quantitative research in mass communication

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=rcmm20

Download by: [University of Alabama] Date: 05 January 2017, At: 12:39

Communication Monographs

ISSN: 0363-7751 (Print) 1479-5787 (Online) Journal homepage: http://www.tandfonline.com/loi/rcmm20

Media's influence on immigration attitudes: An intergroup threat theory approach

Anita Atwell Seate & Dana Mastro

To cite this article: Anita Atwell Seate & Dana Mastro (2016) Media's influence on immigration attitudes: An intergroup threat theory approach, Communication Monographs, 83:2, 194-213, DOI: 10.1080/03637751.2015.1068433

To link to this article: http://dx.doi.org/10.1080/03637751.2015.1068433

Published online: 16 Sep 2015.

Submit your article to this journal

Article views: 1113

View related articles

View Crossmark data

Citing articles: 3 View citing articles

Media’s influence on immigration attitudes: An intergroup threat theory approach Anita Atwell Seatea and Dana Mastrob

aDepartment of Communication, University of Maryland, College Park, MD, USA; bDepartment of Communication, University of California, Santa Barbara, CA, USA

ABSTRACT The current study tests the influence of threatening-mediated intergroup contact on intergroup relations. A 2 (Immigration: Threat, No-threat) × 2 (Ingroup Group Exemplar: Present, Absent) + 1 (Control Condition) experimental design was used to examine both the direct and indirect influence of news exposure on attitudes toward immigration. The moderating role of previous local news consumption was also investigated. Results indicate that exposure to a threatening news story about immigration directly affected attitudes toward immigrants’ human rights, but not attitudes toward immigration policy. News exposure also indirectly influenced both types of immigration attitudes through feelings of intergroup anxiety, particularly for heavy news consumers. The discussion highlights the implications for both intergroup threat theory and cultivation research.

ARTICLE HISTORY Received 28 January 2014 Accepted 21 April 2015

KEYWORDS Cultivation theory; intergroup threat theory; intergroup emotions; immigration

The media landscape is replete with threatening themes and images (Hoffner & Cohen, 2013). From distressing messages about crime, to menacing coverage of terrorism, to a wide range of violent acts, threatening messages are commonplace in both news and enter- tainment programming. Notably, research suggests that members of racial/ethnic min- ority groups are associated with threatening content more so than majority group members (e.g., Whites in the US; Hoffner & Cohen, 2013; Mastro, 2009). Because the media can serve as a socializing force when it comes to learning about one’s own and other social groups (e.g., Vergeer, Lubbers, & Scheepers, 2000), the pairing of these mess- ages with racial/ethnic minority group members has potentially harmful implications for intergroup relations in society.

Empirical evidence demonstrates that exposure to threatening characterizations of racial/ethnic groups can influence perceptions among both outgroup (i.e., majority) and ingroup (i.e., minority) audiences (e.g., Busselle & Crandall, 2002; Fujioka, 2005). Although the work in this area provides important insights into a range of cognitive impli- cations of exposure, including stereotype formation and application, as well as identity- management (e.g., Dixon, 2006; Mastro, 2003; Mastro, Behm-Morawitz, & Ortiz, 2007), the role of affect in this process has received much less scrutiny, albeit with notable

© 2015 National Communication Association

CONTACT Anita Atwell Seate aseate@umd.edu.

COMMUNICATION MONOGRAPHS, 2016 VOL. 83, NO. 2, 194–213 http://dx.doi.org/10.1080/03637751.2015.1068433

exceptions (e.g., Ramasubramanian, 2010; Schemer, 2012). Scholars have long suggested that media exposure has the potential to “cultivate fears” of minority group members (Mastro & Robinson, 2000, p. 394), but intergroup communication scholarship has been slow to explicitly integrate this proposition into a broader understanding of media processes and effects. This previous research provides valuable insights into the impact of media on intergroup dynamics; however, the current study seeks to make three distinct contributions to this area of inquiry. First, building on survey-based findings that highlight the importance of affect in group-based outcomes (e.g., Ramasubramanian, 2010), the present study provides causal evidence for the role of media in producing intergroup emotions. Second, this research examines how particular message features, specifically the use of ingroup exemplars, are implicated in generating intergroup emotions. Third, with its focus on threat, this study helps to fine-tune our understanding of the role of affect in media-based, intergroup contexts. Specifically, this study examines whether one-time exposure to threatening images of minorities activates intergroup anxiety, med- iating the relationship between news exposure and intergroup attitudes. In addition, because previous media-based intergroup research finds that individual differences as well as message factors impact the extent to which audience members are influenced by the media (e.g., Dixon, 2006), we examined the moderating roles of both previous local news exposure and the presence of an ingroup member reinforcing the group-based threat in the media.

An intergroup threat theory-based cultivation approach

Contact between members of different groups has been touted as a salve for improving intergroup relations, under the appropriate conditions (Allport, 1954). Indeed, there is sub- stantial support for the assertion that positive intergroup contact can produce favorable attitudes toward the outgroup, in part by decreasing anxiety (Harwood & Joyce, 2012). Unfortunately, research in the domain of intergroup threat theory (ITT) suggests that con- temporary intergroup contact is typically associated with anxiety due to perceptions of out- groups as (a) physical, (b) economic, and/or (c) symbolic threats to the well-being of one’s own group (Stephan & Stephan, 2000). Rather than improving intergroup dynamics, such anxiety-ridden encounters have the potential to antagonize intergroup bias (Harwood, 2010). This association is further complicated when mediated intergroup interactions are considered. Research shows that media exposure is an important source of contact with various social groups (Ortiz & Harwood, 2007). Given this, if the media persistently characterize particular groups as physical, economic, or symbolic threats to other groups, exposure is likely to have negative implications for intergroup relations.

Media depictions of racial/ethnic threat. In terms of sheer numbers, content analyses consistently document that racial/ethnic minorities (with some exceptions) tend to be scarce in media fare that may contribute positively to intergroup relations, such as prime- time entertainment television (Mastro & Behm-Morawitz, 2005), and overemphasized in content with potentially negative intergroup implications, such as crime news reports (e.g., Dixon & Linz, 2000a, 2000b). Although there have been some strides in the quantity of racial/ethnic minorities in entertainment media, these characters are typically found in the latest and least watched hours of primetime (Children Now, 2004), begging the

COMMUNICATION MONOGRAPHS 195

question of how they positively contribute to intergroup relations if they are rarely seen by viewers.

When it comes to the manner in which racial/ethnic groups are portrayed, research indicates that mediated intergroup contact likely embodies the physical, economic, or symbolic threats described by ITT (Stephan & Stephan, 2000). In terms of physical threat, Blacks and Latinos have long been characterized in the media as a corporal danger to the White majority (e.g., Entman, 1990, 1992). The news media depicts these groups as violent aggressors targeting White victims (Dixon & Linz, 2000a, 2000b). Thus, the news informs us that minorities are a physical threat, specifically to majority group members. To illustrate, in their examination of local television news in Los Angeles, Dixon and Linz (2000b) found that although only 13% of real-world homicide victims in the region were White, 43% of the homicide victims featured in the news were White. Alternatively, 54% of real-world homicide victims in Los Angeles were Latino, but only 19% of homicides shown on the news depicted Latino victims. Although contemporary entertainment media portrayals have strayed from this stereotypical charac- terization to some degree, crime and violence typified depictions of minorities for decades (see Mastro, 2009).

From an ITT perspective, an economic threat refers to a danger to a group’s wealth and financial security but can also include perceiving other outgroup members as competition for jobs or as a burden to shared social systems, such as welfare (Stephan & Stephan, 2000). Like physical threats, racial/ethnic groups also are disproportionately linked with econ- omic threats in the media (Rodgers & Thorson, 2000). Gilens’s (1999) investigation into news portrayals of the poor demonstrates that the issue of poverty has become increasingly (and inaccurately) racialized over the decades, with the least sympathetic messages consistently linked with Blacks. As a consequence, his work reveals that message exposure is associated with negative attitudes about Blacks and welfare programs. Additionally, the consistent pairing of Latinos with stories about undocumented immigra- tion (Rodgers & Thorson, 2000) and the negative connotations of immigration on the economy (cf. Kim, Carvalho, Davis, & Mullins, 2011) also marks this group as a threat, as undocumented immigration has become associated with straining resources serving US citizens (e.g., emergency room care) and taking jobs away from native-born citizens in the United States (Pew Research Center, 2011).

Finally, research suggests that racial/ethnic minorities are portrayed as a symbolic threat to mainstream values and morals (Conway, Grabe, & Grieves, 2007). Symbolic threats refer to challenges to a social group’s system of meaning, such as their language, values, and religious customs (Stephan & Stephan, 2000). Such threats are commonplace in popular media. For example, Conway et al. (2007) found that Bill O’Reilly consistently frames his news segments about foreign governments and foreign citizens as a danger to US moral values and a violation of US social norms. This study also found that “illegal aliens” are predominantly discussed as threats to US value systems, including liberty, democracy, and safety. Although effects cannot be determined from such content analytic work, theory and empirical evidence suggest that these portrayals have the potential to influence majority group members’ perceptions of these groups.

An integrated approach. ITT argues that intergroup encounters are more likely to result in negative affect because these interactions (compared to intragroup interactions) typi- cally include the antecedents needed to produce threat perceptions, such as a history of

196 A. ATWELL SEATE AND D. MASTRO

competition over scarce resources (Stephan & Stephan, 2000). Indeed, research by Stephan et al. (2002) reveals that the more individuals recount having threatening interracial contact, the higher levels of intergroup anxiety they report. This suggests that increasing threatening intergroup contact is associated with greater levels of anxiety. It stands to reason that increased media exposure to threatening images of minority groups would also be associated with higher levels of negative intergroup affect. Cultivation research provides further evidence in support of this claim.

According to cultivation theory, the media (particularly television) presents a largely homogeneous and repetitive set of messages that serves as a dominant socialization force in society (Morgan, Shananhan, & Signorielli, 2009). Although no one is immune to the impact of television, heavy media consumers (vs. light) are more likely to internalize these themes, affecting attitudes and perceptions (Morgan et al., 2009) specific to audience member’s individual patterns of media use (Lee & Niederdeppe, 2010). Exposure exerts itself by influencing the accessibility and applicability of cognitions (Shrum, 2009; Tewks- bury & Scheufele, 2009). Taken together, consuming a media diet that is rife with threa- tening images of racial/ethnic minorities would be expected to lead majority group members to develop and maintain perceptions of minorities that are in line with the mediated version of reality.

Although not explicitly aimed at testing exposure to threat, existing empirical evidence supports these theoretical assertions. For example, employing a mental models approach to cultivation theory, Mastro et al. (2007) found that as television consumption increased, so too did perceptions of Latino criminality in society (i.e., a physical threat) and low work-ethic (i.e., symbolic threat). Similarly, increased exposure to TV news programming (which overrepresents Blacks as criminals) is associated with perceptions of Blacks as violent and unlawful (Dixon, 2008). Although the cultivation approach is not without cri- ticism (e.g., Potter, 1993), particularly with regard to causal order, there is compelling evi- dence that heavy consumption of threatening media depictions of minorities helps to create and reaffirm negative schemata that these groups are threatening. Indeed, it is fair to say that the link between exposure to unfavorable media messages about race/eth- nicity and negative attitudes and cognitions about minority groups has been well docu- mented (e.g., Dixon, 2008; Mastro et al., 2007; Ramasubramanian, 2010). In light of such evidence, it also seems reasonable that media exposure would cultivate emotional responses, coupling certain groups with specific, negative feelings including fear, distrust, and anxiety.

Cultivating emotional responses to intergroup threat. Of course, the idea that media exposure may prompt either short- or long-term emotional reactions is not new. Indeed, there is a growing body of research showing that media portrayals can arouse feelings of threat (Hoffner & Cohen, 2013). For example, cultivation’s “mean world syndrome” recog- nizes that enhanced perceptions of victimization, anxiety, and insecurity may emerge from long-term exposure to media violence (e.g., Morgan et al., 2009). Also notable in the current context is recent theory and research on the extended parallel processing model (EPPM; Witte, 1992), typically applied to health and persuasive communication contexts. A recent extension of the EPPM (So, 2013) suggests that attitude change (or lack thereof) is dictated by the level of fear and anxiety people experience when exposed to fear appeals, along with individual-level factors (e.g., perceptions of self-efficacy). It is these feel- ings of anxiety and fear (alongside individual difference factors) that produce attitude or

COMMUNICATION MONOGRAPHS 197

behavioral change. This argument is congruent with the premise of ITT. However, what has not been recognized in these existing theoretical perspectives is the inherently group-based nature of some threat responses (Mastro & Atwell Seate, 2012). That is to say, threat may arise from media content addressing unique/individualistic contexts (e.g., exposure to fear appeals that highlight an individual’s susceptibility to a health issue) or from messages that involve intergroup relations (e.g., exposure to a news story that highlights Black and Latino criminals targeting White victims).

Thus, we argue that although these two emotional responses may be physiologically comparable, their consequences are profoundly distinct. Accordingly, appreciating the unique dynamic posed by intergroup threat (as expressed in ITT) enhances our under- standing of the range of possible implications of exposure to several common thematic messages in our media landscape. Specifically, integrating ITT’s notions of physical, econ- omic, and symbolic threat into known cultivation-based processes and effects broadens our understanding of (a) what constitutes “thematic” messages and (b) the types of effects that can be expected based on exposure.

When the assumptions of ITT are coupled with insights from cultivation theory, it illu- minates the possibility that media exposure can promote unfavorable emotional, inter- group responses. This is not to say, however, that the role of affect in media processes and effects has been ignored. Rather, important insights have been offered in studies addressing the mediating influence of prejudicial feelings on media models of policy reasoning (see Ramasubramanian, 2010) and the self-reinforcing effect of emotions on attention to media messages (see Schemer, 2012).

Schemer’s (2012) work in this context is particularly notable as his research indicates that exposure to media messages (i.e., political ads) that induce negative affective responses toward minority groups (i.e., asylum seekers in Switzerland) not only produces and reinforces negative affect but also encourages attention to comparable messages in the future. The upshot is that this process becomes self-perpetuating in media contexts that provide constant access to such messages, including those highlighted in the present study. Although the notion of reinforcing spirals has not consistently met with support (i.e., Lee & Niederdeppe, 2010), the significance of Schemer’s findings in the current context cannot be overstated. Here, the implication is that the persistent minority-as- threat characterization found in US mass media offerings has the potential to become “a driving force in maintaining negative sentiments toward and the derogation of ethnic groups” (Schemer, 2012, p. 427). Results from Ramasubramanian (2010) cautiously support this conclusion. In particular, her work suggests that overarching prejudicial feel- ings (measured as an index of racial feelings including anger, discomfort, dislike, etc.), prompted from television exposure, serve to mediate the influence of television consump- tion on support for affirmative action (under certain conditions and distinct from cogni- tions). Although this work does not link specific patterns in media content with particular group-level emotional responses (or their associated attitudinal and behavioral correlates), it nonetheless bolsters arguments regarding the uniquely important role of affect in group- based cultivation effects.

These studies provide strong support for claims that media can influence intergroup attitudes through affective processes. It appears that the types of media cues most likely to promote harmful self-perpetuating spirals (e.g., Schemer, 2012) are precisely those found in an alarming array of interracial/ethnic messages offered across the media

198 A. ATWELL SEATE AND D. MASTRO

landscape. Further, repeated exposure is likely to prompt detrimental group-level emotions identified by ITT (e.g., anxiety), which antagonize intergroup relations and impede the promotion of supportive intergroup dynamics (e.g., Ramasubramanian, 2010). In other words, taken together, these studies provide strong evidence that the tenets of ITT can contribute to our understanding of the role of the media in cultivating and provoking undesirable affective intergroup responses.

Priming emotional responses to intergroup threat. Of course, a substantial volume of research also exists that examines the effects of even a single exposure to media messages intersecting race and crime (see Mastro, 2009). Although not explicitly aimed at testing affective responses to “minority-as-threat” stereotypes, this work lays the foundation for such an empirical examination. For example, Gilliam and Iyengar (2000) report that Whites are more likely to support punitive remedies for crime (e.g., three-strikes law) after exposure to a Black perpetrator of violent crime. This research suggests that message factors that are inherently group-based (i.e., the presence or absence of a group member) can signal threat-related responses, with the preponderance of this work indicating that individuals will react in ways that protect their group.

Hypotheses

Given the research findings examining the role of media in interracial/ethnic processes, the current study’s ITT-based cultivation approach would suggest that the media can affect intergroup attitudes and anxiety in ways that are deleterious to intergroup relations. Specifically, media that emphasize an outgroup’s threatening nature toward other groups are likely to produce negative intergroup attitudes and increase intergroup anxiety. In no media context is this more likely to be relevant than in coverage of immigration. Accord- ing to the Pew Research Center (2011), immigration in society is associated with physical threats (perceptions of increased crime), symbolic threats (challenging Americans’ customs and way of life), and economic threats (taking jobs away from US citizens). Research examining news framing of this topic indicates that these themes are reflected prominently in the reporting on this issue (Kim et al., 2011). As such, it is hypothesized that:

H1: Exposure to an immigration news story featuring intergroup threat leads to negative atti- tudes toward immigration, more so than to exposure to a news story addressing immigration without mention of intergroup threat, and a story that does not feature immigration.

H2: Exposure to an immigration threat leads to greater intergroup anxiety compared with exposure to a story featuring immigration without mention of intergroup threat, and a story that does not feature immigration.

ITT argues that perceiving intergroup contact as threatening works together with feel- ings of intergroup anxiety in producing negative intergroup attitudes (Stephan & Stephan, 2000). From this perspective, intergroup attitudes are determined by the coupling of per- ceptions of outgroups as a threat with the anxiety that arises from that perception. For example, research by Stephan et al. (2002) not only found that threatening contact between social groups impacted intergroup attitudes, but that it does so both directly and indirectly through intergroup anxiety. Similar relationships have been identified with regard to mediated contact. Research by Fujioka (2011) demonstrates that when

COMMUNICATION MONOGRAPHS 199

consumers evaluate the affective tone of immigration news coverage as negative, they are more likely to perceive the outgroup as threatening. Higher levels of perceived economic threat also were associated with more negative attitudes toward this group. Taken together, these studies provide evidence that threatening contact affects intergroup atti- tudes both directly and indirectly, as mediated by intergroup anxiety. Hence, the following prediction is offered:

H3: The relationship between exposure to a threatening news story about illegal immigration and immigration attitudes is mediated by intergroup anxiety.

Of course, the way that a message is presented/framed in the media influences compre- hension and interpretation of the information (Tewksbury & Scheufele, 2009). Although numerous culturally based symbols and message features can influence how receivers understand information, group membership has been found to play a particularly critical role in such interpretive processes (e.g., Esposo, Hornsey, & Spoor, 2013; Schwartz & Bless, 2007). Not surprisingly then, research indicates that issues are commonly framed in the news through the use of exemplars espousing the group’s perspective on the given issue (Zillmann & Brosius, 2000). Not only is information provided by such exemplars more easily recalled (Zillmann & Brosius, 2000), but these exemplars also serve as a visual frame, influencing how consumers interpret this message (e.g., Gibson & Zillmann, 2000), which likely includes both cognitive and affective components. Indeed, the use of such group exemplars has been found to activate cognitive schema that influence how audiences think about an issue (e.g., Abraham & Appiah, 2006). It is argued that exposure to group exemplars also affects how consumers feel about the message, particularly when group-level emotional reactions are highlighted (e.g., Entman, 1990, 1992). However, empirical work has not tested whether or not ingroup members endorsing the intergroup threat influences group-level emotions. This is surprising given that people assimilate norms expressed by ingroup members, but not norms expressed by outgroup members (e.g., Schwartz & Bless, 2007). In other words, the presence of an exemplar from the ingroup (in this case, White citizens as the victim of the immigration threat), reinforcing the feelings and/or emotional responses of their group should enhance the influence of the threat on intergroup judgments. Hence, the following hypothesis is proposed:

H4: Featuring an ingroup exemplar in a news segment strengthens the relationship between intergroup threat and (a) intergroup attitudes and (b) intergroup anxiety.

Because it is proposed that the ingroup exemplar will strengthen the relationship between exposure to intergroup threat, intergroup attitudes (H4a), and intergroup anxiety (H4b), it stands to reason that it should also strengthen the mediational relation- ship hypothesized in H3. Stated formally:

H5: The presence of the ingroup exemplar strengthens the mediational relationship pre- sented in H3.

Consistent with genre-specific cultivation theorizing, long-term exposure to news (vs. overall television use) has been found to exacerbate intergroup bias, particularly when made immediately accessible by one’s current media environment (e.g., Dixon, 2006). For example, Dixon (2006) found that heavy news consumers (vs. light) exposed to a crime news segment featuring a majority of Black suspects (or race-unidentified) perceived

200 A. ATWELL SEATE AND D. MASTRO

the world to be a more dangerous place compared to those viewing a majority of White suspects. Although affective responses were not measured in his work, the overwhelming evidence suggests that processes consistent with ITT are at play in such cultivation-based outcomes.

Content analytic work provides strong evidence that local television news commonly portrays non-dominant group members as threats to majority group members (e.g., Dixon & Linz, 2000a). In terms of immigration reporting, specifically, research indicates that whereas national news coverage of the topic surges and dwindles in-line with socio- political events (Pew Research Journalism Project, 2008), local news coverage (specifically in border-states) offers consistently higher coverage (Dunaway, Branton, & Abrajano, 2010). Accordingly, the following is predicted:

H6: The mediational relationships predicted in H3 are moderated by local television news consumption. More specifically, the path between exposure to immigration threat and inter- group anxiety is moderated by local television news consumption, such that the relationship between experimental condition and intergroup emotion is strongest when local television news consumption is high.

Method

Participants

Four hundred forty-four participants from communication courses at a large public uni- versity in a state on the United States–Mexico border took part in this study. Because this study examined intergroup perceptions regarding immigration on the United States– Mexico border, all non-US citizens and Latinos were excluded from the analyses (N = 70)1,2. Participants were compensated with extra course credit. Most participants self- identified as White (88%), whereas 4.8% identified as Asian/Pacific Islander, 4.8% as African-American/Black, and 2.4% as other. The average age of the participants was 20.14 years (SD = 1.24), and most were female (71.1%). Other data from this project are reported elsewhere (Atwell Seate & Mastro, 2015a, 2015b).

Procedure

Participants came to a computer lab in groups of 1–10 to complete the experiment and were randomly assigned to experimental conditions (see below). First, participants watched a news segment that contained a distracter segment (about the weather) and the experimental manipulation. After watching the news segment, participants completed a questionnaire ascertaining demographic information and the measured variables. After completing the questionnaire, the participants were debriefed about the purpose of the study.

Experimental manipulation

Participants were randomly assigned to a 2 (Immigration: Threat, No-threat) × 2 (Ingroup Exemplar: Present, Absent) + 1 (Control Condition) experimental design. The news segment used for the experimental manipulation was adapted in part from actual

COMMUNICATION MONOGRAPHS 201

immigration news reports (e.g., Lamm & Harrison, 2010; Solís, 2010). The news segment was produced by the local public media affiliate news studio. The actor who played the news anchor was the same in all conditions. In the threat condition, the story discussed the negative implications of illegal immigration including increases in crime, threats to US culture, and threats to the economy, which research suggests are common in this type of news (Kim et al., 2011). The no-threat condition discussed immigration in a non-threatening way. Specifically, this story discussed the positive cultural consequences of immigrants to the local community (see Appendix). The second factor, presence of target, ingroup exemplars (i.e., White, US citizens; one male and one female), was manipu- lated by either featuring, or not featuring, ingroup members reinforcing the emotional tone of the story (see Appendix). The control condition watched the first distracter segment and then a second story about the weather.

Pilot testing of experimental condition. To ensure that the news segments created for use in the current study represented the desired frames regarding immigration, three undergraduate coders (blind to the goals of the study) coded the news segments for perceptions of threat and presenceofexemplar,amongotherdistractorvariables.Toassessthreat,thecodersresponded to a semantic differential item, with “1” representing a threatening frame, “7” representing a non-threatening frame, and “4” representing neither threatening nor non-threatening. A threatening frame was defined as a frame that emphasized the negative, harmful, and unde- sirable consequences of immigration to US citizens, whereas a non-threatening frame was defined as a frame that emphasized the positive and beneficial consequences of immigration to US citizens. The coders were highly reliable (Krippendorff’s α = .98).

Next, the threatening and non-threatening frames were separately averaged across exemplar presentations. In other words, the threatening/exemplar-present condition was averaged with the threatening/exemplar-absent condition, so that a single threatening frame score was created for each coder. The same was done for the non-threatening news stories. The resulting means for each frame—threatening (M = 1.35), non-threatening (M = 6.84), and control (M = 4.00)—reflected the desired manipulation of threat for the news segments used in the current study. To test the appropriateness of the news conditions further, the ideal/desired scoring of each frame (threating = 1, control = 4, non-threaten- ing = 7) was used to predict the coder’s actual ratings. The regression results indicated that the coders perceived the frames as intended in the manipulation, β = 1.00, t = 30.25, p < .001. Finally, all coders correctly identified the presence (or absence) of the ingroup members endorsing the emotional norms in each of the news segments.

Perceptions of the news anchor and news segment

Study participants rated how professional they perceived both the news segment (M = 3.88, SD = 1.65) and the news anchor to be (M = 3.99, SD = 1.51) on a 7-point scale (1 = strongly disagree; 7 = strongly agree).

Covariate

Political orientation. Because of the political nature of immigration attitudes (e.g., Lee, McLeod, & Shah, 2008), we assessed political orientation (using two items) for use as a covariate. On a 7-point scale from conservative (1) to liberal (7), participants rated

202 A. ATWELL SEATE AND D. MASTRO

their political orientation on financial and social issues. We averaged the two items together to form a composite (r = .82, p < .01, M = 3.80, SD = 1.64).

Moderator

Local news consumption. To measure television news consumption, participants identified how often they watched the local news (M = 2.11, SD = 1.01). Responses ranged from “Never” to “Multiple Times a Day” on a 7-point scale.

Mediator

Intergroup anxiety. Participants reported how anxious, uneasy, and threatened they felt regarding undocumented immigrants on a 7-point scale. Scores on these items were aver- aged, with higher scores indicating more intergroup anxiety (α = .86, M = 3.07, SD = 1.48).

Dependent variable

Immigration attitudes. We asked participants a series of questions about their attitudes toward undocumented immigrants adapted from Ommundsen and Larsen (1997). A factor analysis (principal axis factoring) was used, and two separate factors emerged (based on .60/.40 loadings): a human rights factor and a policy factor. The human rights factor consisted of three items (α = .75, M = 3.92, SD = 1.57). An example item is “Undo- cumented immigrants have rights, too.” The policy factor consisted of three items (α = .77, M = 5.14, SD = 1.36). An example item is: “Undocumented immigrants are breaking the law.” Responses ranged from “Strongly Disagree” to “Strongly Agree” on a 7-point scale.

Results

Preliminary analyses

Perceptions of the news segment and news anchor. We calculated two separate one-way ANOVAs to ensure that individuals perceived the news segment and the anchor as equally professional across all five conditions. Neither the news segment, F (4, 369) = 1.43, p > .05, nor the anchor’s professionalism, F (4, 369) = 1.46, p > .05, differed signifi- cantly by experimental condition.

Hypothesis testing

Direct effects of experimental condition on immigration attitudes. To test the assertion that experimental condition influenced attitudes toward immigration and intergroup anxiety (H1, H2, and H4), we conducted a series of 2 × 2 ANCOVAs and OLS regressions. Because the control condition does not fit into the factorial design, the two-step approach is appropriate. We statistically controlled for political orientation in both the ANOVCAs and OLS regressions. First, the 2 × 2 ANCOVAs allowed mean comparisons to be made between the fully factored experimental conditions. When results from the ANCOVAs indicated that there were mean differences, 95% confidence intervals of the estimated

COMMUNICATION MONOGRAPHS 203

means were used to determine which means differed significantly. Because we used the estimated marginal means to determine statistical significance, we also report the standard errors of these estimates. Second, we used OLS regression analyses to compare the mean differences between the fully factored experimental conditions and the control condition. For these regression analyses, we entered the covariate (i.e., political orientation) along with the four dummy-coded variables (coded such that the control condition served as the reference category) as a block.

Hypothesis 1 predicted that exposure to an immigration threat in the news would nega- tively impact immigration attitudes. Hypothesis 4 predicted that this relationship would be strengthened by the presence of an ingroup exemplar reinforcing the threatening tone of the news segment. Supporting H1, those who were exposed to the immigration threat con- dition reported lower levels on the human rights measure (M = 3.60, SE = .12) compared to those in the no-threat condition (M = 4.17, SE = .12), F (1, 309) = 11.48, p < .01, partial η2 = .04. However, the interaction between the immigration threat and the presence of an ingroup exemplar was not significant, F (1, 309) = 1.89, p > .05, partial η2 = .01. To compare the fully factored experimental conditions with the control condition, an OLS regression analysis was used. The block of experimental conditions was significant, F(5, 373) = 13.05, p < .001. Although the block of variables was statistically significant, none of the experimental conditions were significantly different from the control. However, it should be noted that it appears that those in the immigration threat/ingroup exemplar condition reported lower levels on the human rights factor compared to the control (β = −.12, t = −1.92, p = .06). ANCOVA results indicate that neither exposure to the immi- gration threat, F(1, 309) = 2.06, p > .05, nor the interaction between the exposure to the immigration threat and the presence of the exemplar, F(1, 309) = .70, p > .05, predicted individual’s policy attitudes. Similarly, although the block of variables was statistically sig- nificant, none of the experimental conditions were significantly different from the control, F(5, 373) = 9.16, p < .05.

Hypothesis 2 predicted that exposure to an immigration threat in the news would impact intergroup anxiety. Hypothesis 4 predicted that this relationship should be strengthened by the presence of an ingroup exemplar endorsing the emotional tone of the intergroup threat. Results demonstrate that those in the immigration threat condition (M = 3.33, SE = .12) reported more anxiety towards undocumented immigrants compared to those in the no-threat condition (M = 2.86, SE = .12), F (1, 309) = 8.21, p < .01, partial η2

= .03. Moreover, the interaction between the threat factor and the presence of the exemplar was significant, F (1, 309) = 4.51, p < .05, partial η2 = .01. A comparison of the estimated marginal means reveals that those in both the immigration threat/exemplar-present con- dition and the immigration threat/exemplar-absent condition reported significantly more anxiety compared with those in the no immigration threat/ exemplar-present condition (see Table 1). As a block, the experimental conditions accounted for a significant pro- portion of the variability in anxiety, F(5, 373) = 4.38, p < .01, but none of the experimental conditions differed significantly from the control. Taken together, these results generally support for H2 and H4.

Conditional indirect effects. To test the conditional indirect effects (H3–H6) we used the PROCESS macro (Hayes, 2013). This macro is advantageous as it tests conditional indirect effects of the independent, mediator, and moderator variables in one model and uses boot- strapping techniques for the standard errors, which is the most appropriate way to

204 A. ATWELL SEATE AND D. MASTRO

compute these values. Again, we used political orientation as a covariate. For the current analyses, the 95% confidence intervals for all effects used 5000 bootstrapped samples. Specifically, model 4 was used to test the mediational relationship in H3 and model 9 was used to test the conditional indirect effects proposed in H4–H6.

Hypothesis 3 predicted that anxiety would mediate the relationship between exposure to an intergroup threat and attitudes toward undocumented immigrants’ human rights and immigration policy. Hypothesis 5 predicted that the mediated relationship would be moderated by previous news consumption. Supporting H3, results indicate an indirect effect of the threat factor on human rights attitudes through feelings of anxiety, b = −.14, CI 95% [−.27, −.04], such that those who were exposed to the immigration threat were less supportive of immigrants’ human rights. See Table 2 for both mediation models. There was a two-way interaction between the threat factor and local television news consump- tion in predicting intergroup anxiety, b = .35, t = 2.13, p < .05. To interpret this interaction effect, we conducted two separate regressions explicating the relationship between inter- group anxiety and local news consumption (controlling for political orientation) at each level of threat. Results demonstrate that for those in the immigration threat condition, local news consumption was associated with increased levels of anxiety, b = .25, t = 2.08, p < .05, whereas for those in the no-threat condition previous local news consumption was not related to intergroup anxiety, b = −.10, t = −.88, p > .05. Analyses additionally indicate that there were conditional indirect effects of the threat factor on the outcome (see Table 3). Specifically, there was an indirect effect of immigration threat on human rights attitudes only for those who were exposed to the exemplar endorsing the intergroup threat and consumed average or above average levels of television news. This occurred at both the mean level of local television news consumption, b = −.26, CI 95% [−.46, −.11], and at 1 SD above the mean, b = −.37, CI 95% [−.64, −.18].

With regard to attitudes toward policy, results revealed an indirect effect of the threat factor on policy rights attitudes through feelings of anxiety, b = .09, CI 95% [.02, .19]. Similar to that of human rights attitudes, there was a conditional indirect effect of the threat factor on policy attitudes (see Table 4). Again, exposure to the threat factor resulted

Table 2. Direct and indirect effects of experimental condition on attitudes toward immigration.

X–M M–Y Direct X–Y Indirect X–Y (mediation) X–M moderated by local news

X–M moderated by exemplar

Human rights −.58 −.31** −.41* −.14* .35* .67* Policy attitudes −.58 .19** .12 .09* .35* .67* Notes: *p < .05, **p < .01. Political orientation is a covariate in both models. The meditator is anxiety towards undocumen- ted immigrants. Undocumented human rights attitudes is coded such that higher scores indicate endorsement of undo- cumented immigrants’ human rights; whereas policy attitudes is coded such that higher scores indicate endorsement of more restrictive immigration policy.

Table 1. Means for intergroup anxiety toward undocumented immigrants as a function of experimental conditions. Variable Exemplar present Exemplar absent

Intergroup threat 3.39a 3.27a No intergroup threat 2.56b 3.15ab Note: Political orientation is a covariate. Means with different subscripts differ at p < .05 based on 95% confidence intervals. Scores could range from 1 to 7, with higher scores indicating more intergroup anxiety toward undocumented immigrants.

COMMUNICATION MONOGRAPHS 205

in more restrictive policy attitudes for individuals with higher levels of television con- sumption and who were exposed to the exemplar reinforcing the threatening emotions. This occurred at both the mean level of local television news consumption, b = .16, CI 95% [.06, .31], and at 1 SD above the mean, b = .23, CI 95% [.10, .42]. These results provide evidence in support of H3–H6.

Discussion

ITT argues that intergroup contact produces negative affect when outgroups are perceived to be physical, economic, or symbolic threats to the ingroup’s well-being (Stephan & Stephan, 2000). Alarmingly, content analytic evidence indicates that all three of these forms of threat are recurring themes across the media landscape, particularly in news cov- erage of ethnic and racial minorities (e.g., Mastro, 2009). Accordingly, by merging insights from ITT with tenets rooted in cultivation and exemplar-based models of media effects, the current work sheds light on unrecognized themes in media content while testing new or under-explored mediators and moderators of media effects. In doing so, this study enhances the scope of cultivation in three ways. First, it extends the conceptualiz- ation of thematic media messages to recognize “intergroup threat” as one such persistent message in media content. As Gerbner (1998, p. 181) asserts, “patterns common to most programming, such as casting, social typing, and the ‘fate’ of different social types can be expected to cultivate stable and widely-shared images of life and society.” For dominant

Table 3. Conditional indirect effect of exposure to immigration threat as a function of exemplar and local television news consumption on human rights attitudes.

Exemplar Local news consumption

Unstandardized coefficient

Standard error

Lower limit confidence interval

Upper limit confidence interval

Absent 1.09 (−1 SD) .06 .10 −.11 .27 Absent 2.10 (M) −.05 .07 −.21 .09 Absent 3.11 ( + 1 SD) −.16 .09 −.37 .001 Present 1.09 (−1 SD) −.15 .09 −.34 .03 Present 2.10 (M) −.26 .09 −.46 −.11 Present 3.11 ( + 1 SD) −.37 .11 −.64 −.18 Note: Political orientation is entered in the model as a covariate. Local television news consumption was a quantitative moderator. The above values are at 1 SD below the mean, at the mean, and 1 SD above the mean. Undocumented immi- grants’ human rights attitudes are coded such that higher scores indicate endorsement of undocumented immigrants human rights.

Table 4. Conditional indirect effect of exposure to threat as a function of exemplar and local television news consumption on policy attitudes.

Exemplar Local news consumption

Unstandardized coefficient

Standard error

Lower limit confidence interval

Upper limit confidence interval

Absent 1.09 (−1 SD) −.04 .06 −.16 .07 Absent 2.10 (M) .03 .05 −.05 .14 Absent 3.11 ( + 1 SD) .10 .06 −.003 .25 Present 1.09 (−1 SD) .09 .06 −.008 .24 Present 2.10 (M) .16 .06 .06 .31 Present 3.11 ( + 1 SD) .23 .08 .10 .42

Note: Political orientation is entered in the model as a covariate. Local television news consumption was a quantitative moderator. The above values are at 1 SD below the mean, at the mean, and 1 SD above the mean. Immigration policy attitudes are coded such that higher scores indicate endorsement of more restrictive immigration policy.

206 A. ATWELL SEATE AND D. MASTRO

group members (e.g., White Americans), the results found here imply that the character- ization of minorities as a threat likely constitutes one such coherent message.

Second, this study broadens the array of outcomes typically associated with cultivation effects to include affective intergroup consequences. Although fear of racial/ethnic groups is an implied outcome in a number of cultivation studies (particularly those addressing news coverage intersecting race/ethnicity and crime), overall the work in this domain has focused on individual-level responses such as punitive action and cognitive outcomes such as stereotyping (see Mastro & Tukachinsky, 2012 for review). As such, questions remain about the features of media messages relevant to emotional responses (both indi- vidual and group-based) and the intergroup implications when group-level emotions are aroused. Accordingly, our results offer much needed specificity, indicating that anxiety towards undocumented immigrants can be influenced by just a single news exposure to group-based threat and further exacerbated by overall local news consumption.

Given ITT findings demonstrating that stereotyping, discrimination, and outgroup bias are driven in part by the anxiety provoked by perceived intergroup threat (e.g., Riek, Mania, & Gaertner, 2006), expanding our understanding to include how media is impli- cated here is of tremendous consequence. Our data suggest that intergroup anxiety is an important mediator in this process, as exposure to mediated intergroup threat indirectly influenced both policy and human rights attitudes via intergroup anxiety. These findings build on theorizing in the domain of communication and emotions (e.g., Nabi, 2009), including work using the E-EPPM, which suggests that anxiety mediates the relationship between threat perceptions and attitude change (e.g., So, 2013). Given this, the current study’s results allude to the possibility that media use may have disturbing and previously overlooked consequences for intergroup relations in society ranging from avoidance and segregation to physical aggression and violence depending on perceptions of efficacy, sus- ceptibility, and severity.

Third, the present study provides an important test of the role that exemplars play in moderating the direct effects of intergroup emotions and the indirect effects of media exposure on intergroup attitudes. Despite the known ability of exemplars to aggravate (e.g., Abraham & Appiah, 2006) or alleviate (e.g., Mastro & Tukachinsky, 2011) intergroup tensions and influence audience member’s general perceptions and attitudes about the world around them (Zillmann & Brosius, 2000), few studies have incorporated exem- plar-based approaches into our understanding of the role of media in intergroup processes (cf. Ramasubramanian, 2011). Our work finds that exemplars did not directly moderate the effect of threat on intergroup attitudes, but it did so indirectly by moderating the med- iating influence of intergroup anxiety. This is particularly relevant to the integrative frame- work employed here, as it suggests that exemplars enhance the impact of the group-based emotional norms espoused in the news segment. Specifically, when exposed to an ingroup exemplar endorsing intergroup threat, those watching average and above average levels of local news consistently reported unfavorable attitudes toward immigration policy and human rights. Taken together, our findings indicate that the features of media messages (i.e., ingroup exemplars) can influence intergroup relations by altering the emotions people feel toward the outgroup.

Altogether, then, four general conclusions follow from this study’s results. First, exposure to threatening intergroup messages directly influences attitudes that dehumanize immigrants (but not restrictive immigration policy) and indirectly affects both types of

COMMUNICATION MONOGRAPHS 207

attitudes via intergroup anxiety. Second, viewing such messages influences feelings of anxiety toward outgroup members more so when an ingroup exemplar is featured. Third, the greater the (local) news consumption, the more pronounced the experience of intergroup anxiety following exposure to the group-based threat. Finally, long-term news consumption and presence of an exemplar endorsing the threat moderate the impact of exposure on policy positions (under certain conditions).

Accordingly, our results indicate that alongside influencing cognitions, television exposure may also have important intergroup affective consequences. Considering exist- ing evidence demonstrating that threatening themes are common in media characteriz- ations of non-dominant groups and research indicating that negative group-based affect enhances attention to subsequent, comparable media messages (Schemer, 2012), the implication of the current findings is that, over time, the pairing of racial/ethnic minority groups with messages that threaten majority/White audiences will have even more pro- nounced negative intergroup consequences than the results reported here (e.g., Ableson, 1985). Not only do audience members appear to acquire group-level emotions from media exposure but these emotions (in combination with audience and media features) shape attitudes about and positions affecting outgroups (e.g., Ramasubramanian, 2010).

Limitations

Results must be considered in light of the study’s limitations. One limitation of the current study was the inability to recruit more racial/ethnic minority participants. The concern here is twofold. First, although a diverse set of racial/ethnic groups took part in the study, the overwhelming majority of the sample was White. This ultimately limits the gen- eralizability of the study’s findings; though the results still speak to important aspects of intergroup dynamics. Second, although a sizable proportion of Latinos (n = 50) was recruited, this number was insufficient to treat ethnicity as a between-subjects factor in the statistical analyses. Additionally, preliminary analyses indicated that Latinos were sys- tematically different from non-Latinos (including other racial minorities), precluding their inclusion in the current design. Future research would be greatly benefitted by examining how Latinos perceive and respond to immigration news coverage.

Conclusions and Considerations

As Reik, Mania, and Gartner (2006, p. 345) note in their meta-analytic review of ITT, when “ … people perceive more intergroup competition, more value violations, higher levels of intergroup anxiety, more group esteem threats, and endorse more negative stereo- types, negative attitudes toward outgroups increase.” Media coverage of immigration is an important context to examine these relationships due to the threatening messages pre- sented in this content, specifically in terms of intergroup job competition, depletion of social services, and “traditional” value erosion. The findings from the current study suggest that when individuals experience intergroup anxiety stemming from the media’s depiction of immigration threats, they report attitudes that are congruent with ingroup protection mechanisms. Intergroup anxiety appears also to be associated with both per- ceiving immigrants as deserving of fewer human rights and supporting stricter immigra- tion policy. This is consistent with media-related intergroup research suggesting that

208 A. ATWELL SEATE AND D. MASTRO

media can promote social identity salience in ways that exacerbate intergroup differences and promote (social) identity protection (Mastro & Atwell Seate, 2012), which could undermine real-world contact with this group. Put differently, the media’s depiction of immigration can profoundly impact people’s perceptions of and support for immigration policy through chronically activating threat-based emotions. Such affective responses may additionally limit willingness to engage in real-world, intergroup communication. Future tests of such outcomes are certainly in order.

Finally, this work suggests that media can play a role in racial/ethnic societal dynamics beyond the range of individual and cognitive outcomes typically considered. It seems clear that the thematic presentation of racial/ethnic threat in the media (at least in the context of immigration), cultivates negative intergroup emotions, which can harmfully affect inter- group attitudes and actions. Unfortunately, decades of content analyses of mass media offerings demonstrate that the intersection of ethnicity and threat is not isolated to the topic of immigration (e.g., Dixon & Linz, 2000a, 2000b). Instead, racial/ethnic minorities are commonly linked with threats to dominant/majority group norms, values, safety, resources, and the like (see Mastro, 2009). Given this, the potential for negative affect prompted by media use to have even broader and more damaging consequences for real-world intergroup contact becomes even more pronounced. Of course, media use is only one of a number of factors contributing to intergroup processes and societal relations. Nonetheless, the measurable influence of media exposure should not be trivialized, par- ticularly in light of the amount of time spent with media in our daily lives.

Notes

1. The news segment used in the current study discussed immigration specifically from Mexico to the United States. Being an American and being Latino are not orthogonal categories and this news story could have activated either identity for Latino participants. However, research and theory provide evidence that race is one of the most salient categories for members of minority groups because of its frequent activation (Devine, 1989; Fujioka, 2005). Hence, it is not surpris- ing that Latinos in this sample were systematically different than other racial groups because their American identity was not likely most salient.

2. Fifty Latinos (see above) and 20 non-US citizens were removed prior to data analyses.

References

Abelson, R. P. (1985). A variance explanation paradox: When a little is a lot. Psychological Bulletin, 97, 129–133. doi: 10.1037/0033-2909.97.1.129

Abraham, L., & Appiah, O. (2006). Framing news stories: The role of visual imagery in priming racial stereotypes. The Howard Journal of Communications, 17, 183–203. doi: 10.1080/ 10646170600829584

Allport, G. W. (1954). The nature of prejudice. Reading, MA: Addison-Wesley. Atwell Seate, A., & Mastro, D. (2015a). The effect of media exposure on perceptions of group enti-

tativity: A preliminary examination. Communication Research Reports, 32, 29–34. doi: 10.1080/ 08824096.2014.989972

Atwell Seate, A., & Mastro, D. (2015b). Exposure to immigration in the news: The impact of group- level emotions on intergroup behavior. Communication Research. Advance online publication. doi: 10.1177/0093650215570654

COMMUNICATION MONOGRAPHS 209

Busselle, R., & Crandall, H. (2002). Television viewing and perceptions about race differences in socioeconomic success. Journal of Broadcasting & Electronic Media, 46, 265–282. doi: 10.1207/ s15506878jobem4602_6

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

Conway, M., Grabe, M. E., & Grieves, K. (2007). Villains, victims, and the virtuous in Bill O’Reilly’s “No-Spin Zone”: Revising world war propaganda techniques. Journalism Studies, 8, 197–223. doi: 10.1080/14616700601148820

Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5–18. doi: 10.1037/0022-3514.56.1.5

Dixon, T. L. (2006). Psychological reactions to crime news portrayals of black criminals: Understanding the moderating roles of prior news viewing and stereotype endorsement. Communication Monographs, 73, 162–187. doi: 10.1080/03637750600690643

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

Dixon, T. L., & Linz, D. (2000b). Race and the misrepresentation of victimization on local television news. Communication Research, 27, 547–573. doi: 10.1177/009365000027005001

Dunaway, J., Branton, R., & Abrajano, M. (2010). Agenda setting, public opinion, and the issue of immigration reform. Social Science Quarterly, 91, 359–378. doi: 10.1111/j.1540-6237.2010. 00697.x

Entman, R. M. (1990). Modern racism and the images of blacks in local television news. Critical Studies in Mass Communication, 7, 332–345. doi: 10.1080/15295039009360183

Entman, R. M. (1992). Blacks in the news: Television, modern racism and cultural change. Journalism Quarterly, 69, 341–361. doi: 10.1177/107769909206900209

Esposo, S., Hornsey, M., & Spoor, J. (2013). Shooting the messenger: Outsiders critical of your group are rejected regardless of argument quality. British Journal of Social Psychology, 52, 386–395. doi: 10.1111/bjso.12024

Fujioka, Y. (2005). Emotional TV viewing and minority audience: How Mexican Americans process and evaluate TV news about in-group members. Communication Research, 32, 566–593. doi: 10. 1177/0093650205279210

Fujioka, Y. (2011). Perceived threats and Latino immigrant attitudes: How White and African American college students respond to news coverage of Latino immigrants. Howard Journal of Communications, 22, 43–63. doi: 10.1080/10646175.2011.546743

Gerbner, G. (1998). Cultivation analysis: An overview. Mass Communication & Society, 1, 175–194. doi: 10.1080/15205436.1998.9677855

Gibson, R., & Zillmann, D. (2000). Reading between the photographs: The influence of incidental pictorial information on issue perception. Journalism & Mass Communication Quarterly, 77, 355–366. doi: 10.1177/107769900007700209

Gilens, M. (1999). Why Americans hate welfare: Race, media, and the politics of antipoverty policy. Chicago, IL: University of Chicago Press.

Gilliam, F., & Iyengar, S. (2000). Prime suspects: The influence of local television news on the viewing public. American Journal of Political Science, 44, 560–573. doi: 10.2307/2669264

Harwood, J. (2010). The contact space: A novel framework for intergroup contact research. Journal of Language and Social Psychology, 29, 147–177. doi: 10.1177/0261927X09359520

Harwood, J., & Joyce, N. (2012). Intergroup contact and communication. In H. Giles (Ed.), Handbook of intergroup communication (pp. 167–180). New York, NY: Routledge.

Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis. New York, NY: The Guilford Press.

Hoffner, C., & Cohen, E. L. (2013). Short-term and enduring consequences of fright. In E. Scharrer (Ed.), Media effects/media psychology (pp. 374–401). Boston, MA: Wiley-Blackwell.

210 A. ATWELL SEATE AND D. MASTRO

Kim, S., Carvalho, J. P., Davis, A. G., & Mullins, A. M. (2011). The view of the border: News framing of the definition, causes, and solutions to illegal immigration. Mass Communication and Society, 14, 292–314. doi: 10.1080/15205431003743679

Lamm, R. D., & Harrison, L. (2010, October 15). A bold plan to solve America’s illegal immigration problem: We can end the political stalemate if we summon the courage to end illegal immigra- tion, provide amnesty at a price, and be more selective about who we welcome into the country. The Christian Science Monitor. Retrieved from: http://www.csmonitor.com/Commentary/ Opinion/2010/1015/A-bold-plan-to-solve-America-s-illegal-immigration-problem

Lee, C., & Niederdeppe, J. (2010). Genre-specific cultivation effects: Lagged associations between overall TV viewing, local TV news viewing, and fatalistic beliefs about cancer prevention. Communication Research, 38, 731–753. doi: 10.1177/0093650210384990

Lee, N-. J., McLeod, D. M., & Shah, D. V. (2008). Framing policy debates: Issue dualism, journalistic frames, and opinions on controversial policy issues. Communication Research, 35, 695–718. doi: 10.1177/0093650208321792

Mastro, D. (2003). A social identity approach to understanding the impact of television messages. Communication Monographs, 70, 98–113. doi: 10.1080/0363775032000133764

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). New York, NY: Routledge.

Mastro, D., & Atwell Seate, A. (2012). Group membership in race-related media processes and effects. In H. Giles (Ed.), The handbook of intergroup communication (pp. 357–369). New York, NY: Routledge.

Mastro, D., & Behm-Morawitz, E. (2005). Latino representation on primetime television. Journalism & Mass Communication Quarterly, 82, 110–130. doi: 10.1177/107769900508200108

Mastro, D., Behm-Morawitz, E., & Ortiz, M. (2007). The cultivation of social perceptions of Latinos: A mental models approach. Media Psychology, 9, 347–365. doi: 10.1080/15213260701286106

Mastro, D., & Robinson, A. L. (2000). Cops and crooks: Images of minorities on primetime televi- sion. Journal of Criminal Justice, 28, 385–396. doi: 10.1016/A0047-2352(00)00053-2

Mastro, D., & Tukachinsky, R. (2011). The influence of exemplar versus prototype-based media primes on racial/ethnic evaluations. Journal of Communication, 61, 916–937. doi: 10.1111/j. 1460-2466.2011.01587.x

Mastro, D., & Tukachinsky, R. (2012). Cultivation of perceptions of marginalized communities. In J. Shanahan, M. Morgan, & N. Signorielli (Eds.), The cultivation differential: State of the art research in cultivation theory (pp. 38–60). New York, NY: Peter Lang Publishers.

Morgan, M., Shananhan, J., & Signorielli, N. (2009). Growing up with television: Cultivation pro- cesses. In J. Bryant & M. B. Oliver (Eds.), Media effects: Advances in theory and research (3rd ed., pp. 17–33). Hillsdale, NJ: Lawrence Erlbaum Associates.

Nabi, R. (2009). Emotion and media effects. In R. Nabi & M. B. Oliver (Eds.), The Sage handbook of media effects (pp. 205–221). Thousand Oaks, CA: Sage.

Ommundsen, R., & Larsen, K. (1997). Attitudes toward illegal aliens: The reliability and validity of a Likert-type scale. The Journal of Social Psychology, 137, 665–667. doi: 10.1080/ 00224549709595488

Ortiz, M., & Harwood, J. (2007). A social cognitive theory approach to the effects of mediated inter- group contact on intergroup attitudes. Journal of Broadcasting & Electronic Media, 51, 615–631. doi: 10.1080/08838150701626487

Pew Research Center (2011, February 24). Public favors tougher border controls and path to citi- zenship. Retrieved from http://pewresearch.org/pubs/1904/poll-illegal-immigration-border- security-path-to-citizenship-birthright-citizenship-arizona-law.

Pew Research Journalism Project (2008, September 25). Every now and again—A study on news coverage of immigration. Retrieved from http://www.journalism.org/2008/09/25/every-now- and-again-a-study-on-news-coverage-of-immigration/

Potter, W. J. (1993). Cultivation theory and research: A conceptual critique. Human Communication Research, 19, 564–601. doi: 10.1111/j.1468-2958.1993.tb00313.x

COMMUNICATION MONOGRAPHS 211

Ramasubramanian, S. (2010). Television viewing, racial attitudes, and policy preferences: Exploring the role of social identity and intergroup emotions in influencing support for affirmative action. Communication Monographs, 77, 102–120. doi: 10.1080/03637750903514300

Ramasubramanian, S. (2011). The impact of stereotypical versus counterstereotypical media exem- plars on racial attitudes, causal attributions, and support for affirmative action. Communication Research, 38, 497–516. doi: 10.1177/0093650210384854

Riek, B., Mania, E., & Gaertner, S. (2006). Intergroup threat and outgroup attitudes: A meta-analytic review. Personality and Social Psychology Review, 10, 336–353. doi: 10.1207/s15327957pspr1004_4

Rodgers, S., & Thorson, E. (2000). “Fixing” stereotypes in news photos: A synergistic approach with the Los Angeles Times. Visual Communication Quarterly, 7, 8–11. doi: 10.1080/ 15551390009363436

Schemer, C. (2012). Reinforcing spirals of negative affects and selective attention to advertising in a political campaign. Communication Research, 39, 413–434. doi: 10.1177/0093650211427141

Schwarz, N., & Bless, H. (2007). Mental construal processes: The inclusion/exclusion model. In D. Stapel & J. Suls (Eds.), Assimilation and contrast in social psychology (pp. 119–141). Philadelphia, PA: Psychology Press.

Shrum, L. J. (2009). Media consumption and perceptions of social reality: Effects and underlying processes. In J. Bryant & M. B. Oliver (Eds.), Media effects: Advances in theory and research (3rd ed., pp. 50–73). Hillsdale, NJ: Lawrence Erlbaum Associates.

So, J. (2013). A further extension of the extended parallel process model (E-EPPM): Implications of cognitive appraisal theory of emotion and dispositional coping style. Health Communication, 28, 72–83. doi: 10.1080/10410236.2012.708633

Solís, D. (2010, August 16). Hundreds rally against amnesty. The Dallas Morning News, p. B01. Stephan, W. G., Boniecki, K. A., Ybarra, O., Bettencourt, A., Ervin, K. S., Jackson, L. A. … & Renfro,

C. L. (2002). The role of threats in the racial attitudes of Blacks and Whites. Personality and Social Psychology Bulletin, 28, 1242–1254. doi: 10.1177/01461672022812009

Stephan, W. G., & Stephan, C. W. (2000). An integrated threat theory of prejudice. In S. Oskamp (Ed.), Reducing prejudice & discrimination (pp. 23–46). Hillsdale, NJ: Lawrence Erlbaum Associates.

Tewskbury, D., & Scheufele, D. A. (2009). News framing theory and research. In J. Bryant & M. B. Oliver (Eds.), Media effects: Advances in theory and research (3rd ed., pp. 17–33). Hillsdale, NJ: Lawrence Erlbaum Associates.

Vergeer, M., Lubbers, M., & Scheepers, P. (2000). Exposure to newspapers and attitudes toward ethnic minorities: A longitudinal analysis. The Howard Journal of Communication, 11, 127– 143. doi: 10.1080/106461700246661

Witte, K. (1992). Putting the fear back into fear appeals: The extended parallel process model. Communication Monographs, 59, 329–349. doi: 10.1080/03637759209376276

Zillmann, D., & Brosius, H. B. (2000). Exemplification in communication. Mahwah, NJ: Lawrence Erlbaum Associates.

Appendix 1. Description of the Experimental Manipulations

The news segment was produced by the local public media affiliate news studio. The actor who played the news anchor was the same in all conditions. All camera angles and b-roll footage were the same in both conditions. The script below removes identifying information regarding the location of the study.

Despite these record temperatures, issues surrounding immigration drew hundreds to an (pro/ anti) immigration rally at the [name of school] mall today. The rally drew an estimated 300 people to discuss what could be done about immigration issues in [geographic region]. Individuals attend- ing the rally were (supportive/against) the presence of undocumented workers in [location of the school], and all (agreed about/were against) mass deportations and new laws that make it harder to stay in the U.S and eventually gain citizenship. Feelings of those at the rally are supported by

212 A. ATWELL SEATE AND D. MASTRO

a new study published this week by researchers at the [school]. According to this study, [geographic region] has been (positively/negatively) impacted by the increased presence of undocumented workers.

Threat condition. [Location] has seen an increase in violent crime, including murder, since the immigration of undocumented workers reached its peak in 2000. The [location] workforce has also suffered from the cheap manual labor these individuals provide—jobs that native-born citizens just aren’t competitive for. [Location’s] once vibrant art community has been affected by Mexican art being brought over the border and being sold at much cheaper prices, which they can’t compete with.

Cut to rally member 1: “I don’t think people really understand how angry we all are about seeing the number of immigrants coming into [our community]. Our school systems and social services are already extremely overburdened and we just can’t take it anymore. We are getting angry.”

[School] students themselves are also impacted, with [school] reporting a 10% decrease in scho- larships because of the reduction in tax money [school] receives from the state. Some argue this reduction is based on the burden of undocumented workers to the state’s finances.

Cut to rally member 2: I was shocked at the burden these people put on the student body. I am really angry about how the students are being affected.

Non-threat condition. [Location] residence benefitted from the affordable manual labor these individuals provide—jobs that native-born citizens will no longer take. [Location’s] vibrant artist community has also flourished due to the rich culture that undocumented individuals bring with them from Mexico, with many local artists an increase in sales.

Cut to rally member 1: “I think that people assume we are all angry about immigrants coming into [our community], but we aren’t. On the contrary, most of us are proud and happy to share our community with those individuals who bring so much cultural diversity to our great city.”

[School] students themselves are also impacted, with [school] reporting a 10% increase in scho- larships because of research grants [school] receives due to the diversity these individuals bring to campus.

Cut to rally member 2: “I was shocked at the benefits these individuals give to the student body. I am really happy about how the students are being affected.”

COMMUNICATION MONOGRAPHS 213