Prelim pages dissertation augmented

THE FALSE CONSENSUS EFFECT: PROJECTION OR CONFORMITY? A DISSERTATION SUBMITTED TO THE DEPARTMENT OF COMMUNICATION AND THE COMMITTEE OF GRADUATE STUDIES OF STANFORD UNIVERSITY IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY LORI D. GAUTHIER MARCH 2010 © 2010 by Lori Donay Gauthier. All Rights Reserved. Re-distributed by Stanford University under license with the author.

This work is licensed under a Creative Commons Attribution- Noncommercial 3.0 United States License.org/licenses/by-nc/3.0/us/ This dissertation is online at: http://purl.edu/fp259hn7562 ii I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy. Jon Krosnick, Primary Adviser I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy. James Fishkin I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy. Shanto Iyengar I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy.

Byron Reeves Approved for the Stanford University Committee on Graduate Studies. Gumport, Vice Provost Graduate Education This signature page was generated electronically upon submission of this dissertation in electronic format. An original signed hard copy of the signature page is on file in University Archives. iii Abstract The false consensus effect (FCE) is one of the most frequently cited phenomena in the social sciences and is widely presumed to illustrate how attitudes can shape perceptions.

Yet, rarely have researchers directly tested this presumption. The purpose of this research is to address this gap in the literature. I begin with a review of the mechanisms presently thought to account for the FCE, formalizing them in causal diagrams that make explicit the mediational and moderational stories implicit in the explanations. Then, I describe what evidence exists for each of these explanations.

And, finally, I report the results of four studies that tested the causal direction of the FCE. Study 1 manipulated participants’ attitudes and found no impact of the resulting attitude change on perceptions of others’ attitudes. Study 2 employed instrumental variable analysis to estimate the impact of participants’ attitudes on their perceptions of others’ attitudes and found no such impact. Study 3 tested an alternative explanation of the FCE: conformity to perceived social norms.

Consistent with this claim, a manipulation of perceptions of others’ attitudes influenced people’s own attitudes. Applying covariance structure modeling to longitudinal data, Study 4 found evidence that perceptions of others’ attitudes influence participants’ own attitudes but no evidence that people’s own attitudes shape perceptions of others’ attitudes. Thus, instead of documenting attitude-distorted perceptions of others, the false consensus “effect” appears to result from one of the oldest social psychological phenomena in the literature: conformity to perceived social norms. iv Table of Contents Chapter 1 pp.

1 – 8 Mass Media, Individual Attitudes, and Perceptions of Public Opinion Chapter 2 pp. 9 – 12 The False Consensus Effect Chapter 3 pp. 13 – 27 Causal Mechanisms of the False Consensus Effect Chapter 4 pp. 28 – 36 Evidence for Causal Mechanisms of the False Consensus Effect Chapter 5 pp.

37 – 77 Direct Tests of the Causal Direction of the False Consensus Effect Chapter 6 p. 78 Implications for Communication and Public Opinion Research Chapter 7 pp. 79 – 82 Implications for Society Chapter 8 pp. 83 – 88 Suggestions for Future Research References pp.

89 –111 v List of Table and Figures Tables Table 5.1: Demographic distributions, Studies 1 – 4 pp.2: Number of stories mentioning poll questions pp.1: Word primes pp.2: Predicted FCE effect size patterns pp.1: Persuasive press inference model of public opinion pp.2: Spiral of silence model of public opinion pp. Attitude salience model pp.2: Selective affiliation model pp.3: Self-protection motivation model pp.4: The evaluation principle model pp.5: Attributional reasoning model pp.6: Rational induction model pp.7: De facto selective exposure model pp.8: Differential object construal model pp.9: Protocentric perception model pp.1: Study 4: Simultaneous effects model pp.2: Study 4: Lagged effects model pp. 67 vi Chapter 1: Mass Media, Individual Attitudes, and Perceptions of Public Opinion Mass media researchers have considered perceived public opinion as a critical concept in several public opinion models, including the persuasive press inference theory (Gunther, Christen, Liehbart, & Chia, 2001) and the spiral of silence theory (Noelle-Neumann, 1974). The former is a projection-based theory, where individuals’ attitudes influence perceived public opinion, and the latter is a conformity-based theory, where perceived public opinion influences individuals’ attitudes.

Both theories share a connection with the phenomenon known as the false consensus effect (FCE)—the tendency to of people who hold a particular attitude to generate higher public opinion consensus estimates compared to people whose attitudes are different (Ross, Greene, & House, 1977). The persuasive press inference theory conceptualizes individuals’ attitudes as the cause of perceived public opinion, whereas the spiral of silence theory conceptualizes perceived public opinion as the causal force driving individual attitudes. Both theories also posit that news media coverage mediates the FCE. Specifically, perceptions of mass media are thought to act as a projection conduit from individuals’ opinions to perceived public opinion in the persuasive press inference model; whereas, in the spiral of silence model, perceptions of mass media are thought to serve as a conformity conduit from perceived public opinion to individuals’ opinions.

Persuasive Press Inference Model of Perceived Public Opinion Generally speaking, the persuasive press inference asserts that people believe media content significantly affects others’ opinions on issues covered by the media; 1 thus, it follows that perceived public opinion can be accurately inferred from media coverage ―What mass media are saying today must be what the public will be thinking tomorrow‖ (Gunther, 1998, p. In experimental studies, people exposed to news stories with a clear favorable or unfavorable bias not only perceived a corresponding difference in public opinion but also noted that the public’s opinion would have changed ―in the last few days,‖ reflecting the release of the news stories (Gunther 1998; Gunther & Christen, 1999). As early as 1954, Berelson, Lazarsfeld, and McPhee investigated individual’s perception of how certain groups would vote. They found that when people felt a group to be close to them, they expected group members to vote in the same way as themselves.

Likewise, Fields and Schuman (1976) inferred from survey data that assessments of general public opinion showed evidence of projection effects, or, in their terms, the looking-glass perception. People ―look out onto the world and somehow see their own opinions reflected back‖ (p. More specifically, the persuasive press inference model explicates five relationships, as illustrated in Figure 1.1: 1) Relationship 1, a projection (or looking-glass) effect (path a): People’s own opinions are thought to bias their estimates of public opinion. 2) Relationship 2, the hostile media effect (path b): Partisans perceive media coverage to be biased against their opinions.

When asked to rate the objectivity of the same news item, partisans on opposing sides of the issue report that the coverage is biased against them and in favor of their opponent’s view (Vallone, Ross, & Lepper, 1985). A relative hostile media effect may also be at work, 2 wherein partisans on both sides of an issue may not necessarily perceive media as biased against their views, but each will perceive media as more hostile to their own side relative to how the opposing side perceives it (Gunther et al, 2001). 3) Relationship 3, an extrapolation effect (path c): People fall prey to the law-of- small-numbers bias (Tversky & Kahneman, 1971) and see their own individual exposure to media content, however limited or idiosyncratic, as indicative of media content more generally (Gunther et al, 2001). 4) Relationship 4, the persuasive press inference (path d): People tend to believe that others’ opinions are significantly influenced by media content.

Thus, their perceptions of public opinion will reflect their perceptions of media content (Gunther, 1998). 5) Relationship 5, the perceived reach of media (path e): People believe that the media content they consume is the media content others consume. A higher media reach signifies a wider audience, with larger numbers of others exposed to the same media content as the individual (Gunther et al, 2001).1: Persuasive press inference model of public opinion, including (a) projection, (b) hostile media effect, (c) extrapolation, (d) persuasive press inference, (e) moderating effect of perceived reach (Gunther et al, 2001). Thus, the persuasive press inference model of perceived public opinion is essentially outlining a projection effect (or FCE) mediated by perceptions of press coverage.

Projection heightens perceived public opinion consensus, with individuals likely overestimating consensus; whereas perceptions of a relatively hostile media suppress perceived public opinion consensus, with individuals likely underestimating consensus. The two effects could conceivably cancel one another. Spiral of Silence Model of Perceived Public Opinion Generally speaking, the spiral of silence theory, as illustrated in Figure 1.2, suggests that people who believe they hold a minority viewpoint on a public issue will remain in the background where their communication will be restrained. Those who believe they’re a member of the opinion majority will be more encouraged to speak (Noelle-Neumann, 1974).

Moreover, the media will focus more on the majority views, underestimating the minority views. Those in the minority will be less assertive in 4 communicating their opinions, thereby leading to a downward spiral of communication. Interestingly, those in the majority will overestimate their influence and may become emboldened in their communication. Subsequently, the media will report on their opinions and activities (Noelle-Neumann, 1983).2: Spiral of silence model of public opinion (Noelle-Neumann, 1983).

Although Noelle-Neumann’s theory is pre-Internet age, her interpretation of the media’s influence still holds true. Today, given the popularity in usage of blogs, email, Listservs, and YouTube, we are bombarded with mediated messages. The 5 messages emanating from websites, television news, and e-commentary have contributed to our cultural discourse. Indeed, these media have even affected the direction of public discourse on socially significant issues (e.

The spiral of silence theory is predicated on three assumptions (Noelle- Neumann, 1991, 1993): 1) Society threatens deviant individuals with isolation; fear of isolation is pervasive and represents the centrifugal force that accelerates the spiral of silence. Noelle- Neumann based this claim on the famous conformity research of Solomon Asch (1951), who demonstrated that people will ignore the plain evidence of their senses and yield to perceived group pressure, because thoughts of isolation are very real to the person who considers standing firm. 2) This fear of isolation causes individuals to try to assess the climate of opinion at all times. Individuals receive information about public opinion from two sources: personal observation and the media.

Direct observation gives us only a small proportion of the information we use; thus, the media provide most of our knowledge about the world around us. Mistaken ideas of public opinion reflect the media failing to present a mix of viewpoints proportionate to their strength in society. 3) Public behavior is affected by public opinion assessment. Public behavior takes the form of either speaking out on a subject or keeping silent.

If individuals sense support for a topic, then they are likely to communicate about it; if they feel that others do not support a topic, then they maintain silence. To illustrate, consider how the spiral of silence would characterize the 6 Reagan/Carter presidential election campaign: Prevailing public opinion during the early fall of 1980 favored Reagan, but Carter supporters at first perceived only a slight discrepancy between their position and the prevailing public opinion. They felt comfortable expressing their views in public, perhaps even displaying campaign buttons or bumper stickers. But then the nagging fear of isolation convinced them to be more circumspect in what they said.

Bumper stickers disappeared, and they avoided arguments with Reagan supporters.