Alan F. Kay 1999
Public-interest polling is about issues – resolving community, regional, state, national, and international problems. What is needed is governance that will help make the world, or our part of it, work with consensus support as easily realizable as a small town-meeting can find a consensus should it wish to. We are now a nation of 300 million, a world of 6-plus billion. With such big numbers, finding consensus, when one exists, presents a problem of scale. What worked for the Founding Fathers when we were a nation of 3 million no longer works today, and if the truth be told, not then either. But it is worse today.
Public-interest polling requires a healthy mixture of common sense and the knowledge of experts. Questions must be fair. The interviewer reading the questions should not "lead the witness," to borrow a legal term or in polling lingo, should not "cue the respondent." Question-sets (not necessarily every question) should be balanced. The survey design team, consisting of polling and issue experts, should collectively represent a wide range of backgrounds and points of view. These teams should believe that the respondents understand quite precisely what they are being asked, a belief to be tested with later questions to make sure that the teams and the respondents both have it right.
We all know how to do this in a different context. If you get a confusing, ambiguous or unexpected answer when you ask anyone, even a colleague or a spouse, an important question and you really want to know what’s on their minds, what do you do? You ask some more questions until you get it straight. That leads to putting follow-on questions into a survey – or splitting samples to ask the same question a little differently in two half samples. That leads to serial surveys, always for ATI with different samples. The questions in follow-up (or serial) surveys may cover much of the same ground endeavoring to clear up the ambiguities. Once you’ve gotten the hang of it, you can clear up all the ambiguities in one or two additional surveys. In 33 surveys, over and over again, it worked every time for my company, ATI, from 1987 to1999.
Sponsors Role, Avoiding Spin
A sponsor usually pays a lot – $50,000 or $100,000 or more – for a high quality survey and a number of people are involved in it: pollsters, issue experts, interviewers, analysts, public relations people, etc. There is another big requirement for true public-interest polling. The sponsor has to figure out if the pollsters and others working on the survey are putting any spin on the product, and to stop the spin if they are. To get repeat business and referrals, pollsters like to please their clients. They are generally right as to what answers sponsors would prefer. Many policy organizations, politicians running for office, even major news media, enjoy a bit of supportive spin especially if they are not too aware that there is any. Spin can be introduced by small changes in question meaning at many points: in the selection, grouping, and wording of the questions, in survey design, in analysis of responses, in the wording and layout of the report that covers findings and conclusions, and in presentations made to the media. Though few in the public or the media seem to know this, it is easy to spot most of the spin to be found in many routine polls if you know a few things about polling. It may actually be quite difficult to make sure that pollsters and others working on analysis and presentation, out of habit or whatever, do NOT slip in a little spin – or sometimes – a lot. For high profile polls, designed to change the agenda of the nation, spin is frequently introduced by the many cooperating folks behind the scenes in such a way that none among them can be considered responsible and most may not even be aware that spin is present.
High Profile Polls Betray the Nation
For high profile polls, designed to change the agenda of the nation, spin is frequently introduced by the many cooperating folks behind the scenes in such a way that none among them can be considered the responsible party and most may not even be aware that spin is present. Examples of high profile polls that betrayed the nation are: (1) The Republicans Contract with America in the 1994 presidential campaign, (2) The National Issues Convention (NIC) Surveys, (3) The Washington Post Strikes Back Campaign in 1996, and (4) Ross Perot's Electronic Town Meeting in the 1992 presidential campaign (See Chapter 13 of Locating Consensus for Democracy where all of these are examined.).
Ongoing Testing of Findings
A further requirement of public-interest polling is that findings of highly supported policies may need to be considered tentative until tested repeatedly to determine whether support holds up. When first asked, the policy may be supported at consensus levels (by at least a 67% supermajority), but such support may not hold up under testing by wording variations, by news developments, by consilience with previous findings, by exposing respondents to pro and con arguments and by obtaining the reasons respondents, themselves, choose to explain their support. When ATI says a policy is a consensus finding, we mean that, including preferred features when these are critical, the policy has passed a substantial but reasonable number of these tests.
In thirty-three surveys the Americans Talk Issues Foundation found, on all of the issues we have looked at, overwhelming evidence that the legislation and policy choices most supported by Americans are characterized by four key adjectives: stable, consistent, pragmatic, and principled. Let’s look at what these words mean as applied to public-interest polling.
The public’s expressed desires vary slowly over the years, never abruptly unless a major event relevant to the issue occurs, at which point support may move in the expected direction, depending on the event, and then sometimes less than one might think. Even more amazingly – and this comes from the remarkable work of Page and Shapiro¹ – although the degree that different demographic groups: age, gender, income, etc., support a given policy may vary quite a lot, when one of those events occurs for which the whole populations’ support moves up (or down), the demographic groups all move together, in the same direction, roughly a proportional amount, so that on a trend chart (support vs. time), the various demographic groups move in parallel. Page and Shapiro¹ calls them "parallel publics." One exception sometimes occurs when a demographic group itself is a part of the question, e.g. a "race issue" which divides whites and African-Americans. For all other issues demographic groups do generally move in parallel.
¹Page and Shapiro, University of Chicago press, 1992.
This means first that no matter how you get to the question, as long as it has a little context to get respondents to focus on the issue, you will get the same answer and beyond that, the conclusions that can be drawn from data from a new survey never contradict the conclusions that were drawn from earlier surveys on the same issue (i.e. consilience) – again unless a major event relevant to the issue has occurred between surveys, and then in this case the change due to the event can be readily estimated.
Solutions the people prefer are pragmatic, not ideological. The public is much less polarized on policy than the ideological extremes of the two parties, Republican and Democrat, which is not to say they are somewhere in-between the two. Usually the public’s most preferred policies have other dimensions which neither the Republicans nor the Democrats nor the Reform party, mention, largely because it interferes with the financial support of special interests.
The people want real solutions for the larger problems. They do not opt for just the things that each person figures would help him/herself economically or otherwise, they want solutions that are good for everybody and comprehensive. Given an opportunity to choose visionary solutions, they often do. People are more idealistic than politicians and much more so than reporters expect. When assigned to cover politics and government, reporters usually cover government as politics. Their stories ignore or dismiss those views of the public that are not strongly promoted by a top national figure. They become cynical or at least skeptical, and expect little from "the people."
Elitists tend to believe that experts informing officials is the process to be used to make good governance decisions. I started polling, as a pretty good elitist, believing that. Experience, including experimenting with many of the variables of polling, always seeking to improve the quality of results, has convinced me that a kind of wisdom comes out of sensible collaboration of the people and the experts, which employs the principles of public-interest polling laid out above. Neither the people, nor the experts, alone can produce this result. I have seen that the input of the American people is essential to making governance and democracy work properly.
Elites Confirm Consensus Findings
at Low Cost
Here is a most remarkable and important conclusion: once consensus findings are established by the methods of public-interest polling, it turns out to be true empirically that such consensus findings can be readily and inexpensively duplicated by any competent pollster, without including information beyond the minimal context needed to make a question in any poll intelligible to respondents. This means that the consensus findings of public-interest polling do not apply just to those tiny percentages who have gone through a deliberation process but to the entire public. This ought to be a mind-boggling fact when it finally penetrates the thinking of pundits, political and social scientists, and organizations promoting public participation in political and public affairs.
ATI has found that the policies most wanted by people in issue after issue are startlingly at odds with the views of national leaders. Confronted repeatedly with credible and substantial evidence of the disconnect between their views on governance and the desires of the people, presidential candidates Gingrich, Bush (Sr.), Gephardt, Gore, Clinton, Perot, and virtually all of congress and the mainstream news media – just turn away. Each of these political leaders I’ve named specifically, and many others in the book, were approached in multiple ways with ATI survey results. None of them want to know that the reasonable preferences of supermajorities (67 + percent) of Americans differ from the desires of one or another special interest that officials across the political spectrum routinely enact into law. Politicians, pundits, and media editors believe that they can simply ignore poll results that differ from their understanding of political reality. Since the beginning of the republic they have been able to live comfortably with an unclear and, in hindsight, often distorted view of the public’s view. That is, until now. A consequence of unprecedented, massive and ubiquitous polling in 1998 on the Clinton sex scandals and impeachment has been that the stability and coherence of the public’s views on these matters have become clear to all. Politicians, pundits, and editors for the first time have been forced to deal with poll findings that contravene their reality. In view of their assumption that the public’s views are often fickle and foolish, these "opinion leaders" are very confused on the significance of the stability, coherence, and disconnect of the public’s view from their own "more informed" view of the matter.
The very existence of propositions that ATI has found to be consensus policies previously unknown or largely ignored by elected officials and the news media, may seem mysterious, exotic, and dubious. Several surveys may be required to prove out consensus findings and ascertain that preferred versions have been teased out and verified. Since this process has almost never been done in the past, most such consensus proposals so far remain little known. The book, Locating Consensus for Democracy contains dozens of examples of consensus proposals and how they were located.
When the public seems to favor a policy inordinately, ATI expert teams keep challenging the people to clarify how their preferred choices will handle all the real-world challenges to any policy. At each stage it is up to the expert teams to articulate a broad range of choices, including but not limited to choices that one or more of the team think meritorious, and let the public make its selection. This sometimes takes many surveys and several years. Without the persistence and stability of responses, this process would be hopeless. Examples of policies that illustrate the stability, pragmatism, principle and consistency of the disconnected consensus follow.
Next real examples>>> 2.2.1 The Economy