Documenting Portrayals of Race/Ethnicity on Primetime Television over a 20-Year Span and Their Association with National-Level Racial/Ethnic Attitudes

Riva Tukachinsky∗ Chapman University

Dana Mastro University of California, Santa Barbara

Moran Yarchi Interdisciplinary Center Herzliya

The current study content analyzes the 345 most viewed U.S. television shows within 12 separate television seasons spanning the years 1987 to 2009. Us- ing multilevel modeling, the results from this comprehensive content analysis then are used to predict national-level racial/ethnic perceptions (between the years 1988 and 2008) with data from the American National Election Studies (ANES). Content analysis results reveal severe underrepresentation of Latinos, Asian Americans, and Native Americans, and a tendency to depict ethnic minorities stereotypically (e.g., overrepresentation of hyper-sexualized Latino characters). Multilevel-modeling analysis indicates that both the quantity and quality of ethnic media representations contributes to Whites’ racial attitudes.

∗Correspondence concerning this article should be addressed to Riva Tukachinsky, Chapman University, One University Dr. Orange, CA 92866. Tel: 714-516-4685 [e-mail: tukachin@chapman. edu].

These materials are based on work supported by the National Science Foundation under Grant Nos.: SBR-9317631, SBR-9707741, SES-0118451, SES-0535334, SES-0651271, SES-0720428, SES- 0840550, SES-8207580, SES-8341310, SES-8808361, SES-9009379, SES-9209410, SES-9707741, SOC77-08885, the Russell Sage Foundation under grant 82-00-01 Stanford University, and the University of Michigan. Any opinions, findings and conclusions or recommendations expressed in these materials are those of the authors and do not necessarily reflect the views of the funding organizations.

18 Tukachinsky, Mastro, and Yarchi

As advances in digital technology and trends in media convergence have revolutionized access to and creation of media content, headlines such as “TV is Dead: Long Live the Internet” (Davidson, 2011) have become a staple in the popular press. Yet despite the promise of its imminent demise, television viewing has remained the dominant source of media in our lives for decades. The average American today is tuned-in to television for 5 hours and 15 minutes per day (Short, 2013), with the lion’s share of viewing devoted to primetime programming (Nielsen, 2013). Moreover, growth in Internet and mobile device usage does not necessarily subtract from exposure to television content. Rather, the same content is being consumed differently, as television shows account for a large share of entertainment watched on new media platforms such as Netflix, Hulu, and TV.com (Pew Research Center, 2013). Accordingly, the quality of the messages contained in primetime television content is a consequential issue, as research has long demonstrated the small but significant influence of media exposure on the attitudes and beliefs of audience members (e.g., Morgan & Shanahan, 1997). This impact is meaningful to many social domains, including race-relations in society. Indeed, evidence linking media exposure with viewer’s racial attitudes has been demonstrated in both cross-sectional surveys and experiments (e.g., Dixon, 2008; Mastro, Behm-Morawitz, & Ortiz, 2007). Specifically, this work reveals that viewing the limited and often stereotypical characterizations of race and ethnicity offered in the media influences the attitudes, beliefs, and behaviors of audience members, Whites in particular, as well as the self-concept of ethnic minority group viewers (see Mastro, 2009, for review).

Despite the important contributions of this existing research, there are limita- tions to its ability to demonstrate the broad, societal-level implications of viewing such content. First, cross-sectional studies cannot disentangle societal influences on the media from the media’s influence on society. Second, although experi- ments can demonstrate cause-and-effect relationships, they typically explore the implications of short-term exposure to a specific media message, and therefore may lose sight of the broad-based picture reflected in one’s natural media diet. Moreover, although experiments can reveal the psychological mechanisms under- lying media effects, because they often use small, unrepresentative samples it is impossible to infer the actual magnitude of media effects in the general population. Furthermore, experiments typically employ a limited number of stimuli carefully selected to manipulate particular theoretically meaningful variables, rather than a broad, representative sample of media messages. Thus, despite the wealth of research in the realm of media effects on racial/ethnic attitudes, relatively little is known about how societal-level attitudes evolve and interact with overall media representations over time. To this end, the present study examines the association between the prevalence and quality of ethnic minority characters on television and White’s corresponding ethnic attitudes. Employing a repeated cross-sectional survey design and expansive content analysis, the study reconciles the tension

TV Portrayals and National-Level Attitudes 19

between internal validity and generalizability. As such, this work provides a unique and far-reaching assessment of media portrayals of race/ethnicity and the societal implications of exposure to this content.

Television Representations of Race and Ethnicity

Media portrayals of race/ethnicity vary over time and across groups (see Mastro, 2009). When it comes to representations of Blacks on television, the 1980s can be seen as a decade of elevated inclusion both in terms of the quantity and quality of roles—which remains the norm today. Currently, Blacks constitute between 14% and 17% of the primetime population (Children Now, 2004; Mastro & Greenberg, 2000) and approximately 13% of the U.S. population (U.S. Census Briefs, 2011). Despite this numeric parity, Black Americans are disproportionately featured in sitcoms and crime dramas. When in dramas, Black Americans are fea- tured in mixed-race casts (Children Now, 2004) whereas on sitcoms they appear in predominately Black casts (which are less likely to draw diverse audiences). Although some longstanding stereotypes linger in this content (e.g., criminality, aggression), when taken as a whole, the contemporary primetime television land- scape has been found to offer a more respectable array of portrayals of Blacks than offered in previous decades (Mastro, 2009).

Latinos have not achieved the same degree of inclusion on television, despite being the largest ethnic minority group in the United States, at approximately 16% of the population (U.S. Census Briefs, 2011). In fact, both the quantity and the quality of characterizations of this group have been criticized by media scholars and advocacy groups. Currently, Latinos comprise a mere 4–6.5% of the primetime TV population (see Mastro, 2009). When they are seen, they are relegated to a fairly limited set of roles (and have been for several decades), which often revolve around themes of sexuality, criminality, subservience, or intellectual ineptitude.

Very little beyond the sheer rate of appearance is known about the portrayal of Asian and Native Americans on television. Because they have been depicted so infrequently over the decades, they have often been excluded from quantitate analyses of content. Currently, Asian Americans comprise approximately 5% of the U.S. population (U.S. Census Briefs, 2011) and 3% of the primetime population (and only 1% of characters appearing in the opening credits, Children Now, 2004). When they are depicted it is often in minor and nonrecurring roles. However, in these infrequent roles they are often high-status figures. Native Americans face an unprecedented form of invisibility on television, often entirely absent from the TV landscape. Across the most recent content analyses, Native Americans (approximately 1% of the U.S. population) are found to represent between 0.0% and 0.4% of the characters in primetime television.

Although individual content analytic studies such as the ones discussed pre- viously have documented the major trends in the depiction of ethnic minorities

20 Tukachinsky, Mastro, and Yarchi

in broad strokes, the present study expands on this body of knowledge by using a systematic, longitudinal analysis of primetime television content. Utilization of consistent sampling and measurement procedures provides a more accurate basis for comparison between representations of various social groups over time. Thus, the present study allows a more consistent and nuanced understanding of the evo- lution in the quality and quantity of depictions of ethnic minorities on primetime television. To this end, the following research questions are posed:

Research Question 1: How frequently do ethnic minority characters appear on primetime television and does this frequency change over time?

Research Question 2: What are the qualities associated with ethnic minority char- acters on primetime television and do these qualities change over time?

Media representations of racial/ethnic groups and public attitudes. Learn- ing about various social groups from media is not limited to informational media such as news. Although news exposure has been consistently linked to percep- tions of ethnic minorities (e.g., Dixon, 2008), ample research indicates that ethnic portrayals in entertainment media (including fictional, scripted media such as dra- mas and sitcoms) are integrated into consumers’ mental representations of these groups. For example, exposure to situation comedies (but not dramas) has been found to be related to viewers’ perceptions of Blacks’ educational attainment and income levels (Busselle & Crandall, 2002). Similarly, exposure to reality-based crime programs (e.g., Cops), although not exposure to fictional crime shows, has been linked to viewers’ elevated perceptions of crime rates among African American (Oliver & Armstrong, 1998).

The results are even more consistent when examining effects of media on stereotyping and prejudice. For instance, to the extent that viewers believe that television portrays Latinos in a negative light, overall television viewing has been found to be associated with greater endorsement of negative stereotypes of Latinos (Mastro, Behm-Morawitz, & Ortiz, 2007). Similarly, Dong and Murrillo (2007) found that media consumers who report learning about race/ethnicity from television were more likely to endorse negative stereotypes of Latinos.

Importantly, however, media influence is not limited to promoting stereo- types. Positive media representations of ethnic minorities can have pro-social effects, fostering egalitarian beliefs and positive intergroup attitudes. Even brief exposure to positive and likable media figures of color, such as Oprah Winfrey or Jimmy Smits can enhance audience members’ racial attitudes, at least temporarily (Bodenhausen, Schwarz, Bless, & Wanke, 1995; Mastro & Tukachinsky, 2011).

Taken together, the quality and quantity of ethnic and racial representations can influence a wide range of cognitive, affective, and behavioral outcomes ranging from how audiences perceive, feel about, and treat different groups (see Mastro, 2009, for review). However, most existing research in this area has focused on

TV Portrayals and National-Level Attitudes 21

specific television shows or genres at a single point in time. The present study examines these processes on a macro level, investigating the relationship between the natural media landscape and attitudes toward ethnic minorities on a societal level, over two decades. Based on findings from cross-sectional and experimental research, it would be expected that national-level attitudes about race and ethnicity would reflect the variable representations seen, over time, on TV. Specifically, the following hypothesis is posed:

Hypothesis 1: The quantity and quality of representations of ethnic minorities will be positively associated with White Americans’ attitudes toward these groups.

Method

The present study combines two data corpuses. First, White Americans’ attitudes toward ethnic minorities in the United States were assessed in six cross-sectional American National Election Studies (ANES) surveys conducted from 1988 through 2008. Second, media representations of ethnic minorities were examined through a content analysis of the most viewed U.S. primetime television shows between 1987 and 2008. Television seasons are scheduled September through May, whereas ANES surveys take place in proximity to Election Day. Thus, each wave of survey data collection overlaps with one television season. These seasons, and the seasons preceding each interview wave (i.e., the Fall television season a year before each survey was conducted) comprising a total of 12 television seasons, were content analyzed. Examination of two consecutive seasons per interview wave allows capturing more cumulative, long-term rather than immediate media effects. Furthermore, analysis of television seasons preceding each interview wave increases the validity of inferences of causal relationships between media representations and public opinion.

Survey Data

Sample and inclusion criteria. Only data from respondents who identified themselves as non-Hispanic Whites were included in the current study. The survey included questions about respondents’ perceptions of various social groups (e.g., feminists, Catholics) including Blacks and Latinos. Due to survey length consid- erations, not all participants were asked about all social groups. Rather, they were randomly assigned to only report their feelings toward some groups but not oth- ers. Given the focus of the present study, only respondents who were asked about both Latinos and Blacks were included in the sample. These inclusion criteria resulted in a total of 5,299 respondents across the six interviews (see Table 1 for demographic details).

22 Tukachinsky, Mastro, and Yarchi

Table 1. Demographic Characteristics of the Survey Respondents

Years of N % male Age (SD) education (SD) % republican Median income

1988 1,310 45.7% 46.30 12.96 48.5% 25,000–29,999 (17.63) (2.68)

1992 284 43.7% 48.40 13.26 39.1% 30,000–34,999 (17.53) (2.67)

1996 1,178 45.6% 49.10 13.65 32.8% 35,000–39,999 (17.49) (2.46)

2000 671 45.5% 48.75 13.77 35.9% 50,000–64,999 (17.95) (2.54)

2004 760 46.4% – 13.91 46.8% 45,000–49,999 – (2.37)

2008 1,059 44.4% 49.71 13.73 41.4% 45,000–49,999 (17.47) (2.20)

Total 5,262 45.4% 48.33 13.53 41.2% (17.64) (2.50)

Attitudes toward Blacks and Latinos. Attitudes were assessed using feel- ing thermometer questions. Individuals were asked to indicate on a 101-point scale how favorably/warm (closer to 100 degrees) or unfavorable/cold (closer to zero) they feel about Latino-Americans and Blacks. In the 2008 survey wave, the response options range was changed to a 30-point scale. To ensure consistency across interviews, responses in the last wave were rescaled to a 101-point scale (score – 1 / 29*100).

Control variables. Demographic variables including sex, years of educa- tion completed, income (z-transformed due to changes in income categories), and political leaning (as determined by voting for a republican presidential candidate) were used as control variables.

Content Analysis

Coding reliability. Three coders were trained on a sample of shows from seasons other than those included in the reported study. To ensure reliability throughout the coding process, approximately 10% of the shows in the study sample were coded by all three coders. Reliabilities (Cohen’s kappa) are reported for each of the variables in the following section.

TV Portrayals and National-Level Attitudes 23

Sample. The regular cast of the 40 most viewed television shows in each primetime season was analyzed based on episode guides and synopses. Sports, movies, and animated shows were eliminated from the sample. This resulted in a total sample of 345 television shows. These included sitcoms (n = 140, 40.6%, e.g., The Cosby Show), crime series (n = 68, 19.7%, e.g., Law and Order), dramas (n = 50, 14.5%, e.g., E.R.), reality TV and game shows (n = 42, 12.2%, e.g., Who Wants to be a Millionaire), current affairs (n = 25, 7.2%, e.g., 20/20), and dramedy (n = 11, 3.2%, e.g., Desperate Housewives). Nine shows (2.6%) were coded as “other” (e.g., Unsolved Mysteries) (К = .85).

In scripted programs, such as dramas and sitcoms, regular characters were defined as those appearing in at least 50% of the episodes in a given season. For reality TV shows, the program’s hosts and recurring contestants were coded. In reality TV shows with a large number of contestants, such as American Idol, up to 24 finalists were coded, whereas for game shows and current affair programs that did not feature returning participants, only the hosts were included in the analysis. The final sample included a total of 2,575 characters.

Variables. Each regular (human) television character was coded for eth- nicity. The ethnic categories included: White (European, Asian Indian, Mid- dle Eastern), Black (African American, Jamaican, African, Haitian), Latino (Mexican, Cuban, Puerto Rican, Central American, South American), Asian (East Asians, Pacific Islanders), Native American, or mixed minority ethnicity (both of the character’s parents are Black/Latino/Asian/Native American, but the parents are not of the same group). When information about the character’s ethnicity was ambiguous, or when the character was not fictional (e.g., reality show contestants and hosts), the media persona’s actual race was coded. For an actor to be coded “Black,” “Latino,” “Asian” or “American Indian” at least one of the actor’s parents had to be of this descent (К = .97).

Once the characters’ ethnicity was identified, ethnic minority characters were coded for five indicators of quality of representations. These variables were as- sessed on three-point scales. First, each ethnic minority character was coded for moral disposition (bad/ambivalent/good) (К = 1.00) and likability (unlik- able/ambivalent/likable) (К = 1.00). Additionally, ethnic characters were coded for sexual objectification defined as the extent to which the character’s sexual appeal (e.g., portrayal as “eye candy”) and the character’s sexuality were cen- tral to the show (К = .95). Scores ranged from not at all sexual to very sexual (i.e., most of the role in the plot is sexual) with the middle point somewhat sex- ual (i.e., occasional references to the character’s sexuality). Next, social status (low/medium/high) was determined based on the character’s status relative to other characters on the show (e.g., a popular girl vs. a misunderstood geek). If the char- acter was typically the object/target of other’s jokes and mockery the character was scored low on the social status variable. If other characters sought this character’s

24 Tukachinsky, Mastro, and Yarchi

advice and guidance, the character was rated as having high social status (К = 1.00). Finally, professional status was coded based on the character’s professional authority (e.g., partners in a legal firm have higher status than associates in a legal firm, who in turn have higher status than unemployed/unskilled labor workers). The status categories included low, medium, high, and unknown (К = .85).

Results

Frequency of Representations of Ethnic Minorities in Primetime Television

The first research question explored the prevalence of ethnic minority char- acters in primetime television over the decades. To examine this question, data from the two consecutive television seasons were combined to reflect changes in 2-year intervals. Chi-square statistics and analysis of variance were used to examine differences in representations of various groups on primetime television over time. Table 2 presents the distribution of characters’ race and ethnicity across the coded period of time. Overall, the distribution of characters’ ethnicity over the years varied significantly (χ 2(25) = 91.11, p <. 001 φc = .08).

In the late 1980s, Whites comprised 78.1% of the regular characters in top viewed primetime shows. In the following two decades, their share slightly in- creased and remained relatively stable around 82–84% with the exception of the 1995–1997 television seasons when White characters’ share rose sharply to 88%.

The prevalence of African American characters fluctuated dramatically over the years. Due to the popularity of comedies with predominately Black cast such as The Cosby Show and Amen in the 1980s and The Fresh Prince of Bel-Air and Hangin’ with Mr. Cooper in the early 1990s, Black characters were highly prevalent on television (21.6% in 1987–1989 and 16.8% in 1991–1993). Then, the share of Black characters plummeted to 9.6% and remained consistent at around 10–14%.

Other ethnic groups were nearly absent from the small screen. Although the prevalence of Latino characters was found to increase (from less than 1% in the 1980s to over 3% in the 2000s), they remained grossly underrepresented on TV. A similar trend was observed in the case of Asian American characters. No reoccurring Asian characters appeared in the top primetime shows of 1987–1989 and 1991–1993 seasons, but their share gradually rose to 2.8% in 2007–2009.

Finally, Native Americans constituted the most severally underrepresented group. Out of 2,336 regular characters in 12 television seasons, merely three characters were coded as Native American. Two of them are accounted for by Marilyn Whirlwind’s character in two seasons of Northern Exposure. The third Native American character was a contestant on the reality show Survivor—a son of a European Caucasian father and a Quechan mother.

TV Portrayals and National-Level Attitudes 25

Table 2. Ethnic Composition of Characters in Top Prime-Time Television by Season

Native Mixed White Black Latino Asian American minority Total

1987–1989 253 70 1 0 0 0 324 78.1% 21.6% .3% .0% .0% .0% 10.0%

1991–1993 286 58 0 0 2 0 346 82.7% 16.8% .0% .0% .6% .0% 10.0%

1995–1997 294 32 5 2 0 1 334 88.0% 9.6% 1.5% .6% .0% .3% 10.0%

1999–2001 367 53 7 6 1 2 436 84.2% 12.2% 1.6% 1.4% .2% .5% 10.0%

2003–2005 487 70 23 11 0 1 592 82.3% 11.8% 3.9% 1.9% .0% .2% 10.0%

2007–2009 453 54 18 15 0 3 543 83.4% 9.9% 3.3% 2.8% .0% .6% 10.0%

Total 2,140 337 54 34 3 7 2,575 83.1% 13.1% 2.1% 1.3% .1% .3% 10.0%

Together, the results of the content analysis indicate that ethnic minorities are largely excluded from primetime television. Most strikingly, Native Americans are almost entirely absent. These results are consistent with past content analyses, which reported very few, if any, Native American characters in their sample (e.g., Children Now, 2004; Mastro & Greenberg, 2000). Whereas past studies focused on smaller samples of shows over a shorter period of time, the findings from the current study are illuminating in that they demonstrate how consistent and pervasive the alienation of Native Americans from primetime television is.

Despite a trend of gradual increase in the presence of Asians and Latinos, the representation of these groups continues to be strikingly low. In particular, the incongruity between Latino characters’ prevalence on the screen and their share of American population is remarkably discrepant. According to the U.S. Census, in 2000 Latinos comprised over 12% of the American population, yet Latinos constituted a mere 1.5% of regular primetime characters in 1999–2001. By 2010, according to the Census data, Latinos made up over 16% of the U.S. population, but they comprised only 3.3% of the primetime characters in 2007–2009.

As discussed in the introduction, appropriate media depictions can provide vicarious, positive intergroup contact and improve intergroup relationships. It is therefore conceivable that even the slight increase in the frequency of Latino characters in primetime television could contribute to improvements in Whites’ attitudes toward Latinos. However, not only the quantity but also the quality of

26 Tukachinsky, Mastro, and Yarchi

media representations of ethnic groups matter. Exposure to stereotypical and neg- ative ethnic minority characters can reinforce and increase the accessibility of negative intergroup feelings (e.g., Valentino, 1999). Thus, any socially desirable effects of exposure to ethnic minority characters depend on the existence of favor- able, quality representations of ethnic minorities. The second research question examines these aspects of the characters’ representations.

Quality of Representations of Ethnic Minorities on Primetime Television

The second research question addressed the quality of ethnic minority pre- sentations on primetime television. Given the small number of Native American characters, they had to be excluded from these analyses. Thus, the following analyses refer only to Blacks, Latinos and, whenever possible, Asian Americans.

Overall, all ethnic minority characters tended to be presented as highly pos- itive, but Blacks were slightly more positive than Latinos (on a 3-point scale, Black: M = 2.94, SD = .23; Latino: M = 2.85, SD = .36, Asian: M = 2.91, SD = .29, F(2,416) = 3.40, p < .05, η2 = .02, I-J = .10, p <.05). In fact, in the entire sample, only one ethnic minority character was coded as “bad” and 93.6% of the characters were coded as “good.” This is not surprising given that the present content analysis examined only regular characters (which are likely to be either positive or ambivalent), whereas most villains are not part of television shows’ regular cast. For instance, in crime dramas such as Law and Order the regular cast includes law enforcement agents (“good characters”) whereas the law offenders (“bad characters”) change from episode to episode (and thus are not included in the current sample).

Similarly, most (74%) of the ethnic minority characters were likable, almost a quarter (24.6%) were ambivalent, and only 1.4% were unlikable. Ethnic groups varied in terms of their likability. The majority of Black (76.7%) and Latino (74.1%) but less than a half (47.1%) of Asian characters were likable. Comparisons of means of characters’ likability suggest that on average, Asian characters were presented as significantly less likable (M = 2.47, SD = .51) than Blacks (M = 2.75, SD = .47) and Latinos (M = 2.74, SD = .44), F(2,416) = 5.39, p < .01, η2 = .03, Bonferroni post hoc tests: Latino-Asian I-J = .27, p < .05, Black-Asian I-J = . 27, p < .01).

Overall, most of the ethnic minority characters (64.2%) were not sexually objectified. But whereas hyper-sexualized portrayals of Blacks and Asians were relatively scarce (7.6% and 11.8%, respectively), almost one in four Latino char- acters were very sexual (24.1%). A comparison of mean levels of characters’ objectification suggests that only the difference between Latinos (M = 1.78, SD = .82) and Blacks (M = 1.40, SD = .63) was significant (I-J = . 38, p < .001, F(2,416) = 8.22, p < .01, η2 = .04).

TV Portrayals and National-Level Attitudes 27

A small majority of the characters enjoyed high social status (56.6%) and very few had low social status (2.9%). Ethnic groups did not significantly dif- fer in terms of their average social standing (F(2,416) = 1.02, p > .10). Sim- ilarly, groups did not significantly differ in terms of their professional status (F(2,350) = 1.51, p > .10). However, there was more variability in characters’ professional status. About half Blacks (55.9%) but most Asians (71.4%) and Lati- nos (58.7%) occupied a highly professional status. Yet, Latinos were more likely than any other ethnic minority to have low professional status (23.9% vs. 11.8% of Blacks and 7.1% of Asians).

Although gender differences are not central to the present study, it is important to note some meaningful variations in representations of men and women. First, Latina and Black female characters had a significantly lower professional status compared to their male counterparts (Black: M women = 2.32, SD = .75 vs. M men = 2.52, SD = .65, t(277) = 5.74, p < .05, η = .15; Latinos: M women = 1.83, SD = .92 vs. M men = 2.68, SD = .61, t(44) = 3.74, p < .001, η2 = .02). Additionally, though not surprisingly, women were, overall, more sexually objectified than men, however the effect was significant only for Black characters (M women = 1.47, SD = .69 vs. M men = 1.32, SD = .57, t(303) = 2.07, p < .05, η2 = .01) but not for Latinos (M women = 1.86, SD = .89 vs. M men = 1.72, SD = .77, t(52) = .64, p = .86).

In all, then, recurring ethnic minority characters on primetime are presented as moral and likable (or at least ambivalent, but not dislikable). They enjoy high social status, but occupy a variety of professional statuses. Although the majority of the characters are not sexually objectified, a substantial number of Latinos are hyper-sexualized. These findings suggest that exposure to primetime media has the potential to improve attitudes toward ethnic minorities by providing viewers with vicarious encounters with positive, likable and esteemed out-group members. However, the limited roles in which Latinos are presented, as hyper-sexual and less professional, might support social stereotypes such as “the Latin lover.” The following section examines the extent to which these attributes are static or change over time.

Changes in Qualities of Representations over Time

Figures 1 and 2 depict changes in the quality of representations of Black and Latino characters over time. There were no significant changes in representations of Black characters as good (F(5,325) = .19, p > .10) and likable (F(5,325) = 1.99, p = .08). However, over the years, representations of Blacks varied in terms of sexualization (F(5,330) = 7.26, p < .001, η2 = .10). Hyper-sexualized portrayals of Blacks were particularly prominent in the 1991–1993 seasons when about 1 in 5 Black characters (22.8%) were depicted as very sexual. Similarly, the mean levels of social status of Black characters fluctuated tremendously over the years

28 Tukachinsky, Mastro, and Yarchi

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

87-89 91-93 95-96 99-01 03-05 07-09

Good

Likable

Sexual

Social

Professional

Fig. 1. Mean levels of Black characters’ morality, likability, social and professional status and sexual objectification over the years.

(F(5,325) = 6.63, p < .001, η2 = .09) ranging from a low of 2.26 (SD = .56) in 2007–2009, when only 31.5% of Blacks enjoyed high social status, to as high as 2.74 (SD = .44) in 2003–2005 with 74.3% of the characters having high social standing.

Blacks’ average professional status also changed significantly over the years (F(5,273) = 8.12, p < .001, η2 = .13). A significant quadratic trend emerged for Blacks’ professional status (F(1,278) = 6.71, p < .01, deviation: F(3,249) = 2.20, p > .05). Blacks’ professional status consistently increased, peaking in 2003–2005 (M = 2.84, SD = .48) then sharply declining in 2007–2009 to its levels in the 1990s (M = 2.50, SD = .70). The change in averages reflects a decline in the number of highly professional characters (from 85.2% down to 62%) and tripling the share of low status characters (from 3.7% to 12%)—the highest share of Black low status characters since 1993.

Univariate ANOVA reveals a significant time by gender interaction for Blacks (F(5,279) = 3.07, p < .05, η2 = .05) accounted for by the fact that in the 1980s, Black women held substantially lower professional status (M women = 1.69, SD = .68 vs. M men = 2.33, SD – .71), however, the gender gap was narrowed or even reversed in subsequent seasons.

For Latinos, very small cell sizes for the earlier television seasons left statistical analyses underpowered. Nonetheless, the results of the content anal- ysis suggest that Latino representations shifted toward greater sexualization

TV Portrayals and National-Level Attitudes 29

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

91-93 95-96 99-01 03-05

Good

Likable

Sexual

Social

Professional

Fig. 2. Mean levels of Latino characters’ morality, likability, social and professional status and sexual objectification over the years.

(F(3,49) = 2.24, p = .09, η2 = .12), with a significant linear trend (F(1,49) = 3.24, p < .01, deviation: F(2,45) = 1.05, p > .10). To illustrate, the percent of Latino characters whose sexuality was not an important component of the character and the show, dropped from 100% in 1995–1997 to 66.7% in 1999–2001, down to mere 27.8% in 2007–2009. Gender by time interactions could not be tested given the small sample size and the fact that all the recurring Latino characters until the mid-1990s were male. A significant trend of change in Latino likability also emerged (F(3,49) = 2.49, p = .07, η2 = .13) with a quadratic trend (F(1,45) = 3.95, p < .05, deviation: F(2,49) = .01, p > .10). In 1995–1997 a sizeable majority of Latino characters (80%) was likable, but the percent dropped to around 60% in late 1990s and early 20003. However, in 2007–2009 Latino characters were, again, almost uniformly likable (94.4%).

To summarize, although the morality and likability of ethnic minority characters remained consistent over the years, other indicators of the quality of representations shifted from season to season, albeit not always in a systematic fashion. Certain television seasons depict Blacks and Latinos more/less favorably. Specifically, for Blacks, the 1991–1993 and 1999–2001 seasons are marked by particularly unflattering representations involving relatively high levels of sexual objectification and lower social status. Conversely, in 2003–2005, Black characters were depicted exceptionally positively; as less sexual, more professional, and higher in social status. For Latinos, the increased presence on primetime programming over time was associated with deterioration in the

30 Tukachinsky, Mastro, and Yarchi

quality of representation. In 2007–2009, media consumers were more likely than ever before to encounter Latino characters, but these characters were also more sexualized and less professional than Latino characters in previous years. It is, therefore, hypothesized that changes in the public’s perception of ethnic minorities will be explained in part by these shifts in the media climate, such that at times when Blacks and Latinos are depicted more sympathetically, viewers’ attitudes toward these groups become more favorable and vice versa.

White Americans’ Perception of Ethnic Minorities

The research hypothesis asserts that the quality and quantity of media depic- tions of Blacks and Latinos will be associated with public perceptions of these groups (regrettably, Asian Americans and Native Americans could not be included in the analyses given how infrequently they appear on primetime television). To test this hypothesis, a multilevel analysis approach was used. The analysis in- cluded two levels of measurement: the individual level and the media’s quantity and quality representations level. To examine both levels of measurement, SPSS MIXED model was used. This multilevel procedure is advantageous in that it fa- cilitates analysis of data collected at multiple levels simultaneously (Hayes, 2006). This analysis allows us to examine the contribution of both the respondents’ de- mographics and media representations of ethnic minorities on the respondents’ attitudes toward those groups. Although the analysis does not examine changes over time, it enables us to learn about the impact of variables measured on different levels on a dependent variable—in the current study the respondents’ level, and the television portrayal surrounding the time of the survey level, on the respon- dents’ attitudes toward Latinos and Blacks. This approach is consistent with the multilevel modeling approach employed in past studies linking changes in media content with fluctuations in public opinion (Nisbet & Myers, 2012).Specifically, the individual level variables included information about the respondents’ age, sex, income, education and political leaning (based on voting preferences). The media representation level included information about the way the media represented Latinos and Blacks in the television season and the seasons preceding each survey wave (the variables include the information about the characters appearing in both seasons combined). This level of measurement examines both the quantity and quality of the representation. The variables used in the analysis were: the number of Latinos and Blacks appearing in the content; the number of highly liked and “good” Latino and Black characters (i.e., coded “3” on the three-point scale); the number of Latino or Black characters with high social or professional status (i.e., coded “3” on the three-point scale), and the number of hyper-sexualized Latino and Black characters (i.e., coded “3” on a three-point scale). Measures of likability and goodness and measures of social and professional status were grouped based on theoretically driven logic, supported by Principal Component Analysis (with

TV Portrayals and National-Level Attitudes 31

Table 3. Multilevel Models Predicting White Americans’ Attitudes Toward Latinos and Blacks between 1988 and 2008

Model 1: Attitudes Model 2: Attitudes toward Latinos toward Blacks

Fixed effects

Sex (male = 1) 1.68* * 2.94* * Political leaning (Republican = 1) .93 −.16 Age .02 .03 Income −.16 −.18# Education .81

* * * .57

* * *

N of Latino characters −6.35# N of Good and Liked Latino

characters .25

N of Professional and social Latino characters

9.36 * *

N of sexual Latino characters −11.45* * * N of Black characters 1.07

* * *

N of Good and Liked Black characters

−1.00* * *

N of Professional and social Black characters

1.01 * * *

N of sexual Black characters −1.17* * * Intercept 39.52

* * * 43.90

* * *

Variance of Random Components Year .00 .00 −2 Restricted Log Likelihood 35,626.52 35,907.93

N 5,262 5,262

* * * p < .001, * * p < .01, * p < .05, #p < .10. Note. The statistical procedure is multilevel (hierarchical) modeling (using SPSS MIXED model). The variance type is Variance Component. SPSS uses the Restricted Maximum Likelihood (REML) method for mix model analysis. Entries are parameter estimates. All the N variables for Latino and Black characters include the information about the characters appearing in the television season and the seasons preceding each survey wave combined.

variable loadings exceeding .70 for the corresponding factors). Table 3 presents the effect coefficients of media’s representation of Latinos and Blacks on public attitudes toward these groups, controlling for demographics.

Model 1 reveals that the number of highly professional and social Latinos characters had a significant positive effect on Whites’ attitudes toward Latinos. The number of Latino characters and the number of hyper-sexual Latino characters were associated with more negative attitudes toward Latinos (however this result did not reach conventional levels of significance [p = .08], possibly owing to the small N at the media representation level). The number of good and liked

32 Tukachinsky, Mastro, and Yarchi

Latinos characters was not found to have a significant effect on attitudes toward Latinos, when controlling for all other variables. As for the demographic variables (measured at the individual level); education was found to significantly contribute to positive attitudes toward Latinos. Moreover, men were found to have more positive attitudes toward Latinos than women. No other demographic variables were found to have a significant effect on attitudes toward Latinos.

Model 2 examined the media’s contribution to public attitudes toward Blacks. The model revealed that both the prevalence of Black characters (i.e., the overall number of Black characters), and Black characters’ professional and social status had a positive and significant effect on attitudes toward Blacks. The number of good and liked Black characters and the number of hypersexual Black charac- ters significantly reduced the support toward Blacks. The demographic variables exhibit similar patterns as those found in model 1 (predicting attitudes toward Latinos); Education was positively associated with attitudes toward Blacks, and men were found to have more positive attitudes toward lacks in comparison to women. Additionally, income trended toward significance (p = .098) in diminish- ing attitudes toward Blacks. No other demographic variables were found to have a significant effect on attitudes toward Blacks.

Discussion

The current study examined the quality and quantity of representations of ethnic minorities in primetime television, and found a relationship between these media portrayals and White Americans’ perceptions of Blacks and Latinos. Over- all, the results of the content analysis are consistent with past, smaller-scale studies, revealing a disturbing underrepresentation of Native Americans, Asian Americans and Latinos on primetime television. Although ethnic minority tele- vision characters are (generally) presented as good and likable, their social and professional status as well as sexualization varies over the years. Women of color are particularly likely to be presented as hyper-sexual and less professional. These representations are consistent with longstanding stereotypes of ethnic minorities, such as the “Latin lover” (Latinos as passionate and seductive), and those asso- ciating Blacks and Latinos with lower abilities and work ethic. Consequently, as media-based exemplars are incorporated in the viewers’ mental representation of the group as a whole (Busselle & Shrum, 2003; Mastro & Tukachinsky, 2011), sex- ual and unprofessional ethnic minority characters can cultivate racial stereotypes and/or make them more salient and readily available.

Indeed, the juxtaposition of the content analysis results with data from public opinion surveys revealed that television representations of Blacks and Latinos were associated with subsequent public perceptions of these groups. Most notably, controlling for demographic variables, the prevalence of hyper-sexualized Black and Latino characters was associated with more negative attitudes toward Blacks

TV Portrayals and National-Level Attitudes 33

and Latinos in the United States. Conversely, as more Blacks and Latinos on television occupied high social and professional status, White Americans tended to hold more favorable views of these social groups. These results illuminate the importance of improving the quality of representations of ethnic minorities, rather than merely increasing the sheer number of these characters in the media.

The results concerning likability and morality of the characters are less in- tuitive and can be interpreted in a number of ways. The general premise of the parasocial contact hypothesis is that likable and good minority characters will facilitate more positive perceptions of the outgroup. In the present study, however, no such effect was found for Latino characters, and the contrary was true for Black characters. These results resonate with past research that anecdotally addresses the unsuccessful generalization of parasocial contact effects. In a qualitative study of views of The Cosby Show, Jhally and Lewis (1992, p. 95) found that “[a]lthough they [White viewers] happily welcomed the Huxtables into their homes, careful examination of their discussions made it clear that this welcome would not be extended to all black people.”

These seemingly theory-inconsistent results highlight the fact that likability of ethnic minority media figures is a necessary albeit insufficient condition for generalizing the positive response to media characters to the group as a whole. First, as previous studies have revealed, to promote overall positive attitudes toward the outgroup, a likable media persona need also be perceived as typical of the group (Bodenhausen, Schwarz, Bless, & Wänke, 1995). It is conceivable that many of the positive and likable characters in this study were deemed nonrepresentative of Latinos and Blacks as a whole, therefore limiting the potential for positive feelings to extend to the group.

On a related note, the contact hypothesis asserts that group membership should be salient for attitudes to shift (Pettigrew, 1998). If viewers do not perceive the mediated contact to be an interracial/ethnic experience, positive contact will not be transferred to perceptions of the outgroup. For example, in the case of The Cosby Show, Jhally and Lewis (1992) suggest that many White viewers felt that the Huxtables were characteristically “White.” If this is the case, it appears that in some instances, the race/ethnicity of minority characters (e.g., Anita Van Buren and Ed Green on Law and Order) is almost incidental, and although viewers can form positive relationships with these figures, these relationships are not experienced at an intergroup level. It is possible that likability and morality affect attitudes only when the character’s racial/ethnic identity is central to the story line (e.g., Will Smith in The Fresh Prince of Bel-Air, or Dr. Eric Foreman in House M.D.).

Furthermore, different aspects of character representations can work in con- cert to influence viewers’ perceptions in ways that cannot be identified in the current study. For example, it is conceivable that a likable buffoon character will promote racial stereotypes rather than counteract them. The current study can- not provide a more nuanced examination of the contextual complexity of various

34 Tukachinsky, Mastro, and Yarchi

representation variables. Without a more detailed examination of the specific roles played by individual characters, morality and likability per se are not sufficient to guarantee prosocial outcomes.

Finally, it is important to consider the ways in which media portrayals interact with viewers’ psychological variables such as group identity and preexisting atti- tudes. For example, Mastro and Tukachinsky (2011) found that mere exposure to a highly likable Latino actor (Jimmy Smits—who was in fact coded in the current study for his appearances in LA Law, NYPD and The West Wing) does not always result in improved attitudes toward Latinos. Rather, the effect is contingent on viewers’ preexisting racial attitudes. Specifically, the positive and likable Latino character promoted favorable attitudes only among viewers who were already positively predisposed toward members of other ethnic groups.

Relevancy beyond the U.S. Context and Social Implications

Although the present study examined only television representations and public opinion in the United States, the results can be relevant to intergroup media representations in other countries as well. The tendency to negatively depict and underrepresent ethnic, national and religious minorities in media has been documented in many countries, including for example, non-Whites in Canada (Media Watch, 1994), Muslims in the United Kingdom (Saeed, 2007), Turkish immigrants in Germany (Yalçin-Heckmann, 2002) and Arabs in Israel (The Second Authority for Television and Radio, 2006). These findings, replicated around the world, reflect the socio-political dynamic underlying media representations. Minority social groups with low group vitality lack political power and access to the media industry and are therefore symbolically alienated and marginalized by the majority group that controls the media. These representations, in turn, can have implications for media consumers’ attitudes toward outgroup members. Theories that explicate the mechanisms underlying such effects, including social identity theory, parasocial contact, and exemplar management theories, imply that these are universally human psychological processes. Thus, the results of the current research and other media effects studies (such as Hurley, Jensen, Weaver, & Dixon, 2015; Ramasubramanian, 2015; Schmader, Block, & Lickel, 2015) suggest that shifts in the quality of any minority representations can have implications for the majority group audience’s attitudes toward the minority group.

As discussed by Tukachinsky (2015), it is often not conceivable to impose media regulations on racial/ethnic depictions. Thus, the findings of the current study and other studies in the domain of media stereotyping highlight the need for social policies promoting education of media consumers about stereotypical representations of race/ethnicity. Currently, media literacy is rarely a priority in K- 12 education, and even when media literacy programs are incorporated in schools, they typically focus on media violence and health concerns such as body image

TV Portrayals and National-Level Attitudes 35

(for review, see Jeong, Cho, & Hwang, 2012). Few programs, such as the one reported by Scharrer and Ramasubramanian (2015), heighten media consumers’ awareness of racial issues in the media. More interventions should be implemented in educational settings and additional research should examine their effectiveness in reducing susceptibility to antisocial media effects.

Study Limitations

The current study reports the relationship between primetime television and attitudes toward Latinos and African Americans. It was not possible to examine the effects of media on attitudes toward other racial/ethnic groups (e.g., Native Americans, Asian Americans) due to lack of data about such attitudes and the negligible number of characters on television. However, that the results of the two models predicting attitudes toward Latinos and Blacks yielded almost identical results increases confidence in the generazability of these findings to other social groups. In other words, it is logical to assume that shifts in representations of Asian Americans, and possible evolution of the depiction of Native Americans would result in similar changes in audience’s perceptions of these groups.

Another limitation of the study stems from reliance on secondary data— constraining our assessment of attitudes toward Latinos and Blacks to a very crude, single-item measure. The ability to include a more expansive set of measures of attitudes toward ethnic minorities (such as symbolic and modern racism scales, measures of endorsement of specific stereotypes such as laziness, violence and sexuality, etc.) would provide important insights. Additionally, the ANES polls did not include measures of media exposure (other than political campaigns). Although the primetime shows included in the content analysis were widely popular among Americans, it would be ideal if exposure to these media portrayals was not assumed but assessed directly.

The current content analysis revealed gendered ethnic stereotypes. Due to the relatively small number of ethnic minority characters, it was not possible to analyze, in greater depth, the effects of female and male ethnic minority characters. However, experimental studies could follow up on these findings and explore the meaningful intersection between gender and ethnicity, and advance the field’s understanding of sex-based subtyping processes.

Finally, the scope of the present study was limited to reoccurring characters in primetime television. Other media, including newspapers, magazines, movies, user-generated content such as YouTube, and television sports broadcast were not included in the current sample because they do not feature reoccurring characters (as defined in this study). It is important to note, however, that these media platforms can provide meaningful encounters with outgroup members and impact media consumers’ attitudes toward ethnic minorities. For example, ample research documents the overrepresentation of Blacks in local news coverage of crime

36 Tukachinsky, Mastro, and Yarchi

(e.g., Dixon & Linz, 2000) and demonstrates the harmful effects of exposure to such news on viewers’ racial attitudes (e.g., Hurley et al., 2015). Similarly, sports news coverage of crimes associated with Black athletes is more frequent and detailed compared with coverage of crimes perpetrated by White athletes (Mastro, Blecha, & Atwell Seate, 2011). In fact, differential coverage of White and ethnic minority athletes is evident in sports broadcasting more generally, with sports commentators commonly attributing White athletes’ failures to a lack of innate ability, whereas Black athletes’ performance is attributed to their natural physical capabilities (e.g., Billings, 2004). Thus, as much as primetime television constitutes an important component of Americans’ media diet, other types of media content are likely to contribute to the formation, maintenance and salience of racial and ethnic stereotypes. Despite these limitations, the current study frames the relationship between primetime TV representations of ethnic minorities and the majority group’s perception of these groups. This macro level relationship can then be examined more closely accounting for specific contexts and moderating factors.

References

Billings, A. (2004). Depicting the quarterback in Black and White: A content analysis of college and professional football broadcast commentary. The Howard Journal of Communications, 15, 201–210. doi: 10.1080/10646170490521158.

Bodenhausen, G., 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. doi: 10.1006/jesp.1995.1003.

Busselle, R., & Crandall, H. (2002). Television viewing and perceptions about race differences in socioeconomic success. Journal of Broadcasting & Electronic Media, 46, 256–282.

Busselle, R. W., & Shrum, L. J. (2003). Media exposure and exemplar accessibility. Media Psychology, 5(3), 255–282.

Children Now (2004). Fall colors, 2003–2004: Prime time diversity report. Oakland, CA: Children Now.

Davidson, C. (2011, June 3). TV is dead: Long live the Internet. The Guardian. Retrieved from: http://www.theguardian.com/media-tech-law/tv-internet-youtube-online-video.

Dixon, T. (2008). Crime news and racialized beliefs: Understanding the relationship between local news viewing and perceptions of African Americans and crime. Journal of Communication, 58, 106–125. doi: 10.1111/j.1460-2466.2007.00376.x.

Dixon, T., & Linz, D. (2000). Overrepresentation and underrepresentation of Blacks and Latinos as lawbreakers on television news. Journal of Communication, 50, 131–154. doi: 10.1111/j.1460- 2466.2000.tb02845.x.

Dong, Q., & Murrillo, A. (2007). The impact of television viewing on young adults’ stereotypes towards Hispanic Americans. Human Communication, 10, 33–44.

Hayes, A. F. (2006). A primer on multilevel modeling. Human Communication Research, 32, 385–410. doi: 10.1111/j.1468-2958.2006.00281.x.

Hurley, R. J., Jensen, J., Weaver, A., & Dixon, T. (2015). Viewer ethnicity matters: Black crime in TV News and its impact on decisions regarding public policy. Journal of Social Issues, 71(1), 155–170.

Jeong, S. H., Cho, H., & Hwang, Y. (2012). Media literacy interventions: A meta-analytic review. Journal of Communication, 62(3), 454–472.

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

TV Portrayals and National-Level Attitudes 37

Mastro, D. (2009). Effects of racial and ethnic stereotyping. In J. Bryant & M. B. Oliver (Eds.), Media effects: Advances in theory and research (3rd ed., pp. 325–341). Hillsdale, NJ: Lawrence Erlbaum Assoc.

Mastro, D., Behm-Morawitz, E., & Ortiz, M. (2007). The cultivation of social perceptions of Latinos: A mental models approach. Media Psychology, 9, 1–19. doi: 10.1080/15213260701286106.

Mastro, D. E., Blecha, E., & Atwell Seate, A. (2011). Characterizations of criminal athletes: A systematic examination of sports news depictions of race and crime. Journal of Broadcasting & Electronic Media, 55(4), 526–542. doi: 10.1080/08838151.2011.620664.

Mastro, D., & Greenberg, B. S. (2000). The portrayal of racial minorities on prime time television. Journal of Broadcasting and Electronic Media, 44, 690–703. doi: 10.1207/s15506878jobem4404_10.

Mastro, D., & Tukachinsky, R. H. (2011). Exemplar versus prototype-based processing of media content and the influence on racial/ethnic evaluations. Journal of Communication, 61, 916– 937. doi: 10.1111/j.1460-2466.2011.01587.x.

MediaWatch. (1994). “Front and Centre: Minority Representation on Television.” MediaWatch Re- search Series, Volume 1. Toronto: MediaWatch.

Morgan, M., & Shanahan, J. (1997). Cultivation research: An appraisal and meta-analysis. In B. R. Burleson (Ed.), Communication Yearbook, 20 (pp. 1–45). Thousand Oaks, CA: Sage.

Nielsen (2013). What a difference the day(part) makes. Nielsen Npower. Retrieved from: http://www.nielsen.com/us/en/newswire/2013/what-a-difference-the-day-part–makes.html.

Nisbet, E. C., & Myers, T. A. (2012). Cultivating tolerance of homosexuality. In M. Morgan, J. Shanahan, & N. Signorielli (Eds.), Living with television now: Advances in cultivation theory and research (pp. 61–81). New York: Peter Lang.

Oliver, M. B., & Armstrong, G. B. (1998). The color of crime: Perceptions of Caucasians’ and African Americans’ involvement in crime. In M. Fishman & G. Cavender (Eds.), Entertaining crime: Television reality programs (pp. 19–35). New York: Aldine de Gruyter.

Pettigrew, T. (1998). Intergroup contact theory. Annual Review of Psychology, 49, 65–85. Pew Research Center (2013). Online video 2013. Retrieved from: http://www.pewinternet.org/2007/

07/25/online-video/. Ramasubramanian, S. (2015). Using celebrity news stories to effectively reduce racial/ethnic prejudice.

Journal of Social Issues, 71(1), 123–138. Saeed, A. (2007). Media, racism and Islamophobia: The representation of Islam and Muslims in the

media. Sociology Compass, 1(2), 443–462. Scharrer, E., & Ramasubramanian, S. (2015). Intervening in the media’s influence on stereotypes

of race and ethnicity: The role of media literacy education. Journal of Social Issues, 71(1), 171–185.

Schmader, T., Block, K., & Lickel, B. (2015). Social identity threat in response to stereotypic film portrayals: Effects on self-conscious emotion and implicit ingroup attitudes. Journal of Social Issues, 71(1), 54–72.

Short, J. (2013). How much media? Report on American consumers. Institute for Communications Technology Management, Marshall School of Business, University of Southern California. Retrieved from: http://www.marshall.usc.edu/faculty/centers/ ctm/research/how-much-media.

The Second Authority for Television and Radio (2006). The present and the absent in prime time television: A follow up study. [Hebrew] Available from: http://www.rashut2.org.il/ editor/uploadfiles/%D7%93%D7%95%D7%97%20%D7%9E%D7%97%D7%A7%D7%A8 %20%D7%A1%D7%95%D7%A4%D7%99%20%D7%9B%D7%95%D7%9C%D7%9C% 20%D7%A9%D7%A2%D7%A8%20%D7%95%D7%93%D7%91%D7%A8%D7%99%20% D7%A4%D7%AA%D7%99%D7%97%D7%94.pdf.

Tukachinsky, R. (2015). Where we have been and where we can go from here: Looking to the future in research on media, race, and ethnicity. Journal of Social Issues, 71(1), 186–199.

U.S. Census Briefs (2011). Overview of Race and Hispanic Origin: 2010 Census Briefs. Retrieved from: http://www.census.gov/prod/cen2010/briefs/c2010br-02.pdf.

Valentino, N. A. (1999). Crime news and the priming of racial attitudes during evaluations of the president. Public Opinion Quarterly, 293–320. doi: 10.1086/297722.

38 Tukachinsky, Mastro, and Yarchi

Yalçin-Heckmann, L. (2002). Negotiating identities: Media representations of different generations of Turkish migrants in Germany. In D. Kandiyoti & A. Saktanber (Eds.), Fragments of Culture: Everyday of Modern Turkey, (pp. 308–322). New Brunswick: Rutgers University Press.

RIVA TUKACHINSKY (PhD, University of Arizona, 2012) is an Assistant Pro- fessor in the Department of Communication Studies and MS Program in Health and Strategic Communication at Chapman University. Her research investigates the psychological mechanisms underlying media effects in the context of social stereotypes, health, and psychological well-being.

DANA MASTRO (PhD, Michigan State University) is a Professor of Commu- nication at the University of California, Santa Barbara. Her research documents depictions of Latinos in English and Spanish-language media and assesses the extent to which exposure to this content influences racial/ethnic cognitions as well as a variety of intergroup and identity-based outcomes.

MORAN YARCHI (PhD, Hebrew University of Jerusalem, Israel, 2012) is a Lecturer at the Sammy Ofer School of Communications, Interdisciplinary Center (IDC) Herzliya, Israel. Her research interests involve the media’s coverage of conflicts and public diplomacy. Integrating theories from communication studies, political science, and international relations, her studies investigate the ability of political actors to promote their messages using media.

View publication statsView publication stats

https://www.researchgate.net/publication/273523463