Opinion polls are becoming an insidious aspect of American political life.
Increasingly political leaders and the media are being guided to decision-making by the tasking and often publication of polls. Any attempt to ban or inhibit them would be a violation of the Constitution, a flagrant violation of free speech. Given any interpretation of the first amendment, it is obvious that they are not only going to be with us permanently but their influence is likely to increase. So American political life is going to have to live with them.
But their nefarious influence is becoming increasingly obvious for a number of reasons. Simply by calling them “scientific” does not make them omniscient. They violate, of course, the first principle of any scientific investigation, that is, the ability to control the conditions under which an experiment is made. Those of us who have polled know all too well such problems as the syntax of the question, the method used to select a sample, or even the tone of voice of the inquisitor if it is done verbally.
It is true, of course, that with the enormous amount of talent and effort put into the development of polling techniques, they have improved over the decades in their ability to give us an image of opinion in an instant of time. But increasingly narrowing down the sample which is used, by “scientific” means, to get a representative opinion from a very small group is fraught. Therefore, first and foremost is the question of their accuracy. The recent total misreading of opinion in the Michigan primary indicates just how off the track a major poll can go whatever the skill of the pollsters.
Our second complaint about polls is perhaps even more important. They are, after all, a reading of a particular set of values of those polled at a given instant in time. In the increasingly fluid world of American public opinion on a growing variety of issues, it is obvious that even were such a reading accurate, it is only so for that particular moment in time. Within minutes or hours, the opinions of those polled – because of events or the progression of their own thinking – may, and indeed, are likely to change.
As important as the question of accuracy is, the even more problematical role of blow-back from the published results of polls is a more worrying consideration. One gets the feeling, and again it is difficult to evaluate, that increasingly our political leadership is being guided in their thinking about policy by polls. This presents a double jeopardy: with the accuracy of the polls always open to question, it is not certain that following the polls is the best way for a leader to appreciate the thinking of his constituents. But an even greater problem arises from the whole concept of leadership: a successful and valued leader must always in a democracy such as ours balance his decision-making. He has to coordinate the duality of being a representative of the people and their opinion but also with what is generally his particular and more extensive knowledge of the subject under consideration. That means at times he must exercise his independent judgment. Leadership, after all, has an obligation to lead, particularly during periods of crisis and public indecision.
All this speculation comes to mind after recently listening to a Fox News interview of their accredited pollster. He repeated the results of recent polling on the Republican and Democratic candidates for president. But rather than leave it at that, he then went on to strategize what would be or, indeed, should be the future actions of the candidates by establishing conflicting poles of opinion on the subject. There is something in the water, apparently, that leads any successful pollster – that is, successful in the sense he gets his bills paid – to go on to attempt opinion-making. That is, as the current language goes, crossing a red line.
Obviously, again, no government legislation should try to restrict political [and commercial consumer] polling. But it is time that a campaign against the too easy acceptance of polls and an extrapolation of their results to indicate policy was begun.