takes scientific samples and engages them in ways that foster serious thinking about an issue—offering balanced and carefully vetted briefing materials, convening small groups with trained moderators, directing questions developed in the small groups to panels of competing experts, and collecting opinions before and after deliberation in confidential questionnaires, avoiding the distortion of social pressures for consensus. A sample of a few hundred people is large enough to be statistically meaningful and small enough to be practical, even when the incentives and expenses for gathering the people in the sample group are considered. This process has been tried about 70 times in 18 countries. The answers to more than two-thirds of the questions in these Deliberative Polls change significantly. Deliberative opinion is most often different from the initial “top-of-the-head opinion” one gets in conventional polls.

Energy choices in Texas

In Texas, beginning in 1996, the Public Utility Commission (PUC) required the eight regulated electric utilities in the state to consult the public about how they were going to provide electricity in their service territories. All eight utilities did Deliberative Polls in conjunction with the PUC. The results were filed with the “integrated resource plans” for each territory. Averaging over the eight projects, the percentage of the public willing to pay more on their monthly electric bills to provide wind and solar power rose to 84 percent from about 50 percent. A similar jump occurred for willingness to subsidize conservation. These results were the basis for major policy changes that allowed Texas to move, step-by-step, from last among the 50 states in 1996 in the amount of wind power produced to first in 2007. The public simply did not have good information on electric-utility matters before deliberation. The Deliberative Polls allowed the public to really consider the trade-offs and then provide policy makers with informed judgments from a representative sample.

Local budgeting in China

In Zeguo Township, Wenling City, China, local government is using Deliberative Polling to make budgeting decisions. They started in 2005, seeking informed feedback on how to prioritize infrastructure projects, and soon expanded the effort to collect opinions on the full range of budgetary decisions. These polls take place on a yearly basis and have spread to other parts of China.

In the first project, those polled were offered a choice of 30 infrastructure projects; local leaders pledged to implement the top ten supported by the public after deliberation. A sample of more than 250 citizens was recruited randomly and gathered for a full day of deliberation in an area school. Local teachers were trained to moderate the discussions, and each project had expert advocates who answered questions in the plenary sessions following the discussions. Local leaders had expected that the public would support the “image projects”—a fancy town square and some of the large highway projects. But the public’s priorities after deliberation were clean water (sewage-treatment plants), a comprehensive environmental plan, and a “people’s park” for recreation. These projects were all built, and the process became well-known in China. Just as it did in Texas, the Deliberative Poll provided representative and informed input and solved the problem of how to consult the public. The local government had previously held kentan, or “sincere, heart-to-heart,” meetings. The Deliberative Poll was attractive because it provided a scientific way of consulting the public that represented the whole population, not just the more intense groups or the more notable citizens. It was, in effect, scientific kentan.

Enlarging the process

In the last three years, I have worked with Reframe It, a start-up headed by my son, Bobby Fishkin, to combine Deliberative Polling with online annotation. The process was a winner of the Harvard Business Review/McKinsey M-Prize for Management Innovation. The idea is simple. A Deliberative Poll of the relevant population is followed by online commentary from the entire population about the options, the briefing materials, and the results. Those online comments are rated by the random sample that has deliberated. In that way, the commentary process is not open to capture. An informed microcosm sifts the priorities. Then the options are reformulated by an advisory group in light of the highly rated comments, and the process culminates in a second Deliberative Poll on the new options.

This process worked well for considering employee issues within a large corporation, and plans are afoot to bring it to the public. It is unique among methods of consultation in providing representative and informed data from random samples with an opportunity for the rest of the population to deliberate. The commentary process helps ensure that the questions are the right questions. The deliberative process helps ensure that the answers are the public’s most thoughtful answers.

Our political system was born with a vision of representatives deliberating and “refining and enlarging the public views,” as James Madison said in Federalist 10. We have since seen that citizen participation has too often devolved into sound-bite manipulation and impression management—in effect, we have gone from Madison to Madison Avenue. With these deliberative experiments, we can show that the public can refine and enlarge its own views, if only we give it a good chance to think about the issues.

About the author(s)

James Fishkin is the Janet M. Peck Chair in International Communication and the director of the Center for Deliberative Democracy at Stanford University. His most recent book is When the People Speak: Deliberative Democracy and Public Consultation (New York: Oxford University Press, 2009).

More on Public Sector

The next-generation operating model for the digital world

Article - McKinsey Quarterly

What makes a CEO ‘exceptional’?

Article - McKinsey Quarterly

Three game changers for energy

Article - McKinsey Quarterly

How functional leaders become CEOs