John W. Williams, Political Science,

Principia College, Elsah, IL 62028

************************************************************************

This paper was presented at the 1997 annual meeting of the Midwest Association of Public Opinion Research.

************************************************************************

Fields and Schuman have stated, "In the absence of strong counter-forces, a large proportion of people feel that the world they live in agrees with their own opinions on public issues" (1976, 445). Sociological theory postulates that individuals perceive and interpret their environment through social interactions (Merton, 1968). Lack of interaction, characterized by inadequate interpersonal communication and/or use of mass media may lead to substantial inaccuracies in the perception of public opinion and behavior, crystallizing in the concept of pluralistic ignorance (Katz & Allport, 1931; Allport, 1937; Scheff, 1975). Research indicates that people tend to draw on external features of others, such as race and gender, to estimate what entire classes of others think (Glynn, Ostman & McDonald, 1995). This process relies on stereotypes, both cultural and social, and often results in widespread misperceptions, or pluralistic ignorance. For example, during the late 1960s the public consistently underestimated shifts in white support for desegregation because of misperceptions of the obstinacy of the whites (OíGorman, 1975). Pluralistic ignorance describes the situation when a minority position in public opinion is incorrectly perceived to be the majority position and vice versa (O'Gorman, 1975, 1979, 1980, O'Gorman & Garry, 1976, Taylor, 1982). Taylor summarized:

The alternative to pluralistic ignorance is the "looking glass effect," the accurate belief that others think the same as oneself (Breed & Ktsanes, 1961; Fields & Schuman, 1976; O'Gorman, 1980, 1979, 1975; O'Gorman & Garry, 1976; Taylor, 1982). The looking glass effect can have the majority accurately perceiving the position of the majority, known as the "true looking glass effect," or the minority accurately perceiving the position of the minority, known as the "false looking glass effect" (Scheff, 1975). Lemert tested the looking glass effect through several surveys prior to votes on two controversial ballot measures in Oregon (1986). The study compared the accuracy of election predictions by politicians with those made by the voting public. He hypothesized and found that "politicians were better than voters at predicting the outcomes of two controversial ballot measures, and they displayed fewer and weaker signs of the Looking-Glass effect" (1986, 219).

Studies have suggested that communications must account for some of the variations in the accuracy of these perceptions (Breed & Ktsanes, 1961; Fields & Schuman, 1976), even though Lazarsfeld, et. al. (1944), discovered that very few voters are converted by political propaganda. Prior to his 1983 article on "The Third Person Effect in Communication," W. Phillips Davison "became involved in the local phase of a national election, serving as a volunteer for his preferred candidate's organization." In reaction to receiving a leaflet of impressive quality from the opposition, Davison "procured a pile of political literature from his own party's local office and spent the rest of the day distributing it door to door." However, he found that "informal post-election analyses suggested that neither set of propaganda materials had exerted much influence on the voters." Davison asked himself why he had assumed that the rival candidate's leaflet would be so effective. The result of his inquiry led to his formulation of the "third person effect hypothesis." (Davison, 1983) In his landmark article, Davison reported a series of minor and informal efforts to test whether "an individual who is exposed to a persuasive communication via the mass media will see this communication as having a greater effect on other people than on himself or herself." (1983, p. 4) Davison wrote:

People selectively perceive media messages (Sears & Freedman, 1967). The "third person effect" hypothesis suggests that people also selectively perceive how others interpret media messages. The hypothesis predicts that individuals systematically perceive others to be more susceptible to media influence than themselves (Davison, 1983; Gunther, 1991, 1995; Lasora, 1992; Mutz, 1989; Perloff, 1993; Schoenbach & Becker, 1995). Several studies have explored various aspects of the hypothesis and found that individuals overestimate the impact that media exerts on others (Perloff, 1989; Lasorsa, 1989; Mutz, 1987), although these findings have not been universal (Glynn & Ostman, 1988). The opposite of the third person effect, in which others are believed to be less influenced or hold less liberal positions, has been called the "conservative bias" (Fields & Schuman, 1976).

Some of the testing of the third person effect has been experimental. For example, Cohen, et. al. (1988), in an experiment involving audience and jury assessments of the effects of defamatory communications, found the existence of a third person effect. They found that defamatory communications are perceived to affect others more than oneself and that as the definition of others is broadened the perceived effect on others increased. Tyler and Cook (1984) noted, in their investigation of the "impersonal impact hypothesis," that people distinguish between two possible levels of judgment, societal and personal. In a cross-sectional survey, Glynn and Ostman found "little evidence of a societal-level [as opposed to a personal level] phenomenon of a 'third person effect,'" although they cautioned that the effect should not be dismissed (1988, p. 305). Based on a series of experiments, Tyler and Cook concluded:

Some third person effect studies indicate that as geographic distance between self and others increases people perceive the media to exert greater influence on others than on themselves (Cohen, Price, Mutz & Gunther, 1988; Glynn, 1987; Gunther, 1991). Cohen and Davis (1991), however, found no evidence for geographic distance. Conners (1994), in a study that examined social distance as well as geographic distance, found some though limited evidence of a perceptual bias. Perceptual bias is measured by pairs of questions: a question about their own belief in the media, and a question about others belief in the media. Perceptual bias is the difference between "self" and "others." Respondents can be classified as third-person (those who perceive greater media effects on others than on themselves), first-person (those who perceive greater media effects on themselves than on others), and no difference (looking glass or false looking glass effects).

The results of Cohen's (1988) defamation experiment suggested the presence of a "distance" element within the third person effect. Cohen, et. al., found that their subjects perceived that defamatory communications affected others more as the definition of others broadened. Distance, whether geographic or cultural, or the absence of distance (proximity) could explain the results of a 1990-91 study by Williams (1991). The study was based on the results of exit polling during the 1990 elections in Illinois. The results revealed little third person effect. The respondents indicated close physical and social proximity.

What none of these studies have done is to assess the nature of pluralistic ignorance and third person effect over time. Does pluralistic ignorance increase or diminish over time? Likewise, how does the third person effect change? This study is an exploratory study of the stability of these two effects--pluralistic ignorance and third person--over time.

This study involved two exit polls. The first survey was conducted on the day of the 1992 Illinois general election. Surveys were offered to every person who had just voted at two polling stations in Jersey County, Illinois. The same study was repeated at the same two polling stations four years later on the day of the 1996 Illinois general election. The results are based on respondent self-reporting. The survey consisted of thirty-two statements with a four-element Likert scale. Respondents responded if they strongly agreed, agreed, disagreed or strongly disagreed with each statement. Scattered among the thirty-two statements were six pairs of statements designed to elicit information useful in understanding pluralistic ignorance and the third person effect. The six pairs of questions, though not in order presented, were:

* I believe the media is biased. * Most people believe the media is biased.

* I trust the press in this election. * Most people trust the press in this election.

* I trust campaign ads. * Most people trust campaign ads.

* I believe polls. * Most people believe polls.

* Campaign ads never influence me. * Campaign ads never influence most people.

* Polls never influence me. * Polls never influence most people.

For reasons related to other goals of the study, two polling places were selected (Elsah 2, Elsah Township Building; Mississippi 1, Mississippi Township Hall, McClusky). The two precincts were selected, in part, because of large and high turnout and because the precincts have a strong record of competitive voting. With formal approval of the Jersey County Election Authority (the local authority charged by the State of Illinois to conduct elections), trained undergraduate pollsters were stationed at both polling places from 7 am (one hour after opening of the polls) to 7 PM on each election day. They politely asked every person exiting the polling places to complete the one page study. On at least one occasion, the process was observed by officials from the Secretary of Stateís staff.

Number of people registered 1339 1371

Number of people voting 1106 967

Number of people responding 457 489

Response rate 41.32% 50.57%

Mean year born 1949 1948

Mean length of time in county 29 years 24 years

Gender split (% female/male) 52/48 55/45

Party identified (% Demo/Rep) 55/45 53/47

Pluralistic ignorance and third person effect were identified and measured by a 2x2 crossbreak and chi-square test. As Chart I indicates, agreement between majority opinion of self and others is termed Looking Glass while agreement between minority opinion of self and others is termed False Looking Glass. Disagreement between self and others is either pluralistic ignorance or third person effect (if we are speaking about media influence). As noted earlier, the third person effects literature also assigns terms to the direction of disagreement.

I believe YES I believe NO

Others believe YES Looking Glass Pluralistic Ignorance

Effect /Third Person Effect

Others believe NO Pluralistic Ignorance Looking Glass

/Third Person Effect Effect

The 1992 study revealed a looking glass effect for four of the categories and pluralistic ignorance or third person effect for two of the categories. The 1996 study revealed a looking glass effect for three of the categories, pluralistic ignorance or third person effect for two of the categories, and no pattern or discernible effect for one category. In every instance, the strength of the effect declined albeit slightly, as indicated by chi-square values and significance. The results are listed in Table II.

(n, chi-square, significance)

Belief of media bias Strong Looking Glass Effect Strong Looking Glass Effect

417, 54.246, p<.000 429, 41.168, p<.000

No trust in press Strong Looking Glass Effect Strong Looking Glass Effect

441, 15.744, p<.000 461, 8.197, p<.005

No trust in campaign Strong Looking Glass Effect Looking Glass Effect

ads 421, 12.770, p<.000 434, 6.218, p<.013

Belief in polls Pluralistic Ignorance Pluralistic Ignorance

420, 34.154, p<.000 433, 14.856, p<.000

Influence of Looking Glass Effect No pattern or effect

campaign ads 427, 6.777, p<.01 442, 2.540, p=.111

Influence of polls Third Person Effect Third Person Effect

423, 6.463, p<.02 438, 5.926, p<.02

The first four items pertain to pluralistic ignorance. Both in 1992 and 1996, respondents accurately viewed others as sharing (looking glass effect) their belief of media bias, of distrust of the press, and of distrust of campaign ads, though the strength of the effect appears to have weakened. On the other hand, both in 1992 and 1996, respondents inaccurately (pluralistic ignorance) believed that others believed the polls while the respondents did not, though the strength of this effect also appears to have weakened.

The last two items pertain to the third person effect. Respondents accurately viewed others as not influenced by campaign ads in 1992. A note of caution should be raised at this point because the third person effect does not refer to actual influence but to the disparity of perception between how one is influenced and how others are influenced. Thus, the looking glass effect examines perception rather than actual impact of media influence. Respondents acknowledged they were influenced by campaign ads and stated that others were also influenced by campaign ads. The looking glass effect disappeared over the four year period, though not sufficiently to indicate the existence of a third person effect. On the other hand, respondents inaccurately assessed the claims of influence of polls, thereby asserting a third person effect (respondents claimed not to be influenced, while others were influenced), though there may have been a slight weakening of the effect.

These results indicate little movement in opinion. If, as the above results indicate, there was some stability in effects, whether pluralistic ignorance, third person or looking glass, then the movement had to be in both sets of answers--self and others. In order to identify the direction and strength of the changes, the 1992 and 1996 results were subjected to independent sample t-tests measuring change in means from 1992 to 1996. Of the twelve items, ten revealed significant change over the four years. Within the six pairs--self and others--the change was never opposite. The results of the t-tests are set forth in Table III.

(t-test)

t-value significance nature of change

"I" Belief of media bias 26.72 p<.000 agree>strongly agree

"Others" 16.66 p<.000 agree>strongly agree

"I" Distrust in press 10.18 p<.000 agree>strongly agree

"Others" 1.80 not significant no change

"I" Distrust in campaign ads -10.09 p<.000 agree>disagree

"Others" - .16 not significant no change

"I" Belief in polls - 7.88 p<.000 agree>disagree

"Others" - 2.70 p<.00 agree>disagree

"I" No influence of campaign ads 2.40 p<.02 agree>strongly agree

"Others" .58 not significant no change

"I" No influence of polls - 4.00 p<.000 agree>disagree

"Others" -18.25 p<.000 agree>disagree

This study examines the stability of pluralistic ignorance and third person effect over a four year period. Based on the chi-square values and significance levels, there has been a slight diminution of the various effects--pluralistic ignorance and third person effect, both based on misjudgment of others, as well as looking glass effect, based on accurate assessment of others. The shift strongly correlates (.9559, p=.003) across all six sets of questions. Although there was some diminution in the chi-squares, five of the items maintained their respective effects over time. Only one item, that of influence of campaign ads (related to third person effect), changed. Over the four years, this item shifted from looking glass effect to no discernible pattern or effect.

A comparison of the two sets of statistical tests reveals where the diminution occurred. The change occurred not because of direction of change, but because of the absence of change in how respondents viewed "others." For the items of belief in media bias, belief in polls, and influence of polls, both the self and others shifted significantly in the same directions, thereby maintaining the effects from 1992 to 1996. However, for the items of trust in press, trust in campaign ads, and influence of campaign ads, self shifted significantly while others did not shift. The impact of this situation is revealed in the differences in the chi-square significance levels between 1992 and 1996 for the three items. However, this phenomenon only impacted one item--influence of campaign ads--to any extent.

Conceptually, the looking glass effect is opposite of pluralistic ignorance (with knowledge) and the third person effect (with influence). Thus, one would expect an overall diminution of pluralistic ignorance or third person effect to result in an overall maintenance or strengthening of the looking glass effect. For example, as we believe others are becoming less influenced by polls, one would expect that we would not believe that others would become more influenced by campaign ads. Likewise, as we believe others are becoming less trusting of polls, one would not expect us to become more trusting of campaign ads or the media.

The two surveys reveal shared and continuing belief in media bias and distrust of the press. The questions were phrased as opposites--belief of media bias and trust in media--yet the results were consistent. However, as the t-tests revealed, opinion generally shifted toward greater belief in media bias and greater trust in media over the four year period. This is certainly a confusing message for the media.

The four questions pertaining to campaign ads and election polls (both relating to trust/belief and influence) were phrased as opposites--trust in ads, belief in polls, not influenced by ads, not influenced by polls. In three instances, there were shifts away from trust in ads, belief in polls, and influence of ads. Respondents believe that polls are gaining in influence, especially of others, since this phenomenon revealed a third person effect. Thus, respondents believe that others are become more influenced by election polls.

One of the intriguing revelations is the difference in how respondents approach campaign advertising and election polls. Respondents claim that they do not believe in nor are influenced by campaign ads. And, respondents assert that others do not believe in nor are influenced by campaign ads. This may reflect some strong local revulsion to negative campaigning (Williams, 1996). On the other hand, respondents claim that they do not believe in nor are influenced by election polls. However, respondents assert that others do believe and are influenced by the polls. These two studies reveal, with regard to election polls, the existence of both pluralistic ignorance and third person effect.

This study compared the existence and change of pluralistic ignorance, third person effect and their opposites--looking glass effect--over a four year period. The results indicate stability in these effects over time. This stability can be explained by a number of factors. First, there is stability in the geographic or physical proximity of the respondents. The region is stable in population and the study was repeated in the same election precincts. Furthermore, there was no significant change in the nature of the communication systems. The community has been served by the same television, radio and newspapers, with the single addition of a small weekly community paper, over the four year span. Likewise, there was no indication of instability in the informal, interpersonal or small group communication patterns (churches, social clubs, political parties, etc.). Second, there is stability in social or cultural proximity of the respondents. With a stable population, there has been little change in the socio-economic structure or mix of the community. Although this was not tested, there was probably little shift in how the respondents identified "others."

This study suggests the need to explore the temporal nature of pluralistic ignorance and third person effect. As might be expected, the effects appear stable in a community that is stable, both socially and in communication systems. With this base of data, this study should be continued over time. Future studies of this type should be set in more heterogeneous communities and in communities undergoing various types of changes. These changes, whether in social structures, political attitudes, or media systems, might serve as the independent variables for the future studies. For researchers with the patience and resources for longitudinal studies, this should be a important direction for pluralistic ignorance/third person effect research.

* Allport, Floyd (1937). "Toward a science of public opinion." Public Opinion Quarterly 1(1):7-23.

* Breed, Warren and Thomas Ktsanes (1961). "Pluralistic ignorance in the process of pinion formation." Public Opinion Quarterly 25(3):382-392.

* Cohen, Jeremy, and R.G. Davis (1991). "Third-person effect and the differential impact of negative political advertising." Journalism Quarterly 68(4), 680-688.

* Cohen, Jeremy, Diana Mutz, Vincent Price and Albert Gunther (1988). "Perceived impact of defamation: An experiment on third-person effects." Public Opinion Quarterly 52(2):161-173.

* Conners, Joan (1994). "The third-person effect and social distance: Exploring who are the others." Paper presented at Association for Education in Journalism and Mass Communication, Atlanta, Georgia, August 1994.

* Davison, W. Phillips (1983). "The third-person effects in communication." Public Opinion Quarterly 47(1):1-15.

* Fields, James and Howard Schuman (1976). "Public beliefs about the beliefs of the public." Public Opinion Quarterly 40(4):427-448.

* Glynn, Carroll J. (1987). "The communication of public opinion." Journalism Quarterly 64(4):688-697.

* Glynn, Carroll J., and Ronald E. Ostman (1988). "Public Opinion About Public Opinion." Journalism Quarterly 65(2):299-306.

* Glynn, Caroll J., Ronald E. Ostman, and D.G. McDonald (1995). "Opinions, perceptions, and social reality." In T.L. Glaser & C.T. Salmon (Eds.), Public opinion and the communication of consent (pp. 249-277). New York: The Guilford Press.

* Gunther, Albert C. (1995). "Over rating the X-rating: The third-person perception and support for censorship of pornography. Journal of Communication, 45(1), 28-39.

* Gunther, Albert C. (1991). "What we think others think: Cause and consequence in the third-person effect. Communication Research, 18, 355-372.

* Katz, Elihu and Paul F. Lazarsfeld (1955). Personal Influence. Glencoe, IL: Free Press.

* Katz, Daniel and Floyd H. Allport (1931). Student Attitudes. Syracuse, NY: Craftsmen Press.

* Lasorsa, Dominic L. (1992). "Policy-makers and the third-person effect." In J.D. Kennamer (Ed.), Public Opinion, the Press and Public Policy (pp. 163-175). New York: Praeger.

* Lasorsa, Dominic L. (1989). "Real and Perceived Effects of 'Amerika'." Journalism Quarterly 66(2):373-378, 529.

* Lazarsfeld, Paul, B. Berelson, and H. Gaudet (1944). The People's Choice. New York, NY: Columbia University Press.

* Lemert, James (1986). "Picking the winners: Politician vs. voter predictions of two controversial ballot measures." Public Opinion Quarterly 50:208-221.

* Merton, Robert (1968). Social Theory and Social Structure. New York, NY: The Free Press.

* Mutz, Diana C. (1989). "The influence of perceptions of media influence: Third-person effects and the public expression of opinions." International Journal of Public Opinion Research, 1, 3-23.

* Mutz, Diana C. (1987). "Perceptions of Others in the Public Opinion Process: The Third Person Effect and the Spiral of Silence." Paper presented to the annual convention of the International Communication Association, Montreal.

* O'Gorman, Hubert J. (1980). "False consciousness of kind: Pluralistic ignorance among aging." Research on Aging 2(1):105-128.

* O'Gorman, Hubert J. (1979). "White and black perceptions of racial values." Public Opinion Quarterly 43(1):48-59.

* O'Gorman, Hubert J. (1975). "Pluralistic ignorance and white estimates of white support for racial segregation." Public Opinion Quarterly 39(3):313-330.

* O'Gorman, Hubert J. and Stephen L. Garry (1976). "Pluralistic ignorance -- a replication and extension." Public Opinion Quarterly 40(4):449-458.

* Perloff, Richard M. (1993). Third-person effect research, 1983-1992: A review and synthesis. International Journal of Public Opinion Research, 5, 167-184.

* Perloff, Richard M. (1989). "Ego-Involvement and the Third-Person Effect of Televised News Coverage." Communication Research 16(2):236-262.

* Perloff, Richard M., K. Neuendorf, D. Giles, R.K. Chang, and Leo W. Jeffres (1992). "Perceptions of ëAmerikaí." Mass Comm Review, 19(3), 42-47.

* Rucinski, D., and C.T. Salmon (1990). "The ëotherí as the vulnerable votger: A study of the third-person effect in the 1988 US presidential campaign." International Journal of Public Opinion Research, 2, 345-368.

* Scheff, Thomas J. (1975). "Toward a sociological model of consensus." American Sociological Review 32(1):32-46.

* Schoenbach, K., and Lee B. Becker (1995). "Origins and consequencs of mediated public opinion." In T.L. Glaser & C.T. Salmon (Eds.), Public opinion and the communication of consent (pp. 323-347). New York: The Guilford Press.

* Taylor, D. Garth (1982). "Pluralistic ignorance and the spiral of silence: A formal analysis." Public Opinion Quarterly 46:311-355.

* Tyler, Tom R. and Fay Lomax Cook (1984). "The Mass Media and Judgments of Risk: Distinguishing Impact on Personal and Societal Level Judgments," Journal of Personality and Social Psychology 8:195-200.

* Williams, John W. (1991). "Pluralistic Ignorance and Third Person Effect," paper given at Midwest Association of Public Opinion Research, Chicago, Illinois, November 1991.

* Williams, John W. (1993). "Third Person Effect in Local Political Media," paper given at Midwest Association for Public Opinion Research, Chicago, Illinois, November 1993.

* Williams, John W. (1996). "Negative Campaigns--Who does it hurt more, the target or the source?," paper given at Midwest Association for Public Opinion Research, Chicago, Illinois, November 1996.