integrating theoretical traditions in media effects:...

Integrating Theoretical Traditions inMedia Effects: Using Third-PersonEffects to Link Agenda-Setting and

Cultivation

Leo W. Jeffres, Kimberly Neuendorf, andCheryl Campanella Bracken

Cleveland State University

David AtkinUniversity of Connecticut

In an earlier period of mass communication research, scholars were moreadventuresome in advancing ‘‘new’’ theories and less hesitant to ‘‘create’’theory. The 1970s, in particular, bore witness to the emergence of several suchtheories—from the knowledge gap and agenda-setting to cultivation. Scholarshave generated substantial literatures elaborating work in these and othertraditions. Those contributions are now sufficiently robust that it is time todirect some of our energies toward synthesizing theories. This article nomi-nates third-person perception as a candidate for such integration. Several

Leo W. Jeffres (Ph.D., University of Minnesota) is professor of communication at

Cleveland State University. His research interests include urban communication, media effects,

and political communication.

Kimberly Neuendorf (Ph.D., Michigan State University) is professor of communication at

Cleveland State University. Her research interests include communication technologies, film,

and media effects.

Cheryl Campanella Bracken (Ph.D., Temple University) is associate professor of communi-

cation at Cleveland State University. Her research interests include media effects and psycho-

logical processing of media with a focus on the concept of presence.

David Atkin (Ph.D., Michigan State University) is professor of communication at the

University of Connecticut. His research focuses on media effects, diffusion, audience analysis,

and media economics.

Correspondence should be addressed to Leo W. Jeffres, School of Communication, MU

253, Cleveland State University, Cleveland, OH 44115.

Mass Communication and Society, 11: 470–491

Copyright # Taylor & Francis Group, LLC

ISSN: 1520-5436 print=1532-7825 online

DOI: 10.1080/15205430802375303

470

prominent theories of media effects in the mass communication literature areselected to illustrate how the theories can or have been integrated. Resultsfrom three surveys provided evidence that the theories of third-person percep-tion, agenda-setting and cultivation can be interrelated. The propositionexamined here can serve as a model for further integration of other mediatheories. This integration attempt harkened back to the times when theorybuilding in media effects was more common and perhaps more optimisticabout explaining processes of influence.

Parsimony is one of the hallmarks of good theorizing (e.g., Casmir, 1994),and scholars responding to this criterion have advanced theory in masscommunication by advancing models posed at abstract levels that give direc-tion to researchers. For instance, the work of McCombs, Shaw, and Weaver(1997) in several places locates agenda-setting within a broader frameworkof cognitive effects. Each of these theoretical frameworks is certainly useful,but our research in media effects seldom illustrates how existing theories canbe linked or integrated.

A review of the literature on media effects over the past couple ofdecades shows strong lines of investigation in such ‘‘branded’’ theoriesas agenda-setting, cultivation and mainstreaming, framing and priming,the spiral of silence, diffusion, the knowledge gap, media=channeldependency and even iterations of the venerable two-step flow andhypodermic needle models. In addition, advances in the literature havefocused on effects while employing social categories and uses and grat-ifications as ‘‘mitigating’’ factors or stressing channel media differences(e.g., videomalaise). Others have focused on aspects of the message, thecontext, the medium, the technology, or the audience in studying parti-cular types of effects (e.g., from civic participation to stereotyping andbody image). Of course, we have the application of theories from otherdomains or disciplines that are necessarily employed and have expandedour knowledge of what happens in media effects, for example, sociallearning theory, the theory of reasoned action, exchange theory, the ela-boration likelihood theory, cognitive consistency, and the hierarchy ofeffects. This study explores how such integration might be tackledhead-on by ‘‘cross-tabbing’’ these theories to see what questions mightemerge to help us achieve parsimony. We then discuss how the third-person effects model is a strong candidate for bridging the other‘‘branded’’ theories in media effects. Following that, we extend our dis-cussion with an example applying third-person effects to two of thesetraditions—agenda-setting and cultivation. Finally, we describe threestudies that explore such a possibility.

INTEGRATING THEORETICAL TRADITIONS 471

EXPLORING LINKS BETWEEN THEORIES OF MEDIA EFFECTS

In an earlier period of mass communication research, scholars (e.g., Hage,1972) provided actual tools for creating theory by finding new variables,specifying definitions, ordering concepts, statements, and linkages. Hagesuggested synthesizing theories as well, but because so few existed withinmass communication, creation was the order of the day. Within that decade,we saw the explicit statements forming theories of agenda-setting, cultiva-tion, the spiral of silence theories, uses and gratifications, the knowledgegap, and media dependency.

As the discipline of communication grew, scholars ventured beyond‘‘media effects’’ and into larger systems models. In the early 1970s, lin-kages between media effects and the other social sciences seemed a possi-bility, and this is reflected in the texts that borrow heavily from sociologyand psychology (e.g., hegemony theory; Gitlin, 1980). Later in thatdecade, with the growth of ideological differences and cultural conflicts,discussions within communication became more difficult and groundshifted to emphasize epistemological problems. A special 1983 summerissue of the Journal of Communication, titled Ferment in the Field, is anexample of this (e.g., see Rosengren, 1983). Although conflict can be invi-gorating, stimulating research that otherwise would have failed to materi-alize, the cultural wars and advance of humanistic thought seems to haveretarded both the development of ‘‘branded’’ mass communication the-ories as well as efforts to integrate the literatures. This has been accompa-nied by isolationist practices such as scholars attempting to ‘‘custom’’build theories to fit their own research tracks, often simply putting oldwine into new bottles. These are unfortunate developments, because the1980s and 1990s certainly saw a dramatic expansion of research in mediaeffects and mass communication but little of the explosive development ofnew theories that help us to describe and explain a world of mass commu-nication constantly in flux.

Here, we put aside the epistemological conflict and place boundaries onour exploration into theory integration, focusing only on the area of mediaeffects within mass communication. We are taking a very practical approachto theory integration, and, to avoid the book-length text necessary for a fulltreatment of the following, we do not argue differences based on notions ofparadigms and perspectives, covering laws, the reliance on subjectivity inhuman action approaches, systems and institutional perspectives, or traitsversus states.

Third-person effects work is ideally suited for efforts to integrate othertheories. The notion of third-person effects is based on the differencebetween people’s perceptions of media effects on themselves versus others.

472 JEFFRES ET AL.

Davison (1983) argued that communication effects, which we view as mediaeffects, may be due not to audience reactions to the message itself but totheir perceptions or anticipation of how others react to the same message.Thus, perceptions of or anticipation of media influence on others isseen as a mediating variable in effects. This posits self-conscious behavioron the part of audiences and places third-person perception into the‘‘limited’’ effects camp. It should be noted that Davison was talking about‘‘persuasive’’ messages, which could lead some to argue that third-personperception should operate only with explicitly provocative messages suchas commercials, public service announcements, campaign messages, andsimilar media content rather than apply to news, sitcoms, or radio discus-sions. In another view, all communication is persuasive by definition, seek-ing to create some change in the state of audiences, even if it is justimparting a story or fulfilling some gratification. And all media effects the-ories focus on some form of change in audience beliefs, perceptions, knowl-edge, attitudes, behaviors, emotional states, or values. Table 1 presentsquestions raised by applying third-person effects in the context of nine othertheories or research traditions.

There are numerous approaches to theory integration. Long ago, Stinch-combe (1968) showed how to ‘‘pit’’ theories against each other in crucialtests that could shed light on which would receive empirical support in aDarwinian contest of natural selection. This approach is certainly viablewhen theories contest the same turf and argue for different interpretationsof the same phenomenon. Another approach is suitable when theories oper-ate at different levels, and this is the tactic actually followed by some res-earchers in their theoretical discussions and reviews. Ball-Rokeach (1998)did this in her comparison of assumptions and propositions of media systemdependency theory and uses and gratifications (also see Rubin & Windhal,1986). A third approach largely achieves a synthesis in media theories by‘‘telescoping’’ them into abstract approaches of other disciplines or othersocial sciences, turning communication into fodder for semiotics, (e.g.,Fortner, 1994). As our purpose is to strengthen theory construction andresearch within mass communication, we do not pursue this strategy here.

Another approach is to search the conceptual territory covered by eachtheory, finding common concepts and assumptions, then ask how one the-ory can inform the questions raised by a second theory. Each theory focusesour attention, leading us to raise questions that allow us to integrate the-ories and expand our knowledge of media effects. It is useful to considertwo theories seldom paired—the spiral of silence and agenda-setting; theformer focuses on the expression of opinions in public and the latter onhow media set audience agendas through increasing salience with greatercoverage.


TABLE 1

Applying Third-Person Perception in the Context of Other Effects Theories

Third-person effects

A. Agenda-setting Do people perceive greater agenda-setting by the

media on other people than on themselves? Is this

difference greater for the public ‘‘cognitive’’

agenda than for the ‘‘interpersonal’’ agenda?

B. Cultivation and Mainstreaming Do people perceive long-term effects of attention to

media content (e.g., crime shows, violent movies)

to have more impact on other people than on

themselves? Is this difference the same for media

effect in covering critical events (short term)?

Reversing the direction, do people who are more

extreme on issues perceive the media as having

more effect on the public mainstream than on

themselves?

C. Knowledge Gap If the knowledge gap occurs with regard to a

particular population, is the size of that gap

correlated with actual learning from the media,

thus confirming Davison’s notion that ‘‘actual’’

communication effects may be in part because of

our anticipation of media effects on others?

D. Diffusion Do people believe that media influence their own

awareness and adoption of new innovations more

than they do that of others?

E. Framing=Priming How do framing and priming processes affect third-

person perceptions? Does the existence of third-

person perceptions in a particular domain (e.g.,

politics) lead to greater media effects through

framing?

F. Spiral of Silence Would those with a larger third-person effects gap on

controversial issues be more or less likely to

express their opinion to strangers in public places?

G. Two-Step Flow & Opinion

Leadership

Is opinion leadership (like education) positively

related to the size of third-person effects gaps? Are

people who are among the less active segments of

the public (e.g., discussants or those influenced in

the two-step flow model) more likely to see a

smaller third-person effects gap because they see

themselves as more ‘‘vulnerable’’ or more ‘‘open’’

to influence on positive issues?

H. Media=Channel Dependency Is media=channel dependency on a topic correlated

with third-person perceptions, that is, do those

who rely on nonmedia sources of information see

themselves as being influenced less than others by

the media?

(Continued )

474 JEFFRES ET AL.

Since formulated by Noelle-Neumann (1973) 3 decades ago, the spiral ofsilence has generated volumes of work, with attention to numerous issues:the extent to which there is a pressure to conform in a context or society,whether the reluctance to express opinions is culture bound, whether andhow the reluctance to express opinions in public settings is affected by theclimate of opinion, whether there is one or multiple climates of opinion,how that climate of opinion is affected by media coverage, and the accuracyof people’s perceptions of the opinion climate (also see Noelle-Neumann,1993). Although traced to B. Cohen (1963), the work of McCombs andShaw (1974) has been more important in stimulating research intoagenda-setting, which has focused on numerous issues: the obtrusivenessof issues, models for how issues appear=disappear from public agendas,agenda building, influences on media agendas, intermedia agenda-setting,and processes at the individual level by which salience via attention to mediacoverage is achieved—including attention, priming, media uses and gratifi-cations, and social categories (see McCombs et al., 1997).

Agenda-setting is often linked to cultivation effects, as both are explain-ing an audience’s processing of messages leading to accumulative effectsacross time, but it also can direct us to ask different questions ‘‘within’’

TABLE 1

Continued

Third-person effects

I. Persuasion Selecting the hierarchy of effects from among the

persuasion theories=models, we might ask whether

the third-person effects gap would be greater for

the low involvement sequence than the other two

sequences.

Note. The questions raised in the table are intended to be illustrative, not exhaus-tive. Although not listed as an effects theory, uses and gratifications also can beadded to the list of mass communication theories linked to third-person effects. Stu-dies in uses and gratifications ask audiences to rate how important various mediaare for diversion, personal relationships—including social utility such as conversa-tions; personal identity, which includes personal reference or self-evaluation;exploration (ideas about personal concerns); and value reinforcement (McDonald& Glynn, 1984). When we ask whether media have an impact on us, we are askingfor the extent to which a media behavior has provided a particular use or gratifica-tion—helped us decide for whom to vote, help reinforce our values or decide what’simportant, give us things to talk about with friends. Following the same logic, weare asking for the extent to which media fulfill the same gratifications for otherpeople.


the framework of the spiral of silence. Thus, researchers have identifiednot only media agendas and public agendas but also interpersonal agendas(the relative importance of topics discussed with others) and polling agen-das. The focus on interpersonal agendas has received relatively less atten-tion since being raised years ago (see McLeod, Becker, & Byrnes, 1974),but it directs us to consider how media influence on interpersonal agendasfits within the spiral of silence. If greater media coverage of an issue raisesits salience among audiences, it also increases the likelihood that the topicis promoted for interpersonal discussion. Such coverage is greatest whenthere is the greatest controversy. The spiral of silence argues that coveragefavoring one view will discourage those who see their opinion as decliningfrom expressing their views in public. This creates a dilemma for some citi-zens, where media coverage promotes their participation in interpersonaldiscussions (agenda-setting) but ‘‘unbalanced’’ coverage discourages suchparticipation. Add to this mix the current abundance of polls, and theissue changes further. A number of interesting questions are raised forthe spiral of silence. Because polls ascertain people’s opinions in ‘‘private,’’we would expect minority views to receive their due in the percentagesrather than creating a ‘‘silent majority.’’ Thus, we’d ask whether the poll-ing agenda affects the media agenda, which affects the interpersonalagenda. We might ask under what conditions such agenda-setting wouldforestall the hypothesized dynamic of the spiral of silence, to the extentthat it would exist in a particular context. The next step is to conductempirical research that begins to explore some of these relationships andfind answers to questions raised. That’s what we propose to do in examin-ing three other theories next.

THIRD-PERSON PERCEPTION=EFFECTS AS ANINTEGRATING LINK

The third-person perception effects perspective is particularly well-suitedto bridging other theories about effects, as it is an ‘‘audience perception’’of effects, and we can literally insert variables from different effects mod-els=theories into the slot. In fact, the general third-person literature doesincorporate various versions of direct=limited=mediated effects into itsconsiderations. The third-person literature at its core posits a differencebetween perceived effects on self and others, with the difference betweenthird-person and first-person effects largely being the direction of theeffect, that is, whether it is a positive or negative media impact such asimproving healthy behaviors or inciting aggression. First-person effectsis often used to refer to a tendency to place oneself in a positive light

476 JEFFRES ET AL.

(White & Dillon, 2000), whereas third-person perception refers to placingothers in a more negative light. Reflecting on the research generated overthe years, Davison (1996) noted that third-person effect is a complex reac-tion that varies by type of communication, characteristics of the indivi-dual, and the social context.

Evidence supporting the third-person perception effect has been providedby numerous studies in different countries (Chapin, 1999; Davison, 1996;Y. Kim, Ahn, & Song, 1991; Paul, Salwen, & Dupagne, 2000; Perloff,1993, 1999; Shah, Faber, & Youn, 1999). Paul et al. (2000) examined32 studies in a meta-analysis, finding an overall effect size between estimatedmedia effects on self and others (r¼ .50). However, not all studies havefound third-person perception effects (Glynn & Ostman, 1988).

Consequences of the third-person perception have also been studied inseveral contexts (Hoffner et al., 1999; McLeod, Eveland, & Nathanson,1997; Price, Tewksbury, & Huang, 1998; Salwen, 1998; Salwen & Driscoll,1997; Youn, Faber, & Shah, 2000), but the bulk focuses on explanatoryfactors or variables influencing the phenomenon (e.g., self-enhancement,attribution theory, biased optimism, social distance from the ‘‘others’’;see Paul et al., 2000, for a summary of the literature on these theoreticalfoundations). The basis for third-person perception has been located inself-serving biases (see Eveland & McLeod, 1999), pluralistic ignoranceand status responses. Lopez-Saez, Martinez-Rubio, and Arias (1997)noted that third-person perception has been explained in attribution the-ory as a self-serving bias. Hoorens and Ruiter (1996) argued that peopleperceive their own responses to the media not as weaker but as moreappropriate than other people’s responses. Survey researchers also knowthat respondents often give responses to avoid offending the interviewer,to please the interviewer, or so they won’t appear in a negative light. Thistoo can influence perceptions of third-person effects in studies based onsurveys.

Some explanations for the third-person perception effect havefocused on individual characteristics of those perceiving the effects,for example, education and being knowledgeable (Driscoll & Salwen,1997; Lasorsa, 1989), media use (Peiser & Peter, 2000), and self-esteem(David & Johnson, 1998).

The relationship between ‘‘self ’’ and ‘‘other’’ also affects perceptions ofdifferential media impact. Studies show that the more distant the ‘‘other’’person is, or the more dissimilar she or he is to the perceiver, the greaterthe third-person effect (Gibbon & Durkin, 1995). An experiment by Cohen,Mutz, Price, and Gunther (1988) showed that readers thought defamatorynewspaper articles would affect others more than themselves, and theimpact was magnified as the ‘‘others’’ became more distant.


In addition, message attributes (Gunther & Thorson, 1992; White, 1997),the nature of the issue (Cohen & Davis, 1991; Duck & Mullin, 1995), themedium or channel (Atwood, 1993; Mason, 1995; Ten Eyck, 1998), andthe context (e.g., source bias when people are highly involved, Gunther& Mundy, 1993; political campaign advertising, Cohen & Davis, 1991;and election campaigns, Duck, Hogg, & Terry, 1995; Innes & Zeitz,1988; Salwen, 1998) are mitigating circumstances for third-person percep-tion effects. As Perloff (1999) noted, because people are more likely to seethe reverse—greater impact on self than others—when the message ispositive, or pro-social, self-enhancement biases provide a parsimoniousexplanation.

Applying Third-Person Perception to Agenda-Setting and Cultivation

Most of the literature on third-person perception effects has not made adistinction between different types of effects but has asked the public fora ‘‘general’’ impact assessment on a particular topic or behavior. Thethird-person perception literature has examined the following types of mediaeffects: television effects on thinking aggression is acceptable and viewingthe world as a dangerous place, also called ‘‘mean-world perceptions’’(Hoffner et al., 1999; Salwen & Dupagne, 2001); perceptions of risk ofgetting skin cancer (Morton & Duck, 2001); public service announcementeffectiveness in persuading people on organ donation (White & Dillon,2000); impact of gambling advertising (Youn et al., 2000); impact of TVprograms and commercials on perception of ideal body weight (David &Johnson, 1998); impact of defamatory newspaper articles on people’s opi-nion of a political and public sports figure (Cohen et al., 1988); impact ofcommercials on opinions of products (Gunther & Thorson, 1992); perceivedimpact of three media issues—a drunk-driving advertising campaign, mediaviolence and politics (Innes & Zeitz, 1988), perceived impact of press cover-age on opinions of the O. J. Simpson trial (Salwen & Driscoll, 1997). The listshows a variety of contexts (from politics to violence and body perceptions)and dependent measures (beliefs, opinions, attitudes, and behaviors). It doesinclude many variables that are comfortable in cultivation studies (e.g.,mean world perceptions), but it does not include the key notion in agendasetting—the relative salience of the target subject. We propose to applythird-person perception to not only a traditional impact topic but also toagenda-setting and cultivation, as proposed in Table 1. Because cultivationis based on a long-term notion of media effects, we want to focus on prob-ably the most enduring and hard-to-change variable—the core values thatrepresent what people think is important in life.

478 JEFFRES ET AL.

We ask people to consider the effect of media on themselves versusothers (third-person effects) in three areas: voting choice—a typical andimportant variable from the literature; agenda setting—what’s importantto think about; and values—what’s important in life. The third representsthe cumulative impact of cultivation theory. Thus, people are asked tomake a distinction between agenda-setting effects, media effects on votingchoices, and media effects on values. These three types of effects representa range from the relatively facile cognitive outcome of agenda setting, tothe perceived difficult-to-produce behavioral outcome of voting, to the alsodifficult-to-generate affective=attitudinal outcome of enduring values. Thefollowing hypotheses posit robustness of third-person perceptions forall three:

H1: People will perceive the media as being more important in telling otherpeople what to think about than the media are in telling them what tothink about (third-person agenda-setting effect).

H2: People will perceive the media as being more important in telling otherpeople for whom to vote than the media will be in telling them forwhom to vote (third-person voting effect).

H3: People will perceive the media as being more important in telling otherpeople what’s important in life than the media are in telling them what’simportant in life (media effect on values).

The logic supporting third-person effects generally should apply to all threeexpectations. We are predicting differences not based on direction. Thus,whether people think the behavior is good or bad, a difference is expected,whether it is based on biased optimism (e.g., they are more informed andthus less vulnerable to media influence on their voting decisions), ego invol-vement (e.g., I’m a political activist so of course I’m more independent in myjudgment than others who are less involved), or self enhancement (e.g., I’msmarter and don’t need help in finding something to talk about).

A separate issue focuses on whether we should expect the differences posedin the three hypotheses to be similar. For example, why would media impact onothers (vs. self) be greater for voting effects or values than agenda-setting? Acouple factors from the literature are suggestive. First, research tends to showthat media are more powerful in producing cognitive effects than in changingaffect, values, or behavior (see Jeffres, 1997), and the literature on politicalcommunication differentiates between such media effects as the influence ofendorsements on voting choices (operationalized as an affective or behavioraleffect) and agenda-setting (a cognitive effect; e.g., Kleinnijenhuis, van Hoof,Oegema, & de Ridder, 2007). Similarly, a distinction can be drawn betweenagenda-setting effects and media impact on values through the cumulativeimpact of messages—cultivation (see Gerbner, Gross, Signorielli, Morgan, &


Jackson-Beeck, 1979). We propose to introduce these distinctions in an applica-tion of the third-person effects hypothesis. The notion of agenda-setting in anelection context posits a relationship between media coverage and audience sal-ience, that is, the greater the prominence of particular issues or topics in mediamessages, the more salient the topic is among audiences; thus, the media areseen as setting the election agenda, telling voters what’s important in thecampaign.1

We also might expect the public to make the same distinction betweenperceived cognitive effects and perceived effects on values or political beha-viors. If the media are more successful in telling people ‘‘what to thinkabout’’ (McCombs & Shaw, 1993) than they are in influencing actual votingbehaviors and enduring affective traits, the public might recognize such adifference and demonstrate a smaller ‘‘third-person perception gap’’ foragenda-setting relative to media influence on the direction of one’s voteor on what is important in life. Although untested here, this is an interestingtwist on Davison’s argument that media effects may be due not to themessage but flow instead from people’s perceptions or anticipation ofhow others react to the same message.

Another point arguing for differences in third-person perceptions forvoting decisions, agenda-setting and values stems from the literature on‘‘self-serving’’ bias. One could argue that there would be less of a ‘‘statusresponse’’ or ‘‘self-serving’’ bias for acknowledging a media influence thatcould be viewed as ‘‘learning’’ or ‘‘understanding,’’ compared to acceptingthe notion that the media tell you whom to vote for or what is importantin life. This also is consistent with Hoorens and Ruiter’s (1996) notion thatpeople consider their responses as more appropriate. At the same time, weshould not expect the gap between perceptions of media influence on selfand other to close completely for either type of effect. Missing from the lit-erature are audience perceptions of the appropriateness of various mediaeffects. Thus, learning from the news is likely to be judged favorably, whereasbeing influenced by commercials is less likely to be judged favorably. But thatappropriateness also is likely to be influenced by the variety of factors in thethird-person perception literature; for example, a liberal may think thatbeing influenced by the New York Times editorial page is appropriate, buta conservative would think this entirely inappropriate. Clearly, there is roomfor more explanatory work here, investigating audience judgment of mediaeffects that take into account individual differences. In the spirit of theory

1For a sample of the extensive research on agenda-setting see Lopez-Escobar, Llamas, and

McCombs (1998); McCombs and Shaw (1993); McLeod et al. (1997); and Shaw, McCombs,

Weaver, and Hamm (1999).

480 JEFFRES ET AL.

integration, that line of inquiry might take into account the uses and gratifi-cations that sustain people’s media behavior patterns.

All of these explanatory mechanisms would predict a weaker third-personperception effect for cognitive impacts (i.e., agenda setting) than for eitheraffective=trait (i.e., values) or behavioral (i.e., voting choice) outcomes.Thus, drawing from the confluence of findings in the agenda-setting andthird-person traditions, we hypothesize the following:

H4a: The third-person perception effect for agenda-setting (telling voterswhat’s important in the campaign) will be smaller than the third-person effect for media influence on one’s voting choice (an effectsgap based on differential assessments).

H4b: The third-person perception effect for agenda-setting (telling voterswhat’s important in the campaign) will be smaller than the third-person effect for media influence on values (what’s important in life;an effects gap based on differential assessments).

As noted, the third perception person effects phenomenon has beenapplied across a variety of issues, populations, and contexts. Although expla-nations for the differential perceptions are generally social-psychologicaltendencies, we might find that people’s experience with the environmentand other people—either through interpersonal or mass communication—is also a factor. Indeed, past research has linked such social–psychologicalorientations as perceived knowledge and self-esteem, both related tocommunication activity, to the strength of third-person differentials(David & Johnson, 1998; Lasorsa, 1989). In addition, social categories couldsupplement communication variables to account for some of the differencein third-person effects (e.g., the work of Peiser & Peter, 2000). Because ofthe limited range and fragmented nature of past work on social categories,communication behaviors, and third-person, no hypotheses are offeredat this point, but communication variables and social categories willbe scrutinized as potential contributing explanations of the third-personphenomenon.

METHOD

The effort to integrate third-person perception effects with agenda-settingand cultivation was explored in a series of surveys. Two surveys wereconducted in the 21

2 weeks leading up to the 2000 presidential election.One survey was conducted in a metropolitan area of the Midwest, withrandom digit dialing interviews of 505 adults using a computer-aided


telephone-interviewing system. The second was a national survey of 2,172respondents—conducted over the Internet by a commercial research firmthat sends requests to a diverse set of potential respondents—who loggedonto the survey site to participate.2 The interview schedule for the telephonesurvey and the Internet instrument followed parallel forms, with no differ-ences in wording other than those created by the unique nature of the for-mats. Thus, the Internet interview included some matrices to display itemsthat were administered serially in the telephone survey.

Third-Person Effects Items for Surveys 1 and 2

There were two items operationalizing third-person agenda-setting effects:

How much influence would you say the media have in telling voterswhat’s important in the presidential campaign? (The response scale wasalmost no influence, a little influence, considerable influence, very muchinfluence)

How much influence would you say the media have in telling you what’simportant in the presidential campaign? (The response scale was almostno influence, a little influence, considerable influence, very much influence)

There were two items operationalizing third-person media effects onvoting choice:

How much influence would you say the media have on people’s actualvoting choice in the presidential race? (The response scale wasalmost no influence, a little influence, considerable influence, very muchinfluence)

How much influence would you say the media have on your voting choice inthe presidential race? (The response scale was almost no influence, a littleinfluence, considerable influence, very much influence)

Items referring to people’s voting choice were paired, as were items refer-ring to the respondents being interviewed. In other words, people wereasked how much influence the media have in telling voters what’s importantin the campaign, and that was followed with the item asking how muchinfluence the media have on people’s actual voting choice. This pair of itemsthen was followed by a second pair asking for the same assessments of

2The response or cooperation rate was 44% for the metropolitan study. Because the Internet

survey was not based on a probability sample, no response rate could be computed. This is less

important in situations where the focus is on relationships, not estimating parameters.

482 JEFFRES ET AL.

media effects on the respondent. Interviewers were trained to emphasize youand your in the second pair to be certain that the distinction was clear andrecognized. Numerous other measures were included in the study.3

A third survey was conducted the following summer in the same metro-politan region, using the same random digit dialing sampling proceduresand computer-aided telephone-interviewing system.4 The survey was pre-sented as a general poll with an emphasis on values and what people thinkis important in life. Respondents were asked to rate how important each of30 values were on a 0 (totally unimportant) to 10 (extremely important) scale.Values ranged from independence, participating in the political system, andpursuing wealth to religious beliefs, being physically fit, and having goodfriends.5 After a few intervening items, respondents were asked how oftenthe entertainment media as a whole—on television, in movies, in music,on the radio—portray the same set of values. Just before the demographicitems at the end of the survey, respondents were asked four items tappingthe third-person phenomenon. Thus, there should be no misunderstandingabout what was meant by the concept of ‘‘values.’’

Third-Person Perception Effects Items for Survey 3

There were two items operationalizing third-person agenda-setting effects:

How much influence would you say the media have in giving people ideas ortopics to think about or talk about? (The response scale was almost noinfluence, a little influence, considerable influence, or very much influence)

How much influence would you say the media have in giving you personallyideas or topics to think about or talk about? (The response scale wasalmost no influence, a little influence, considerable influence, or very muchinfluence)

There were two items operationalizing media influence on values:

How much influence would you say the media have in telling people what’simportant in life? (The response scale was almost no influence, a littleinfluence, considerable influence, or very much influence)

3Contact the authors for measures of other variables not reported here.4The response or cooperation rate was 45%.5The roster of 30 values statements was assembled from several sources: All but 3 of the 16

values used by Tan et al. (1997) to study ethnic adolescents were included in the study. In addi-

tion, values were added from the Rokeach (1973) scale and the Schwartz and Bilsky (1987,

1990) value types to represent a broader range of values for the general population.


How much influence would you say the media have in telling you personallywhat’s important in life? (The response scale was almost no influence, alittle influence, considerable influence, or very much influence)

Again, items referring to voters were paired, as were items referring to therespondents being interviewed. In other words, people were asked how muchinfluence the media have in telling people what’s important in life, and thatwas followed with the item asking how much influence the media have in giv-ing people ideas or topics to think about or talk about. This pair of items thenwas followed by a second pair asking for the same assessments of mediaeffects on the respondent. As before, interviewers were trained to emphasizeyou and your in the second pair to be certain that the distinction was clearand recognized. Numerous other measures were included in the study.

ANALYSIS AND RESULTS

Profiles of the three samples show that theory testing could proceed withdiversity on the relevant social categories. The 2000 metro sample was43% male and 57% female. Almost half were married, with 11% divorced,11% widowed, 3% separated, and 24% never having been married.6 One fifthof the sample was African American, and 64% was White, with the restother ethnic=racial backgrounds. The 2000 national Internet sample wasdominated by women (73%), with 62% married, 14% divorced, 2.3%widowed, 1.4% separated, and 21% never having been married.7 The samplewas 88.6% White. The third survey, conducted in 2001, was 45% male and55% female, with 43% married, 14% divorced, 11% widowed, 3% separated,and 29% never having been married.

8

One fifth of the sample (21%) wasAfrican American, and 64% was White.

6Although 27% of the sample was age 30 or younger, 19% were 31 to 40, 21% 41 to 50, 15%

51 to 60, 10% 61 to 70, and 7% 71 or older. Some 7.7% had less than high school education,

whereas 20% were high school graduates, 37% some college, 23% college graduate, and 12%

advanced degrees.7Ages ranged from 27% 30 years or younger, 28% 31 to 40, 27% 41 to 50, 14% 51 to 60, 3.8%

61 to 70, and .6% 71 or older. Less than 1.5% had less than a high school education, whereas

16% were high school graduates, 41% had some college, 30% were college graduates, and 11%

had advanced degrees.8Although 24% were 30 years or younger, 18% were 31 to 40, 21% 41 to 50, 13% 51 to 60, 7%

61 to 70, and 10.5% 71 or older. About 5% had less than a high school education, whereas 20%

were high school graduates, 32% some college, 24% college graduates, and 14% had advanced

degrees.

484 JEFFRES ET AL.

The first hypothesis predicted that respondents would perceive a third-person agenda-setting effect. The hypothesis was supported by data fromboth the metropolitan survey and the national Internet surveys conductedin 2000, as Table 2 shows. On a scale where 1 means the media have almostno influence and 4 means they have very much influence, in the Metropoli-tan survey perceived influence on agenda-setting for respondents was 2.08,compared to a perceived influence of 2.95 on other voters, and the differencewas statistically significant, t(487)¼ 16.9, p< .001. In the Internet survey,perceived influence on agenda-setting for respondents was 1.87, comparedto a perceived influence of 2.60 on other voters, and that difference alsowas statistically significant, t(2, 154)¼ 34.7, p< .001.

The second hypothesis predicted a third-person effect for media influenceon voting decisions. Again, the hypothesis received significant support in

TABLE 2

Testing Third-Person Effects (2000 Surveys)

Metropolitan

survey

National Internet

survey

M SD M SD

Perceived media influence on votinga

Influence on self 1.88 .93 1.70 .81

Influence on other voters 2.89 .88 2.54 .90

Perceived media influence on agenda

(what’s important in the presidential campaign)b

Influence on self 2.08 .96 1.87 .86

Influence on other voters 2.95 .94 2.60 .91

Differences between perceived third-person effectsc

Third-person effect for media influence on voting 1.01 .84

Third-person agenda-setting effect .87 .73

Note. A difference score for third-person effects was computed by subtracting per-ceived effects on ‘‘self ’’ from perceived effects on ‘‘other’’; thus, the higher the score,the greater the perceived third-person effects, that is, the media affect other peoplemore than self. Paired items and difference scores also have been used by other scho-lars examining third-person effects (e.g., Hoffner et al., 1999; McLeod, Eveland, &Nathanson, 1997; Salwen & Driscoll, 1997).

aMetropolitan survey, t(481)¼ 19.9, p< .001 (N¼ 482); national Internet survey,t(2, 148)¼ 38.1, p< .001 (N¼ 2,149).

bMetropolitan survey, t(487)¼ 16.9, p< .001 (N¼ 488); national Internet survey,t(2, 154)¼ 34.7, p< .001 (N¼ 2,155).

cMetropolitan survey, t(478)¼ 2.7, p< .01 (N¼ 479); national Internet survey, t(2,139)¼ 5.5, p< .001 (N¼ 2,140).


both 2000 surveys. In the Metropolitan survey, perceive influence of themedia on respondents was 1.88, compared to a perceived influence of 2.89on other voters, t(481)¼ 19.9, p< .001. In the Internet survey, perceivedmedia influence on voting choices of respondents was 1.70, compared to aperceived media influence of 2.54 on other voters, t(2, 148)¼ 38.1, p< .001.

The 2001 summer survey allowed us to test the first hypothesis a secondtime as well as test the third hypothesis that predicted a third-person effecton values. Table 3 provides results. Again, the first hypothesis predicting athird-person effect for agenda-setting was supported, operationalized thistime as media influence in telling people what to think or talk about (M influ-ence on self¼ 2.67, M influence on others¼ 3.26), t(289)¼ 10.5, p< .001. Thethird hypothesis—predicting a third-person effect of the media on values heldby people—received the same confirmation, with greater perceived mediaeffect on others (M¼ 3.15) than on self (M¼ 2.19), t(289)¼ 14.8, p< .001.

The fourth hypothesis predicted that the perceived third-person effect formedia influence on voting decisions and on values would be greater than thatfor media effects on agenda-setting. Again, the hypothesis received support inboth of the 2000 election surveys, although the difference was a bit morepronounced in the metro sample (14 vs. 11). The differences between the third-person effects were both statistically significant (see Table 2), t(478)¼ 2.7,

TABLE 3

Testing Third-Person Effects (2001 Survey)

M

Perceived media influence on valuesa

Influence on self 2.19

Influence on others 3.15

Perceived media influence on agenda

(what to think about or talk about)b

Influence on self 2.67

Influence on others 3.26

Differences between perceived third-person effectsc

Third-person effect for media influence on values .96

Third-person effect for media influence on agenda .59

Note. A difference score for third-person effects was computed bysubtracting perceived effects on ‘‘self’’ from perceived effects on‘‘other’’; thus, the higher the score, the greater the perceived third-person effects, that is, the media affect other people more than self.

at(289)¼ 14.8, p< .001 (N¼ 290).bt(289)¼ 10.5, p< .001 (N¼ 290).ct(287)¼ 5.3, p< .001 (N¼ 288).

486 JEFFRES ET AL.

p< .01, and t(2,139)¼ 5.5, p< .001. The fourth hypothesis also was confirmed inthe 2001 values survey, where the third-person effect on values was much greaterthan that found for agenda-setting (see Table 3), t(287)¼ 5.3, p< .001.

Next, we investigated whether third-person effects were related to eithersocial categories (achievement, ascriptive) or life cycle, individual differencesreflected in political philosophy, party identification or community links,and both interpersonal and mass communication variables. Despite substan-tial sample sizes for two of the three surveys, there was no clear pattern ofrelationships across the samples, and the handful of statistically significantcorrelations were likely due to chance. Regression analyses that controlledfor social categories did not reveal additional relationships. Thus, it appearslikely the third-person phenomenon is due to psychological and predisposi-tional factors noted in the literature and not measured here.

DISCUSSION

This study provided evidence that the theories of third-person perception,agenda-setting, and cultivation can be interrelated. Drawing from ourtheoretical synthesis, the study tested hypotheses (a) linking third-personperception concepts to other media effects (i.e., predicting personal agendas,voting choice, and cultivation of values), (b) predicting size of the third-person effects among these three (with third-person perceptions of media effectson personal agendas smaller than those on voting and values), and (c) placingthird-person perceptions as a dependent variable (examining possible relation-ships among social categories, political variables, media exposure factors,interpersonal communication variables, and third-person perceptions).

The proposition examined here can serve as a model for further integra-tion of other media theories. This integration attempt harkens back to thetimes when theory building in media effects was more common and perhapsmore optimistic about explaining processes of influence. Certainly there isroom for more focused lines of inquiry into the theories specifically, whichhave been apparent in examinations of differences in the nature of issues(e.g., obtrusive vs. unobtrusive), differences in the nature of ‘‘other’’ versusself (e.g., whether the other is similar or distant), differences in the nature ofthe message (e.g., positive vs. negative, weak vs. strong arguments), andrelationships with social characteristics of respondents (e.g., level of educa-tion). To this list we add the nature of the ‘‘effect’’ itself, and this is the keyto integration in this instance. Others may find alternative ways to integratetheories in media effects or mass communication in general.

Across three surveys, we found support for third-person effects formedia influence—on voting, agenda-setting, and values; all hypotheses were


supported. The fourth hypothesis, in particular, received support in both tel-ephone surveys, with a larger third-person effect for media influence on vot-ing than on agenda-setting and a larger effect for media influence on valuesthan on agenda-setting, suggesting audiences recognize differences in howmedia affect them.

However, subsequent analyses on likelihood of opinion expression uncoverfew significant patterns between third-person effects and social categories,individual differences reflected in political identification or community links,or interpersonal and mass communication variables. We can interpret thisdisconnect by noting that the third-person effects phenomenon is traced topsychological factors rather than social patterns or communication behaviors.

The lack of relationships with other variables is consistent with a focus onpsychological factors as the source of third-person effects. The third-personphenomenon has been predicated on notions of social desirability in respond-ing to interviewers, pluralistic ignorance, and status responses (e.g., Perloff,1999). Our results are consistent with these factors, but the differentialthird-person result for agenda-setting versus voter influence suggests thataudiences may have a more complex view of media effects and this shouldbe investigated further. The fact that audiences respond differently to pro-social versus anti-social messages also suggests that audiences do not treat‘‘effects’’ as unidimensional. People may have different perceptions of whichof the following effects are acceptable, or at least tolerable—media displace-ment of other activities, media effects in keeping them entertained (satisfyingspecific uses and gratifications), media effects that are cognitive—not onlyagenda setting but also learning and image formation, and media effects onchildren’s perceptions, attitudes, and behavior.

This brings up the point that the media environment today is mostcomplex and we shouldn’t expect audiences to be more realistic in theirassessments of how the media affect them. We might extend the researchon risk communication into the third-person effects area, asking whetherpeople differentiate between what they view as serious risks frommedia exposure and those that are seen as inconsequential. Later workmight profitably expand on our call to integrate theoretical traditions inmedia effects by using third-person perception as a link between agenda-setting and cultivation theories.

REFERENCES

Atwood, L. E. (1993). Perceived impact of an earthquake prediction: The third-person effect.

International Journal of Mass Emergencies and Disasters, 11, 365–378.

Ball-Rokeach, S. J. (1998). A theory of media power and a theory of media use: Different

stories, questions, and ways of thinking. Mass Communication & Society, 1, 5–40.

488 JEFFRES ET AL.

Casmir, F. L. (1994). Building communication theories: A socio=cultural approach. Hillsdale, NJ:

Erlbaum.

Chapin, J. R. (1999). Third-person perception and sexual risk taking among minority ‘‘at-risk’’

youth. Mass Communication & Society, 2, 163–173.

Cohen, B. (1963). The press and foreign policy. Princeton, NJ: Princeton University Press.

Cohen, J., & Davis, R. G. (1991). Third-person effects and the differential impact in negative

political advertising. Journalism & Mass Communication Quarterly, 68, 680–688.

Cohen, J., Mutz, D., Price, V., & Gunther, A. (1988). Perceived impact of defamation: An

experiment on third-person effect. Public Opinion Quarterly, 52, 161–173.

David, P., & Johnson, M. A. (1998). The role of self in third-person effects about body image.

Journal of Communication, 48(4), 37–58.

Davison, W. P. (1983). The third-person effect in communication. Public Opinion Quarterly, 47,

1–15.

Davison, W. P. (1996). The third-person effect revisited. International Journal of Public Opinion

Research, 8, 113–119.

Driscoll, P. D., & Salwen, M. B. (1997). Self-perceived knowledge of the O. J. Simpson

trial: Third-person perception and perceptions of guilt. Journalism & Mass Communication

Quarterly, 74, 541–556.

Duck, J. M., Hogg, M. A., & Terry, D. J. (1995). Me, us and them: Political identification and

the third-person effect in the 1993 Australian federal election. European Journal of Social

Psychology, 25, 195–215.

Duck, J. M., & Mullin, B. (1995). The perceived impact of the mass media: Reconsidering the

third-person effect. European Journal of Social Psychology, 25, 77–93.

Eveland, W. P. Jr., & McLeod, D. M. (1999). The effect of social desirability on perceived

media impact: Implications for third-person perceptions. International Journal of Public

Opinion Research, 11, 315–333.

Fortner, R. S. (1994). Mediated communication theory. In F. L. Casmir (Ed.), Building commu-

nication theories: A socio=cultural approach (pp. 204–240). Hillsdale, NJ: Erlbaum.

Gerbner, G., Gross, L., Signorielli, N., Morgan, M., & Jackson-Beeck, M. (1979). The demon-

stration of power: Violence profile no. 10. Journal of Communication, 29(3), 177–196.

Gibbon, P., & Durkin, K. (1995). The third person effect: Social distance and perceived media

bias. European Journal of Social Psychology, 25, 597–602.

Gitlin, T. (1980). The whole world is watching: Mass media in the making & unmaking of the New

Left. Berkeley: University of California Press.

Glynn, C. J., & Ostman, R. E. (1988). Public opinion about public opinion. Journalism

Quarterly, 65, 299–306.

Gunther, A. C., & Mundy, P. (1993). Biased optimism and the third person. Journalism

Quarterly, 70, 58–67.

Gunther, A. C., & Thorson, E. (1992). Perceived persuasive effects of product commercials and

public service announcements: Third-person effects in new domains. Communication

Research, 19, 574–596.

Hage, J. (1972). Techniques and problems of theory construction in sociology. New York:

Wiley.

Hoffner, C., Buchanan, M., Anderson, J. D., Hubbs, L. A., Kamigaki, S. K., Kowalczyk, L.

et al. (1999). Support for censorship of television violence: The role of the third-person effect

and news exposure. Communication Research, 26, 726–742.

Hoorens, V., & Ruiter, S. (1996). The optimal impact phenomenon: Beyond the third person

effect. European Journal of Social Psychology, 26, 599–610.

Innes, J. M., & Zeitz, H. (1988). The public’s view of the impact of the mass media: A test of the

third person effect. European Journal of Social Psychology, 18, 457–463.


Jeffres, L. W. (1997). Mass media effects. Prospect Heights, IL: Waveland Press.

Kim, Y., Ahn, J., & Song, J. (1991, May). Perceived media influence on self and others on a

controversial issue. Paper presented to the Political Communication Division at the annual

conference of the International Communication Association, Chicago.

Kleinnijenhuis, J., van Hoof, A. M. J., Oegema, D., & de Ridder, J. A. (2007). A test of

rivaling approaches to explain news effects: News on issue positions of parties, real-world

developments, support and criticism, and success and failure. Journal of Communication,

57, 366–384.

Lasorsa, D. L. (1989). Real and perceived effects of ‘‘Amerika.’’ Journalism Quarterly, 66, 373–378.

Lopez-Escobar, E., Llamas, J. P., & McCombs, M. (1998). Agenda setting and community

consensus: First and second level effects. International Journal of Public Opinion Research,

10, 335–348.

Lopez-Saez, M., Martinez-Rubio, J. L., & Arias, A. V. (1997). The third-person effect in the

electoral campaign: Analysis from the point of view of the superior conformity of self

[Abstract]. Revista de Psicologia Social, 12, 153–166.

Mason, L. (1995). Newspaper as repeater: An experiment on defamation and third-person

effect. Journalism & Mass Communication Quarterly, 72, 610–620.

McCombs, M. E., & Shaw, E. L. (1974). The emergence of American political issues: The agenda-

setting function of the press. St. Paul, MN: West.

McCombs, M. E., & Shaw, D. L. (1993). The evolution of agenda-setting research—25 years in

the marketplace of ideas. Journal of Communication, 43, 58–67.

McCombs, M. E., Shaw, D. L., & Weaver, D. H. (Eds.). (1997). Communication and democracy:

Exploring the intellectual frontiers in agenda-setting theory. Mahwah, NJ: Erlbaum.

McLeod, J. M., Becker, L. B., & Byrnes, J. E. (1974). Another look at the agenda-setting

function of the press. Communication Research, 1, 131–166.

McLeod, D. M., Eveland, W. P., Jr., & Nathanson, A. I. (1997). Support for censorship of

violent and misogynic rap lyrics: An analysis of the third-person effect. Communication

Research, 24, 153–174.

Morton, T. A., & Duck, J. M. (2001). Communication and health beliefs: Mass and interper-

sonal influences on perceptions of risk to self and others. Communication Research, 28,

602–626.

Noelle-Neumann, E. (1973). Return to the concept of powerful mass media. Studies in Broad-

casting Communication, 9, 69–105.

Noelle-Neumann, E. (1993). The spiral of silence: Public opinion—our social skin (2nd ed.).

Chicago: University of Chicago Press.

Paul, B., Salwen, M. B., & Dupagne, M. (2000). The third-person effect: A meta-analysis of the

perceptual hypothesis. Mass Communication & Society, 3, 57–85.

Peiser, W., & Peter, J. (2000). Third-person perception of television-viewing behavior. Journal

of Communication, 50(1), 25–45.

Perloff, R. M. (1993). The dynamics of persuasion. Hillsdale, NJ: Erlbaum.

Perloff, R. M. (1999). The third-person effect: A critical review and synthesis. Media Psychol-

ogy, 1, 353–378.

Price, V., Tewksbury, D., & Huang, L. (1998). Third-person effects on publication of a

Holocaust-denial advertisement. Journal of Communication, 48(2), 3–26.

Rokeach, M. (1973). The nature of human values. New York: Free Press.

Rosengren, K. E. (1983). Communication research: One paradigm, or four? Journal of Commu-

nication, 33(3), 185–207.

Rubin, A. M., & Windhal, S. (1986). The uses and dependency model of mass communication.

Critical Studies in Mass Communication, 3, 184–199.

490 JEFFRES ET AL.

Salwen, M. B. (1998). Perceptions of media influence and support for censorship: The third-

person effect in the 1996 presidential election. Communication Research, 25, 259–285.

Salwen, M. B., & Driscoll, P. D. (1997). Consequences of third-person perception in support of

press restrictions in the O.J. Simpson trial. Journal of Communication, 47(2), 60–78.

Salwen, M. B., & Dupagne, M. (2001). Third-person perception of television violence: The role

of self-perceived knowledge. Media Psychology, 3, 211–236.

Schwartz, S. H., & Bilsky, W. (1987). Toward a psychological structure of human values.

Journal of Personality and Social Psychology, 53, 550–562.

Schwartz, S. H., & Bilsky, W. (1990). Toward a theory of the universal content and structure of

values: Extensions and cross-cultural replications. Journal of Personality and Social Psychol-

ogy, 58, 878–891.

Shah, D. V., Faber, R. J., & Youn, S. (1999). Susceptibility and severity: Perceptual dimensions

underlying the third-person effect. Communication Research, 26, 240–267.

Shaw, D. L., McCombs, M., Weaver, D. H., & Hamm, B. J. (1999). Individuals, groups and

agenda melding: A theory of social dissonance. International Journal of Public Opinion

Research, 11, 2–24.

Stinchcombe, A. L. (1968). Constructing social theories. New York: Harcourt, Brace, World.

Tan, A., Nelson, L., Dong, Q., & Tan, G. (1997). Value acceptance in adolescent socialization:

A test of a cognitive-functional theory of television effects. Communication Monographs,

64, 82–97.

Ten Eyck, T. A. (1998). Processing information: Interpersonal and mass mediated communica-

tion and the modern self. Current Research in Social Psychology, 3, 35–47.

White, H. A. (1997). Considering interacting factors in the third-person effect: Argument

strength and social distance. Journalism and Mass Communication Quarterly, 74, 557–564.

White, H. A., & Dillon, J. F. (2000). Knowledge about others’ reaction to a public service

announcement: The impact on self persuasion and third-person perception. Journalism &

Mass Communication Quarterly, 77, 788–792.

Youn, S., Faber, R. J., & Shah, D. V. (2000). Restricting gambling advertising and the third-

person effect. Psychology and Marketing, 17(7), 633–649.


integrating theoretical traditions in media effects:...

Documents