Mass media

University of Pennsylvania ScholarlyCommons Departmental Papers (ASC) Annenberg School for Communication 2010 Mass Media Diana Mutz Annenberg School for Communication, mutz@upenn.edu Seth Goldman Follow this and additional works at: https://repository.upenn.edu/asc_papers Part of the Communication Commons Recommended Citation Mutz, D., & Goldman, S. (2010). Mass Media. The SAGE Handbook of Prejudice, Stereotyping and Discrimination, 241-258. https://doi.org/10.4135/9781446200919.n15 This paper is posted at ScholarlyCommons. https://repository.upenn.edu/asc_papers/624 For more information, please contact repository@pobox.upenn.edu. Mass Media Abstract The way outgroup members are portrayed in the media is widely believed to have consequences for levels of prejudice and stereotyping in the mass public. The visual nature of television and its heavy viewership make it a key source of information for impressions that ingroup members may have of other social groups. However, most research to date has focused on documenting the portrayals of various groups in television content, with only a few studies documenting the causal impact of television viewing. To further understanding of this hypothesis, we outline the contributions and limitations of past work, and point to the most promising theoretical frameworks for studying media influence on outgroup attitudes. Disciplines Communication | Social and Behavioral Sciences This book chapter is available at ScholarlyCommons: https://repository.upenn.edu/asc_papers/624 Mass Media Abstract The way outgroup members are portrayed in the media is widely believed to have consequences for levels of prejudice and stereotyping in the mass public. The visual nature of television and its heavy viewership make it a key source of information for impressions that ingroup members may have of other social groups. However, most research to date has focused on documenting the portrayals of various groups in television content, with only a few studies documenting the causal impact of television viewing. To further understanding of this hypothesis, we outline the contributions and limitations of past work, and point to the most promising theoretical frameworks for studying media influence on outgroup attitudes. Mass Media Stereotypes, Gordon Allport wrote, ‘are socially supported, continually revived and hammered in, by our media of mass communication–by novels, short stories, newspaper items, movies, stage, radio, and television’ (1954, p. 200). Yet, Allport provided no direct evidence that media exposure increased stereotyping or prejudice. Today most researchers concur that a systematic agenda examining the nature and consequences of mass media on stereotyping and prejudice is warranted, but lacking – an oversight this chapter hopes to begin to correct. We limit our discussion of mass media to television, the dominant medium in countries with welldeveloped national media systems. Moreover, the audio-visual nature of the medium best approximates faceto-face intergroup contact, and makes the group identities of people and characters on television salient to viewers, thus facilitating potential effects on outgroup attitudes. Overview Fifty years after Allport's observation about the potential importance of media, scores of studies have examined representations of a broad range of social groups in news, entertainment, and advertising, but empirical evidence of effects from exposure has lagged considerably behind descriptive studies of media content. Content analyses have examined portrayals of African Americans, Latinos, gays and lesbians, women, and older people, as well as a smattering of other social groups. Perhaps because of television's heavy emphasis on the criminal justice system in both fictional and non-fictional programming, the most common subjects have been crime and criminality, and media portrayals that link crime to racial or ethnic outgroups. Content is only half of the story, but historically it is where the greatest research effort has been focused. Despite the many studies of outgroup portrayals, scholars have been unable to systematically sample from a universe of media content, or sample from the same programs over a long period of time. In the United States, for example, television archives include only a few news programs that are no longer as widely viewed, and contain no entertainment programs at all. In other countries, systematically collected broadcast archives are even more difficult to come by. For this reason, research tends not to characterize the portrayal of outgroups in a given media environment of a given country or time period, but instead characterizes a particular television program or small set of programs at a particular point in time. Little, if any, systematic evidence of change goes beyond impressionistic accounts to document how televised images of social groups have changed over time. Beyond content analyses, scholars do not even agree about the predicted direction of effects from the same media content. For example, will frequent media portrayals of well-to-do African Americans improve Whites’ attitudes toward Blacks, or only serve to convince Whites that Blacks who have not ‘made it’ are not trying hard enough? This chapter is organized into three parts, progressing from a discussion of research on media content, to a review of evidence of actual effects on stereotyping and prejudice. To date, a relatively small body of evidence bears on the critical issue of impact. Moreover, much of the research is correlational, showing associations between amount of television viewing and prejudice, but leaving causality disappointingly ambiguous. Finally, we review the most promising theoretical frameworks for future examinations of media effects on stereotyping and prejudice. Because of the limited progress that has been made in this area of research, we suggest a reordering of priorities, essentially reversing the emphases to date. Instead of descriptive analyses of media content, we suggest that scholars first direct their efforts toward a theoretical understanding of what kinds of content will influence prejudice and stereotypes and through what process. Without knowing what kinds of content are most important in shaping viewers’ ideas about outgroups, or the process by which media representations exercise influence, scholars studying media content alone are blindly guessing about what is worth analysing. The small number of studies documenting effects is not all that surprising in light of the lack of theoretical frameworks to guide this research. By offering three potentially fruitful theoretical frameworks, we hope to draw related research together in productive ways. Media Portrayals of Outgroups In lieu of an exhaustive list of findings about portrayals of various outgroups in different genres of media content, we focus our discussion on the multiple analytical frameworks used to examine media content, and what they suggest about the need for a greater theoretical understanding of how media exposure affects viewers’ perceptions of outgroups. The varying strategies of comparative analysis employed in contentanalytic studies suggest different implicit theories as to the kinds of content that are likely to influence audiences. Moreover, although the results of a given content-analytic study tend to be specific to the media of a given country, a particular television program, and a historical point in time, content-analytic strategies are not specific to any national boundaries. We illustrate these analyses with examples drawn primarily, though not exclusively, from studies of American media, where content analysis has been a particularly popular approach. However, the same problematic theoretical issues pertain equally well to other media environments. Analyses of media content have generally come in one of three forms (see Dixon & Linz, 2000a, 2000b). One variety, intragroup comparisons, considers how common a certain role, behavior, or characteristic is among members of a social group relative to that same social group in some other role. On local and national network news, for example, Blacks are more commonly portrayed as perpetrators than victims of crime (Dixon & Linz, 2000b; Dixon, Azocar, & Casas, 2003; Romer, Jamieson, & de Coteau, 1998). Along similar lines, a study of reality-based police programs showed that Blacks and Hispanics were more often depicted as perpetrators than police officers (Oliver, 1994). Yet, in prime-time as a whole, Blacks are more likely to be seen as police officers than as perpetrators, and are rarely shown as victims (Tamborini, Mastro, Chory-Assad, et al., 2000). As these illustrative studies suggest, intra-group comparisons do not suggest consistently more negative portrayals across genres or roles. The usual implication drawn from such studies is that the more Blacks are depicted in high-versus low-status roles, the more positive white viewers’ attitudes should become, and vice versa. But it is unclear how one would expect this information to influence attitudes toward outgroups. The world of prime-time television has an unusually high percentage of lawyers, doctors, and law enforcement personnel, regardless of race. So what is influential could instead be the sheer number of Blacks shown in high-status roles. As Whites become accustomed to seeing Blacks as doctors, lawyers, judges, and police, harmful negative stereotypes may change. Moreover, depictions of Blacks as loving parents on sitcoms might likewise alter White viewers’ attitudes. The second content-analytic approach asks whether certain roles, behaviors, and the like are more commonly portrayed among members of one social group relative to members of another group. These intergroup comparisons are most often used to contrast media portrayals of one racial outgroup relative to a majority of ingroup, with the assumption that more positive portrayals of outgroups relative to ingroups will improve attitudes toward outgroups. Focusing, as in the example earlier, on Whites’ attitudes toward Blacks, findings from intergroup comparisons have been inconsistent with respect to whether Blacks or Whites are more commonly shown as the perpetrators of crimes. In studies of local TV news, some have found that most perpetrators were Black (Dixon & Linz, 2000b; Gross, 2006), while others have found that most perpetrators were White (Gilliam & Iyengar, 2000; Klite, Bardwell, & Salzman, 1997). Because these studies differed in many ways – including the time frames, cities, and the number of TV stations analysed within each city – it is impossible to pinpoint the source of variations in the results. On network news, prime-time, and in reality-based shows (e.g., America's Most Wanted), Whites were more likely to be shown as perpetrators than Blacks (Dixon, Azocar, & Casas, 2003; Entman, 1994; Oliver, 1994; Tamborini, Mastro, Chory-Assad, et al., 2000). Still other intergroup comparisons focus on more subtle differences in portrayals of one group relative to another. For example, Black suspects on local TV news were more likely than Whites to be shown poorly dressed (i.e., in jeans and a t-shirt or wearing jail clothing), in mug shots, and without a specified name (Entman, 1992; Entman & Rojecki, 2000). According to the authors of these studies, the implicit message is twofold: that people of color are more likely to be guilty and dangerous than White criminal suspects; and that the ‘individual identity [of a Black suspect] does not matter … the accused is part of a single undifferentiated group of violent offenders: just another Black criminal’ (Entman & Rojecki, 2000: 82). Because prejudice is centrally concerned with intergroup relations, still other intergroup comparisons have considered how members of different social groups interact (or fail to) in the mass mediated world, comparing characteristics of interracial relationships to same-race relationships. Interestingly, most interracial interactions in prime-time television in the United States are hierarchical, occurring in the workplace between a higherranked employee and a subordinate, whereas most White-White interactions take place between peers (Entman & Rojecki, 2000). Perhaps surprisingly, Black characters are more likely to be in the superior than the subordinate position relative to Whites – what Entman and Rojecki (2000) call a ‘utopian reversal’ relative to the likely positions of Blacks and Whites in the real world. Although it is clear that effects on viewers from intergroup and intragroup comparisons would involve different processes and could lead to substantively different conclusions, it is rare for scholars to consider the two side-by-side. One notable exception comes from Gamson's (1998) analysis of portrayals of gays and lesbians on daytime television talk shows. On the one hand, he found that gays on these programs (e.g., Jerry Springer and Ricki Lake) were often portrayed stereotypically (e.g., flamboyant, hypersexual, and incapable of maintaining healthy romantic relationships). On the other hand, heterosexual guests were portrayed in similarly unflattering ways. More important than the stereotyping of gays (i.e., the intragroup comparison) was the appearance of similarity between gays and straights (i.e., the intergroup comparison): ‘Not only are we everywhere, apparently, we are also just as loud, goofy, dysfunctional, funny, nasty, emotional, and combative as everyone else’ (Gamson, 1998: 64). The third type of content-analytic approach rests on the assumption that media's effects on prejudice and stereotyping will be observable only when mediated representations of out-groups fail to mirror the real world. To identify these inconsistencies, television-reality comparisons compare portrayals of social groups in mass media to the real-world frequency of the same characteristic. For instance, compared to government arrest statistics, local TV news over-represents Whites as perpetrators of crime in both portrayals of violent and nonviolent crime; Blacks are slightly over-represented as perpetrators of violent crime (Gilliam, Iyengar, Simon, et al., 1996). On network news, both Blacks and Whites are represented as perpetrators of violent and non-violent crime in the same proportions as they are in national government arrest statistics (Dixon, Azocar, & Casas, 2003). Reality-based police shows also portray both groups accurately as perpetrators of violent crime, although they under-represent Whites and over-represent Blacks in non-violent crime stories (Oliver, 1994). Television-reality comparisons tend to use national or local statistics to make their central points about over-or under-representation. And yet, upon reflection it seems obvious that few, if any, people are in touch with ‘reality’ as it exists statistically at a national, or even a local level. Instead, perceived realities tend to be rooted in people's immediate environments and networks; television is probably most influential when it deviates from those realities rather than from official statistics on abstract entities such as cities, counties, or nations. Still, one might expect heavy television viewers’ images of their nation to have less variance than those of nontelevision viewers because heavy viewers’ perceptions would drift toward the televised version of reality, whereas non-viewers should instead reflect variations in personal networks and local realities (see Mutz, 1992). Through content analyses, scholars have provided numerous points of entry for researchers interested in studying potential effects of media exposure on prejudice and stereotyping. But the perspective used in analysing media representations can lead to divergent findings and contradictory predictions. One particularly illuminating example is the contentious debate over The Cosby Show, a hugely popular primetime hit featuring an upper middle-class black family, in which the mother is a lawyer and the father is a doctor. For Jhally and Lewis (1992), the show fails to represent the true situation of most African-Americans, who are disproportionately likely to be less well-off than Whites. The probable result, they argue, is the impression among Whites that Blacks are no longer economically disadvantaged; rather, the message is that Blacks who try hard can succeed (as the Cosby family does), while those who do not must be lazy. Bogle (2001) argues, alternatively, that the Cosby family is a refreshing example of a counter-stereotypic representation of Blacks in mass media. Examples of middle-class Black families were almost non-existent until the airing of The Cosby Show, thus perhaps this content countered Americans’ tendency to inaccurately stereotype poor people as overwhelmingly Black (e.g., Gilens, 1996). The theories used to predict effects from television content are no less ambiguous in that they do not suggest which aspects of media content are most important to the outcome. For example, two different analyses of Will & Grace, a prime-time show featuring two gay male characters, produced opposing predictions. Schiappa, Gregg, and Hewes (2006) emphasized that the gay characters were likeable, with the resulting hypothesis that exposure to the show would lead viewers to form more positive judgments about gay men. Others predicted more negative attitudes about gay men because of the stereotypically effeminate portrayals of these same characters, and their apparent inability to have healthy, long-lasting romantic relationships (Battles & HiltonMorrow, 2002; Gross, 2001). Unfortunately, the absence of empirical data on the validity of these predictions limits the usefulness of this approach. Indeed, the content-analytic approach more generally is plagued by a wealth of interesting descriptive findings that in the end cannot tell us much about the effects of media on prejudice or stereotyping. The ultimate lesson of our review is that content-analytic studies are, despite their illuminating qualities, inherently speculative. Effects of Outgroup Portrayals Studies of media impact on prejudicial attitudes date back at least to the 1940s, when results most often suggested limited or no impact due to selective perception; that is, viewers rejected the intended premise of the message because it did not mesh with their pre-existing prejudices. For example, some people who read comic strips designed to ridicule a character named ‘Mr. Biggott’ dismissed the cartoon character as so unusual and extreme that they simply ridiculed him without examining the implications of the cartoon for their own prejudices (Cooper & Jahoda, 1947; Kendall & Wolf, 1949). Decades later in the 1970s, a study of the hit prime-time show All in the Family produced similarly disappointing findings. Producers of the program claimed that it ridiculed Archie Bunker, the white family's openly-racist father. Yet, a survey of viewers revealed that many people saw ‘nothing wrong’ with Archie's racial slurs; by the end of a typical episode, these viewers believed that Archie, rather than his anti-racist son-in-law, had ‘won’ (Vidmar & Rokeach, 1974; see also Brigham & Giesbrecht, 1976). Concerns about selectivity in exposure and perception of media messages remain today. And like the studies described above, most of the research conducted since the 1970s has been observational rather than experimental. Surveys have demonstrated significant correlations between self-reported media exposure – including overall recalled television exposure, exposure to particular topics and genres, and exposure to specific programs – and prejudice toward a variety of social groups. So, for instance, three metaanalyses reported correlations between self-reported media exposure and stereotypical beliefs about women, though the type of self-report measures employed were not specified (Herrett-Skjellum & Allen, 1995; Mares & Woodard, 2005; Oppliger, 2007). In the case of race, surveys showed positive correlations between both overall recalled TV viewing and watching All in the Family, on the one hand, and more prejudice toward Blacks, on the other (Gross, 1984; Vidmar & Rokeach, 1974). On the subject of sexuality, one survey showed a correlation between overall recalled TV viewing and more prejudice toward gays (Gross, 1984), while another survey showed a correlation between watching Will & Grace and less prejudice toward gays (Schiappa, Gregg, & Hewes, 2006). As a final example, one survey demonstrated a correlation between self-reported exposure to TV about the homeless and lower levels of prejudice toward the homeless (Lee, Farrell, & Link, 2004). Correlational evidence, however, provides a weak basis for causal inference for a multitude of reasons in this particular case. First, because many of the analyses failed to control for factors related to both media exposure and prejudice, the association between these two variables may have been spurious (e.g., Gross, 1984; Oppliger, 2007; Signorielli, 1989). Second, the association between exposure and prejudice could be accounted for by reverse causation; that is, people selectively exposing themselves to media content congruent with their prejudices (e.g., Ball-Rokeach, Grube, & Rokeach, 1981; see also, Morgan, 1982, 1987). Third, in observational studies media exposure to positive or negative portrayals of outgroups is inferred through selfreports. The well-known weaknesses of self-reported exposure measures in terms of both validity and reliability (see, e.g., Bartels, 1993; Price & Zaller, 1993), combined with the lack of evidence that these respondents were exposed to any prejudice-reducing or enhancing messages when watching, means that many of these studies lack a convincing connection between exposure to media portrayals of outgroups and attitudes toward those same outgroups. For these reasons, we focus our review on studies that are experimental or quasi-experimental in design. Notably, the outcome measures of stereotyping and prejudice used in these studies vary widely – from beliefs about the outgroup as a whole, to judgments about outgroup members in unrelated situations, to behaviors. Nonetheless, collectively these studies make a convincing case that exposure to mass media has the capacity to alter levels of prejudice in both positive and negative directions. For example, using a longitudinal quasi-experimental design to examine the impact of television viewing on adolescents’ sex-role attitudes, greater television viewing produced more sexist attitudes six months to a year later (Morgan, 1982, 1987). An Australian field study evaluating the effects of a campaign designed to reduce the belief that indigenous Australians (Aborigines) were lazy produced similar effects, this time in the direction of reducing prejudice (Donovan & Leivers, 1993). Compared to respondents surveyed before the campaign, the postcampaign sample was more likely to believe that Aborigines remained in their jobs for more than one year. The strongest evidence to date for a causal link between mass media exposure and prejudice comes from five studies employing fully-randomized experimental designs. Exposure to a sympathetic documentary about one of the first openly-gay elected officials in the United States (The Times of Harvey Milk) reduced negative attitudes toward gays (Riggle, Ellis, & Crawford, 1996). Further, a study carried out in Germany exposed adolescents to one talk show episode a day over five days – each including tolerant content about gays. A week after the final exposure, participants in the treatment condition reported stronger pro-gay attitudes (Rossler & Brosius, 2001). In two other experimental studies, watching multiple episodes of programs including gay male characters (Six Feet Under and Queer Eye for the Straight Guy) or a stand-up comedy act performed by Eddie Izzard (Dress to Kill) dressed in women's attire led to more tolerant attitudes toward gay men and transvestites, respectively (Schiappa, Gregg, & Hewes, 2005). In a fifth study, participants viewed a televised comedy skit portraying Blacks stereo-typically (poor, uneducated, and prone to acts of violence and crime) or to a neutral comedy skit featuring Blacks, but not in stereotypical ways (Ford, 1997). They subsequently read a vignette about a student accused of physically assaulting his roommate – with no conclusive evidence, but some circumstantial evidence of guilt. The accused person was named either Tyrone or Todd, to suggest a Black or White suspect. Whereas the perceived guilt of the White student suspect (Todd) did not vary by condition, participants who viewed the stereotypical portrayal of Blacks first were more likely to perceive the Black student (Tyrone) as guilty than were participants who viewed the neutral portrayal. In addition to these five experiments, two additional studies provide evidence claiming that portrayals of outgroup members ‘prime’ prejudice and stereotyping. Although these findings are not framed as direct evidence that media increase or decrease prejudice, given that the results are consistent with either interpretation, we include them as support for this general argument. In the first experiment, participants saw newsletters including autobiographical essays by either a stereotypic Black college student, a counter-stereotypic Black student, or a control. Participants in the stereotypic condition were more likely to endorse the anti-Black stereotypes highlighted in the treatment (lazy, aggressive, unintelligent, and socially destructive) than were participants in the counter-stereotypic condition. But most importantly, participants in the stereotypic condition were more likely to generalize these conclusions to seemingly unrelated people, becoming increasingly likely to suggest that African-American Rodney King brought the highly publicized beating by Los Angeles police on himself (relative to the counter-stereotypic condition), while participants in the counter-stereotypic condition were more likely to say that King was innocent (relative to the control and stereotypic conditions) (Power, Murphy, & Coover, 1996). In a similar experiment, participants who saw newsletters including autobiographical essays by a stereotypic female college student were more likely to endorse negative stereotypes of women (self-centered, weak, overemotional, and unintelligent) than were participants in a counter-stereotypic condition. Participants who read the stereotypic portrayal also generalized these stereotypes to other situations, becoming less likely to believe the sexual harassment allegations Anita Hill made against Clarence Thomas during his US Supreme Court nomination hearings (relative to the control and counter-stereotypic conditions) (Power, Murphy, & Coover, 1996). Thus, despite preceding decades of disappointingly inconclusive or null findings, experimental studies have demonstrated that media exposure to even a single outgroup member can both produce and reduce prejudice toward a variety of social groups. Yet the issue of selective exposure is still yet to be fully addressed. Selective exposure is of particular concern in generalizing from experimental studies because experiments force people to watch television programs that they might not otherwise have chosen to view. Thus, these findings leave us confident that media can, in fact, alter levels of prejudice, but not that media, as it occurs and is widely viewed by the public, often does so in real world settings. Concerns about the potential for selective exposure are heightened by the growth of cable television and the enormous increase in the range of program choices now available to the average viewer. On the one hand, greater choice should enable viewers to more easily avoid content that might contradict their views. But on the other hand, only a small proportion of programming wears its outgroup politics on its sleeve. When people watch a crime drama, for example, they seldom select it for the anticipated race of the victims versus perpetrators. And sitcoms are watched because they are funny or clever, not because of the stereotypes they convey. Moreover, there is an element of voyeurism in television viewing that may attract viewers to precisely the kind of content they find titillating, though repugnant and disagreeable. Jerry Springer and related programs are interesting to watch precisely because they feature people who are unlike those most people know in their everyday lives (e.g., a father who marries his child's grandmother, Ku Klux Klan parents, and so forth). Thus the exercise of selectivity in viewing may be incomplete at best. Overall, our own assessment of the likelihood of positive influence on outgroup attitudes from television is far more optimistic than those of earlier scholars who argued that television merely reflected and reinforced existing prejudices and stereotypes. First, mass media provide a potential source of ‘contact’ that ingroup members can have with outgroup members. The omnipresence of mass media in contemporary life means that the majority of people are exposed to outgroup members more through mass media than through face-to-face contact (e.g., Bowman & Foster, 2006; Charles, 2003; Dixon & Rosenbaum, 2004; Logan, 2001). Thus media constitute an especially important source of information about minority group members with whom majority group members otherwise have limited or no face-to-face contact. Second, although some televised exposure to outgroup members undoubtedly reinforces negative outgroup stereotypes, it also exposes viewers to more positively-valenced stereotypes than they are likely to encounter in everyday life, if only because of their relative isolation from outgroup members. Moreover, blatantly stereotypical portrayals of outgroup members often produce a public outcry that focuses attention on the negative stereotype, thus negating its potential impact (see Mendelberg, 2001). Unfortunately, the relative extent of positive to negative portrayals of a given outgroup in a given culture's television programming or in a given individual's chosen content remains largely unknown and probably highly variable across individuals as well as cultures. However, to the extent that some positively-valenced portrayals reach viewers through media, when they generally do not reach people through other avenues, one might expect media's net contribution to be positive – that is, unless positive portrayals produce negative consequences, as has been argued by some. Ultimately, however, this is an empirical question, and one that is unanswerable without a theoretical framework from which to understand media's impact. Theoretical Frameworks Three theoretical perspectives seem potentially applicable to understanding the role of media in prejudice and stereotyping. However, as we argue later, evidence to date points to one of these theoretical frameworks as particularly well suited to the expansion of our knowledge of the influence of mass media on prejudicial attitudes. For this reason, we focus first and more briefly on the alternatives, and then turn to a more lengthy discussion of the most promising model. Although the empirical studies discussed earlier generally reference at least one of these theoretical perspectives, the evidence itself typically does not allow the reader to distinguish support for one theoretical model from another. Parasocial Interaction First coined in 1956 (Horton & Wohl, 1956), the term parasocial interaction means that viewers feel and react toward people and characters on television just as they do in face-to-face interactions (Kanazawa, 2002). More recently, Schiappa and his colleagues (2005) proposed the ‘parasocial contact hypothesis,’ positing that if viewers get to know and like outgroup members on television, then their attitudes toward the outgroup as a whole will improve. As implied by its title, this perspective proposes that mediated contact fits alongside faceto-face intergroup contact (e.g., Pettigrew & Tropp, 2006) as a viable strategy for reducing prejudice. Yet both the conceptualization and measurement of what constitutes ‘parasocial interaction’ remain highly variable (e.g., Giles, 2002). One study, for instance, included measures of whether viewers felt they knew the characters, found them physically attractive, wanted to be their friend, thought they did their jobs well, or perceived themselves as similar to the characters (Schiappa, Gregg, & Hewes, 2005). Moreover, the impact of parasocial interaction on prejudice reduction has received mixed empirical support, with a preponderance of either unsupportive evidence or evidence that could be interpreted through multiple theoretical frameworks (Schiappa, Gregg, & Hewes, 2005, 2006). Nonetheless, studies of narrative persuasion and transportation – that is, becoming ‘absorbed in the narrative world, leaving the real world, at least momentarily, behind’ (Green & Brock, 2002: 317) – further underline the possibility that viewers may become so immersed in a storyline, and so empathic with respect to characters and their interactions, that they experience the same kind of human contact that fuels the positive effects of intergroup contact (Green & Brock, 2000). Although the extent to which people report experiencing transportation has been associated with more positive evaluations of sympathetic characters (Green & Brock, 2000), to date evidence linking transportation with beliefs about social groups is lacking (Green, 2004). However, to the extent that viewers do form affective bonds with television characters, this perspective opens up the possibility of not only direct parasocial contact effects (i.e., the viewer has a parasocial relationship with an outgroup member), but also indirect parasocial contact effects (i.e., the viewer has a parasocial relationship with an ingroup member who has a positive relationship with an outgroup member). Support for this idea comes from research showing prejudice-reducing effects from either having an ingroup member say that he or she had a friendly interaction with an outgroup member, or by witnessing a friendly intergroup interaction (Wright, Aron, McLaughlin, et al., 1997). If merely witnessing a friendly interaction can produce these effects in interpersonal contexts, then witnessing intergroup contact on television may produce similar influence. Intergroup contact via mass media may be particularly advantageous because it avoids the anxiety that often characterizes face-to-face intergroup interactions (Dovidio, Gaertner, & Kawakami, 2003; Greenland & Brown, 1999; Pettigrew & Tropp, 2006; Stephan & Stephan, 1985). Anxiety and feelings of threat are known barriers to achieving the benefits of intergroup contact (Paolini, Hewstone, Cairns, et al., 2004). To the extent that television, for example, allows people to be exposed to those who are different from themselves, to empathize with their plights, to listen to their stories, without the anxiety associated with in-person contact, then prejudice toward the group may be likely to decline (Dovidio, Gaertner, & Kawakami, 2003; Pettigrew & Tropp, 2006). Although parasocial relationships provide a plausible route through which media exposure could reduce prejudice, the requirements of this theory create a narrow scope of potential impact. In part, it is limited because beneficial effects would require strong and positive emotional bonds with outgroup characters, the kind resulting from repeated exposures. Many people probably have feelings about television characters, but only with a relatively few television characters do viewers form deep bonds (i.e., parasocial relationships). Further, precisely because of pre-existing prejudice, ingroup viewers would be unlikely to perceive an outgroup television character as highly familiar, likeable, and similar to him or herself. Modeling Intergroup Interactions A second theoretical framework, known as modeling theory or social cognitive theory (Bandura, 2002), suggests that viewers emulate the relations between ingroups and outgroups that they observe enacted on television. If, for example, television portrays ingroup and outgroup members resolving their differences through violence, then viewers will follow that example. And if viewers witness peaceful, friendly intergroup interactions, then they will emulate those behaviors instead. In both cases, television provides low-cost opportunities for people to observe ingroup and outgroup members interacting. Viewing intergroup exchanges on television may also affect viewers’ levels of anxiety about future face-to-face interactions. By teaching ingroup members new social skills, or ‘rules of behavior,’ exposure may increase self-efficacy when engaging in real-world intergroup contact (Bandura, 1986: 47; Green, 2006). According to this model, ingroup members should engage in less prejudicial behaviors only if the intergroup interactions they view on television engender more positive than negative outcomes (Bandura, 1986, 2001). Viewers will emulate some television characters more than others, depending upon characteristics of the ingroup member (Bandura, 1977). The more a viewer identifies with a televised person, the more he or she is expected to model that character's behaviors (Eyal & Rubin, 2003). Identification occurs because the viewer sees himself or herself as similar to the ingroup member and vicariously participates in their experiences (Hoffner, 1996). Viewers are expected to develop the same emotional reactions to outgroup members as the ingroup member they identify with on television (Bandura, 1999). The vicarious learning suggested by this model occurs because the viewer is so immersed in the character's perspective that he or she emulates the character's emotional reactions. In this respect, social cognitive theory supports what studies of ‘transportation’ via mass media also have suggested: that influence occurs when viewers are sufficiently absorbed by a narrative that they take on the perspective of a character and truly feel themselves to be personally involved. However, in the case of social modeling, the person must engage in intergroup interactions as part of the narrative. Interestingly, although vicarious learning has long been championed as the mechanism linking television violence to aggressive behavior in children, it has not been well studied as a means of either reducing or producing prejudice (see Graves, 1999; and Ortiz & Harwood, 2007, for exceptions). Many children's television programs are premised on the belief that viewers will model the friendly intergroup interactions they witness, but there is a lack of causal evidence to confirm this. As with the parasocial interaction explanation, social modeling puts limits on potential media influence on outgroup attitudes through its various requirements. Most importantly, influence can occur only when there are intergroup interactions to model. Moreover, the viewer must clearly identify with the ingroup character engaged in the intergroup interaction. Overall, there is probably a limited amount of naturallyoccurring television content that meets all of the requirements for the social modeling process. And regardless, it is difficult to differentiate this process and its predictions from alternative theories. The Media World as Real World As initially suggested, we favor a third, more general information-processing model that both broadens the set of media portrayals with the potential for media influence, and reduces the intensity of affective response required from viewers. In short, this perspective suggests that viewers process televised portrayals of people largely as if they were real-world, first-hand observations. Intergroup attitudes are influenced by salient outgroup exemplars, many of which people observe through mass media. It is a well-worn truism that people do not experience a large proportion of the world first-hand. Instead, ‘the images in our heads’ are often formed from the images and information found in the media, particularly on television (Lippmann, 1922). To the extent that people either perceive media as conveying accurate depictions of the world (as in news consumption) or subconsciously process media content as if it were real (as in fictional dramas), media portrayals of both ingroup and outgroup members should be consequential. We find this theory both plausible and appealing for a number of reasons. First, it predicts that both fictional and non-fictional portrayals of outgroup members have potential for influence. For televised social information to be discounted so that it does not influence social judgments, viewers would need to remember the source of information and then purposefully disregard it – conditions which appear unlikely to co-occur without explicit intervention (Shapiro & Lang, 1991; Shrum, Wyer, & O'Guinn, 1998). Second, research on human-media interaction bolsters the idea that there are minimal differences between the firsthand experience of others and viewing them on television. Although adults clearly understand that all events seen on television did not actually happen, this is a learned reaction (Worth & Gross, 1974). And regardless of their awareness, people's physiological and psychological reactions to television exposure are fundamentally the same as their reactions to real people and events. So, for instance, exposure to a human being who appears larger and closer due to a larger television screen produces more arousal, better memory for the content, and more liking of the content than exposure to the same content on a smaller screen. When a person comes physically closer in real life, and fills more of the viewer's field of vision, the same reactions occur. ‘All of these results are pretty much the same in the real world,’ according to Reeves and Nass (1996: 198). Along similar lines, attention (as measured by brain activity) synchronizes with motion on television – that is, within about a second of televised movement, attention increases (Reeves & Nass, 1996). Again, these reactions are the same as those found in face-to-face contact. Physiological reactions notwithstanding, most existing research on media effects implicitly or explicitly assumes that people process media as informative about the real world. To the extent that this claim is supported, media content has obvious relevance for prejudice and perceptions of social norms. A third argument in favor of this far more encompassing, information-processing approach is that there are fewer necessary conditions required than in other theoretical perspectives. In order for portrayals of out-groups to be influential, exemplars need only be observed by viewers. It is not necessary that viewers identify with the outgroup member, nor that successful intergroup interactions be featured, nor that the viewer feel he/ she has a personal relationship with someone on television. Some evidence suggests that merely imagining contact with an outgroup member may reduce intergroup bias (Turner, Crisp & Lambert, 2007). To the extent that intergroup influence is brought about more easily than was once thought, then loosening requirements for media influence also seems reasonable. How well does evidence support assertions that media supply influential exemplars in social judgment? Correlational studies consistently support the idea that media portrayals affect perceptions of the frequency of events such as crime and, by extension, the prevalence of crime associated with specific outgroup members. However, there is less evidence of influence on personal attitudes and beliefs about social groups (e.g., personal fear of crime) (Gerbner, Gross, Morgan, et al., 1980; Gross, 1984; Hawkins & Pingree, 1981; Signorielli, 1989). Effects have been somewhat stronger when using genre-specific measures of exposure (e.g., crime dramas), rather than global measures of television exposure (e.g., hours of viewing per day), leading many to suggest that perhaps these effects would be clearer if exposure were controlled in a laboratory setting rather than self-reported (Hawkins & Pingree, 1981; Potter, 1993; Shrum, 1996; Shrum, & O'Guinn, 1993; but see Shrum, Wyer, & O'Guinn, 1998). Indeed, some recent quasi-experimental and experimental findings demonstrate that exemplars of outgroup members viewed on television influenced intergroup attitudes, even in the short-term after relatively little exposure (e.g., Morgan, 1982, 1987; Rossler & Brosius, 2001). If we view media as simply one of many potential sources of exemplars that can shape outgroup attitudes, then it is possible to evaluate the effects of atypical exemplars like the Cosby family, and sort out the contradictory predictions that have been made about its effects. For example, if social judgment of outgroups depends upon exemplars that are readily available in people's minds, and media are major suppliers of salient exemplars, then television's ready supply of middle and upper class Black exemplars is noteworthy. Indeed, Bodenhausen, Schwarz, Bless, et al., (1995) find that atypical Black exemplars who are successful and well-liked positively affect judgments and beliefs about Blacks, and increase perceptions of majority discrimination against the minority. If it is pointed out that these exemplars are atypical, however, the positive effects disappear. Integration The three theoretical models that we have discussed suggest quite different kinds of influence processes, and they point to different kinds of content as important to understand the influence of media on prejudice. If one were interested in the potential for media to serve as a source of modeling information, one would want to study content such as the prevalence of interracial marriage on television. If instead one viewed mediated experience as a source of interpersonal contact with outgroups, one would care about the incidence of ingroup members viewing likable outgroup members on television. And if one were studying this hypothesis from the perspective that media simply provide exemplars that prime people's perceptions of the real world, then one would want to know which kinds of exemplars were most relevant to human judgment – portrayals of outgroups, ingroups relative to outgroups, outgroups relative to real world experience, and so forth. Future Directions Our review has focused on the potential for television to serve as a form of intergroup contact, primarily because television's audiovisual stimuli do such an excellent job of simulating the experience of being near another human being (see Reeves & Nass, 1996). This is not to say that novels, radio, or other media should not have an impact, but television's pervasiveness, sensory simulation of reality, and engrossing storylines make it a natural first place to look. Notably, our discussion has not differentiated between studies of media influence on the extent of stereotyping, attitudes toward outgroups, and actual behaviors. Nor has it included an exhaustive review of all potentially relevant studies, including those with media content designed explicitly for purposes of public information campaigns to reduce prejudice (see Paluck & Green, 2009, for a review), or prosocial children's television programming designed to reduce stereotyping and prejudice (for meta-analyses, see Mares & Woodward, 2005; Oppliger, 2007). Indeed, our review has largely neglected the body of research known as ‘educationentertainment’ initiatives, which purposely embed educational messages within entertainment programs (see Singhal & Rogers, 2002). Studies of this kind are not based on any one theoretical model, but rather tend to use a shotgun approach, using all available suggestions about what would increase chances of success. For example, some education-entertainment projects focus on how development of parasocial relationships with positive role models increases the persuasive impact of the educational message (Papa, Singhal, Law, et al., 2000). Others rely on social-cognitive theory to maximize potential impact. For the most part, entertainmenteducation has been applied to public health concerns, with only limited evidence regarding social group attitudes, such as gender equality (Slater, 2002). However, in one recent exception, Paluck (2007, 2009) reports the results of a year-long education-entertainment effort in Rwanda, designed purposely to reduce outgroup prejudice. Those who listened to a radio program emphasizing intergroup reconciliation were more likely than those in the control condition to express positive attitudes about intergroup marriage, though social distance measures did not suggest reduced prejudice. Because attempts at education-entertainment tend to be large-scale and expensive, they simultaneously incorporate many characteristics thought to be potentially beneficial, thus making it difficult to ascertain what aspects of the media content are effective, if any (see Paluck, 2009, for a review). Despite Allport's (1954) admonition about the importance of mass media as a source of intergroup contact, we know surprisingly little about its role in prejudice. Mass media are rich sources of information about outgroup members, and the forms that intergroup interactions may take. Currently, our understanding of these processes is largely limited to speculation, albeit based on thoughtful considerations of media content. The small number of studies that have undertaken empirical verification of the effects of mass media on prejudice demonstrate the potential for mediated contact to influence real-world attitudes and beliefs about social groups, as well as the potential limitations posed by selectivity, both in perception and exposure. In order to improve upon these initial suggestions of impact, however, theoretical frameworks must be advanced and tested. How might such a research agenda proceed? Although the tremendous emphasis to date on studies of media content may seem self-explanatory to a casual observer, to empirical social scientists it should be recognized as putting the cart before the horse: content does not equal effects. Before more scholarly time and energy are devoted to documenting the most prevalent types of content, it is incumbent upon scholars to figure out which kinds of media content comparisons ultimately matter to intergroup attitudes. There are several ways that researchers might go about this, but it seems clear upon reflection that more experimental studies of effects are needed, particularly ones that can differentiate between influence that flows from intragroup and intergroup media portrayals, as well as whether it matters if television differs from the real world, at least as it is perceived by viewers. Because experiments must often rely on one-shot exposure to a media stimulus, or at least on a small number of exposures, within-subject designs may be key to obtaining the statistical power that is necessary to isolate the impact of a tiny number of exposures relative to the enormous amount of ongoing television content consumed by the average person in developed countries. Moreover, because of the sensitive nature of intergroup attitudes, such studies may also require augmenting self-reports with unobtrusive measures, such as the Implicit Association Test (IAT) (see Greenwald & Banaji, 1995). Although the IAT is controversial as a measure of prejudice, it is uncontroversial as a measure of the extent to which people have formed positive or negative associations with members of certain groups (see Arkes & Tetlock, 2004). If media portrayals in an experimental setting consistently link Blacks with crime, for example, these effects may be more easily observed using techniques such as the IAT. If instead (or in addition), what matters is whether a television program associates Blacks with crime more often than it associates Whites with crime, then the presence of ingroup associations will matter as much as outgroup associations. Finally, and perhaps most importantly, scholarly attention needs to be directed toward theoretical development. Because studies documenting effects remain few and far between at this point, scholars have been satisfied with merely demonstrating effects, and offering convenient theoretical frameworks to explain them. Such emphases are natural given the incipient nature of this research. However, what these studies lack is the ability to differentiate between the various theoretical models in order to determine the underlying process of influence. Without understanding process, scholars’ ability to determine the kinds of media content that are beneficial or harmful to intergroup attitudes will remain speculative at best. Summary and Conclusion A long history of interest in mass media's impact on intergroup attitudes has, nonetheless, produced limited evidence of effects. On the one hand, studies of media content have flourished, providing many descriptive accounts of how people of various races, ethnicities, sexualities, and genders are portrayed in various television genres. On the other hand, scholars still know relatively little about the kind of content that is most influential in either encouraging or discouraging prejudice. For this reason, we recommend that the emphasis in future work shift in the direction of first seeking to understand the underlying process of influence, so that subsequent analyses of media content can be guided by knowledge of the specific kinds of media portrayals that matter. To date, only a few experimental studies have established a causal connection between media portrayals of outgroup members and the attitudes that ingroup viewers hold toward them. More such studies are needed to be sure. Once this causal process is well understood, scholars will need to combine evidence of impact with evidence from the audiences viewing such content in naturalistic settings to eliminate the possibility that selective exposure limits media exposure to content that is congruent with people's prejudices. Only by combining experimental work on the process of influence with observational studies of viewing habits will we ultimately be able to address Allport's hypothesis about the importance of media. Diana C.Mutz Seth K.Goldman References Allport, G. W.(1954).The Nature of Prejudice. Reading, MA:Addison-Wesley. Arkes, H., &Tetlock, P.E. Attributions of implicit prejudice, or would Jesse Jackson fail the implicit association test? Psychological Inquiry, 15,257–278. (2004). Ball-Rokeach, S.Grube, J., & Rokeach, M. “Roots: The Next Generation” Who watched and with what effect? Public Opinion Quarterly, 45,58–68. (1981). Bandura, A.(2002).Social cognitive theory of mass communication. In J.Bryant, & D.Zillman (Eds), Media Effects: Advances in Theory and Research. (2nd edition, pp. 121–153).Hillsdale, NJ:Lawrence Erlbaum Associates. Bandura, A. (1977).Social Learning Theory. Englewood Cliffs, NJ:Prentice Hall. Bandura, A. (1986).Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ:Prentice Hall. Bandura, A.(1999).Social cognitive theory of personality. In D. Cervone, & Y.Shoda (Eds), The Coherence of Personality (pp. 185–241).New York:Guilford Press. Bandura, A. Social cognitive theory of mass communication. Media Psychology, 3,265–299. (2001). Bartels, L. M. Messages received: The political impact of media exposure. American Political Science Review, 87,267–285. (1993). Battles, K., &Hilton-Morrow, W. Gay characters in conventional spaces: Will & Grace and the situation comedy genre. Critical Studies in Media Communication, 19,87–105. (2002). Bodenhausen, G. V.Schwarz, N.Bless, H., &Waenke, M. Effects of atypical exemplars on racial beliefs: Enlightened racism or generalized appraisals. Journal of Experimental Social Psychology, 31,48–63. (1995). Bogle, D.(2001).Primetime Blues: African Americans on Network Television. New York:Farrar, Straus, & Giroux. Bowman, K., &Foster, A.(2006).Attitudes about Homosexuality and Gay Marriage. Washington, DC:American Enterprise Institute. Brigham, J., & Giesbrecht, L. “All in the Family”: Racial attitudes. Journal of Communication, 26,69–74. (1976). Charles, C. Z. The dynamics of racial residential segregation. Annual Review of Sociology, 29,167–207. (2003). Cooper, E., & Jahoda, M. The evasion of propaganda: How prejudiced people respond to anti-prejudice propaganda. Journal of Psychology, 23,15–25. (1947). Dixon, J. C., &Rosenbaum, M. S. Nice to know you? Testing contact, cultural, and group threat theories of anti-black and anti-hispanic stereotypes. Social Science Quarterly, 85,257–279. (2004). Dixon, T. L., &Linz, D. Overrepresentation and underrepresentation of African Americans and Latinos as lawbreakers on television news. Journal of Communication, 50,131–54. (2000a). Dixon, T. L., &Linz, D. Race and the misrepresentation of victimization on local television news. Communication Research, 27, 547–573. (2000b). Dixon, T. L.Azocar, C. L., &Casas, M. The portrayal of race and crime on television network news. Journal of Broadcasting and Electronic Media, 47,498–523. (2003). Donovan, R. J., &Leivers, S. Using paid advertising to modify racial stereotype beliefs. Public Opinion Quarterly, 57,205–218. (1993). Dovidio, J. F.Gaertner, S. L., &Kawakami, K. Intergroup contact: The past, present, and the future. Group Processes and Intergroup Relations, 6,5–21. (2003). Entman, R. M.Blacks in the news: Television, modern racism and cultural change. Journalism Quarterly, 69,341–361. (1992). Entman, R. M. Representation and reality in the portrayal of blacks on network television news. Journalism Quarterly, 71,509–520. (1994). Entman, R. M., &Rojecki, A.(2000).The Black Image in the White Mind: Media and Race in America. Chicago:University of Chicago Press. Eyal, K., &Rubin, A. M. Viewer aggression and homophily identification and parasocial relationships with television characters. Journal of Broadcasting and Electronic Media, 47,77–98. (2003). Ford, T. E. Effects of stereotypical television portrayals of African-Americans on person perception. Social Psychology Quarterly, 60,266–275. (1997). Gamson, J.(1998).Freaks Talk Back: Tabloid Talk Shows and Sexual Nonconformity. Chicago:University of Chicago Press. Gerbner, G.Gross, L.Morgan, M., &Signorielli, N. Aging with television: Images on television drama and conceptions of social reality. Journal of Communication, 30, 37–47. (1980). Gilens, M. Race and poverty in America. Public Opinion Quarterly, 60,515–541. (1996). Giles, D. C. Parasocial interaction: A review of the literature and a model for future research. Media Psychology, 4,279–305. (2002). Gilliam, F. D.Jr.Iyengar, S.Simon, A., &Wright, O. Crime in black and white: The violent, scary world of local news. Harvard International Journal of Press/Politics, 1,6–23. (1996). Gilliam, F. D.Jr., &Iyengar, S. Prime suspects: The influence of local television news on the viewing public. American Journal of Political Science, 44,560–573. (2000). Graves, S. B. Television and prejudice reduction: When does television as a vicarious experience make a difference? Journal of Social Issues, 55,707. (1999). Green, M. C. Transportation into narrative worlds: The role of prior knowledge and perceived realism. Discourse Processes, 38,247–266. (2004). Green, M. C. Narratives and cancer communication. Journal of Communication, 56S,163–183. (2006). Green, M. C., &Brock, T. C. The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79,701–721. (2000). Green, M. C., &Brock, T. C.(2002).In the mind's eye: Transportation-imagery model of narrative persuasion. In M. C.Green, J. J.Strange, & T. C.Brock (Eds), Narrative Impact: Social and Cognitive Foundations (pp. 315–342).Mahwah, NJ:Lawrence Erlbaum Associates, Inc. Greenland, K., &Brown, R. Categorization and intergroup anxiety in contact between British and Japanese nationals. European Journal of Social Psychology, 29,503–21. (1999). Greenwald, A. G., &Banaji, M. R. Implicit social cognition: Attitudes, self-esteem, and stereotypes. Psychological Review, 102,4–27. (1995). Gross, K.(2006).Covering crime in Washington D.C.: Examining the nature of local television news coverage of crime and its effect on emotional response. Research paper R-28. Cambridge, MA: The Joan Shorenstein Center on the Press, Politics and Public Policy, Harvard University. Gross, L.(1984).The cultivation of intolerance: Television, blacks and gays. In G.Melischek, K. E.Rosengren, & J.Stappers (Eds), Cultural Indicators: An International Symposium(pp. 345–363).Vienna:Osterreichische Akademie der Wissenschaften. Gross, L.(2001).Up from Invisibility: Lesbians, Gay Men, and the Media in America. New York:Columbia University Press. Hawkins, R. P., &Pingree, S. Uniform messages and habitual viewing: Unnecessary assumptions in social reality effects. Human Communication Research, 7,291–301. (1981). Herrett-Skjellum, J., &Allen, M.(1995). Television programming and sex stereotyping: A meta-analysis. In B. R.Burleson (Ed.), Communication Yearbook 19(pp. 157–185).Troy:NY: Sage. Hoffner, C. Children's wishful identification and parasocial interaction with favorite television characters. Journal Broadcasting and Electronic Media, 40,389–402. (1996). Horton, D., &Wohl, R.R. Mass communication and parasocial interaction: Observations on intimacy at a distance. Psychiatry, 19,215–229. (1956). Jhally, S., &Lewis, J.(1992). Enlightened Racism: The Cosby Show, Audiences, and the Myth of the American Dream. Boulder, CO:Westview Press. Kanazawa, S. Bowling with our imaginary friends. Evolution and Human Behavior, 23,167–171. (2002). Kendall, P. L., &Wolf, K. M.(1949).The analysis of deviant case studies in communications research. In P.Lazarsfeld, & F.Stanton (Eds), Communications Research 1948–1949(pp. 152–179).New York:Harper. Klite, P.Bardwell, R. A., &Salzman, J. Local TV news: Getting away with murder. Harvard International Journal of Press/Politics, 2,102–112. (1997). Lee, B. A.Farrell, C. R., &Link, B. G. Revisiting the contact hypothesis: The case of public exposure to homelessness. American Sociological Review, 69,40–63. (2004). Lippmann, W.(1922).Public Opinion. New York:Harcourt, Brace, & Co. Logan, J.(2001).Ethnic diversity grows, neighborhood integration lags behind. Report by the Lewis Mumford Center. University at Albany, State University of New York. Mares, M., &Woodard, E. Positive effects of television on children's social interactions: A meta-analysis. Media Psychology, 7,301–322. (2005). Mendelberg, T.(2001). The Race Card: Campaign Strategy, Implicit Messages, and the Norm of Equality. Princeton, NJ:Princeton University Press. Morgan, M. Television and adolescents” sex role stereotypes: A longitudinal study. Journal of Personality and Social Psychology, 43,947–955. (1982). Morgan, M. Television, sex-role attitudes, and sex-role behavior. Journal of Early Adolescence, 7,269–282. (1987). Mutz, D. C. Mass media and the depoliticization of personal experience. American Journal of Political Science, 36,483–508. (1992). Oliver, M. B. Portrayals of crime, race, and aggression in “reality-based” police shows: A content analysis. Journal of Broadcasting and Electronic Media, 38,179–93. (1994). Oppliger, P. A.(2007). Effects of gender stereotyping on socialization. In R. W.Preiss, B. M.Gayle, N.Burrell, & M.Allen (Eds), Mass Media Effects Research: Advances Through Meta-Analysis(pp. 199–214).Mahwah, NJ:Lawrence Erlbaum Associates. Ortiz, M., &Harwood, J. A social cognitive theory approach to the effects of mediated intergroup contact on intergroup attitudes. Journal of Broadcasting and Electronic Media, 51,615–631. (2007). Paluck, E. L.(2007). Reducing intergroup prejudice and conflict with the media: Afield experiment in Rwanda. Ph.D. Dissertation. Yale University. Paluck, E. L. Reducing intergroup prejudice and conflict using the media: A field experiment in Rwanda. Journal of Personality and Social Psychology, 96,574–587. (2009). Paluck, E. L., &Green, D. P. Prejudice reduction: What works? A review and assessment of research and practice. Annual Review of Psychology, 60,339–67. (2009). Paolini, S.Hewstone, M.Cairns, E., &Voci, A. Effects of direct and indirect cross-group friendships on judgments of Catholics and Protestants in Northern Ireland: The mediating role of an anxiety-reducing mechanism. Personality and Social Psychology Bulletin, 30,770–86. (2004). Papa, m. J.Singhal, A.Law, S.Pant, S.Sood, S., &Rogers, E., et al. Entertainment-education and social change: An analysis of parasocial interaction, social learning, collective efficacy, and paradoxical communication. Journal of Communication, 50,31–55. (2000). Pettigrew, T. F., &Tropp, L. R. A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90,751–783. (2006). Potter, W. J. Cultivation theory and research: A conceptual critique. Human Communication Research, 19,564–601. (1993). Power, J. G.Murphy, S. T., &Coover, G. Priming prejudice: How stereotypes and counter-stereotypes influence attribution of responsibility and credibility among ingroups and outgroups. Human Communication Research, 23,36–58. (1996). Price, V., &Zaller, J. Who gets the news? Alternative measures of news reception and their implications for public opinion research. Public Opinion Quarterly, 57,133–164. (1993). Reeves, B., &Nass, C.(1996).The Media Equation. New York:Cambridge University Press. Riggle, E. D. B.Ellis, A. L., &Crawford, A. M. The impact of “media content” on attitudes toward gay men. Journal of Homosexuality, 31,55–69. (1996). Romer, D.Jamieson, K. H., &de Coteau, N. J. The treatment of persons of color in local television: Ethnic blame discourse or realistic group conflict? Communication Research, 25,286–305. (1998). Rossler, P., &Brosius, H. Do talk shows cultivate adolescents” views of the world? A prolonged-exposure experiment. Journal of Communication, 51,143–163. (2001). Schiappa, E.Gregg, P. B., &Hewes, D. E. The parasocial contact hypothesis. Communication Monographs, 72,92–115. (2005). Schiappa, E.Gregg, P. B., &Hewes, D. E. Can one TV show make a difference? Will & Grace and the parasocial contact hypothesis. Journal of Homosexuality, 51,15–37. (2006). Shapiro, M. A., &Lang, A. Making television reality: Unconscious processes in the construction of social reality. Communication Research, 18,685–705. (1991). Shrum, L. J. Psychological processes underlying cultivation effects: Further tests of construct accessibility. Human Communication Research, 22,482–509. (1996). Shrum, L. J., &O'Guinn, T. C. Processes and effects in the construction of social reality. Communication Research, 20,436–471. (1993). Shrum, L. J.WyerJr., R. S., &O'Guinn, T. C. The effects of television consumption on social perceptions: The use of priming procedures to investigate psychological processes. Journal of Consumer Research, 24,447–458. (1998). Signorielli, N. Television and conceptions about sex roles: Maintaining conventionality and the status quo. Sex Roles, 21,341–360. (1989). Singhal, A., &Rogers, E. M. A theoretical agenda for entertainment-education. Communication Theory, 12,117–135. (2002). Slater, M. D.(2002). Entertainment education and the persuasive impact of narratives. In M. C.Green, J. J.Strange, & T. C.Brock (Eds), Narrative Impact: Social and Cognitive Foundations(pp. 157–181).Mahwah, NJ:Lawrence Erlbaum Associates, Inc. Stephan, W. G., &Stephan, C.W. Intergroup anxiety. Journal of Social Issues, 41,157–75. (1985). Tamborini, R.Mastro, D.Chory-Assad, R., &Huang, R. The color of crime and the court: A content analysis of minority representation on television. Journalism and Mass Communication Quarterly, 77,639–654. (2000). Turner, R. N.Crisp, R. J., &Lambert, E. Imagining intergroup contact can improve intergroup attitudes. Group Processes and Intergroup Relations, 10,427–41. (2007). Vidmar, N., &Rokeach, M. Archie Bunker's bigotry: A study in selective perception and exposure. Journal of Communication, 24,36–47. (1974). Voci, A., &Hewstone, M. Intergroup contact and prejudice toward immigrants in Italy: The meditational role of anxiety and the moderational role of group salience. Group Processes and Intergroup Relations, 6,37–54. (2003). Worth, S., &Gross, L. Symbolic strategies. Journal of Communication, 24,27–39. (1974). Wright, S. C.Aron, A.McLaughlin, T., &Ropp, S. A. The extended contact effect: Knowledge of cross-group friendships and prejudice. Journal of Personality and Social Psychology, 73,73–90. (1997).