Our benefits

24/7 customer support

Professional writers

No plagiarism

Privacy guarantee

Affordable prices

94% of return customers

Free extras

Free title page

Free bibliography

Free formatting

Free of plagiarism

Free delivery

Race Representation as "White" Accommodation
Television and Social Identity: Race Representation as "White" Accommodation.


by Gail E. Coover



In recent years, Americans have sought to redress racial inequity through changes in the ways that race is spoken about (e.g., Ward, 1985) or portrayed (e.g., Entman, 1990; Gray, 1995; van Dijk, 1990). In television especially, pressure from activist groups outside the media industry and professional groups inside the industry has led to an increase in the number of Blacks on television who are portrayed in a positive light (Montgomery, 1989). Situation comedies and dramas that feature Blacks as lead characters or continuing characters have become pervasive (Gray, 1995).(1)


Analyses of these changes have suggested inconsistent consequences with respect to how many White viewers in particular respond to these representations independently of program content. On the one hand, researchers have suggested that popular, positive portrayals of many Blacks in the media are marketable because they affirm White audience members' self-concepts as non-racist, as well as many Whites' negative attitudes toward Blacks (Campbell, 1995; Entman, 1990; Gray, 1989; Jhally & Lewis, 1992). These analyses characterize many White viewers' positive responses to programs like The Cosby Show as a form of "enlightened racism" whereby White viewers' fears of appearing racist are allayed but their negative assumptions or racist attitudes about Blacks in general go unchallenged (Gray, 1989; Jhally & Lewis, 1992). A consequence of enlightened racism is that Black characters are well-liked, not in spite of their race, but because of their race.


An alternative approach to research on many White viewers' responses to representations of Blacks in the media suggests that positive representations of Blacks may facilitate the gradual attenuation of racial bias on the part of many Whites against Blacks (Bodenhausen, Schwarz, Bless, & Wanke, 1995; Power, Murphy, & Coover, 1996). These formulations of White viewers' responses to portrayals of Blacks suggest that for many Whites the problem of racial prejudice is based on limited interpersonal contact and limited access to directly experienced information about Blacks. Consequently, repeated exposures to positive portrayals of Blacks on television should gradually chip away at the negative stereotypes of Blacks that most Whites adopt at an early age.


The present study addresses the contradiction manifested in these two approaches by adopting a theoretical framework that considers racial identity rather than racial prejudice as an underlying mechanism which guides White viewers' responses to race representation. In adopting this approach, this study explicitly addresses the assumed norm of White identity (Dyer, 1988). This approach is not intended as an "apology" for the unintentional or non-conscious aspects of many Whites' racial prejudices. Rather, it seeks to empirically demonstrate the dynamic role played by media representations of Blacks in sustaining many Whites' racial prejudice with a view toward informing media literacy and race awareness education. A similar focus on the role of the media on Black audience members' racial prejudices, while valuable, is beyond the scope of this study. The next section reviews research on White audience responses to race representation in the media.


Review of Audience Responses to Race Representation


Race is portrayed in the media in two ways: through the content of a message or program, and through the race representation of sources included in a program or story. Previous research has demonstrated that race representation on television tends to affirm, and possibly galvanize, White viewers' racial attitudes. In other words, racist and non-racist Whites interpret positive and negative representations of Blacks in ways that confirm their pre-existing pro- or anti-Black prejudices (e.g., Armstrong, Neuendorf, & Brentar, 1992; Vidmar & Rokeach, 1974). This process suggests that the media play a relatively minimal role with respect to influencing individuals' personal racial attitudes apart from reinforcing them.


More recent research has indicated that positive representations of Blacks can have the implicitly or explicitly intended effect of diminishing racial bias on the part of Whites against Blacks (Bodenhausen et al., 1995; Power et al., 1996). Such an effect is typically the result of cognitive priming. For example, Power et al. (Study 1, 1996) found exposure to a stereotype-disconfirming (counter-stereotypic) exemplar of a Black male led White participants to make more external attributions of blame for negative situations that involved Black men. Conversely, those participants exposed to a stereotypic exemplar made more internal attributions. Similarly, Bodenhausen et al. (1995) found that White participants were more sensitive to racial issues (e.g., were less likely to endorse the statement that racism is no longer a problem) if they had recently been thinking about a well-liked Black celebrity (e.g., Oprah Winfrey or Michael Jordan). This effect disappeared, however, when the atypicality of the positive exemplar prime was made salient or when the well-known Black exemplar (e.g., Jesse Jackson or Spike Lee) was not particularly well-liked (Bodenhausen et al., 1995). Consequently, exposure to stereotype-disconfirming media representations of Blacks can influence Whites' subsequent racial attitudes and beliefs, but the influence is contingent upon White viewers holding positive attitudes toward the individual who serves as the exemplar as well as viewers not questioning the atypicality of the exemplar.


The results of Bodenhausen et al.'s (1995) study raise the question of what leads Whites to prefer certain Black celebrities, and by association, certain representations of race. The present study argues that preferences are biased toward representations that accommodate White racial attitudes without confronting them. Accommodation is typically used to characterize patterns of convergence and divergence in interpersonal interactions (see Giles, Mulac, Bradac, & Johnson, 1987 for a review). Specifically, accommodation refers to the subjective experience communication partners have when the social identities they bring to an interaction (e.g., gender identities, sexual identities, ethnic or racial identities) are supported or affirmed. Individuals use verbal and non-verbal strategies to indicate solidarity and convergence with their partners or to indicate difference and divergence from their partners. Accommodation in interactions between individuals from different social groups is associated with more positive feelings toward and evaluations of the interaction and/or the conversation partner (Aune & Kikuchi, 1993; Jones, Gallois, Barker, & Callan, 1994; Lepoire, Ota, & Hajek, 1997). In this study, accommodation is applied to the relationship between a television viewer and a message.


Accommodation of White identity through the use of Black representation can be realized through a variety of strategies. Black characters and celebrities might project a highly assimilated personality or appearance (Gray, 1995). Such representations provide viewers with an image of inter-racial harmony and equality but assiduously avoid topics related to racism and racial inequity (Campbell, 1995; Gray, 1995; Jhally & Lewis, 1992). Coverage of issues like racism and inter-racial conflict which implicate White identity can be framed in terms that deflect the stigma of being racist to other non-White ethnic groups (e.g., Shah & Thornton, 1994) or specific bigoted individuals (Campbell, 1995). These analyses of accommodation of White identity indicate that research should consider the implicit representation of the relationship between Whites and Blacks that representations of Blacks convey. Such a consideration requires an understanding of how racial identity might operate as a social identity for Whites.


Racial Identity as a Social Identity


Tajfel and Turner (1986) specify three levels at which identity is experienced. The personal level of identity refers to the identities individuals possess as unique to them and their life experiences. These identities are idiosyncratic and specific to the individual. The group level of identity refers to the identities individuals possess by virtue of membership in a group (i.e., race/ethnic groups or gender groups). Finally, Tajfel and Turner (1986) specify a supra-ordinate level of identity at which individuals experience a collective identity and relate to all people on the basis of their shared humanity.


Social identity theory proposes that people are motivated to maintain positive social (group) identities (Tajfel, 1982; Tajfel & Turner, 1986). For example, people who are divided into groups in an explicitly random way will still demonstrate a favorable bias toward their fellow group members, or the ingroup, and discriminate against those who are not group members, or the outgroup (e.g., Billig & Tajfel, 1973). Factors such as the "status" or power of one group relative to another, the relative size of the groups, and the permeability of group boundaries determine the strategies group members adopt in order to maintain positive social identity (e.g., Brewer, Manzi, & Shaw, 1993; Ellemers, Doosje, van Knippenberg, & Wilke, 1992).


In the case of racial identity, Whites are defined (culturally) relative to other racial or ethnic groups as having more power and possessing majority status (Gray, 1995; van Dijk, 1990). Most people learn at a very early age their racial identity and its significance (Allport, 1954; Hartmann & Husband, 1974; Ward, 1985). For most Whites, this "racial socialization" is taught; it does not occur through direct contact with members of other racial groups (Allport, 1954, Hartmann & Husband, 1974; Judd, Park, Ryan, Brauer, & Kraus, 1995; Ward, 1985). Unlike a group membership defined on an arbitrary dimension, social identities based in "natural" group memberships, like race or gender, are loaded with social and historical significance (Ethier & Deaux, 1994; Helms, 1990; Smedley, 1993).


Most Whites do not acknowledge their racial or ethnic identity as an important social identity (Helms, 1990; Katz, 1978). Further, most Whites do not recognize this social identity as one that is relevant to their relationships, even relationships with members of other ethnic/racial groups (Collier, 1996). There are two reasons why this would be so. Firstly, the attenuated importance of racial identity for Whites in part is indicative of the relative lack of distinctiveness of the identity. Brewer (1991) proposes that individuals actually choose the identities by which they wish to be known according to the extent to which the identity affords a feeling of optimal distinctiveness. Optimal distinctiveness refers to individuals' desire to experience a sense of belonging to a group and still retain a feeling of uniqueness (Brewer, 1991).


A second and more complex reason for many Whites' reluctance to overtly acknowledge the importance of their racial identities concerns an underlying motivation to preserve a positive social identity. The Civil Rights movement led to a shift in social norms such that overt expressions of racism, and by extension individuals who engage in overtly racist rhetoric, are not granted legitimacy (and are even condemned) within public discourse (Gaertner & Dovidio, 1986; McConahay, Hardee, & Batts, 1981; Sears, 1988). The consequent aversion to White identity may reflect a form of aversive racism (Gaertner & Dovidio, 1986). Aversive racism refers to the phenomenon whereby some Whites avoid the appearance of being racist through extremely favorable evaluations of Blacks when such ratings do not involve any direct cost or loss of status for themselves either personally or as group members (e.g., Jackson, Sullivan, & Hodge, 1993). Although some have characterized this shift as a sign that racism is declining (e.g., Schuman, Steeh, & Bobo, 1985), other research suggests that it would be more accurate to characterize this shift as a change rather than a decline in racial prejudice on the part of Whites toward Blacks (e.g., Gaertner & Dovidio, 1986).


An application of social identity theory to the evaluation of race representation effects allows for a focus on the group aspect of racial identity for Whites. This perspective provides a basis for understanding how the media influence the White audience's feelings of affiliation with their own racial ingroup (i.e., Whites) through an evaluation of (1) feelings toward a racial ingroup and/or outgroup members (i.e., a Black commentator and/or a White commentator); (2) perceptions of the intra- and intergroup relationship represented by ingroup and/or outgroup members; (3) attitude-based affiliations with the racial group or race relationship portrayed. This study does not explicitly test whether individuals see themselves in the situation of a character portrayed. Rather, it uses racial identity is a basis for interpreting responses to race representation as reflecting greater or lesser feelings of accommodation.




Three dimensions on which ingroup identification is manifested are described above. The first concerns feelings directed toward ingroup or outgroup members. Individuals derive a sense of positive social identity by comparing their ingroups favorably to outgroups (Tajfel & Turner, 1986). Turner (1987) notes that one consequence of this process is a feeling of attraction to fellow group members (e.g., Moreland, 1985). The minimal group paradigm indicates that this attraction is a consequence and not a cause of group identity (Turner, 1987). These positive feelings associated with "groupness" provide group members with a sense of positive social identity (Tajfel, 1982; Tajfel & Turner, 1986; Turner, 1987).


Ingroup identification is not a sufficient condition for intergroup discrimination to occur (Turner, 1987). In other words, just because ingroup identification is cued, group members will not necessarily discriminate against outgroup members. Turner (1987) notes that "... when social identity in terms of some group membership is unsatisfactory, members will attempt to leave that group (psychologically or in reality) to join some more positively distinct group and/or make the existing group more positively distinct" (p. 30). These strategies for preserving positive social identity have been reflected in past research on the role of group status or the permeability of group boundaries in predicting ingroup identification (Brewer et al., 1993; Ellemers, Wilke, & van Knippenberg, 1993).


With respect to White racial identity and race representation, research suggests that Whites use particular representations of Blacks to psychologically "leave" their racial group (Gray, 1995). For example, Jhally and Lewis (1992) suggest that the media assuage Whites' racial ambivalence by providing a non-threatening opportunity for inter-racial affiliation. In particular, representations that accommodate White identity are those in which inter-racial harmony (DeMott, 1995; Gray, 1995; Jhally & Lewis, 1992) is portrayed, or those in which intra-racial conflict is portrayed. (Gray, 1989; Jhally & Lewis, 1992). Consequently, positive affect directed toward ingroup members will be low compared to that directed toward outgroup members.


H1a: White viewers will indicate the most liking for outgroup-only representations that portray a relationship of disagreement or,


H1b: for ingroup-outgroup representations that are characterized by agreement.


This general hypothesis of outgroup attraction and the specified interaction reflect the ways in which many White viewers of representations of Blacks might use these representations to maintain a positive social identity. Yet, the absence of a strong ingroup identification for many Whites does not mean that intergroup discrimination does not take place. Gaertner and Dovidio's (1986) analysis of aversive racism has suggested that intergroup biases are still present.


One of the more pervasive biases demonstrated between groups concerns how groups are differentiated from each other. Social identity also has implications for how group members perceive their group in relationship to other groups. In particular, ingroup members evaluate minority outgroup members as more similar (e.g., Wilder, 1984). Perceptions of differences are most extreme when comparisons are made between ingroup and outgroup members (Turner,


1987). These perceptions of similarity and difference are amenable to influences introduced on dimensions other than the one defining the groups relative to each other. For example, Biernat and Vescio (1993) found that the beliefs expressed by different group members also serve as categories. Consequently, the extent to which group members are portrayed as agreeing or disagreeing with each other should interact with the effect of group representation on viewers' perceptions of similarity between different group members.


H2: The portrayal of the relationship should interact with race representation such that a mixed, ingroup-outgroup pair is evaluated as especially dissimilar when they disagree with each other and the outgroup-only pair is perceived as especially similar when they agree.


Media portrayals of conflict between two specified groups can lead to opinion polarization between group members exposed to the portrayal (Price, 1989). This suggests that opinions expressed in response to media representations of intergroup conflict reflect the group affiliations cued by the representation. Specifically, issues presented in the context of intergroup disagreement lead individuals exposed to the message to formulate attitudes expressed by the ingroup as opposed to the outgroup (Price, 1989; Biernat & Vescio, Study 3, 1993).


H3: Opinions will reflect greater agreement with ingroup versus outgroup sources in the context of intergroup disagreement.


An alternative hypothesis can also be formulated, however. If opinions reflect an attitudinal basis for affiliation, then representations that are most accommodating of White identity might also lead to opinions that reflect this accommodation. In this study, a discussion of land use policy served as the basis for a series of videos in which race representation was manipulated. The issue of land use is not generally understood as a race-related issue (see deHaven-Smith, 1988, for a review of beliefs associated with land-use regulation). Positions on environmental issues might be explained by individuals' underlying value orientations on the dimensions of freedom and equality (Braithwaite, 1994). However, the media typically present such issues in terms of a single dimension--namely, liberal (pro-regulation) versus conservative (pro-economic growth) political orientation (Braithwaite, 1994). The frame imposed on land use policy in the present study reflects the more general media style of presenting such issues. Consistent with Braithwaite's observation, the issue was presented in terms of a pro-growth/anti-regulation or pro-regulation/limited growth perspective.


Liberalism versus conservatism is also a dimension along which race-related policies are typically presented (Gray, 1989; van Dijk, 1990). Moreover, political conservatism is associated with anti-Black prejudice (Sears, 1988). Consequently, representations of Blacks that do not accommodate a positive White identity might lead White viewers to engage in outgroup affiliation on the basis of their political positions by taking a more liberal position than might otherwise be expected. This liberal shift should occur independently of viewers' stated political orientations. If liberalism-conservatism does serve as an alternative dimension for affiliation, the following pattern of results should occur:


H4: The most liberal (pro-regulation) opinions should be expressed by individuals exposed to intergroup disagreement or outgroup agreement.


Design and Method


The first two hypotheses described differences based upon the representation of race (whether a given pair of commentators is White, Black, or mixed) and the portrayal of the relationship between two commentators (whether two commentators agree or disagree). These manipulations form a 3 (race representation--ingroup ingroup, outgroup-outgroup, and mixed, ingroup-outgroup) by 2 (relationship portrayal--commentators agree, commentators disagree) design. In order to test the extent to which race representation and portrayal of the inter- or intra-racial relationship influenced viewers' perceptions and evaluations of what they were watching, different versions of the same video segment were used.


Stimulus Materials


Because this study focused on how race representation in and of itself influences viewers' perceptions of what they are watching, the news segment used featured a presentation of land use policy issues rather than on a race-related issue. In response to the statement "Land development should be limited in southern California," a convenience sample of 42 students in two undergraduate communications classes at a large university indicated on a scale from -9 (strongly disagree) to +9 (strongly agree) that they did not have strong opinions about this topic (M = 2.4, Mdn = 0). Nor was there a great deal of certainty about the topic (M = 3.0, Mdn = 0) as indicated by ratings on a scale of -9 (very unsure) to +9 (very sure). Finally, the topic of land development policy was relatively unimportant (M = 2.3, Mdn = 0) as indicated by ratings on a scale of -9 (extremely unimportant) to +9 (extremely important).


A three-minute script, written by the researcher with technical assistance from journalism alumni and an attorney with experience in land use issues, was the basis for the video. The video featured two commentators (Mark Adams and Bill Warner). In the version of the script that portrayed an agreeable or supportive relationship between the two commentators, Mark Adams and Bill Warner exchanged information about land use issues and agreed with each other about the importance of each point considered. In the version of the script that portrayed a relationship characterized by disagreement, Mark Adams and Bill Warner essentially exchanged the same information contained in the first script, but they disagreed at the end as to which side of the issue was more important-more growth or more regulation. All the information contained in the script was factual and current. Both commentators provided information about both sides of the issue discussed.


The actual stimulus was a three-minute "public affairs" program. The segment had an SVHS format to ensure the appearance of a program that might actually have been broadcast over a local station. The commentators were shown seated at a news desk, angled toward each other so that their comments were easily directed toward each other and toward the camera. The background of the set showed bookcases and a paneled screen illuminated by colored lights. One camera angle was employed. Such a format eliminated the role of the camera in directing viewers' attention toward one commentator over the other. At one point early in the segment, the commentators' names appeared at the bottom of the screen as they were talking to facilitate participants' identification of them.


Procedure and Variables


A total of 175 individuals (58% men, 42% women) were included in the study sample. All participants were students at a large research university. Participants' ages ranged from 17 years to 50 years with a mean age for the sample of 22 years. Participants were recruited from liberal arts and business classes. Twenty-seven percent of the participants were majors in communication arts, 17% in political science, 18% in business, 15% humanities, 10% natural sciences, and 13% were undeclared. Compensation for participation was provided in the form of a small amount of extra credit or for $5.00. All those included in the sample described themselves as White, correctly identified the race of the commentators, and indicated that they did not personally know either of the commentators featured in the video. Participants were randomly assigned to conditions.


Upon arriving for the study, each participant was seated alone in a room with a television monitor and a VCR. Participants were told that they would view a brief part of a program put together by some students. Participants were informed that the purpose of the program was to provide a casual presentation of public affairs issues. Participants were instructed to stop the tape and turn over the questionnaire in front of them once the video segment had ended. They were informed that the purpose behind the questionnaire was to provide feedback for the students who made the tape in addition to providing some information about how people respond to "this sort of programming generally."


The first independent variable, race representation, was manipulated by casting Black and/or White actors in the roles of the two commentators featured in the video. One version of the tape featured two White actors, a second version of the tape featured two Black actors and a third version featured a Black and a White actor. In order to remove variance introduced by specific qualities of the roles, each of the three pairings (White-White, Black-Black, and Black-White) were duplicated with the actors switching roles.


The relationship between the two commentators was the second independent variable. In the version of the script portraying a harmonious relationship, the commentators signaled their agreement with each other with nods and brief introductory statements (i.e., "Right," "That's true"). Each commentator considered both sides of the issue; and as they reached the end of the script, each agreed with the other that both sides were important. In the condition portraying disagreement, the two commentators listened to one another but did not provide the nonverbal and verbal signals contained in the previous script. As the commentators neared the end of their discussion Mark Adams took a position in support of more regulation and Bill Warner took a position in support of more economic growth.


To determine target-directed affect, the first dependent variable, participants were asked to indicate on a scale ranging from 1 to 11 their liking for, agreement with, and similarity to each of the commentators. In order to counter-balance order effects, half of the participants indicated their liking for Mark Adams first, whereas the other half indicated their liking for Bill Warner first.


To determine similarity of ingroup and outgroup members, the second dependent variable, participants indicated on an 11-point rating scale the extent to which each commentator appeared to agree with and like the other commentator with ratings of 1 indicating no liking or agreement and ratings of 11 indicating extreme liking and agreement.


Finally, to determine opinion, participants indicated their opinion about land use regulation policy on an 11-point scale with respect to whether they favor "lots of growth" (1) or "lots of regulation" (11).




The present study tested the extent to which race representation and the portrayal of intra- or intergroup relationships influenced viewers' reactions to a brief news segment in which two commentators discussed land use policy. The result was a 3 (race representation: ingroup-ingroup, outgroup-outgroup, ingroup-outgroup) by 2 (portrayal of relationship: agreement, disagreement) design. The design was counter-balanced for the role played by each actor and the order in which commentators were evaluated.(2)


Manipulation Checks


Participants who viewed the version of the video in which the commentators' relationship was portrayed as harmonious were more likely to indicate that the commentators agreed (M = 7.45) than participants who watched the version of the video in which the commentators disagreed with each other (M = 5.16, t(173) = 6.56, p [is less than] .001). This indicates that the manipulation of the intra- or inter-group agreement or disagreement was effective.


Hypotheses 1a and 1b specified that race representation and the portrayal of the intra- or inter-racial relationship would interact in their influences on participants' liking for the commentators. In order to test this hypothesis, participants' ratings of liking for, agreement with, and similarity to each commentator were averaged to form a measure of target-directed affect for each commentator (Cronbach's alpha = .70 for ratings of Bill Warner and .80 for ratings of Mark Adams). These ratings were entered in a 3 (race representation: ingroup only, outgroup only or mixed ingroup-outgroup) by 2 (portrayal of relationship: commentators agree, commentators disagree) by 2 (role of commentator: Mark Adams or Bill Warner) analysis of variance. Race representation and the portrayal of the relationship were between-subjects factors. The role of the commentator was a within-subjects factor.


The hypothesized relationship was only partially supported. There was a significant main effect for race representation (F(2, 167) = 7.02, p = .001, [[Eta].sup.2] = .08). The pattern of means reflected a consistent preference for outgroup over ingroup representations. Participants who watched a version of the video in which both commentators were Black (outgroup only) or a version in which one commentator was Black and one was White (mixed, ingroup-outgroup) indicated considerably more liking for the commentators (M = 5.54) compared to those participants who watched a version of the program in which both commentators were White (M = 4.95, t (173) = 2.70, p = .008, [[Eta].sup.2] = .04). Race representation did not interact with the portrayal of the intra- or inter-racial relationship (F(2, 167) [is less than] 1.0, n.s.). Within the mixed, ingroup-outgroup condition, participant ratings of target-directed affect did not vary on the basis of the commentators' race (F(1, 52) [is less than] 1.0, n.s.). In other words, the participants in the mixed representation condition reported liking the White and Black commentators equally.


Perceptions of Similarity


The second hypothesis stated that race representation and the portrayal of the relationship between the commentators would interact. Specifically, ratings of the commentators' similarity to each other were expected to be highest for outgroup representations (e.g., two Black commentators) that were portrayed as harmonious (e.g., commentators agree with each other). Conversely, ratings of commentators' similarity to each other were expected to be lowest in a mixed ingroup-outgroup representation that showed a relationship characterized by disagreement.


An average of participants' ratings of each commentator's liking for and agreement with the other commentator was used to measure perceived similarity. The resulting measure of perceived similarity (Cronbach's alpha = .88) was comprised of four items--two items for judgments about each commentator's relationship to the other commentator. A 3 (race representation: ingroup only, outgroup only, mixed) by 2 (portrayal of relationship: commentators agree, commentators disagree) analysis of variance revealed significant differences between experimental conditions (F(5, 169) = 10.59, p [is less than] .001). Not surprisingly, the portrayal of the relationship significantly influenced perceived similarity (F(1, 169) = 48.81, p [is less than] .001, [[Eta].sup.2] = .22). Commentators who agreed with each other were rated as more similar (M = 7.28) than commentators who disagreed with each other (M = 5.46).


Consistent with the hypothesis, there was a marginally significant interaction between the portrayal of the relationship and race representation (F(2, 169) = 2.56, p = .08, [[Eta].sup.2] = .03). As shown in Table 1, participants who watched a version of the video in which a Black and a White commentator disagreed with each other indicated that the commentators were the least similar (M = 5.06). The conditions where either a Black and a White commentator agreed with each other or where two Black commentators agreed with each other received the highest ratings of similarity (M = 7.6). Interestingly, the difference in similarity ratings between the condition where two White commentators agreed with each other (M = 6.79) and the condition where a Black and a White commentator agreed with each other (M = 7.64) was marginally significant (t(62) = 1.67, p [is less than] .10).


Table 1                                                              
Perceived Similarity of Commentators(*)

Relationship Portrayal

Race Representation Agreement Disagreement

Ingroup only
M [6.79.sub.b] [5.64.sub.a]
SD 1.93 1.62
Outgroup only
M [7.55.sub.b] [5.63.sub.a]
SD 1.62 1.48
M [7.64.sub.b] [5.06.sub.a]
SD 2.08 1.66

(*) Higher numbers reflect greater similarity as a function of race
representation and the portrayal of the relationship. Different
subscripts indicate significant differences between means at the
.05 level of significance.
  Race Representation Influences on Opinion The last hypothesis stated that opinions expressed about the issue discussed in the video subsequent to viewing the program might reflect a bias in favor of opinions expressed by an ingroup member relative to an outgroup member. In the condition where the commentators disagreed with each other, the actor playing the role of Mark Adams presented a pro-regulation position while the actor playing the role of Bill Warner presented a pro-growth position. A t-test comparing the mean opinion of participants who watched a version of the video in which a White commentator took a pro-growth position and a Black commentator took a pro-regulation position (M = 7.50), with the mean opinion of participants who watched a version of the video in which a White commentator takes a pro-regulation position against a Black commentator who took a pro-growth position (M = 8.27), indicated that the race of the source did not directly influence participants' opinions (t(25) = 1.06, p = .30). However, this comparison of mean opinions within the mixed, ingroup-outgroup race representation condition did not separate the influence of each actor's performance generally from the influence of each actor's race.


The fourth hypothesis specified an interaction effect for the racial pairing and the portrayal of the relationship between the commentators. Participants' opinion ratings were evaluated in a 3 (race representation: ingroup only, outgroup only, mixed) by 2 (portrayal: commentators agree, commentators disagree) analysis of variance (F(6, 168) = 3.68, p = .002, [R.sup.2] = .12). A measure of political orientation (with 1 indicating "strong liberal" and 15 indicating "strong conservative") was entered as a covariate. The inclusion of this covariate permitted an evaluation of the influence of race representation on opinion outcomes above and beyond participants' preexisting political inclinations. As hypothesized, both race representation and the portrayal of the relationship influenced participants' opinions (see Table 2). There was a significant main effect for race representation such that participants who viewed a version of the video that featured two White commentators reported more conservative or pro-growth/anti-regulation opinions (M = 6.20) than participants who viewed a version of the video in which either one or both commentators were Black (M = 7.11, F(2, 168) = 3.54, p = .03, [[Eta].sup.2] = .04). There was also a significant interaction between race representation and portrayal of the relationship between the two commentators (F(2, 168) = 3.36, p [is less than] .04, [[Eta].sup.2] = .04). In the conditions where two Black commentators (outgroup only) agreed with each other, or when a Black and a White commentator (ingroup-outgroup) disagreed with each other, participants indicated significantly more liberal opinions (M = 7.76), than when the portrayal of the relationship was reversed with two Black commentators disagreeing and a Black and a White commentator agreeing (M = 6.52, t(102) = 2.65, p = .009).


Table 2                                                              
Opinions about Land Use Policy(*)

Relationship Portrayal

Race Representation Agreement Disagreement

Ingroup only
M [6.29.sub.a] [6.11.sub.a]
SD 2.49 2.19
Outgroup only
M [7.57.sub.bc] [6.48.sub.ab]
SD 2.25 3.11
M [6.55.sub.ab] [7.93.sub.c]
SD 2.28 1.88

(*) Higher numbers reflect more liberal opinions, as a function of race
representation and the portrayal of the relationship. Different
subscripts indicate significant differences between means at the .05
level of significance.
  Race representation influenced opinions. However the pattern of influence did not reflect a preference for ingroup members positions over outgroup members' positions. Instead, participants' opinions were most liberal when White racial identity might have been most salient--namely, when two outgroup members agreed, or when an outgroup member and an ingroup member disagreed. Discussion and Conclusion


The purpose of this study was to assess the ways in which White viewers' responses to race representation correspond to strategies of White racial identity accommodation. The overall pattern of results suggests that this may be the case. The three dependent measures considered in this study (target-directed affect, perceptions of group member similarity, and opinions on land use) revealed biases consistent with Gaertner and Dovidio's (1986) concept of aversive racism, whereby the appearance of racism is avoided by an overcompensating, pro-Black bias. Participants who viewed versions of the program in which the racial outgroup is represented reported that they liked the commentators more than did participants who viewed a version in which only the ingroup was present.


These responses might reflect participants' self-presentation efforts to appear nonracist (e.g., Gaertner & Dovidio, 1986; Jackson et al., 1993). Such a response is consistent with the population from which the sample was drawn--namely, young White college students. Certainly there is evidence that this population is socialized to avoid the appearance of racial prejudice or stereotyping (Judd et al., 1995). However, the goal of this study was not to link racial prejudice per se with exposure to specific representations of race. Rather, this study explores the extent to which certain aspects of race representation are instrumental in facilitating or attenuating outgroup affiliation for White viewers. The patterns of responses in participants' expressed opinions about the issue discussed as well as their judgments of the commentators' similarity to each other indicate that this is the case. Participants perceived the greatest difference between a Black and a White commentator who disagreed. If this difference were only due to perceptions biased by the category memberships of the commentators, then the condition where the Black and the White commentator agreed should have received low ratings of similarity. Yet, the condition where inter-racial agreement was portrayed received one of the highest ratings of similarity. This condition also elicited the most favorable ratings of target-directed affect.


One possible explanation for this result is that White audiences are pleasurably engaged by portrayals of inter-racial harmony. The commercial success of the action "buddy" movies in which a Black character and a White character form a platonic partnership that requires they risk their lives for each other (i.e., Mel Gibson and Danny Glover in the Die Hard series, or John Travolta and Samuel Jackson in Pulp Fiction) is consistent with this explanation (see Nakayama, 1994, for an analysis of such films). In part, some forms of race representation on television--particularly the portrayal of harmonious, platonic, inter-racial relationships--might be an opportunity for vicarious affiliation with the outgroup for Whites. Apart from appearing antiracist to others, the outgroup affiliation afforded by portrayals of inter-racial harmony may allow some White viewers to affirm "antiracist" identities for themselves.


Another related possibility is that viewers are oriented toward race representations in terms of conflict. This expectation is consistent with Rokeach's (1960) theory of belief congruency. Rokeach (1960) demonstrated that individuals use social categories (i.e., race or religion) as proxies for categories formed on the basis of beliefs. Once individuals were confronted with an outgroup member who shared their belief, the group membership loses its importance (Rokeach, 1960). Recall, however, that the mixed, ingroup-outgroup pair (a Black and a White commentator) who were in agreement were evaluated as equally similar to the outgroup-only pair (two Black commentators) who were in agreement. This extreme evaluation of similarity for the mixed pair might reflect a response to an expectancy violation (e.g., Jussim, Coleman & Lerch, 1987). Jussim et al. (1987) tested expectancy violation with respect to racial stereotypes. The possibility that individuals' expectations about race relations are dominated by perceived conflict (e.g., Bobo, 1988) might influence how representations of inter-racial relationships are viewed.


Differences in opinions about land use policy between groups also indicate that race representation was a meaningful aspect of how viewers responded to the issue presented. The most liberal opinions were reported by participants who viewed either a version of the program in which two outgroup members agreed with each other, or in which an outgroup and an ingroup member disagreed with each other. An analysis of the dimension on which differences in opinion were expressed might provide some explanation for why race representation influenced opinions as it did. Braithwaite (1994) demonstrated that individuals' value orientations are mapped on two dimensions simultaneously--namely, freedom and equality. However, representations of social debate in the media tend to collapse these two orientations into a liberal-conservative dimension (Braithwaite, 1994). This is certainly how the issue was presented to participants in the present study.


The pattern of differences in opinion indicate that participants associated liberalism-conservatism with race. Such an association is revealed in the rhetoric of the Reagan and Bush administrations (Gray, 1995). The most illustrative example might be George Bush's successful presidential campaign in which Michael Dukakis was stigmatized by both the "L" word and Willie Horton. That the most liberal opinions were expressed by those for whom racial identity might have been most salient suggests the use of the liberal-conservative dimension, artificial though it may be, as a dimension on which participants could distance themselves from a White conservative identity.


Although the pattern of results described above is suggestive of race representation influences, the magnitude of the effects due to race representation is small. This indicates that other factors exerted an influence over participants' responses to the video segment. Information about participants' value orientations (e.g., Braithwaite, 1994) may have helped explain more of the variance in participants' responses. Aspects of the video segment itself may also have introduced some of this unexplained variance. The videos are not as polished as segments produced by a professional news organization. The relatively low means for target-directed affect show that the participants didn't have strong feelings of liking for the commentators.


A further limitation of the study concerns the generalizability of this particular set of messages in which race representation and intergroup relationships were systematically manipulated to other representations of inter-racial or intra-racial agreement or disagreement (Burgoon, Hall, & Pfau, 1991; Jackson, 1992). For the purposes of this study, the primary concern was whether a White sample would respond at all to race representation in ways that reflect the strategic management of an identity that is typically not even considered an identity. The pattern of results indicate that participants did affiliate or distance themselves from the commentators on the basis of race representation and relationship portrayal apart from their political orientations or perceptions of how well the commentators performed. This suggests that viewers may use media representations of race as an opportunity for outgroup affiliation. In this study, outgroup affiliation might have occurred through liking for the commentators, perceived similarity between an outgroup and an ingroup member, or through the projection of another dimension (political orientation) on which affiliation might occur. Such responses may foster an "enlightened racism"; however, such a conclusion requires a more direct evaluation of race representation influences on viewers' opinions about race-relevant topics.


The author gratefully acknowledges Sheila T. Murphy, Sandra Ball-Rokeach, Mark Wehrly, and Norman Miller for their helpful comments on earlier versions of this manuscript. Correspondence concerning this article should be addressed to Gail E. Coover, M3C Consulting, 4413 Somerset Lane, Madison, Wisconsin, 53711, or via e-mail at: gcoover@wccf.org




(1) The use of the terms "Whites" and "Blacks" as opposed to other more ethnically specific labels (e.g., African American, Haitian, Cape Verdean, Black Hispanic, Anglo American, Americans of Western European descent, etc.) is deliberate. The terms Black and White refer more directly to the historically created and socially and culturally perpetuated inter-racial hierarchy.


(2) Each race representation condition was counter-balanced by the actors playing the roles of the commentator (Mark Adams and Bill Warner). For example, half of the participants in the mixed, ingroup-outgroup condition watched a version of the video in which Mark Adams is White and Bill Warner is Black. The other half of the participants in the same race representation condition watched a version of the video in which Mark Adams is Black and Bill Warner is White. The same White actor in the mixed, ingroup-outgroup condition was paired with another White actor in the ingroup only condition. Similarly, the same Black actor in the mixed, ingroup-outgroup condition was paired with another Black actor in the outgroup only condition. This systematic pairing mitigated the effect of any specific actor on participants' responses across the effect(s) of each commentator's role. The order in which evaluations of the commentators' performances were solicited within the questionnaire was also manipulated. Half of the participants were asked to evaluate Mark Adams before Bill Warner, while the other half of the participants evaluated Bill Warner before Mark Adams. The effects of questionnaire ordering or the role of the commentator on participants' judgments are noted in analyses in which they are relevant to the measure being evaluated.




Allport, G. W. (1954). The nature of prejudice. Reading, Mass.: Addison-Wesley.


Armstrong, G. B., Neuendorf, K. A., & Brentar, J. E. (1992). TV entertainment, news and racial perceptions of college students. Journal of Communication 42(3), 153-175.


Aune, R. K., & Kikuchi, T. (1993). Effects of language intensity similarity on perceptions of credibility, relational attributions, and persuasion. Journal of Language and Social Psychology, 12, 224-238.


Biernat, M. & Vescio, T. K. (1993). Categorization and stereotyping: Effects of group context on memory and social judgment. Journal of Experimental Social Psychology, 29, 166-202.


Billig, M. G., & Tajfel, H. (1973). Social categorization and similarity in intergroup behaviour. European Journal of Social Psychology, 3, 27-52.


Bobo, L. (1988). Group conflict, prejudice, and the paradox of contemporary racial attitudes. In P. A. Katz & D. A. Taylor (Eds.), Eliminating racism: Profiles in controversy (pp. 85-114). New York: Plenum Press.


Bodenhausen, G. V., Schwarz, N., Bless, H., Wanke, M. (1995). Effects of atypical exemplars on racial beliefs: Enlightened racism or generalized appraisals? Journal of Experimental Social Psychology, 31, 48-63.


Braithwaite, V. (1994). Beyond Rokeach's equality-freedom model: Two-dimensional values in a one-dimensional world. Journal of Social Issues, 50, 67-94.


Brewer, M. (1991). The social self: On being the same and different at the same time. Personality and Social Psychology Bulletin, 86, 307-324.


Brewer, M. B., Manzi, J. M., & Shaw, J. S. (1993). In-group identification as a function of depersonalization, distinctiveness, and status. Psychological Science, 4, 88-92.


Burgoon, M., Hall, J., & Pfau, M. (1991). A test of the "messages-as-fixed-effect fallacy" argument: Empirical and theoretical implications of design choices. Communication Quarterly, 39, 18-34.


Campbell, C. P. (1995). Race, myth, and the news. Thousand Oaks, CA: Sage.


Collier, M. J. (1996). Communication competence problematics in ethnic friendships. Communication Monographs, 63, 314-336.


deHaven-Smith, L. (1988). Environmental belief systems: Public opinion on land use regulation in Florida. Environment and Behavior, 20, 176-199.


DeMott, B. (1995). The trouble with friendship: Why Americans can't think straight about race. New York, NY: Atlantic Monthly Press.


Dyer, R. (1988). White. Screen, 29, 44-64.


Ellemers, N., Doosje, B., Van Knippenberg, A., & Wilke, H. (1992). Status protection in high status minority groups. European Journal of Social Psychology, 22, 123-140.


Ellemers, N., Wilke, H., & Van Knippenberg, A. (1993). Effects of legitimacy of low group or individual status on individual and collective status-enhancement strategies. Journal of Personality and Social Psychology, 64, 766-778.


Entman, R. (1990). Modern racism and the images of lacks in local television news. Critical Studies in Mass Communication, 7, 332-345.


Ethier, K., & Deaux, K. (1994). Negotiating social identity when contexts change: Maintaining identification and responding to threat. Journal of Personality and Social Psychology, 67, 243-251.


Gaertner, S. L., & Dovidio, J. (1986). The aversive form of racism. In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 61-89). New York: Academic Press.


Giles, H., Mulac, A., Bradac, J. J., & Johnson, P. (1987). Speech accommodation theory: The first decade and beyond. Communication Yearbook, 10, 13-48.


Gray, H. (1989). Television, black Americans and the American dream. Critical Studies in Mass Communication, 6, 376-386.


Gray, H. (1995). Watching Race: Television and the struggle for the sign of blackness. Minneapolis: University of Minnesota Press.


Hartman, P., & Husband, C. (1974). Racism and the mass media. Totowa, NJ: Roman & Littlefield.


Helms, J. E. (1990). Black and White racial identity: Theory, research, and practice. New York: Greenwood Press.


Jackson, L. A., Sullivan, L. A., & Hodge, C. N. (1993). Stereotype effects on attributions, predictions, and evaluations: No two social judgments are quite alike. Journal of Personality and Social Psychology, 65, 69-84.


Jackson, S. (1992). Message effects research: Principles of design and analysis. New York: The Guilford Press.


Jhally, S., & Lewis, J. (1992). Enlightened racism: The Cosby Show, audiences, and the myth of the American dream. San Francisco: Westview Press.


Jones, E., Gallois, C., Barker, M., & Callan, V. J. (1994). Evaluations of interactions between students and academic staff: Influence of communication accommodation, ethnic group, and status. Journal of Language and Social Psychology, 13, 158-191.


Judd, C. M., Park, B., Ryan, C. S., Brauer, M., & Kraus, S. (1995). Stereotypes and ethnocentrism: Diverging interethnic perceptions of African American and White American youth. Journal of Personality and Social Psychology, 69, 460-81.


Jussim, L., Coleman, L. M., & Lerch, L. (1987). The nature of stereotypes: A comparison and integration of three theories. Journal of Personality and Social Psychology, 52, 536-546.


Katz, J. H. (1978). White awareness: Handbook for anti-racism training. Norman, OK: University of Oklahoma Press.


Le Poire, B. A., Ota, H., & Hajek, C. (1997). Self-disclosure responses to stigmatizing disclosures: Communicating with gays and potentially HIV+ individuals. Journal of Language and Social Psychology, 16, 159-190.


McConahay, J. B., Hardee, B. B., & Batts, V. (1981). Has racism declined in America? It depends on who is asking and what is asked. Journal of Conflict Resolution, 25, 563-579.


Moreland, R. L. (1985). Social categorization and the assimilation of new group members. Journal of Personality and Social Psychology, 48, 1173-1190.


Montgomery, K.C. (1989). Target prime time: Advocacy groups and the struggle over entertainment television. New York: Oxford University Press.


Nakayama, T. K. (1994). Show-down time: "Race," gender, sexuality, and popular culture. Critical Studies in Mass Communication, 11, 162-179.


Power, J. G., Murphy, S. T., & Coover, G. (1996). Priming prejudice: How stereotypes and counter-stereotypes influence attribution of responsibility and credibility among ingroups and outgroups. Human Communication Research, 23, 36-58.


Price, V. (1989). Social identification and public opinion effects of communicating group conflict. Public Opinion Quarterly, 53, 197-224.


Rokeach, M. (1960). The open and closed mind. New York: Basic Books.


Schuman, H., Steeh, C., & Bobo, L. (1985). Racial attitudes in America: Trends and interpretations. Cambridge, Mass.: Harvard University Press.


Sears, D. O. (1988). Symbolic racism. In P. A. Katz & D. A. Taylor (Eds.), Eliminating racism: Profiles in controversy (pp. 53-84). New York: Plenum Press.


Shah, H., & Thornton, M.C. (1994). Racial ideology in U.S. mainstream news magazine coverage of black-Latino interaction, 1980-1992. Critical Studies in Mass Communication, 11, 141-161.


Smedley, A. (1993). Race in North America: Origin and evolution of a world view. Boulder: Westview Press.


Tajfel, H. (1982). The social psychology of intergroup relations. Annual Review of Psychology, 33, 1-39.


Tajfel, H., & Turner, J., (1986). The social identity theory of intergroup behaviour. In S. Worchel & W.G. Austin (Eds.), Psychology of intergroup relations (pp. 7-24). Chicago: Nelson-Hall.


Turner, J. (1987). Rediscovering the social group: A self-categorization theory. New York: Basil Blackwell.


van Dijk, T. (1990). Elite discourse and racism. Newbury Park: Sage.


Vidmar, N., & Rokeach, M. (1974). Archie Bunker's bigotry: A study in selective perception and exposure. Journal of Communication, 24(1), 36-47.


Ward, D. (1985). Generations and the expression of symbolic racism. Political Psychology, 6, 1-18.


Wilder, D. A. (1984). Predictions of belief homogeneity and similarity following social categorization. British Journal of Social Psychology, 23, 177-194.


Gail E. Coover (Ph.D., University of Southern California, 1995) is an Assistant Professor in the Department of Communication Arts, University of Wisconsin. Her research examines the role of communication in self-identity development and change.
Next >

Service features

24/7 customer support

Written from scratch papers only

Any citation style

Fully referenced

Never resold papers

300 words per page Courier New font