PUBLIC OPINION

You have probably heard the saying, “perception is reality.” In much of our lives, what we perceive is the only reality that matters. If you perceive a homeless man on the street corner to be dangerous, you will behave accordingly, perhaps crossing the street to stay at a safe distance from him. Your perception, not the homeless man’s actual intentions, determines your behavior. Our perceptions are subjective, even when we might think we are simply observing objective physical reality. (Recall, from Chapter 6, the experiment in which participants apparently altered their standards for the colors green and blue.) As research psychologist Drew Westen notes, people essentially “twirl the cognitive kaleidoscope until they get the conclusions they want.”¹ We do not simply take in facts through our eyes and ears; we make inferences and derive meanings, and those inferences affect what we “see.”

We make judgments about other people all the time, and we base these judgments on our perceptions. Think of the many times you rely on your perceptions. If a recent acquaintance walks past you without saying hello, you may perceive the action as coming from either rudeness or preoccupation. A fellow student who expounds loudly and extensively in class may be seen as either obnoxious or extremely intelligent. You may perceive that people are angry or sad, that they are dishonest or sincere. Naturally, such perceptions are likely to affect your actions. We all continually try to interpret other people’s motives so we can explain and predict their behaviors. In many ways, these perceptions provide the foundation for a public’s opinion. And these perceptions can be very important. In trials, perceptions of the accused as well as the victim can influence jury members and have important implications for the outcome. When you apply for a job, the employer’s perception of you can provide you with a job opportunity or keep you in the ranks of the unemployed.

Our perceptions help us understand the world around us. We try to predict what others will do or say and rationalize their motives to provide explanations for their behaviors. We constantly perceive our world, and we rationalize what is going on in that world based on our perceptions.

As Walter Lippmann describes, there is a world outside, and there are the pictures in our heads. He argues:

Whatever we believe to be a true picture, we treat as if it were the environment itself. … For the real environment is altogether too big, too complex, and too fleeting for direct acquaintance. … The analyst of public opinion must begin then, by recognizing the triangular relationship between the scene of action, the human picture of that scene and the human response to that picture working itself out upon the scene of action.²

This chapter concentrates on perceptions of other people’s opinions and how they influence the public opinion process. Psychologists tend to construe “perception” as a process: the processing of information received through the five senses. We use the term in a way closer to its usual conversational meaning: a “perception” is a summary attitude based (perhaps not very accurately) on our past and present sensory information. We now turn our attention to this fascinating world of perception—a world that is both internal and external to ourselves, one that not only influences what we think or say but how we act. In essence, the perceptual world is the world of public opinion.

Human beings face physical and cognitive limitations that constrain our perceptions, including our perceptions of other people’s actions and motives. Some of these are obvious. As George McCall and J. L. Simmons note, the “physical characteristics of our sense organs impose limitations on both the breadth and the acuity of our perceptions.”³ We can only see so far, and we cannot clearly see small objects at a distance. In addition, our perceptions are limited by our physical perspective: if we are looking straight ahead, we cannot see behind us. If we look to our left, we cannot see to our right.

But as McCall and Simmons argue, our cognitive limitations matter even more. Our perceptions are subject to our selective attention and inattention. For example, people tend to pay attention to social objects (people and events) that are most relevant to their lives and to the obviously unfamiliar. Attending to these things, we may be unable to “see” other people and events, as surely as if we turned our heads away from them.

Our perceptions of others are always incomplete and are generally less than accurate.⁴ As Jerome Bruner comments, we tend to “‘recode’ into simpler form the diversity of events that we encounter” and so “not only is information lost, but also misinformation is added.”⁵ People actually tend to fill in events that “should” have happened, given their overall perceptions. Magicians often rely on these processes to convince audiences that they have seen the impossible. Most of the time, the processes are ubiquitous yet almost unnoticeable.

Suppose we are talking with a friend while attending a soccer game. We will have to narrow our focus so that we can alternate our attention, in order to carry on the conversation but also keep track of the main occurrences during the game. We will then alter the information we receive from the game in order to simplify this perceptual challenge. For example, we will monitor crowd noise to get an idea of what is happening on the field. We may begin to clap whenever the people around us clap, with only the vaguest sense of what they are clapping about. If they seem really excited, we may turn our full attention to the field, where (say) a player is sprinting toward the goal, and semiconsciously attempt to reconstruct the last few seconds of play that led to this scoring chance. Bruner adds that people often turn to “technological aids” to manage their attention processes.⁶ Indeed, we will probably “keep an eye” (or some fraction of an eye) on the scoreboard to judge how close the game is, how much time is left, and whether we should pay closer attention. We may even think that we have “multitasked” almost flawlessly, focusing on the conversation and also taking in the game—although a few well-chosen questions could reveal just how selective our attention was. People are always multitasking, far from flawlessly. Our perceptual limitations—and our mechanisms for coping with it—can restrict, confine, and help define public opinion.

Crucially, yet often subtly, people’s opinions, as well as their behaviors, are shaped by comparisons with others, including perceptions of others’ opinions. Even the influence of group norms (discussed at length in Chapter 6) largely depends on how people perceive those norms.

One way our opinions and behaviors are shaped is through socialization: the lifelong process by which people learn norms, beliefs, and behaviors from the people around us. In childhood, parents and close adults play a large role in socialization. There are as many forms of socialization as there are kinds of beliefs and behaviors. In this book we are especially (but not exclusively) interested in political socialization: how people develop particular ideological orientations and policy preferences. Political socialization is enormously difficult to study because it is very hard to track individuals over time, from their earliest political experiences (e.g., shaking a congressperson’s hand at age eight) to their adult voting choices. Moreover, researchers cannot readily distinguish between the social influence of trusted adults and the independent impact of political events themselves.

One of the authors of this book recalls a specific example of the impact of political socialization on his life in 1973, when he heard his first political joke. His father said, “Behind every Watergate there is a Milhous.” “Milhous” was President Richard Nixon’s middle name, so the pun meant that Nixon was responsible for the Watergate scandal (he resigned the following year). In one sense, the coauthor’s family and other people around the country were reacting to the emerging revelations about Nixon’s involvement. In another sense, the actual events were somewhat irrelevant. Everything the coauthor learned about Watergate he learned from his family. His perceptions and opinions depended in large part on theirs.

Several scholars have tried to document the political socialization process.⁷ For example, political scientist Paul Abramson has studied how people’s party identification, feelings of political efficacy, and trust in government change during their lives. Some observers have supposed that people tend to become more conservative, and therefore to move toward the Republican Party, as they age. Abramson concluded that on the contrary, people tend to keep their party affiliations even as they grow older, find partners, and advance in their careers. Similarly, in recent research Yair Ghitza and Andrew Gelman conclude that people’s propensity to vote for Democratic or Republican presidential candidates is most strongly influenced by events that occurred when they were between ages fourteen and twenty-four; events that happened at age forty or older make far less difference. (Ghitza and Gelman use presidential approval ratings to measure events, so their model focuses on when people were socialized, not with whom they were socialized.)⁸

Interestingly, Abramson found that young adults feel more efficacious (able to influence the course of government) than do older adults. Perhaps parents and teachers try to express more efficacy to their children than they actually feel, or perhaps youths tend to be more optimistic. However, later research finds that young adults in recent years tend to have less efficacy and trust in government than young people did in the 1970s and 1980s.⁹ This result could mean that young people are gradually learning cynicism from their parents. It could also mean that Americans of all ages are becoming more skeptical about government, for whatever reasons. As we said before, political socialization is difficult to study.

Leon Festinger’s discussion¹⁰ of how social comparison informs opinions is so important for our purposes that we summarize his major points in Box 7.1. Festinger argues that when we cannot readily test our opinions against physical reality, we resort to comparing ourselves, our opinions, and our beliefs with the opinions and beliefs of others around us. How we make these social comparisons depends on various factors, some of which are mentioned in Box 7.1. Festinger argues that social communication results from this desire to test our opinions, and further, that people strive for consensus: agreement on what opinion is correct or valid. Of course in some cases, people comfortably accept some disagreement, but they prefer to have the widest possible range of shared social facts. People generally react to disagreement either by changing their opinions to achieve the desired consensus or by changing their relationships—ceasing to use those people or groups who disagreed as a basis of comparison. In social comparison theory, social influence does not depend on the existence of a social norm. Rather, we influence each other (and norms tend to emerge) as we look to others for comparative social appraisals, even when there is no real social comparison information.¹¹

Kruglanski and Mayseless¹² present a more complex view of motives for social comparison. They identify three distinct motives: fear of holding an invalid (inaccurate) opinion; need for cognitive structure, or the “desire to possess some knowledge … , any knowledge as opposed to confusion or ambiguity”; and the wish to maintain beliefs that are pleasing, even if not entirely accurate. Kruglanski and Mayseless find that people seek out particular others for comparison based on their goal. For example, people who seek an accurate opinion on a controversial issue (perhaps the economic effects of tax cuts) may actively seek out dissimilar people most likely to hold opposing views. However, people who want to hold a pleasing belief will probably seek out similar people who can be expected to tell them what they want to hear. People who seek cognitive structure with regard to an issue also are likely to seek out similar people to tell them what to think. If you think that accuracy should matter most, two of these three motives may seem rather shallow to you. Like it or not, probably everybody holds each of these motives toward some issues, sometimes.

BOX 7.1

Festinger’s Social Comparison Theory

Here are some of the hypotheses and corollaries from Leon Festinger’s original statement of social comparison theory that have proved especially useful (we have paraphrased for clarity):

• Hypothesis 1: People have a drive to evaluate their opinions and their abilities.

• Hypothesis 2: To the extent that objective, nonsocial means are not available, people evaluate their opinions and abilities by comparing them with those of others.

• Hypothesis 3: The more different another person is in opinion or ability, the less likely people are to compare themselves to him or her.

• Hypothesis 6 and corollary: The cessation of comparison with others (because they differ too much) will be accompanied by hostility or derogation in the case of opinions, generally not in the case of abilities. (People tend to “write off” people with whom they sharply disagree, not so much people who are more or less talented.)

• Corollary 7a: The stronger the attraction to a comparison group, the stronger will be the pressure toward uniformity concerning abilities and opinions within that group.

• Corollary 7b: The greater the relevance of the opinion or ability to the comparison group, the stronger will be the pressure toward uniformity concerning that opinion or ability.

• Hypothesis 9: Where there is a range of opinion or ability in a group, people close to the mode (most common position) of the group will have (1) stronger tendencies to change the positions of others, (2) relatively weaker tendencies to narrow the range of comparison, and (3) much weaker tendencies to change their own position, compared to those who are distant from the mode.

This last hypothesis has been modified by other researchers, such as Elisabeth Noelle-Neumann.¹ Some studies have found that to maintain a distinctive group identity, groups sometimes adopt an opinion that is more extreme than that of the mode of the group. This more extreme opinion helps the group differentiate itself from other groups.² However, many of Festinger’s hypotheses have been confirmed in multiple studies that show the importance of social comparison in opinion formation.³

SOURCE: Leon A. Festinger, “A Theory of social Comparison Processes,” Human Relations 7 (1954): 117–140.

¹Elisabeth Noelle-Neumann, The Spiral of Silence. Public Opinion—Our Social Skin (Chicago: University of Chicago Press, 1984).

²Serge Moscovici, La psychanalyse, son image et son publique (Paris: Presses Universitaires de France, 1961); Serge Moscovici, “The Coming Era of Representations,” in Cognitive Analysis of Social Behavior, ed. Jean-Paul Codol and Jacques-Philippe Leyens (The Hague: Martinus Nijhoff, 1982), 115–119.

³A summary of the hypotheses and recent support for them is found in K. Corcoran, J. Crusius, and T. Mussweiler, “Social Comparison: Motives, Standards, and Mechanisms,” in D. Chadee, ed., Theories in Social Psychology (Oxford, UK: Wiley-Black-well, 2011), 119–139.

Since Festinger, several theoretical frameworks have treated the role of perception in public opinion formation. These perspectives share basic assumptions: that individuals care what others think about public issues; form perceptions of what others think; and to an extent, modify their own opinions or behaviors, or both, on the basis of these perceptions.¹³ Together, these perspectives construe our everyday lives as rich tapestries of “interwoven roles, groupings and social norms to which the individual continually adapts, obtains information and provides information for others.”¹⁴

This section examines seven public opinion perspectives: (1) looking-glass perception, (2) pluralistic ignorance, (3) disowning projection, (4) conservative/liberal/ideological biases, (5) false consensus effect, (6) impersonal impact and unrealistic optimism, and (7) the third-person effect. (Self-censorship, a burgeoning area of research, is explored in more detail later in this chapter, as part of the spiral of science discussion.) These perspectives examine different parts of the process and, in fact, define public opinion in different ways. Some perspectives appear to present public opinion as the majority opinion; others present it as generated within individuals.

Because public opinion research spans several academic disciplines, including sociology, social psychology, political science, and communication, some of these perspectives have emerged from parallel investigations in different fields. Where different terms have very similar meanings, we have discussed them together or have used the most commonly employed label. We urge you, in your own research, not to limit your literature explorations to one field. At first glance it may seem as if the discussion below is simply a potpourri of terms, definitions. and descriptions. In fact the various approaches taken toward understanding public opinion can be seen as building blocks in a young field as we attempt to understand the complexities of perceptions and opinions. Much rich understanding can come from incorporating the ideas of scholars working in multiple disciplines.

The typology developed by Glynn, Ostman, and McDonald,¹⁵ shown in Figure 7.1, summarizes the differences among several of these approaches. One corner of this typology deals with people to whom Glynn and associates refer as “hard core.” These are people who hold minority views and are aware of the fact that most people disagree with them. Because this chapter is primarily concerned with how others’ opinions affect our own, we do not consider “hardcore” people here.

Looking-glass perception can be simply defined as “the belief that others think the same as oneself.”¹⁶ Research has found that in many situations and for many public or social issues, most people believe that most other people have opinions similar to their own. Often this perception is correct; on uncontroversial issues, most people may guess—rightly—that theirs is the majority opinion. Indeed, in the typology, the term applies to these correct inferences. However, on divisive issues, some people in the minority may wrongly perceive that most people agree with them. In this case, looking-glass perception becomes “false consensus,” which we discuss further below.¹⁷

SOURCE: Adapted from Carroll J. Glynn et al., “Opinion, Perception, and Social Reality,” in Theodore L. Glasser and Charles T. Salmon, eds., Public Opinion and the Communication of Consent (New York: Guilford Press, 1995).

The term “looking-glass perception” does implicitly suggest a particular source of this belief: people examine their own opinions and project them on other people. As Ruben Orive¹⁸ argues, when individuals cannot conveniently compare their opinions with actual others, they may engage in “social projection” to create a “self-generated” consensus. That is, people (unconsciously) assume that other people agree with their opinion and use that putative agreement as another reason to hold the opinion. Of course, social projection is only one way that people might decide that others agree with them.

A modern-day example of the looking-glass perception is the ever-present “selfie” on social media. The photos we take of ourselves and then share online and use as our profile pictures enable us to develop a sense of self based on the perceptions of those with whom we interact.¹⁹ Did your last selfie get a lot of likes? How did that impact your perception of self?

According to Glynn, Ostman, and McDonald, pluralistic ignorance is among the most extensively researched concepts in the perceptual influence literature.²⁰ In their framework, pluralistic ignorance occurs when people who hold the majority position on an issue incorrectly perceive it to be the minority position (see Figure 7.1). Several scholars use the term pluralistic ignorance to describe all misperceptions of others’ opinions.²¹ Richard Schanck originally labeled this phenomenon “misperceived consensus” or “misperceived sharing.”²² For example, suppose you believe that abortion is wrong yet perceive that your friends think abortion is appropriate in most situations. If, in fact, most of your friends actually think abortion is wrong in most situations, this is a case of pluralistic ignorance. Thus, pluralistic ignorance entails people underestimating the proportion of others who think, feel, or act as they themselves do.

Other definitions of pluralistic ignorance refer to either underestimating or overestimating other people’s agreement with one’s opinions.²³ (Glynn, Ostman, and McDonald would typically call the latter case “false consensus.”) For example, Hubert O’Gorman and Stephen Garry describe pluralistic ignorance as the “false ideas held by individuals regarding the groups, social categories and collectivities to which they belong.”²⁴ O’Gorman and Grady assert that “even members of small and relatively cohesive groups frequently misjudge the values and attitudes of other members,” but that pluralistic ignorance is more common in larger, more impersonal settings. They add that in times of “accelerated social change it tends to become extensive.”²⁵ (O’Gorman and Grady do not elaborate, but presumably in times of change, it is difficult to tell how many other people have or haven’t changed their opinions.)

Often, pluralistic ignorance is tested by asking someone’s opinion on a particular issue and then asking the same individual his or her perception of majority opinion. For example, in their 1970 study, O’Gorman and Garry²⁶ asked white respondents around the country whether they favored “desegregation, strict racial segregation, or something in between.” Then they asked, “How about white people in this area? How many of them would you say are in favor of strict segregation of the races—all of them, most of them, about half, less than half, or none of them?” They also asked a similar question about “desegregation.” As shown in Table 7.1, most white respondents, even in the South, said they did not personally support segregation. However, white respondents tended to markedly overestimate support for segregation. Pluralistic ignorance appeared in every demographic group O’Gorman and Garry examined, although some variables such as education were associated with both people’s opinions and their perceptions of other people’s opinions.

TABLE 7.1. White Support for Racial Segregation and Whites’ Perception of White Support for Racial Segregation in Area, by Region and Education

SOURCE: Adapted from Hubert O’Gorman and Stephen L. Garry, “Pluralistic Ignorance—A Replication and Extension,” Public Opinion Quarterly 40 (1976): 452, 454, Tables 1 and 2.

¹Responded that “half,” “most,” or “all” whites in the area favored racial segregation.

²Responded that “most” or “all” whites in the area favored racial segregation.

⁵Excludes respondents who did not answer at least one of the two questions reported here.

⁶Alabama, Arkansas, D.C., Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia.

A recent study by Leviston and colleagues published in Nature Climate Change demonstrated a strong false consensus effect for people’s estimates of others’ beliefs in climate change. The authors conducted two surveys twelve months apart in Australia with over 5,000 participants, who were asked their own opinion about the nature of climate change and then asked to estimate the general population’s opinion. The results demonstrate that people overestimate the number of people who reject climate change in the general public. The authors also conclude that people with high false consensus are less likely to change their opinions.²⁷

Another recent study of pluralistic ignorance at work examined college students’ perceptions of “hooking up.” Both women and men rated their peers as being more comfortable with hooking up than they were themselves. Men expressed more comfort with hooking up than women, but both overestimated the other gender’s level of comfort with these behaviors.²⁸

Glynn and associates note that research in pluralistic ignorance generally has provided little or no explanation for it. “Pluralistic ignorance merely describes the coincidence of an issue with inaccurate perceptions.”²⁹ Dale Miller and Cathy McFarland suggest that fear of embarrassment is a key motivation behind some cases of pluralistic ignorance.³⁰ If you believe your opinion is socially unacceptable, you may state the opposite opinion, and other people may do exactly the same, not realizing that all of your opinion expressions are equally insincere. You may remember the folk story “The Emperor’s New Clothes,” in which the emperor supposedly is wearing a spectacular suit of clothes invisible to stupid people. In fact, he is naked, but nobody wants to say so for fear of being thought stupid—until a young child calls out the truth, that he is wearing nothing at all. Miller and McFarland produce experimental evidence for a similar dynamic, in which people conform to what they believe is majority opinion, while wrongly believing that other people are much less conformist than themselves. A similar account may help to explain O’Gorman and Garry’s findings. Some whites may have privately turned against racial segregation while being reluctant to say so; they may sometimes even have expressed support. But this is only one possibility. Another is that supporters of segregation were more vocal than opponents, creating the illusion that they were more numerous. And yet another possibility is our next topic: disowning projection.

Disowning projection is the pot calling the kettle black, when the kettle is shinier than the pot. More formally, as Norman Cameron defined it in 1947, disowning projection occurs when an individual “tends to attribute selfish motives, evil intent and stupid attitudes to others and to disclaim them for oneself, even though objectively the reverse of this may seem true to an impartial observer.”³¹ As Cameron describes:

A child accused by a playmate of some act, attitude, motive or characteristic, which his peers condemn or ridicule, commonly denies it and immediately ascribes it to his accuser—”I am not! You are!”—or shifts the accusation at once to a scapegoat whom everyone dislikes. … The average adult still uses direct denial and counteraccusation when he is surprised into it, and when he is frightened or angry. When a person is normally on guard and neither angry nor frightened, his denials, counteraccusations and scapegoating are carried out by indirect techniques—by implication, by stressing opposite trends, by distracting others, etc.³²

Thus, disowning projection is common, although how common depends on the person and the circumstances. You may know a classmate who habitually complains about how competitive other students are, when he or she is the most competitive student of all. Cameron’s discussion focuses on pathological cases: people suffering from schizophrenia and paranoia often exhibit disowning projection, “reacting to their own behavior as if it were the behavior of others.”

O’Gorman and Garry’s evidence of pluralistic ignorance on segregation might be partly explained by disowning projection, or at least by social desirability bias. By 1970 many whites had seen violent attempts to defend segregation and had decided that it was wrong—or at least socially unacceptable—to side with the defenders. Yet many whites remained deeply uncomfortable with black people. Some may have dealt with this dissonance by saying, in effect, “Of course I don’t support segregation, but those other people. …” Such people may actually have exaggerated other whites’ support for segregation (true disowning projection), or they may have accurately reported widespread support for segregation while refusing to admit to it themselves.³³ In either of these related accounts, some people’s expressions of opposition to segregation reflect social desirability bias—a desire to give the socially acceptable answer—rather than true opinion. At first glance, it may seem nonsensical to suggest that “social desirability” explains why many white respondents cast other whites as more supportive of segregation than they themselves were. But this sort of moralizing is not uncommon. The classmate who hypocritically condemns other students’ competitiveness is appealing to a widely shared value that people should not be so competitive, even if they often are.

This explanation is almost the opposite of the pluralistic ignorance account given above, which posited that many white people were less supportive of segregation than they seemed to be. Which account is correct? We suspect that both accounts have some validity for various people, but we cannot go back to 1970 to test them. Even if we could, we would be hard pressed to learn what people were saying about segregation in ordinary conversation or to sort out their ambivalent opinions and attitudes based on their answers to survey questions.

A series of studies³⁴ conducted in Detroit, Michigan, between 1956 and 1971 provides further evidence about white racial attitudes at the time. The researchers, James Fields and Howard Schuman, presented various scenarios such as the following:

One day a six-year-old asks her mother if she can bring another girl home to play. The mother knows that the other girl is a Negro, and that her own daughter has only played with white children before. What should the mother do? Here are three possible responses:

B. The daughter should be told that she may play with Negro children in school, but not at home; or

(Today, the word “Negro” is no longer an appropriate term to use to describe African Americans, but even as late as 1971, it was still considered a respectful term for referring to blacks.) The researchers went on to ask whether “most people in the Detroit area” and “most people in this neighborhood” would agree, and if not, what answer they would give. In the 1969 study, the vast majority of respondents (76 percent) answered that the child should be permitted to come to the home. However, only 33 percent believed this to be the majority opinion in the Detroit area, and only 38 percent believed it to be the majority opinion in their own neighborhoods.

Fields and Schuman argue that this contradiction probably does not reflect disowning projection or social desirability bias. They reach this conclusion in part by analyzing a separate measure of people’s behaviors. Respondents in the 1969 survey who said they favored open housing—that is, prohibiting discrimination against would-be homebuyers and renters on the basis of race—were subsequently asked to sign a petition stating this position. One might expect that people who gave lip service to open housing but refused to sign the petition would be more likely to attribute pro-segregation views to others than people who did sign the petition. But Fields and Schuman found otherwise: the two groups’ estimates of other people’s opinions were indistinguishable. While some people may have misstated their own views about segregation, the evidence for pluralistic ignorance in Detroit is probably best explained by real misperception.

More recent research on racial attitudes concludes that when blacks and whites are asked about the relevance of race in everyday issues, each group expects to agree more with its own race than with the other. Yet the groups reported remarkably similar scores. Perceptions matter.³⁶

Fields and Schuman describe their findings on segregation—that people exaggerate the racial conservatism of others—as an example of conservative bias. They and others have pointed to other instances of apparent conservative bias or liberal bias. For example, Fields and Schuman note that in 1956, 79 percent of respondents said that a child should be allowed to refuse to recite the Pledge of Allegiance in school, but only 65 percent perceived this as the majority opinion. People who said that a child should not be allowed to refuse to recite the pledge were far more likely to think that theirs was the majority opinion than the people actually in the majority.³⁷ Fields and Schuman describe this result as another example of conservative bias. Glynn³⁸ notes that the meanings of “conservative” and “liberal” can change over time, and suggests using the broader term ideological bias to describe apparent misperceptions of majority opinion that have some ideological content.

In a study of neighbor perceptions of others’ opinions on local, state, and national issues, Glynn³⁹ found evidence of ideological biases in both directions. In Glynn’s study, individuals in thirty-one neighborhoods in a midwestern city were asked their own opinions, and their perceptions of other people’s opinions, on eight issues: two each from the national level, the state level, the city level, and the neighborhood level. People were asked separately what they thought people in the city, and in their neighborhood, thought about each issue (except for the neighborhood issues, for which they were only asked about people in the neighborhood). Glynn summarizes the results:

[O]n every issue without exception, respondents viewed individuals living in the city as having more “liberal” opinions than their own, regardless of whether it was a national, state or local issue. On the other hand, with only one exception, respondents viewed their neighbors as having more “conservative” opinions on issues than their own … [providing] considerable evidence for … the existence of a “conservative bias” at least at the local level. … [The findings indicate] that respondents can, and do, have differing perceptions about different groups of people.⁴⁰

Some of these differences were large. For example, 39 percent of respondents thought their neighbors were more supportive of increasing military spending (here considered a conservative opinion), while only 21 percent thought they were less supportive. Conversely, 35 percent thought others in the city were less supportive of increasing military spending, while only 24 percent thought they were more supportive. The largest biases appeared in the neighborhood questions. Fully 61 percent thought their neighbors were less supportive of (more “conservative” about) building a home for the mentally retarded, while only 5 percent thought the neighbors were more supportive.

Glynn’s findings are remarkable because the people she interviewed were residents of both the neighborhoods and the city. They believed opinion in their city was more liberal than their own and that opinion in their neighborhood was more conservative than their own, even though they themselves were the elements composing the “public” they were speculating about.

A more recent example of ideological biases at work concerns beliefs about public officials, like the president of the United States. A 2014 study found that one in five Americans still believed the incorrect assumption that President Barack Obama was a Muslim, even after he had been in the White House for years. Although race and religious affiliation influenced this belief, so did ideology.⁴¹

There is also a lot of discussion of ideological bias in the media. Do a search online for “media bias,” and you are sure to come across a plethora of cases and examples deriding media (fairly or not) for favoring one side over the other. Research demonstrates that more liberal Americans deem FOX News to be biased, and conservatives view CNN and the New York Times as biased.⁴² And these perceptions can have real, behavioral effects. Rojas (2010) concluded that perceptions of media bias can lead to “corrective” political participation; those who perceive bias will often seek to rectify the unfair portrayal by engaging in political activity.⁴³

False consensus is the tendency for individuals “to see their own behavioral choices and judgments as relatively common and appropriate to existing circumstances while viewing alternative responses as uncommon, deviant and inappropriate.”⁴⁴ Richard Schanck⁴⁵ originally labeled this phenomenon “misperceived consensus” or “misperceived sharing.” False consensus can occur on a variety of issues. One study examining the death penalty, gun regulation, and teaching morality in schools found that those who strongly favored the policies estimated public support to be higher. Moreover, if respondents reported engaging in disagreement with others about the issues, the association between the individual opinion and false consensus declines.⁴⁶

False consensus also means that people who engage in a given behavior, or who hold a given opinion, believe it to be shared by a larger proportion of people in the group than those with an alternative behavior or a differing opinion.⁴⁷ Lee Ross, David Greene, and Pamela House, in a 1977 study of college students, found false consensus on a variety of beliefs, preferences, and traits.⁴⁸ For example, students who preferred brown bread to white estimated, on average, that about 52 percent of students agreed with them; students who preferred white bread estimated that only about 37 percent of students preferred brown bread. On some issues, there was practically no difference: both students who considered themselves “politically left of center” and those who didn’t estimated, on average, that just under 60 percent of students were left of center. The largest difference appeared on a question about whether there would be a woman on the Supreme Court within a decade. Students who thought that there would estimated that 63 percent of students agreed with them; students who thought that there wouldn’t estimated that only 35 percent of students thought there would be. Thus, students on each side thought that nearly two out of three students agreed with their views. (As it happened, Sandra Day O’Connor joined the court in 1981.)

Other studies have found even more extreme examples of false consensus. For example, a 2013 study on youth delinquency shows a strong false consensus effect. Young people who report engaging in and supporting delinquent activities are more likely to exaggerate how much delinquent behavior their friends engage in. This perception has potential behavioral effects, as it may encourage more delinquent behavior. Indeed, the study showed that over time, overestimating friends’ delinquency led to more delinquent behavior and friends.⁴⁹

Why does false consensus occur? Lee Ross suggests two sets of reasons. On the one hand, false consensus can reflect a perceptual distortion that emerges from selective exposure to and recall of other people’s opinions. If we disproportionately spend time with people who agree with us (or who politely downplay their disagreements), it’s no wonder we tend to exaggerate the extent to which other people do agree. On the other hand, false consensus can arise from motivated bias: the desire to feel or to appear normal, appropriate, and rational.⁵⁰

False consensus can have a major impact on public opinion formation and on public debate. In the case of the abortion debate, public discourse was radically altered by a case of false consensus (see Box 7.2).

BOX 7.2

A Case of False Consensus: Origins of the Pro-Life Movement

The origins of the pro-life movement provide a remarkable example of the effects of false consensus. Prior to the Supreme Court’s 1973 decision in Roe v. Wade, which protected a woman’s right to an abortion in the first six months of pregnancy, many activists were working to make abortions easier to obtain legally, but there was almost no grassroots movement opposed to abortion. Kristin Luker argues that most people who opposed abortion believed that theirs was the majority view:

Perhaps the most important reason why pro-life activists were not well prepared to resist abortion reform was that they simply couldn’t believe such a movement would get very far. They tacitly assumed that the unsavory connotation of abortion rested on a deep belief in the sacredness of embryonic life, and they found it hard to understand how such a belief could be changed so quickly. They counted on public opinion to be outraged and were stunned when most of the public was either unaware or unconcerned.

Luker’s qualitative interviews with pro-life activists show that this case of false consensus stemmed in part from their selective exposure to the social world. Most pro-lifers had never had abortions themselves, nor did they know anyone who had experienced an abortion (or who admitted to it). Thus, they believed that it was an uncommon event. These people also misperceived the “unsavory connotation of abortion”:

In particular, they interpreted the relative invisibility of abortion prior to the 1960s as proof that their opinion was the common one. And in a way, their assumption was plausible. If people didn’t talk about abortion very much (or talked about it only in hushed tones in back rooms), wasn’t that because most people believed it was the taking of an innocent life, hence morally repugnant? What these early pro-life activists did not understand was that for many people abortion was “unspeakable” not because it represented the death of a child but because it represented “getting caught” in the consequences of sexuality. Sex, not abortion, was what people didn’t talk about.

Because of this false consensus effect, many abortion opponents were shocked not only by the Roe v. Wade decision itself, but also that so few people shared their horror about it. They had difficulty formulating persuasive arguments to support their opinions:

Their belief that everyone accepted a common definition of the meaning and moral nature of abortion left these pro-life people with few arguments to use against the abortion reformers. They tried to appeal to what they thought was the commonly shared value, but when it turned out to be not so common after all, they were literally at a loss for words.

As you can see, false consensus can have both psychological and social consequences. In this case, the shape of public debate was altered by people’s perceptions of majority opinion.

SOURCE: Kristin Luker, Abortion and the Politics of Motherhood (Berkeley: University of California Press, 1984), quoted passages from pp. 128–131.

A very different, but important, perceptual influence on public opinion is called unrealistic optimism. The basic idea is as simple as it sounds: people tend to overstate the probability that good things will happen to them and to understate the probability that bad things will happen to them.⁵¹ As Neil Weinstein says, people “expect others to be victims of misfortune, not themselves.”⁵² Similarly, they think that good things are more likely to happen to them than to others. Researchers have found that this phenomenon occurs across a wide range of topic areas, from perceptions of being murdered to perceptions of winning the lottery.⁵³

James Shepperd and colleagues note that the bias toward favorable outcomes occurs for many negative events, including diseases, natural disasters, unwanted pregnancies, and even the end of a romantic relationship.⁵⁴ These scholars further distinguish between two types of unrealistic optimism: unrealistic absolute optimism and unrealistic comparative optimism. Simply put, the former occurs when a person’s estimate of his or her personal risk is too low relative to some specific standard. The latter, on the other hand, occurs when a person incorrectly judges how his or her risk compares with that of other people.

The related concept of impersonal impact holds that mass media information influences people’s perceptions of risk to others, but not (or more weakly) their perceptions of risks to themselves.⁵⁵ For example, Tyler and Cook found that people who had seen a television newsmagazine story about fraud in home health care services were more likely afterward to view the problem as serious—but were no less confident in their own ability to secure good home health care for themselves or their relatives.⁵⁶ Impersonal impact has been considered a separate area of research, but it dovetails with unrealistic optimism: impersonal impact suggests that people will tend to perceive social risks as other people’s problems.

Weinstein conducted a study that looked at college students’ perceptions of future life events and the extent to which students expressed unrealistic optimism about these life events.⁵⁷ As shown in Table 7.2, unrealistic comparative optimism was widespread among these students. When you read the table, note that a positive value in the first column indicates that the students tended to believe chances were greater than average that a good thing would happen to them; a negative value indicates that students believed the chances were less than average that a bad thing would happen to them. The numbers in the second column indicate the ratio of optimistic to pessimistic responses.

Students in the study thought they were, on average, 50 percent more likely than other students at their college to like their first job after graduation, 44 percent more likely to own a home, and 35 percent more likely to travel to Europe. They also thought they were 58 percent less likely than others to have a drinking problem, 56 percent less likely to attempt suicide, and 49 percent less likely to be divorced a few years after marriage. (The optimism was not universal; for example, people were not relatively optimistic about being the same weight in ten years or about not being burglarized.) Individually, some of these optimistic judgments may have been reasonable, but it isn’t reasonable for so many students to have above-average prospects compared to each other. Although researchers have raised questions about Weinstein’s measures,⁵⁸ many studies with various measurement strategies have found broadly similar results.⁵⁹

^a In making a comparative judgment, students estimated the difference in percent between the chances that an event would happen to them and the average chances for other students at their college of the same sex. N = 123 to 130, depending on rating form and missing data. Student’s t was used to test whether the mean is significantly different from zero.

^b For positive events, the response that one’s own chances are greater than average is considered optimistic, and the response that one’s own chances are less than average is considered pessimistic. For negative events, the definitions of optimistic and pessimistic responses are reversed.

^c Significance levels refer to a chi-square test of the hypothesis that frequencies of optimistic and pessimistic responses are equal.

SOURCE: Adapted from Neil D. Weinstein, “Unrealistic Optimism About Future Life Events,” Journal of Personality and Social Psychology 39 (1980): 806–820.

How might unrealistic optimism affect public opinion? That depends on how it interacts with other aspects of people’s thinking. For example, if people’s policy preferences tend to coincide with their perceived self-interest, then unrealistic optimism could bias them toward weaker “safety net” programs than they otherwise would support. However, if people tend to take a society-wide view of policy impacts, then unrealistic optimism may not bias their policy preferences, although it may influence their private choices. (For example, people may tend to underinvest in insurance for themselves, but may still be supportive of government health care for the unfortunate even if they dismiss the prospect of being so unfortunate themselves.)

The concept of impersonal impact ultimately suggests that people assume media are powerful in influencing others, which can influence cognitions and behavior. A study by Hoffman concluded that political discussions in the media influenced citizens’ perceptions of which presidential candidate would win their state in an election.⁶⁰ Mutz noted that the mechanism driving impersonal influence is likely to be different for “citizens with differing levels of information,”⁶¹ which is borne out by Hoffman’s study. Hoffman found that citizens who read more content in their local papers and talked more about the Democratic candidate were more likely to be influenced in their perception that the candidate would win.

The third-person effect describes the fact that when individuals are exposed to a persuasive message, such as an advertisement on television, they often perceive greater effects on others (i.e., the “third person”) than on themselves. The term was established by sociologist W. Phillips Davison in 1983.⁶² Using his own personal experiences with journalists, he examined the phenomenon in which some journalists seemed to be convinced that editorials had an effect on others’ attitudes, but not on people like them. Or, you might think to yourself, that ad for sugary soda might influence other people to buy it, but definitely not me!

The third-person effect is similar to the idea of unrealistic optimism, only instead of referring to the chances of good or bad events, the concern is about the perceived effects of the mass media. Davison’s ideas have generated a good deal of research. For example, Jeremy Cohen, Diana Mutz, Vincent Price, and Albert Gunther studied the third-person effect by asking people to read defamatory newspaper articles. They found that readers who were exposed to these newspaper articles estimated that others would be significantly more affected by the messages than they themselves had been.⁶³

The third-person effect has two components: perceptual and behavioral. The perceptual component is just that: the perception that others are more influenced by content. The behavioral component proposes that biased third-person effects will result in behavioral action. The most commonly studied behavioral outcomes are support for censorship and willingness to speak out.

Why do people engage in third-person effects? One analysis suggested that “people project negative effects onto others in order to avoid the discomfort caused by admitting that such content affects themselves.”⁶⁴ There are other motivations for this perception. Hoffman and Glynn also note that in addition to unrealistic optimism, whereby people are motivated to believe that unfortunate events are more likely to happen to others than to themselves, individuals are also motivated to protect their self-concept from threatening messages and maintain general control over their environment.

In addition to the third-person effect, some research has found a “reverse first-person effect,” wherein people perceive a greater effect on themselves than on others, as long as the content is positive and desirable. So if a person were watching a media message encouraging healthy eating habits, he or she might be more likely to say he or she is affected by the content than are others.⁶⁵

In the age of microtargeted messaging, when an online search can result in tailored ads displayed for different users, the third-person effect might take on a different role. Even in the “old days” of mass-targeted messages, we saw that ads can influence us even when we don’t think they do. But now, the likelihood of being impacted by highly targeted ads should arguably be much higher.

The theories we have considered so far in this chapter operate at the individual level, although they involve people’s opinions about social groups. Such theories are useful, but they have limited capacity to describe or explain some of the ways in which public opinion changes The theories we turn to next describe public opinion more as a social process: how groups, or even the entire public, behave and change collectively. There are a number of public process models; we briefly discuss several in this section.

Following Herbert Blumer’s lead, Foote and Hart⁶⁶ construe a public as created and defined by its members’ discussion of some issue. (Thus, many publics exist simultaneously, and they change size and shape as the discussion evolves.) First comes the problem phase, in which some people—a nascent public—come to agree that a problem exists. Next comes the proposal phase, in which members of the public consider various ways to deal with the problem. Then come the policy and program phases, by the end of which a course of action is decided on, typically with broad public support. The final phase is the appraisal stage, in which the public periodically reevaluates the policy.

This model is idealized; issues and publics do not have to develop in this linear manner, or at all. (And it may be hopelessly unclear who belongs to a particular public.) It may be most useful for the questions it prompts us to ask about “public opinion” on an issue, such as: Who is thinking about it? Who is talking about it, and to whom? How specific are those conversations, and what kinds of information (accurate or inaccurate) do they incorporate? How many people are saying “Do something,” and how many know what they want to have done? The common idea of an issue public has roots in Foote and Hart’s model.

W. Phillips Davison’s model⁶⁷ is grounded in individual behaviors, but it ultimately portrays broader social dynamics. Davison starts from a famous discussion of public opinion by the German poet Christoph Wieland in 1798. Wieland wrote that he understood “public opinion” to mean

an opinion that gradually takes root among a whole people, especially among those who have the most influence when they work together as a group. In this way it wins the upper hand to such an extent that one meets it everywhere. … It is an opinion that without being noticed takes possession of most heads, and even in situations where it does not dare to express itself out loud it can be recognized by a louder and louder muffled murmur. It then only requires some small opening that will allow it air, and it will break out with force. Then it can change whole nations in a brief time and give whole parts of the world a new configuration.⁶⁸

In this view, then, a public is defined by the opinion it shares, an opinion that may not often be openly expressed but nonetheless can have momentous consequences.

Somewhat like Foote and Hart, Davison emphasizes the role of issues in public opinion formation. For Davison, the formation process begins with issue presentation, in which people communicate ideas—potential issues—to each other. When an issue is passed to a third and fourth and fifth person, it begins to take root. Most potential issues disappear from attention before this “human chain” grows to an appreciable length, but the few that remain form the basis for public opinion. A contemporary analogy might be how a YouTube video can either “go viral” or remain obscure, depending on how many people share it.

Davison emphasizes the role of groups in determining how opinions are communicated. Opinions initially spread within a primary group—a group of people who converse together on an issue. Most opinions never progress beyond a primary group, but some find “leadership transcending the original primary group.”⁶⁹ The most influential opinion leaders have the means to publicize their views, as well as the skill to simplify and generalize their ideas so that they will appeal to the largest possible audience. If many in that audience accept the new idea, a widely held public opinion may develop. Face-to-face discussion proceeds in other primary groups, and these new discussions lead to more public communication about the issue.

Davison’s model also integrates people’s perceptions of others’ opinions. As public discussion of an issue increases, individuals begin to form expectations about what others’ opinions are likely to be. As we have seen, those expectations may rely on particular conversations, on social projection, on assumptions about other people’s thought processes, and so on. Individuals decide what to say about their opinions—whether to speak, remain silent, or even express the opposite of their private views—considering the opinions of salient others. Davison argues, “A process is set up in which expectations produce behavioral adjustments, and these in turn reinforce expectations. When this has happened, public opinion has been formed.”⁷⁰

Price and Roberts’s 1987 model of public opinion processing⁷¹ shares many basic elements with Davison’s approach. Price and Roberts describe public opinion as a “process of social organization through communicating.” In their model, (1) a public is not organized in any fixed fashion until forced to communicate in resolving issues; and (2) public opinion is decidedly not the distribution of opinions within a public, but is instead a complex function of processes in which disparate ideas are expressed, adjusted, and compromised en route to collective determination of a course of action.

According to Price and Roberts, the public opinion process may be conceptualized as communication between political actors who are pursuing public recognition and support for views and members of the interested public who are trying to understand the issue and decide whom to support. Price and Roberts observe that many people “deciding their stand on a public issue are not so much deciding their own opinion (or where they stand on the matter) but instead deciding on their social loyalties (in other words, with whom they stand).”⁷² They call public opinion a social process, and more specifically, they emphasize the important role of communication, suggesting that the relationship between cognitions and behavior is a “continuing dialectic,” in which individuals incorporate new information with preexisting opinions.⁷³ The public, then, is formed by the issue, not the other way around. A public is composed of various groups, some of them interpersonal, others broadly social, “attempting through discursive means to resolve a common issue.”⁷⁴ The process of forming and changing public opinion is a complex one that involves political actors, media, and an interested public, and all these elements attempt to ascertain and influence opinions on social issues. Thus, the public opinion process is a process of social accommodation.

Hoffman and her colleagues built on this model by identifying sources of influence: (1) predispositions and interests (i.e., intrapersonal filters); (2) the media, which disseminate public opinion and issue specific information (i.e., media filters); and (3) interpersonal political discussion, which gives an issue momentum and strengthens opinion, as well as supplying additional relevant information (i.e., social filters).⁷⁵ They concluded that interpersonal factors largely accounted for how public opinion was formed around a community issue.

Whereas Davison’s model considers both how opinions spread and how contrary opinions may go unspoken, Elizabeth Noelle-Neumann’s spiral of silence theory⁷⁶ focuses squarely on the latter. Noelle-Neumann, a German political scientist and pollster, posits that our relationships with others are so important to us that we will willingly change or repress our opinions in order to be accepted by others. She states that to the individual, “not isolating himself is more important than his own judgment.”⁷⁷

Perhaps surprisingly, spiral of silence theory began with a voting behavior puzzle. In 1965 Noelle-Neumann polled and studied the West German federal elections, in which the ruling Christian Democratic Union coalition was challenged by the Social Democrats.⁷⁸ She found that for many months before the election, the polls showed the parties essentially tied, yet they also showed a steady shift toward believing that the Christian Democrats would remain in power. Then, just before the election, about 3–4 percent of voters suddenly switched their support to the Christian Democrats, who won by a decent margin. During the 1972 election, Noelle-Neumann noticed a similar pattern: a late shift to support the expected winners. Surprised, Noelle-Neumann decided to investigate, and her research on the spiral of silence began.

Spiral of silence theory posits that one’s perception of the distribution of public opinion affects one’s willingness or unwillingness to express opinions.⁷⁹ Individuals who notice that their personal opinions are spreading will voice these opinions self-confidently in public; those who notice their opinions are “losing ground” will be inclined to adopt a more reserved attitude and remain silent. These perceptions of reality become reality as people’s opinions are swayed by the increasingly one-sided content of public discourse. The result is a spiral of silence, as the expression of dissenting opinion becomes smaller and fainter.⁸⁰

The theory is complex, and it is not possible to present the details here. However, it is important to note that four key elements enter into the theory: (1) one’s own opinion on an issue; (2) one’s perception of the predominant public opinion; (3) one’s assessment of the likely future course of public opinion; and (4) one’s willingness to support one’s opinion with action, verbal statements, or other signs of commitment.⁸¹ One’s own opinion, and one’s perception of predominant opinion, both are assumed to influence one’s judgment of the future course of opinion. One’s willingness to speak out then depends on the interaction among these variables—basically, on whether one expects to end up in the majority or the minority.

Noelle-Neumann’s full conception of the public opinion process contains several important aspects, including (1) societal functions of public opinion, (2) moral and behavioral components of public opinion, (3) the importance of perception in the public opinion process, and (4) the importance of communication in opinion outcomes. The spiral of silence theory attempts to describe impacts on public opinion that go far beyond interpersonal interaction.

Since Noelle-Neuman’s initial research, more public opinion research has been conducted on the spiral of silence than on almost any other single public opinion approach, thesis, or question.⁸² An updated meta-analysis concluded with essentially the same findings: perceived opinion climate continues to have an effect on opinion expression.⁸³ Yet several of the most recent studies in this review demonstrated small, sometimes not significant, effects when other key variables were included. More recent research suggests that opinion expression is affected by the climate of opinion only when opinions are of low or moderate certainty, but it doesn’t matter for those with high certainty about their opinions.⁸⁴ Another study concluded that communication apprehension (anxiety about communicating), along with fear of isolation, can also impact opinion expression.⁸⁵ Carroll Glynn and Jack McLeod⁸⁶ argue that the spiral of silence theory is important because it describes how, through social interaction, people influence each other’s willingness to express opinions.

Other scholars have identified a personality characteristic that might explain opinion expression better than fear of isolation: self-censorship. “Willingness to Self-Censor” was developed by Glynn and colleagues to address conflicting results in the spiral of silence literature.⁸⁷ The eight-item validated scale draws from research in social psychology, specifically on conformity, which suggests there is a lot of variability in individuals’ willingness to offer an opposing opinion in a confronting opinion climate. It is also rooted in research on individuation, which suggests that a person’s willingness to engage in behaviors that make him or her be distinctive from others in a group could impact opinion expression.

Noelle-Neumann also ascribes importance to the mass media’s impact on the formation and presentation of the public’s opinions. According to Noelle-Neumann, the media play a crucial role in that their messages are “ubiquitous” (everywhere) and “consonant” (they repeat the same messages over and over).⁸⁸ Essentially, the media continually bombard us with information that frames our social reality—and regardless of the reporter or the medium, the messages tend to be the same. According to Noelle-Neumann, we learn most of our societal norms, customs, and so forth from the media. That is, media messages continually reinforce accounts of what we should believe about any numbers of issues.

It is clear that the spiral of silence is an important theory that the student of public opinion should understand. Although many researchers have found problems with the theory,⁸⁹ it is widely regarded as a major contribution to public opinion theory and research. If anything, the most simple of its premises attest to its importance: we are social, others are important to us, we want others to like us, and we want to fit in. These factors clearly influence public opinion formation. Of all the approaches discussed in this chapter, Noelle-Neumann’s spiral of silence theory may be the most broadly applicable.

These theories are exciting yet incomplete. What is fascinating for the student of public opinion is that we know that psychological, social, and political factors can change or motivate public opinion outcomes, but we are not sure what these factors are or exactly when these outcomes will occur.

We know that the media are important in the formation of public opinion, but we do not know enough about how people are influenced by these powerful transmitters of information. Politicians know there are ways to manipulate the public through the media, yet public opinion scholars do not know enough about the public opinion process to be able to determine how and when this manipulation will occur. The public is arguably more vulnerable and more susceptible because of this lack of research in this important area.

Perception has been defined throughout this chapter as a summary attitude based on all of our past and present sensory information. Perceptions are limited because they are selective. Perceptions also play a comparative role; we compare our perceptions to those of others. Socialization, the influence of what others important to us think, and social comparison, the comparison of one’s opinion to others to test its validity, have become important theories used to explain how individuals apply their perceptions to the larger social context.

Public opinion is clearly more than responses to public opinion polls. It is a verbal expression of culture, of social interactions, of psychological processes. Students of public opinion should understand the approaches described in this chapter, but should also make sure they have a solid grasp of theories developed in other fields, especially sociology, social psychology, and psychology. It is important that we understand how public opinion works so that we can go beyond mere speculation or description. The field is exciting, and there continues to be much to learn. Public opinion scholars of the future can help us understand this important and fundamental social process that is vital to our very survival.

1. Drew Westen, Pavel S. Blagov, Keith Harenski, Clint Kilts, and Stephan Hamann, “Neural Bases of Motivated Reasoning: An fMRI Study of Emotional Constraints on Partisan Political Judgment in the 2004 U.S. Presidential Election,” Journal of Cognitive Neuroscience 18, no. 11 (2006): 1947–1958.

3. George J. McCall and J. L. Simmons, Identities and Interactions: An Examination of Human Associations in Everyday Life, 2nd ed. (New York: Free Press, 1978), 104.

5. Jerome S. Bruner, “Social Psychology and Perception,” in Eleanor E. Maccoby, Theodore M. Newcomb, and Eugene L. Hartley, eds., Readings in Social Psychology (New York: Holt, Rinehart and Winston, 1958), 86.

7. Paul R. Abramson, Political Attitudes in America: Formation and Change (San Francisco: W. H. Freeman and Co., 1983); M. Kent Jennings and Richard G. Niemi, Generations and Politics: A Panel Study of Young Adults and Their Parents (Princeton, NJ: Princeton University Press, 1981); Roberta L. Sangster and Robert W. Reynolds, “A Test of Inglehart’s Socialization Hypothesis for the Acquisition of Materialist/Postmaterialist Values: The Influence of Childhood Poverty on Adult Values,” Political Psychology 17 (1996): 253–269.

8. Yair Ghitza and Andrew Gelman, “The Great Society, Reagan’s Revolution, and Generations of Presidential Voting” (working paper, July 7, 2014).

9. Erica Weintraub Austin, “Apathy, Voter,” in Lynda Lee Kaid and Christina Holtz-Bacha, eds., Encyclopedia of Political Communication (Thousand Oaks, CA: Sage Publications, 2007), 37.

10. Leon Festinger, “A Theory of Social Comparison Processes,” Human Relations 7 (1954): 117–140.

11. Hyun Soon Park and Charles T. Salmon, “A Test of the Third-person Effect in Public Relations: Application of Social Comparison Theory,” Journalism & Mass Communication Quarterly 82 (2005): 25–43.

12. Arie W. Kruglanski and Ofra Mayseless, “Motivational Effects in the Social Comparison of Opinions,” Journal of Personality and Social Psychology 5, no. 53 (1987): 834–842.

13. See, e.g., Albert C. Gunther and Stella Chih-Yun Chia, “Predicting Pluralistic Ignorance: The Hostile Media Perception and Its Consequences,” Journalism and Mass Communication Quarterly 78, no. 4 (2001): 688–701; Diana C. Mutz, Impersonal Influence: How Perceptions of Mass Collectives Affect Political Attitudes (Cambridge, UK: Cambridge University Press, 1998); Jordon M. Robbins and Joachim I. Krueger, “Social Projection to In groups and Out groups: A Review and Meta-Analysis,” Personality and Social Psychology Review 9 (2005): 32–47; Jounghwa Choi, Myengja Yang, and Jeongheon J. C. Chang, “Elaboration of the Hostile Media Phenomenon: The Roles of Involvement, Media Skepticism, Congruency of Perceived Media Influence, and Perceived Opinion Climate,” Communication Research 36 (2009): 54–75; Lindsay H. Hoffman and Carroll J. Glynn, “Media and Perceptions of Reality,” in Wolfgang Donsbach, ed., The International Encyclopedia of Communication, (Malden, MA: Blackwell, 2008), VII:2945–2959; Tien-Tsung Lee, “The Liberal Media Myth Revisited: An Examination of Factors Influencing Perceptions of Media Bias,” Journal of Broadcasting & Electronic Media 49 (2005): 43–64; Tien-Tsung Lee, “Why They Don’t Trust the Media: An Examination of Factors Predicting Trust,” American Behavioral Scientist 54 (2010): 8–21; Jonathan S. Morris, “Slanted Objectivity? Perceived Media Bias, Cable News Exposure, and Political Attitudes,” Social Science Quarterly 88, no. 3 (2007): 707–728.

14. Glynn, Ostman, and McDonald, “Opinions, Perceptions and Social Reality”; Carroll J. Glynn, “Public Opinion as a Normative Opinion Process,” in Brant R. Burleson, ed., Communication Yearbook 20 (Thousand Oaks, CA: Sage Publications, 1996), 157–183.

15. Glynn, Ostman, and McDonald, “Opinions, Perceptions and Social Reality.”

16. James M. Fields and Howard Schuman, “Public Beliefs About the Beliefs of the Public,” Public Opinion Quarterly 40 (1976): 445.

17. Lauren J. Human and Jeremy C. Biesanz, “Through the Looking Glass Clearly: Accuracy and Assumed Similarity in Well-adjusted Individuals’ First Impressions,” Journal of Personality and Social Psychology 100, no. 2 (2011): 349–364; Slavko Splichal, “The Public in the Private, Privately about the Public: Opinion Polls in the Political Process,” Javnost-The Public 16 (2009): S81; Serena Chen, Helen C. Boucher, and Molly P. Tapias, “The Relational Self Revealed: Integrative Conceptualization and Implications for Interpersonal Life,” Psychological Bulletin 132, no. 2 (2006): 151–179.

18. Ruben Orive, “Social Projection and Social Comparison of Opinions,” Journal of Personality and Social Psychology 54, no. 6 (1988): 953–964.

20. Classic influential discussions include Frank H. Allport, Social Psychology (Cambridge, MA: Riverside Press, 1924); Richard Louis Schanck, “A Study of a Community and Its Groups and Institutions Conceived of as Behaviors of Individuals,” Psychological Monographs 43, no. 2 (1932): entire issue; Hubert O’Gorman and Stephen L. Garry, “Pluralistic Ignorance—A Replication and Extension,” Public Opinion Quarterly 40 (1976): 449–458; Dale T. Miller and Cathy McFarland, “Pluralistic Ignorance: When Similarity Is Interpreted as Dissimilarity,” Journal of Personality and Social Psychology 53, no. 2 (1987): 298–305; and D. Garth Taylor, “Pluralistic Ignorance and the Spiral of Silence: A Formal Analysis,” Public Opinion Quarterly 46 (1982): 311–355.

21. William P. Eveland Jr., “The Impact of News and Entertainment Media on Perceptions of Social Reality.” in James Price Dillard and Michael Pfau, eds., The Persuasion Handbook: Developments in Theory and Practice (Thousand Oaks, CA: Sage, 2002), 691–727; Fields and Schuman, “Public Beliefs about Beliefs of the Public;” Hoffman and Glynn, “Media and Perceptions of Reality.”

23. Patricia Moy, “Pluralistic Ignorance and Nonattitudes,” in Wolfgang Donsbach and Michael W. Traugott, eds., SAGE Handbook of Public Opinion Research (Thousand Oaks, CA: SAGE Publications, 2007), ch.15.

27. Z. Leviston, I. Walker, and S. Morwinski, “Your Opinion on Climate Change Might Not Be as Common as You Think,” Nature Climate Change 3 (2013): 334–337.

28. Tracy A. Lambert, Arnold S. Kahn, and Kevin J. Apple, “Pluralistic Ignorance and Hooking Up,” Journal of Sex Research 40, no. 2 (2003): 129–133.

29. Glynn, Ostman, and McDonald, “Opinions, Perceptions and Social Reality,” 262.

31. Norman Cameron, The Psychology of Behavior Disorders: A Biosocial Interpretation (Boston: Houghton Mifflin, 1947), 168.

33. Fields and Schuman discuss both these possibilities in “Public Beliefs about the Beliefs of the Public,” 434.

36. Diana E. Betz, Laura R. Ramsey, and Denise Sekaquaptewa, “Perceiving Race Relevance in Everyday Events: Target Race Matters, Perceiver Race Does Not,” Group Processes & Intergroup Relations 16, no. 6 (2013): 699–716.

37. Fields and Schuman, “Public Beliefs about the Beliefs of the Public,” 432. (The 65 percent figure is calculated from numbers reported in the text.)

38. Carroll J. Glynn, “Perceptions of Others’ Opinions as a Component of Public Opinion,” Social Science Research 18 (1989): 64.

41. Michael A. Cacciatore, Sara K. Yeo, Dietram A. Scheufele, Michael A. Xenos, Doo-Hun Choi, Dominique Brossard, Amy B. Becker, and Elizabeth A. Corley, “Misperceptions in Polarized Politics: The Role of Knowledge, Religiosity, and Media,” PS—Political Science & Politics 47, no. 3 (2014): 654–661.

42. Carroll J. Glynn and Michael E. Huge, “How Pervasive Are Perceptions of Bias? Exploring Judgments of Media Bias in Financial News,” International Journal of Public Opinion Research 26, no. 4 (2014): 543–553; Lee, “The Liberal Media Myth Revisited” and Lee, “Why They Don’t Trust the Media.”

43. Hernando Rojas, “‘Corrective Actions’ in the Public Sphere: How Perceptions of Media and Media Effects Shape Political Behaviors,” International Journal of Public Opinion Research 22, no. 3 (2010): 343–363.

44. Lee Ross, David Greene, and Pamela House, “The ‘False Consensus Effect’: An Egocentric Bias in Social Perception and Attribution Processes,” Journal of Experimental Social Psychology 13 (1977): 280.

46. Magdalena Wojcieszak and Vincent Price, “What Underlies the False Consensus Effect? How Personal Opinion and Disagreement Affect Perception of Public Opinion,” International Journal of Public Opinion Research 21 (2009): 25–46.

47. Brian Mullen et al., “The False Consensus Effect: A Meta-Analysis of 115 Hypothesis Tests,” Journal of Experimental Social Psychology 21 (1985): 262–283.

49. Jacob T. N. Young and Frank M. Weerman, “Delinquency as a Consequence of Misperception: Overestimation of Friends’ Delinquent Behavior and Mechanisms of Social Influence,” Social Problems 60, no. 3 (2013): 334–356.

50. Lee Ross, “The Intuitive Psychologist and His Shortcomings: Distortions in the Attribution Process,” in Leonard Berkowitz, ed., Advances in Experimental Social Psychology (New York: Academic Press, 1977), 150–195.

51. Roger A. Drake, “Lateral Asymmetry of Personal Optimism,” Journal of Research in Personality 18, no. 4 (1984): 497–507.

52. Neil D. Weinstein, “Unrealistic Optimism About Future Life Events,” Journal of Personality and Social Psychology 39, no. 5 (1980): 806.

53. Hugh M. Culbertson and Guido H. Stempel, “Media Malaise: Explaining Personal Optimism and Societal Pessimism About Health Care,” Journal of Communication 35 (1985): 180–190.

54. James A. Shepperd, William M. P. Klein, Erika A. Waters, and Neil D. Weinstein, “Taking Stock of Unrealistic Optimism,” Perspectives on Psychological Science 8, no. 4 (2013): 395–411.

55. Patricia Moy, “Social Perception: Impersonal Impact,” in Donsbach, ed., International Encyclopedia of Communication.

56. Tom R. Tyler and Fay L. Cook, “The Mass Media and Judgments of Risk: Distinguishing Impact on Personal and Societal Level Judgments,” Journal of Personality and Social Psychology 47 (1984): 693–708.

58. Adam J. L. Harris and Ulrike Hahn, “Unrealistic Optimism About Future Life Events: A Cautionary Note,” Psychological Review 118 (2011): 135–154.

59. See, e.g., the references cited in Neil D. Weinstein and William M. Klein, “Unrealistic Optimism: Present and Future,” Journal of Social and Clinical Psychology 15 (1996): 1–8, and in Harris and Hahn, “Unrealistic Optimism.”

60. Lindsay H. Hoffman, “When the World Outside Gets Inside Your Head: The Effects of Media Context on Perceptions of Public Opinion,” Communication Research 40, no. 4 (2013): 463–485.

62. W. Phillips Davison, “The Third-Person Effect in Communication,” Public Opinion Quarterly 47 (1983): 1–15.

63. Jeremy Cohen et al., “Perceived Impact of Defamation: An Experiment on Third–Person Effects,” Public Opinion Quarterly 52 (1988): 161–173.

65. Anita G. Day, “Out of the Living Room and into the Voting Booth: An Analysis of Corporate Public Affairs Advertising under the Third-person Effect,” American Behavioral Scientist 52, no. 2 (2008): 243–260.

66. Nelson N. Foote and Clyde W. Hart, “Public Opinion and Collective Behavior,” in Muzafer Sherif and M. O. Wilson, eds., Group Relations at the Crossroads (New York: Harper and Row, 1953), 301–332; Vincent Price, Public Opinion (Newbury Park, CA: SAGE Publications, 1992), 29–30.

67. W. Phillips Davison, “The Public Opinion Process,” Public Opinion Quarterly 21 (1957): 103–118.

71. Vincent Price and Donald F. Roberts, “Public Opinion Processing,” in C. R. Berger and S. H. Chaffee, eds., Handbook of Communication Science (Beverly Hills, CA: Sage Publications, 1987), 781–816.

75. Lindsay H. Hoffman, Carroll J. Glynn, Michael E. Huge, Tiffany Thomson, and Rebecca Border Seitman, “The Role of Communication in Public Opinion Processes: Understanding the Impacts of Individual, Media, and Social Filters,” International Journal of Public Opinion Research 19, no. 3 (2007): 1–26.

76. Elisabeth Noelle-Neumann, “Return to the Concept of a Powerful Mass Media,” Studies of Broadcasting 9 (March 1973): 67–112; Elisabeth Noelle-Neumann, The Spiral of Silence: Public Opinion—Our Social Skin (Chicago: University of Chicago Press, 1984).

77. Elisabeth Noelle-Neumann, “Turbulences in the Climate of Opinion: Methodological Applications of the Spiral of Silence Theory,” Public Opinion Quarterly 41 (1977): 143–158.

78. The Christian Democratic Union was allied with the much smaller Christian Social Union. For simplicity, we use “Christian Democrats” as shorthand for the two parties together.

79. Carroll J. Glynn, “The Communication of Public Opinion,” Journalism Quarterly 64 (1987): 688–697.

82. Carroll J. Glynn, Andrew F. Hayes, and James Shanahan, “Spiral of Silence: A Meta–Analysis,” Public Opinion Quarterly 61, no. 3 (1997): 452–463.

83. Carroll J. Glynn and Michael E. Huge, “Speaking in Spirals: An Updated Meta-analysis of the Spiral of Silence,” in Wolfgang Donsbach, Charles T. Salmon, and Yariv Tsfati, eds., The Spiral of Silence: New Perspectives on Communication and Public Opinion (New York: Routledge, 2014).

84. Jörg Matthes, Kimberly Rios Morrison, and Christian Schemer, “A Spiral of Silence for Some: Attitude Certainty and the Expression of Political Minority Opinions,” Communication Research 37, no. 6 (2010): 774–800, doi: 10.1177/0093650210362685.

85. K. Neuwirth, E., Frederick, and C. Mayo, “The Spiral of Silence and Fear of Isolation,” Journal of Communication 57, no. 3 (2007): 450–468.

86. Carroll J. Glynn and Jack M. McLeod, “Implications of the Spiral of Silence Theory for Communication and Public Opinion Research,” in Keith R. Sanders, Lynda Lee Kaid, and Dan D. Nimmo, eds. Political Communication Yearbook (Carbondale: Southern Illinois University Press, 1985), 43–68.

87. Carroll J. Glynn, Andrew F. Hayes, and James Shanahan, “Perceived Support for One’s Opinions and Willingness to Speak Out: A Meta-analysis of Survey Studies on the “Spiral of Silence,” Public Opinion Quarterly 61 (1997): 452–461.

88. Noelle-Neumann, The Spiral of Silence; Noelle-Neumann, “Return to the Concept of a Powerful Mass Media”; Andrew F. Hayes, Carroll J. Glynn, and James Shanahan, “Willingness to Self-censor: A Construct and Measurement Tool for Public Opinion Research,” International Journal of Public Opinion Research 17 (2005): 299–323; Andrew F. Hayes, Carroll J. Glynn, and James Shanahan, “Validating the Willingness to Self-censor Scale: Individual Differences in the Effect of the Climate of Opinion on Willingness to Express an Opinion,” International Journal of Public Opinion Research 17 (2005): 443–445.

89. Glynn, “Perceptions of Others’ Opinions as a Component of Public Opinion”; Dominic L. Lasorsa, “Real and Perceived Effects of ‘Amerika,’” Journalism Quarterly 66 (1989): 373–378; Charles T. Salmon and Chi-Yung Moh, “The Spiral of Silence: Linking Individual and Society Through Communication,” in J. David Kennamer, ed., Public Opinion, the Press, and Public Policy (Westport, CT: Praeger, 1994), 145–161; Glynn, Hayes, and Shanahan, “The Spiral of Silence: A Meta-Analysis”; Glynn and McLeod, “Implications of the Spiral of Silence Theory”; Carroll J. Glynn and Jack M. McLeod, “Public Opinion du Jour: An Examination of the Spiral of Silence,” Public Opinion Quarterly 48, no. 4 (1984): 731–740; Charles T. Salmon and F. Gerald Kline, “The Spiral of Silence Ten Years Later,” in Sanders, Kaid, and Nimmo, eds., Political Communication Yearbook, 3–30; Charles T. Salmon and Hayg Oshagan, “Community Size, Perceptions of Majority Opinion and Opinion Expression,” Public Relations Research Annual 2 (1990): 157–171; Wolfgang Donsbach and Robert L. Stevenson, “Challenges, Problems and Empirical Evidence of the Theory of the Spiral of Silence” (paper presented at the International Communication Association Conference, San Francisco, 1984); Frank L. Rusciano, Isolation and Paradox: Defining “the Public” in Modern Political Analysis (New York: Greenwood Press, 1989).

	Favors Segregation	Perceives at Least Half Support¹	Perceives Majority Support²
*North*³
Grammar school⁴	23%	65%	32%
High school	12%	52%	28%
College	4%	36%	16%
All respondents⁵	12%	50%	25%
*South*⁶
Grammar school	49%	84%	54%
High school	27%	76%	48%
College	18%	70%	36%
All respondents	30%	77%	47%

	Measures of Optimism
Abbreviated Event Description	Mean Comparative Judgment of Own Chances vs. Others’ Chances (%)^a^,^b	Number of Optimistic Responses Divided by Number of Pessimistic Responses^b^,^c
POSITIVE EVENTS
Liking your postgraduation job	50.2^f	5.93^f
Owning your own home	44.3^f	6.22^f
Traveling to Europe	35.3^f	2.25^f
Starting salary > $15,000	21.2^e	1.56^d
Graduating in top third of class	14.2	1.02
Your work being recognized with an award	12.6^d	1.72^d
Not spending a night in the hospital for 5 years	8.5	1.23
Having a gifted child	6.2^d	2.26^e
In 10 years, earning > $40,000 a year	-0.7	0.64^d
Marrying someone wealthy	-9.1	0.36^d
NEGATIVE EVENTS
Having a drinking problem	-58.3^f	7.23^f
Getting divorced a few years after getting married	-48.7	9.50^f
Being fired from a job	-31.6^f	7.56^f
Dropping out of college	-30.8^f	3.49^f
Having a heart attack	-23.3^f	3.18^f
Not finding a job for 6 months	-14.4^f	2.36^f
Buying a car that turns out to be a lemon	-10.0^d	2.12^e
Tripping and breaking a bone	-8.3^d	1.66^d
Being sued by someone	-7.9	2.38^f
Being a victim of mugging	-5.8	3.17^f