Polling Matters

CHAPTER 1

The Lure of Knowing What
Other People Think

At some personal level, it is pretty obvious that an awful lot of people care what other people around them are thinking. Almost every one of us does informal polling. We ask friends, neighbors, even people next to us on the bus what they think or feel about an issue or problem. We share opinions, listen to gossip, and get a general feel for the lay of the land—opinionwise—of those around us. Everybody likes to talk about their opinions, and we listen back as others give us their thoughts. In fact, gossip, discussion, and verbal interaction have been the mainstays of the human species since speech first evolved. There's a lot of speculation about why this should be the case, but it is probably correct to say that we as a species benefit from our drive to hear and understand what other people are doing and thinking. It keeps us in tune with our environment and helps us stay alert to developments that may affect us. People may deride gossip as negative, nasty, and counterproductive, but scholars tell us that gossip is a very important element of human social interaction.

The bottom line is that knowing what other people feel or think appears to be of basic importance to the species. Humans live with and around other people. Acquiring a knowledge of these people is an important way in which humans manage to survive, get along, and come together to accomplish common goals. Thus, I think one of the most important rationales for polling is the fundamental interest that humans have in the opinions of those around them.

Indeed, a social psychologist named Leon Festinger—one of the great minds in the development of social psychology in the 1950s and 1960s—developed a "theory of social comparison" which attempted to explain the interest humans have in the opinions of others. He argued that humans have an innate drive to compare themselves to others. Festinger said that we constantly seek a reference standard against which to analyze our own thinking. "Festinger postulated that there is a basic drive in human beings to evaluate their opinions and abilities; he stated once again that when physical reality checks are not available in making these evaluations the person will use others as a point of reference . . ." ¹ In other words, when it is not possible to check our attitudes, opinions, and feelings against a concrete reality (as is the case most of the time when it comes to attitudes and opinions), we are interested in comparing them to the attitudes, opinions, and feelings of others. As Festinger said, "An opinion, a belief, an attitude is 'correct,' 'valid' and 'proper' to the extent that it is anchored in a group of people with similar beliefs, opinions and attitudes." ² We are driven to want to know what other people think in order to put our own opinions in context.

At previous times in history, most residents of small villages or towns had little trouble following through on this drive. They essentially knew what everyone in their restricted social world was thinking. There was enough gossiping and sharing of opinions that most people were fairly knowledgeable about where those around them stood on the key issues of the day.

But there have been changes over the years in the ability of humans to compare themselves to people in the social systems around them. Human societies have gotten bigger. It is impossible, for the most part, to know what everyone in our social sphere is thinking. We don't have the social networks and highly frequent face-to-face interaction that we once did. Instead, there's been movement toward surrogate interaction brought about by technology—mainly radio, television, and the Internet.

Harvard political scientist Robert Putnam documents this transition in his fascinating 2000 book, Bowling Alone. He makes the case that Americans are increasingly less likely to engage in activities that bring them into contact with their fellow humans (exemplified by the decline in group participation in bowling leagues that forms the rationale for the title). Putnam amasses evidence to show that

across a very wide range of activities, the last several decades have witnessed a striking diminution of regular contacts with our friends and neighbors. We spend less time in conversation over meals, we exchange visits less often, we engage less often in leisure activities that encourage casual social interaction, we spend more time watching (admittedly, some of it in the presence of others) and less time doing. We know our neighbors less well, and we see old friends less often. ³

In other words, we don't spend as much personal, experiential time to find out what others are thinking as we may have had in the past, for a variety of reasons.

However, I don't think these facts of life suggest that there's less interest in being with and finding out about friends and neighbors than there has been in the past. On the contrary, the drive is still there. But in many ways, the structure of our society today encourages people to seek to fulfill their social comparison drive in different (and perhaps less satisfying) ways than the old face-to-face patterns that dominated in the past.

Mass media are a big factor here. Much of television, a lot of talk radio, and a good deal of the Internet, in one way or another, are an expanded version of old-style face-to-face talk and discussion. Indeed, as we move into the electronic digital age, maybe it is not surprising that aspects of the media that appear to fascinate us most are those that give us the chance to hear from or about other people. Much television programming today, from sitcoms to cable news, is a window into the lives and thoughts of others. Television (and movies) provide surrogate neighbors, friends, and families. In recent years, television has turned increasingly to talk programming and "reality" shows that allow us to observe and hear from and about other "real" humans. Radio talk shows have become important mechanisms by which Americans get their news and information about the world around them (particularly the political world). Some of the most popular features of the Internet are e-mail, chat rooms, and instant messaging that allow us to talk back and forth with others. People apparently still thrive on getting to know other people and like to tune in to find out just what other people are doing, how they are doing it, and what they're thinking about. They're just doing it in a different way.

I'm fascinated with local television newscasts—which in today's American society can be a prominent way in which we figure out what our friends and neighbors are doing and thinking. Television consultants point out that the on-air crew of the typical evening newscast in many ways represents a family setting to viewers: the father figure (typical male anchor), the mother figure (female anchor), the bratty brother or sister (weathercaster), and the visiting uncle (sportscaster). We tune in to the 6 and 11 p.m. news as much to spend time with these surrogate family members as we do to find out about the latest murder, fire, or car wreck.

In other words, the electronic mass media have helped meet the need for learning about others in a world in which there are millions of people and in which many individuals no longer live in the intense, highly networked, smaller social environments of the past.

Polling performs a parallel function in a different way. It compiles and compresses the opinions of millions of people. Polling gives us the ability to understand—fairly precisely—what the people around us think and feel about the key issues of the day. It provides the same types of insights into our neighbors that we might have obtained in days gone by from gossip at the village pub, but on an expanded basis that involves literally all of our neighbors.

When we polled people about polls (which pollsters do) in June 2001, for example, we found significant support for the idea that people like the content of polls:

•  76 percent of Americans were interested in polls about political campaigns and elections, including the presidential election (34 percent said they were very interested, and 42 percent said they were somewhat interested). Only 23 percent said they were not too interested or not at all interested.

•  There was an even higher interest in hearing about the results of polls "which measure how Americans feel about the major political issues of the day, including those on which Congress is debating and voting": 77 percent of those polled said they were interested in these types of polls, with only 22 percent not too interested or not at all interested.

•  64 percent of Americans were interested in polls about Americans' religious attitudes and behaviors, 85 percent were interested in polls measuring Americans' feelings about the economy and business and industry, and 66 percent were interested in polls measuring Americans' attitudes about the entertainment industry.

•  The highest interest level of all was in polls measuring Americans' attitudes about enduring social issues such as gun control, abortion, and affirmative action. A whopping 88 percent were interested in these types of polls, including 57 percent who said they were very interested. Only 12 percent were not interested.

This human drive to want to know about the opinions and feelings of others is certainly the reason why newspaper editors and broadcast producers use polls as a significant part of their daily news coverage. Most media gatekeepers are fairly cold-blooded when they make decisions on the content of their publications and broadcasts. They want material that will interest their readers and viewers and increase circulation and ratings. Thus, it's significant that these gatekeepers seem to be committed to the idea of getting the views of the common people into their news coverage. In the old days this was done with "man in the street" interviews, by which reporters provided flavor and texture to news coverage.

Polling today simply provides information from all of the "men in the street." The fact that polls have moved to a prominent position in the media firmament is confirmation of their interest to the average consumer. In a big, mass world, polling provides a shorthand way to figure out what our fellow humans are thinking and feeling.

As we will discuss later in this book, this interest on the part of humans to know about others has its perverse side. We often don't like it if we find that other people do not share our personal opinions and views. It is, I think, a love-hate relationship. We want to know what others are thinking, but we may not like what we find. Fundamentally, however, the fact remains that much of the reason we have polling today is that humans find it interesting and fascinating to understand the people around them.

THE SCIENTIFIC RATIONALE

Understanding things is the role of science. Scientists study their subject matter—insects, trees, molecules, asteroids, rock formations—because it exists. Mathematicians study the properties of numbers because they are there to be studied. By studying "things" (that is, matter, nature, natural processes, etc.), scientists add to the fund of human knowledge about the world. Scientists assume that this is a true and noble goal. The scientific desire to understand what goes on around us has been at the forefront of progress of the human species as far back as we have written and oral records.

The motivation of the social scientists, psychologists, and pollsters who study human beings for a living most certainly reflects this same sentiment. The human species forms a fascinating subject of study. For many, in fact, humans are the single most fascinating topic in the world.

My own initial interest in sociology and polling came about when I was in high school in Texas and became more and more interested in the ways the people around me were behaving. I found the status hierarchies and social patterns at school to be weirdly compelling. I wasn't interested in bugs or the planets or chemical reactions, but in people. What interested me most was the extraordinarily powerful impact that social categorization had on the daily lives of all of us in high school. There was no printed list or official rules that designated students as members of the jocks, nerds, cool kids, rejects, cowboys, and so forth, yet these informal social categories (and who belonged in each one) were well known and well understood by everyone at the school. One's positioning on the subjective ladder of popularity was so important that it could be a make-or-break factor in one's enjoyment of the entire high school experience (as we learn when school shootings give tragic witness to the power of rejection and feelings of isolation on the part of student loners).

Thus, for me—and most social scientists and survey researchers—the drive to study and understand human beings is part and parcel of the same motivation to acquire knowledge and understanding that has propelled science forward over the ages.

There are a wide variety of ways to study individual humans on a one-by-one basis. But there are very few ways to study large numbers of humans without developing some system for systematically collecting information about them. That's particularly true in modern societies, when we're talking about the analysis of tens and hundreds of millions of people. Polling is thus of particular interest to scientists who study people: it provides an effective, quick, and cost-efficient way to analyze very large groups of humans without having to extract measures from each one of them individually. It would take an army of anthro- pologists to find and interview all of the residents of a state or country (something the U.S. government attempts only once every ten years). Polling short-circuits that process, and thus provides great practical value to social scientists.

Polling also takes advantage of another very powerful fact of life. Humans have the unique ability to talk about themselves. (After all, language is one of the key things that separate us from our close cousins the apes.) Humans can self-report their own behavior and save the scientist/observer the time and trouble of having to constantly observe human actions him- or herself. This includes reports of actual behavior ("I went to church last Sunday") and the emotional orientations to objects which we usually call attitudes (one's reaction to the question "How do you react when I say the word 'abortion'?"). Humans can report on their own history and—with varying degrees of precision—predict their behavior in the future. Humans can also introspect and report on what they perceive to be the reasons behind their behavior.

Polling thus provides the scientist interested in studying large groups of humans a decided advantage in the scientific process of measurement and discovery. Rocks, asteroids, ants, and neutrons cannot cooperate directly with an investigator and talk about their own history, why they are doing or feeling certain things, or inform others on what they intend to do in the future. Humans can. Humans reflect, examine, remember, and project. Humans study themselves and—of course—know themselves better than anyone else. This opens up enormous possibilities. A subject that cooperates and can summarize and analyze itself—on demand—provides amazingly fertile possibilities for investigators.

Polling, which for the most part consists of asking people questions about their feelings, opinions, past behavior, and future behavior, takes advantage of this uniquely human ability. And because of the miracle of sampling, polling allows these measures to be obtained in ways that generalize to literally millions of people. Polling is in many ways uniquely situated as a major component of any scientific effort to study and understand the human species.

Polling thus has two primary benefits: it allows us to generalize—with a good deal of precision—to very large groups of people without having to study each of them individually, and it takes advantage of the ability of humans to self-report.

A tour through the journals of most branches of social science, particularly sociology and political science, reveals the degree to which polling forms the methodological basis for a great deal of what these sciences are about. The study of the fundamentals of politics and governance, race relations, gender differences, power, status, inequality, sexual behavior, child rearing, health, and so forth is greatly enhanced by polls that provide insights from large groups of people. Historians can only drool at the valuable information we would have if there had been accurate polls throughout history. What did the people of the Roman Empire really feel about condemning and crucifying Jesus Christ? How did the French populace feel about Napoleon? Did the people of France and England wholeheartedly support the idea of opposing the Germans in the trenches of World War I? Did the people of Japan support the expansionist dreams of their government in the 1930s and early 1940s? Did the Chinese people support Mao Tse-tung or Chiang Kai-shek? Even in recent years, scholars wonder what the people who live in countries with totalitarian regimes think about their leaders and the structure of their societies.

In the most general sense, the basis for science is mea-surement and description. That's exactly what polling does: it measures and describes the feelings, opinions, and projected behavior of the people living in specific social groups. Polling provides us a way of summarizing or typifying human societies based on what the people who live in those societies think and feel. Polling is thus an invaluable tool for those interested in studying humans and the ways in which they organize themselves and live their lives.

1. Edward E. Jones and Harold B. Gerard, Foundations of Social Psychology (New York: John Wiley & Sons, 1967), 312.

2. Leon Festinger, "Informal Social Communication," in Classic Contributions to Social Psychology, ed. Edwin P. Hollander and Raymond G. Hunt (New York: Oxford University Press, 1972), 340.

3. Robert D. Putnam, Bowling Alone (New York: Simon & Schuster, 2000), 115.

CHAPTER 2

The Bottom-Up Approach

We can turn now to what I believe is the single most important reason why polls are so valuable. Polling provides us with a way of collecting and collating a distilled, summarized wisdom based on the input of the members of a society. This wisdom can be used as the basis for altering and improving the direction of the societies in which we live.

This is based on the key proposition that groups of people often have more wisdom among them than does any one person alone. The collected opinions, observations, and attitudes of all of the individuals in a population provide a distillation of knowledge and insights that is more likely to be sensible and useful than the insights and knowledge of single individuals or small groups of people independently. There is a great deal that can be done with that wisdom—more than just measuring it, studying it, and talking about it. It can be used to help guide a society forward.

In a mass society, polling is the only practical way to bring this knowledge together. The power of polling lies in its ability to harness wisdom and apply it to the problems of governing societies and making decisions about what those societies should do.

This is a pretty radical thesis. It implies that insight, wisdom, and knowledge levels increase with larger and larger numbers of people. This flies in the face of the feelings of many who hold just the opposite thesis: that wisdom and knowledge decrease as groups become larger and larger. There is in fact a long history of distrust of the mob or group, with a concomitant reverence accorded the contributions made by individuals and brilliant, single-minded leaders. Here I'm arguing that the views of the mob, as it were, have the potential to be extremely valuable to all concerned, and that the views of individuals or small elites often leave much to be desired.

One of the ways to approach this issue, I think, is to look around us. There is something very natural about the derivation of great value from many discrete actions collected together. It appears as if nature itself operates such that the combined actions of millions of entities in a social system can ultimately work to produce the most adaptive and useful structural patterns and actions for that system. Similarly, while the attitudes of single individuals in human social systems may seem insignificant in and of themselves, one attitude coupled with another and another and another ultimately brings together a totality of thought that is much more than the sum of its parts. Every person in a human social system is distinct in many ways and has a different genetic inheritance. In addition, by adulthood, humans have lived through and experienced life in distinctly different ways, reflecting the results of their cultural exposure. When the results of all of these differences in background and exposure are brought together, it constitutes the basis for an extraordinarily powerful body of knowledge. And that knowledge is gathered and processed by polls.

The core principle here is straightforward. The bringing together of all of the experiences and knowledge of a group of individuals allows for a distillation of truth that is more profound than an alternative that involves only the experience and knowledge of a few. It is my conviction that in many situations no individual or small group knows as much about the real world in which problems originate (and in which they must be solved) as larger groups and populations. No one physician knows as much about a disease or treatment as do all physicians combined, and no one rocket scientist knows as much about the space shuttle as do all rocket scientists put together.

No single football coach or sportswriter can decide on the best college football team in the nation as fairly as a group of many coaches or sportswriters. No single corporate purchasing agent's views on the progress of the economy are as likely to be accurate as are the views of hundreds of purchasing agents amalgamated together. No central economic authority can determine the value of companies as efficiently as the actions of millions of stock buyers and sellers acting individually on the major stock exchanges. No juror is as likely to produce a fair decision in a court case as are twelve jurors with their collectively combined views. And, in the most general sense, no individual can make as effective and efficient a decision on the broad direction a society should take as the collected views of all that society's citizens.

It is this last point that seems to generate the most resistance from observers. Many well-meaning citizens feel that powerful or smart people—rather than the public and its collected insights—are in the best position to provide the information and understanding that a society needs to rely on for direction.

Of course, nothing is absolute. There are certainly situations in which individual guidance is exactly what is needed. No one argues that the opinions of a broad cross section of society can provide the same insights into the treatment of cancer as can the judgment of trained specialists. No one argues that the views of all of the people in a society are as valuable in making a decision on the course of a hurricane as are the insights of meteorological specialists, or that random samples of average citizens can provide meaningful insights into decisions on the selection of the proper flu vaccine at the beginning of the influenza season.

But the broad principle here is that the thoughts, opinions, and insights of larger groups of people in many cases have the potential to be more valuable than the thoughts, opinions, insights, or wisdom of just one person or a small number of people. Or at the least, they add significant value to the decision making of whoever is in power. And as I will discuss later, even in the realm of specialists, it is increasingly evident that the knowledge of larger groups of individuals in a given area (for example, all diabetes specialists across the world) can be more powerfully wise than the opinions of just one or two brilliant people by themselves (one or two diabetes specialists).

There is no shortage of examples. One of the most contentious issues facing the United States today is health care. Certainly, Harvard professors and legislative committee staffers who focus on health care can have encyclopedic knowledge of health care statistics and the intricacies of how health payment systems work. But these experts may never have set foot in a charity hospital, have probably never had to sit in an emergency waiting room for hours seeking diagnosis and treatment, and have never gone without medical help as a result of not knowing where to go or how to pay for it. Average Americans, on the other hand, have collectively seen it all: hospitals, bad doctors, bureaucracy, HMOs, Medicaid, and ridiculously expensive drugs. Their combined experiences could provide the basis for a textbook of health care wisdom far exceeding that of the experts.

It is thus no surprise that the health care plan proposed by the Clinton administration in the 1990s, guided by experts meeting behind closed doors, failed miserably. What was missing? At least in part, a strong reliance on the wisdom of the people. Hillary Clinton and her task force gave short shrift to the tremendous expertise lodged in the "collected together" insights of the people, and proposed a system that the people were unwilling to accept. Imagine how much better the proposed health care reform system might have been if every aspect of it incorporated comprehensive polling of the people who were expected to live with it, examining how individuals dealt with health care issues in the real world and what they thought might most effectively be changed to make the system work to the greater benefit of all involved.

Another example was suggested to me by the fascinating book Ghost Soldiers, by journalist Hampton Sides. ¹ The book deals with the daring World War II commando raid that rescued Allied prisoners being held by the Japa-nese in the Philippines. Some of the most intriguing parts of the book were Sides's descriptions of the ways in which the prisoners over their three- or four-year captivity created their own small-scale social structures and interactive patterns that maximized chances of survival. Small, closely bonded social groups of no more than three men gradually evolved. These small groups seemed to provide the most effective way of dealing with the exigencies of getting food and shelter and surviving interactions with the Japanese authorities and guards.

The prisoners, in other words, essentially evolved their own social structures that were crucial in maximizing their chances of survival. There was no autocratic control from the Japanese—in fact, there were few leadership directives at all. The prisoners were free to arrange their daily lives and social structures any way they wanted in all of the camps in which they were held.