a cognitive processing model of information source use and...
TRANSCRIPT
A Cognitive Processing Model ofInformation Source Use and
Stereotyping: African-AmericanStereotypes in South Korea
Alexis Tan, Francis Dalisay, Yunying Zhang, Eun-Jeong Han,and Mariyah M. Merchant
This study proposes a cognitive processing model to explain that stereotypes
are the result of a two-way interaction between receiver evaluation of the
valence of accessed information from the media, and receiver evaluation of
the realism and believability of the information. This model is applied to ana-
lyze American media use and stereotyping of African-Americans among 378
high school students in South Korea. Findings partially support the proposed
model: positive media portrayals perceived to be real and believable result
in positive stereotyping, while negative media portrayals perceived to be real
and believable result in negative stereotyping. Respondents perceived that
messages disseminated by the American media were somewhat realistic and
believable, and that American media portray African-Americans negatively.
Research suggests that the media can form and reinforce stereotypes of African-
Americans in the United States (e.g., Appiah, 2002; Dixon, 2006; Oliver, 1999;
Rada, 2000; Tan, Fujioka, & Tan, 2000). Supporting the U.S. findings, a recent study
Alexis Tan (Ph.D., University of Wisconsin-Madison) is a professor in the Edward R. Murrow Collegeof Communication and University Faculty Diversity Fellow at Washington State University. His researchinterests include communication and prejudice.
Francis Dalisay (Ph.D., Washington State University) is a term assistant professor in the School ofCommunication at Cleveland State University. His research interests include the socio-cultural impacts ofmass and interpersonal communication, particularly within intercultural/international contexts.
Yunying Zhang (Ph.D., Washington State University) is an assistant professor in the Department ofCommunication at Austin Peay State University. Her research interests include stereotypes, media effectsand campaign effects, and new technology.
Eun-Jeong Han (M.A., University of South Alabama) is a doctoral candidate in Communication atWashington State University. Her research interests include diversity in organizations, globalization andracial/ethnic minority issues, and critical cultural study.
Mariyah M. Merchant (M.B.A., M.S., and M.S., University of Houston Clearlake) is a doctoral candidateat Washington State University. Her Research interests include international student’s experience in aU.S. university environment, student research, best methods of data reporting and universities outcomeimprovement.
© 2010 Broadcast Education Association Journal of Broadcasting & Electronic Media 54(4), 2010, pp. 569–587DOI: 10.1080/08838151.2010.519810 ISSN: 0883-8151 print/1550-6878 online
569
570 Journal of Broadcasting & Electronic Media/December 2010
(Tan, Zhang, Zhang, & Dalisay, 2009) showed that Chinese high school students’ use
of American media leads to negative stereotyping of African-Americans. There are
no recent studies that investigate whether foreign audiences’ exposure to American
media influence their perceptions of African-Americans or any other American
minority racial group. This gap in the literature is worth noting, considering that
media audiences abroad increasingly are exposed to American media via new
communication technologies (Internet World Stats, 2008; Tan et al., 2009), and
that more American racial minorities, particularly African-Africans, are portrayed in
American news and entertainment (Dates & Stroman, 2001). With amplified glob-
alization of business, education, and diplomacy (Friedman, 2006; Gerdes, 2006),
and the increased participation of African-Americans and other American racial
minorities in U.S. government and diplomacy, foreign populations increasingly
interact with American racial minorities. It is important to understand how American
racial minorities are perceived in foreign countries and how those perceptions are
formed.
The present study explores information source use and stereotyping of African-
Americans among high school students in South Korea. A cognitive processing
model to explain how receiver evaluations of information might influence stereo-
typing is proposed. A preliminary test of a general prediction that stereotypes are
the result of receiver evaluations of the valence of portrayals and the believability
and realism of the portrayals, is provided.
The study examines high school students in a country that figured prominently in
U.S. foreign policy and diplomacy, and developed into one of the world’s largest
economies (World Bank, 2008). Compared with college students and other adults,
high school students are at an earlier stage of socialization (i.e., more likely to be
‘‘novices’’) and more susceptible to media influence (Austin & Freeman, 1997).
South Korean high school students also have limited direct contact with African-
Americans and can likely form impressions of African-Americans vicariously from
the media.
Stereotypes and Cognitive Processing
Most definitions ascribe cognitive and affective dimensions to stereotypes (Schnei-
der, 2004; Snyder & Miene, 1994). Cognitively, stereotypes are oversimplified and
over-generalized beliefs about other groups that are based on simplified information
(Stangor & Schaller, 1996; Tajfel, 1981). Stereotypes can be positive, negative,
or neutral, depending on the valence of information about the stereotyped group
(e.g., Ashmore & Del Boca, 1981; Schneider, 2004). Most theories of stereotype
formation explain stereotypes as resulting from the frequency and valence of in-
coming information (e.g., Stangor & Schaller, 1996). Examples are schema the-
ory (Snyder & Miene, 1994), group prototype theory (Fyock & Stangor, 1994),
and exemplar theory (Stangor & Schaller, 1996). According to these theories, and
considering limited human capacity to absorb and process incoming information
Tan et al./AFRICAN-AMERICAN STEREOTYPES IN SOUTH KOREA 571
(e.g., Lang, 2000; Lang & Friestad, 1993), stereotypes are formed after incoming
information is sorted and organized into simplified cognitive molds to describe
other groups. A common heuristic is to base a group characterization (stereotyping)
on the valence of information (e.g., Gilbert & Hixon, 1991). Exposure to positive
information (e.g., associating a group with traits valued in the receiver’s culture)
leads to positive stereotypes; exposure to negative information leads to negative
stereotypes; exposure to mixed information leads to mixed stereotypes. Since most
information environments include positive and negative information, some theories,
most notably from the communication fields, posit that the preponderance of images
(whether positive or negative) measured by frequency of ‘‘exposure’’ will determine
receiver perceptions of the real world, including stereotypes of other groups (e.g.,
Gerbner, Gross, Morgan, Signorielli, & Shanahan, 2002).
The study proposes that stereotyping is more accurately and fully explained
when the perceived valence, realism, and believability of incoming information
are accounted. Rather than main effects (e.g., positive information leads to positive
stereotypes; frequency of exposure leads to a stereotype), it is proposed that stereo-
types are the result of a two-way interaction between receiver evaluation of the
valence of accessed information (positive or negative with regard to the stereotyped
group) and receiver evaluation of the realism and believability of the information.
Specifically, it is predicted that information perceived to be positive will lead to
positive stereotypes when the information is also perceived as real and believable;
information perceived to be negative will lead to negative stereotypes when it is
perceived as real and believable. This analysis is applied to the specific case of
American media use and stereotyping of African-Americans among South Korean
high school students.
Stereotypes, Media and Information
Stereotypes of out-groups often are acquired through direct contact with out-
group members (e.g., Pettigrew, 1998). When opportunities for direct contact are
lacking, the media serve as important agents in the formation of stereotypes (e.g.,
Appiah, 2002; Dixon, 2006; Fujioka, 1999; Oliver, 1999; Rada, 2000). Cultivation
theory suggests that there is a direct correlation between the media’s portrayals of
social groups and audiences’ perceptions (Gerbner et al., 2002). Similarly, Bandura’s
(2002) social cognitive theory, and its concept of vicarious learning, explains that
stereotypes of racial groups can be learned through observing media portrayals. The
extent to which audience members are influenced by media content is determined in
part by the extent to which they perceive that such contents are real (e.g., Busselle,
2001; Dominick & Greenberg, 1972) and believable (e.g., Cappella, Lerman, Ro-
mantan, & Baruh, 2005; Gibbons, Lukowski, & Walker, 2005). Audiences are more
likely to learn from media portrayals that are perceived to be real and believable
because such portrayals increase audiences’ cognitive involvement (Gibbons et al.,
2005; Rubin & Perse, 1987) and self-efficacy (Austin & Freeman, 1997).
572 Journal of Broadcasting & Electronic Media/December 2010
Valence
Valence is a receiver’s evaluation of an observed event as ‘‘positive’’ or ‘‘nega-
tive.’’ Events depicting behaviors that are valued by the receiver are judged posi-
tively, while events not valued are judged negatively (Rokeach, 1968). With regard
to the relationship between intergroup contact and racial perceptions, Pettigrew
(1998) suggested that it is not merely the frequency of personal contact with out-
group members, but rather the perceived valence of such contact (i.e., whether
the contact experience is positive or negative) that predicts the formation of out-
group stereotypes. There is evidence that this is also the case for vicarious contact
experiences that are an outcome of media exposure and that the valence (positive or
negative) of TV portrayals, rather than mere frequency of exposure, leads to stereo-
typing (Fujioka, 1999; Schiappa, Gregg, & Hewes, 2005). In many foreign countries,
where personal contact with Americans is infrequent, the media may play a powerful
role in influencing stereotypes of Americans (Tan et al., 2009). Vicariously learning
stereotypes through media exposure is facilitated when portrayals of Americans—
particularly, American minorities—are clearly valenced (positive or negative traits
are depicted) and when such portrayals are considered by the receiver to be realistic
and believable.
African-American and Out-Group Stereotypes in Korea
In December 2008, the number of foreign residents in South Korea was just over
1 million, including more than 200,000 illegal residents (The Ministry of Justice,
2008). Although Korea remained a racially homogeneous country for a long time,
it experienced a massive influx of foreign labor workers since the 1980s (Han,
2007). This increased number of foreign residents prompted some Korean scholars
to investigate Koreans’ perceptions, attitudes, and stereotypes toward foreigners.
Cha and Choi (1992) surveyed college students in Seoul about their perceptions
of people from 11 countries. Their respondents reported positive and negative
stereotypes of White Americans, describing them as ‘‘liberal (open-minded),’’ ‘‘in-
dividualistic,’’ ‘‘selfish,’’ ‘‘practical,’’ and ‘‘arrogant.’’ Comparing their results with
those of a similar study conducted by Cha (1992), the authors concluded that Korean
college students’ attitudes toward White Americans were predominantly positive
from the 1960s to the 1980s. In contrast to positive perceptions of White Americans,
Cha and Choi (1992) found that their respondents used negative descriptions of
African-Americans, referring to them as ‘‘violent,’’ ‘‘lazy,’’ ‘‘sloppy,’’ ‘‘stupid,’’ and
‘‘having a sense of inferiority.’’
Jang (2001) surveyed 1,288 college students in South Korea about their perceptions
of social distance and national/ethnic preferences. Respondents rated Western Euro-
peans and White Americans as the most ‘‘positive’’ groups, and African-Americans
and Blacks living in Africa as the most ‘‘negative.’’ Female students reported more
favorable attitudes toward interracial marriage than male students, and considered
White Americans and Western Europeans as their most ‘‘ideal’’ marriage partners.
Tan et al./AFRICAN-AMERICAN STEREOTYPES IN SOUTH KOREA 573
Kim and Oh (1999) compared the perceptions of Americans held by both South
Korean college students and non-college students. They found that Korean college
students held more negative attitudes toward Americans, and speculated that these
attitudes were attributed to anti-Americanism, a dominant ideology among Korean
college students in the 1980s.
Hwang, Kim, Lee, Choi, and Lee (2007) conducted a national survey of 1,203
adults in South Korea to investigate perceptions and attitudes toward foreigners
and immigrants (Chinese, Japanese, South East Asians, Americans, South Asians,
Mongolians, Koreans from China, and North Korean refugees living in South Korea).
Hwang and colleagues found that younger and highly educated Koreans held posi-
tive perceptions of foreigners while older Koreans (over 60-years-old) held negative
perceptions. Americans were ranked as the most favorable foreigners, and Chinese,
South Asians, and Mongolians were ranked as the least favorable. Respondents
rated Americans most positively in almost all dimensions (affection, preference,
power, impact, social distance) and rated South Asians, Mongolians, and Chinese
least positively. Congruent with Kim and Oh’s (1999) findings, South Koreans were
reluctant to accept foreigners as Korean citizens, as family members, and as marriage
partners. However, South Koreans accepted foreigners as their friends, neighbors,
and co-workers. Hwang et al. (2007) like other researchers cited above, assume
that all Americans are White, without examining and making clear the distinction
between White and non-White Americans (e.g., African-Americans).
U.S. Media Portrayals of African-Americans
African-Americans continue to be negatively portrayed by the American media
(e.g., Dixon & Linz, 2000; Entman, 1994; Greenberg, Mastro, & Brand, 2002).
Although African-Americans had more presence in the U.S. media than any other
racial minority group in the past decade, they continue to be overrepresented as
criminals and criminal suspects (e.g., Dixon & Linz, 2000), and underrepresented as
professionals (Greenberg et al., 2002). Studies of American audiences have shown
a link between exposure to such portrayals and negative stereotyping of African-
Americans (e.g., Dixon, 2006; Oliver, 1999; Rada, 2000; Tan et al., 2000). Although
stereotypes of African-Americans in the United States improved through the years,
Madon et al. (2001) conclude that African-Americans are still perceived negatively
in comparison to White Americans.
American Images Abroad
There is considerable interest in American images abroad and speculation, with
limited direct testing, on possible influences on these images. The most recent Pew
(2009) worldwide surveys show a slight increase in positive valence of American
images in a number of countries after the election of President Obama, although the
images are still predominantly negative. DeFleur and Defleur (2003) found mixed
stereotypes of Americans in 12 countries which they attributed to mixed images of
574 Journal of Broadcasting & Electronic Media/December 2010
Americans in the media. Tan, Zhang, Zhang, and Dalisay (2007) found that stereo-
types of Americans held by Chinese high school students were influenced more by
print media and personal sources (e.g., teachers) than by American television and
movies.
U.S. Media in Korea
U.S. media played an influential role in the culture of South Korea. Even before
it opened its markets to international media in 1999, South Korea imported 26–
40 U.S.-made movies each year between 1971 and 1985 (Messaris & Woo, 1991).
U.S.-produced TV programs were available on Korean TV stations and American-
owned TV broadcasting stations such as the American Forces Korean Network,
which started its operations in 1957. The United States continues to be the largest
exporter of movies to South Korea. In 1994, American films had over a million
viewers, and total TV program imports increased from 8,074 units in 1994 to
12,921 units in 1995, mostly from the United States. (Ha & Yang, 2002). In 2003,
TV programs from the United States accounted for 77.8% of all imported foreign
programs in South Korea (Shim, 2006; Yim, 2002; Yoon, 1997).
Research is limited on the effects of U.S.-imported media on the perceptions of
America held by South Koreans. Messaris and Woo (1991) interviewed South Korean
immigrants in Philadelphia about their pre-immigration experiences with images of
the United States in movies, TV programs, and magazines. Respondents reported that
media images led them to immigrate to the United States. The American image they
remembered was of a dreamland, beautiful landscape, luxurious and extravagant
lifestyle, beautiful and fashionable women, quality education, justice, ethics, rich
families, and an ideal country for laborers. A primary motive for immigration was
to gain political freedom with fewer traditional and social constraints. The more
educated respondents endorsed social relationships with people in the United States,
and less educated respondents found the wealth and economics in the United States
attractive.
A Cognitive Processing Model
The basic premise of this model is that receiver evaluations of information are
a more powerful explanation of media influence on stereotypes than frequency of
exposure to information. This premise is based on the fundamental assumption that
human beings are rational and seek to maximize cognitive rewards by utilizing
available information (Bandura, 2002; Elasmar, 2003). The de-emphasis in the
model on frequency of exposure is based on previous research indicating that
even brief and ‘‘one-shot’’ exposures to information may alter stereotypical beliefs,
attitudes, and behaviors (e.g., Appiah, 2002; Dalisay & Tan, 2009; Oliver, 1999).
Personal contact research shows that even a single encounter with a member of
another group can lead to stereotyping (e.g., Pettigrew, 1998). What matters, at
Tan et al./AFRICAN-AMERICAN STEREOTYPES IN SOUTH KOREA 575
Figure 1
Stereotypes of African-Americans: Interaction Between Perceived Realism/
Believability and Valence
least in stereotype development, is not how much information one has but how
one evaluates the information.
The model focuses on information valence and realism/believability, dimensions
of information evaluation that increase cognitive involvement, identification, cred-
ibility and self-efficacy (e.g., Austin & Freeman, 1997; Gibbons et al., 2005; Rubin
& Perse, 1987). The study predicts that the influence of information on stereotypes
is an interaction between evaluation of the valence of the information (positive or
negative portrayals), and evaluation of the realism/believability of the portrayals.
Figure 1 shows the expected interaction effect.
Hypothesis
The following hypothesis regarding the potential influence of American media on
stereotypes of African-Americans in South Korea is tested.
H1: The valence of stereotypes of African-Americans among South Korean high
school students will depend on, a) the perceived valence of portrayals of
African-Americans in American media used as sources of information, and
b) the extent to which these portrayals are perceived to be realistic and
believable. This interaction hypothesis is illustrated in Figure 1.
Figure 1 reveals that perceived negative portrayals lead to negative stereotypes
when these portrayals are perceived to be real and believable. Perceived positive
portrayals will lead to positive stereotypes when these portrayals are perceived to
be real and believable.
576 Journal of Broadcasting & Electronic Media/December 2010
Method
The study’s purposive sample comprised of 378 high school students from three
high schools in South Korea, chosen to include men and women, and rural and
urban students. Students were recruited from one co-ed high school in Young-
dong/Chungbook province, a rural area (n D 149); one women’s high school (n D
119) and one men’s high school (n D 110), both in Incheon city, a metropolitan
area. Because t-tests found no differences across areas for the survey items, the data
from the three samples were merged for final analysis. Purposive sampling was the
only feasible method of obtaining data from high school students in Korea. Lists of
high school students are extremely difficult to obtain. When available, the release
of these lists requires approvals by numerous school and community authorities.
Students selected randomly from lists would be very difficult to reach.
Procedure
Questionnaires were developed in English and translated to Korean using back
translation (van de Vijver & Leung, 1997). Han, a Korean native and co-author of
this article, collected data for this study, traveled to South Korea between May–June
2008, visited three high schools, and distributed the self-administered questionnaires
to student respondents during their class time. Students were provided free ballpoint
pens and candies as incentives to participate in the survey. A consent form, approved
by a human subjects review board in the United States, informed the students that
participation was voluntary. Students also were told in a written consent form, that
the purpose of the study is ‘‘to ask you about your use of the mass media and about
your perceptions of Americans, American culture, Korea and Korean culture. The
information you give us will be used in classes in the U.S., and will be written
up for publication in research journals.’’ The consent form assured confidentiality.
In addition to the variables used in this study, other measures mentioned in the
instructions were included. Students submitted completed questionnaires to the
researcher, and return rates were close to 100%. The researcher collected the
questionnaires and brought them back to the United States. Data were coded,
entered into, and analyzed using, SPSS version 16.0.
Measures
Stereotypes of African-Americans.
A 7-point, 14-item semantic differential scale adapted from Smith (1991) and
Tan et al. (2000) was used to measure stereotypes of African-Americans. This scale
included 14 bipolar adjectives (e.g., Violent/Not Violent, Polite/Not Polite), in which
negative poles were scored ‘‘1,’’ and positive poles were scored ‘‘7’’. Exploratory
Tan et al./AFRICAN-AMERICAN STEREOTYPES IN SOUTH KOREA 577
factor analysis identified four factors. Factor 1 included 5 items, ‘‘hardworking,’’
‘‘honest,’’ ‘‘good morals,’’ ‘‘polite,’’ and ‘‘generous,’’ which explained 27.8% of
the total variance (M D 4.38, SD D .98, ˛ D .78). Factor 1 was called ‘‘Work
Ethic’’ since the adjectives are traits that would likely lead to success in the work-
place. Factor 2 included 3 items, ‘‘arrogant,’’ ‘‘violent,’’ and ‘‘aggressive,’’ which
explained 14.5% of the variance (M D 5.17, SD D 1.05, ˛ D .71). Factor 2 was
named ‘‘Aggression,’’ to describe a propensity to aggressive behavior. Factor 3
included 3 items, ‘‘not prejudiced,’’ ‘‘humorous,’’ and ‘‘hedonic,’’ explained 11.4%
of the variance (M D 3.12, SD D 1.05, ˛ D .64). Factor 3 was called ‘‘Likability’’
to describe the pleasant side to a person, or traits that would lead one to like
him/her. Factor 4, labeled ‘‘Attractiveness’’ included 3 items, ‘‘intelligent,’’ ‘‘beau-
tiful,’’ and ‘‘sloppy,’’ which explained 7.3% of the variance (M D 3.29, SD D .93,
˛ D .60). The adjectives ‘‘intelligent,’’ ‘‘beautiful,’’ and ‘‘sloppy,’’ could lead to
attraction.
Respondents’ Perceptions of Valence of Stereotypes.
Respondents were asked to rate how positive or negative each stereotype trait
was to them on a 7-point Likert scale (1 D very negative, 7 D very positive). These
ratings came at the end of the questionnaire and were separated from the semantic
differential scale measuring stereotypes of African-Americans (in the first part of the
questionnaire) by the media use and media evaluation measures. Stereotype traits
that were rated by all respondents (mean score) to be at least 4.0 were defined
as ‘‘positive.’’ Thus, ‘‘hedonic’’ was rated as positive, and ‘‘arrogant’’ was rated
negative.
Media Sources.
Respondents were asked to indicate how often they used specific media sources
to ‘‘know about Americans living in the United States.’’ The information sources
included American television, movies, newspapers, and magazines. Responses were
measured on a 4-point Likert scale (1 D not very often, 4 D very often). The 4 items
measuring media use were combined to form a single index (M D 1.94, SD D .57,
˛ D .68).
Valence of African-American Portrayals in the Media.
Respondents were asked to indicate how they felt American media portrayed
African-Americans along bi-polar adjectives (bad/good, not likable/likable, not at-
tractive/attractive). Responses were measured along a 7-point semantic differential
scale, where a higher score indicated positive portrayals. The information sources
included American TV, movies, newspapers, and magazines. The 12 items (3 items
for each of the 4 media sources) were combined to form an index of media valence
(M D 3.37, SD D 1.26, ˛ D .95).
578 Journal of Broadcasting & Electronic Media/December 2010
Realism and Believability.
Respondents were asked to rate the realism (‘‘how true to life?’’) and believability
(‘‘whether the information was believable’’) of the information about Americans
they received from specific media sources. Responses were measured along a 7-
point Likert scale, where a higher score indicated that the sources were perceived
to be more real and believable. The information sources included American TV,
movies, newspapers, and magazines. Four items measured realism of the sources
and 4 items measured believability. The correlation between the believability and
realism measures was .85. To prevent multicollinearity in the regression analyses
(Cohen, Cohen, West, & Aiken, 2002), these two constructs were combined to form
a single, 8-item index of realism/believability of American media (M D 3.79, SD D
1.37, ˛ D .90).
Interest.
Respondents were asked how interested they were in knowing about African-
Americans. Responses were measured on a 4-point Likert scale (1 D not at all
interested, 4 D very interested). The mean score of interest was 2.12 (SD D .88).
Demographics.
Respondents were asked to provide demographic information, which included
their age, gender, family household income, and whether they had ever traveled to
the United States.
Results
Stereotypes of African-Americans
South Korean high school students’ stereotypes of African-Americans are shown
in Table 1. On a 7-point scale (1 D most negative, 7 D most positive), half of the
items had means greater than 4, while the other half had means less than 4. This
indicates that their stereotypes were mixed, with half being positive (means greater
than 4) and half negative (means less than 4). Among the positive stereotypes,
African-Americans were regarded as humorous (M D 5.42, SD D 1.25), hedonic
(M D 5.35, SD D 1.26), hardworking (M D 4.88, SD D 1.44), not prejudiced (M D
4.79, SD D 1.58), generous (M D 4.44, SD D 1.27), honest (M D 4.30, SD D
1.29), and polite (M D 4.23, SD D 1.35). Among the negative stereotypes, African-
Americans were regarded as violent (M D 2.65, SD D 1.28), aggressive (M D 2.80,
SD D 1.28), sloppy (M D 3.09, SD D 1.21), not beautiful (M D 3.30, SD D 1.32),
not intelligent (M D 3.49, SD D 1.24), arrogant (M D 3.90, SD D 1.42), and having
poor morals (M D 3.95, SD D 1.31).
Tan et al./AFRICAN-AMERICAN STEREOTYPES IN SOUTH KOREA 579
Table 1
African-American Stereotypes by South Korean High School Students
African-American Mean Std. Deviation
Humorous 5.42 1.25
Hedonic 5.35 1.26
Hardworking 4.88 1.44
Not prejudiced 4.79 1.58
Generous 4.44 1.27
Honest 4.30 1.29
Polite 4.23 1.35
Good morals 3.95 1.31
Arrogant 3.90 1.42
Not intelligent 3.49 1.24
Not beautiful 3.30 1.32
Sloppy 3.09 1.21
Aggressive 2.80 1.28
Violent 2.65 1.28
In addition to the semantic differential scales, respondents were asked to list the
five adjectives that they thought were most descriptive of African-Americans. They
could add adjectives that were not included in the semantic differential scales. The
adjective most often mentioned was ‘‘athletic,’’ followed in descending order by
‘‘passionate,’’ ‘‘open-minded,’’ ‘‘violent,’’ and ‘‘aggressive.’’
Media Use by South Korean Students
Respondents averaged 1.88 in American media use on a 4-point scale (1 D almost
never, 4 D very often) (see Table 2), indicating that they didn’t use American media
Table 2
Media Use by South Korean High School Students
American Media Mean Std. Deviation
American movies 2.88 .84
American TV 1.98 .96
American books 1.70 .78
American magazines 1.50 .78
American newspapers 1.35 .60
580 Journal of Broadcasting & Electronic Media/December 2010
often to learn about Americans. American movies were the most often used (M D
2.88, SD D .84), followed in descending order by TV (M D 1.98, SD D .96), books
(M D 1.70, SD D .78), magazines (M D 1.50, SD D .78) and newspapers (M D
1.35, SD D .60).
Realism, Believability and Valence of American Media
Respondents evaluated American media as somewhat realistic (M D 3.89 SD D
1.38) and somewhat believable (M D 3.83, SD D 1.31) (see Table 3) (both means
were slightly below the midpoint of the 7-point scale). They believed that African-
Americans were portrayed somewhat negatively in American media (mean of 3.39
on 7-point scale).
Tests of Hypothesis
To test the hypothesis, each of the four African-American stereotype factors were
hierarchically regressed on the following variables: demographics (block 1); inter-
est in Americans and American culture (block 2); main effect variables (block 3:
media use, believability, valence of media portrayals); two-way interactions (block
4: media use by valence; media use by perceived realism/believability; valence
by perceived realism/believability); three-way interaction (block 5: media use by
valence by perceived realism/believability).
Factors 1 (work ethic) and 4 (attractiveness) were significantly predicted by the
regression models. Factor 1 (R-square D .13, p < .001) was significantly predicted
by realism and believability of American media (ˇ D �.12, p < .05), and the
interaction of perceived valence and the perceived realism/believability of American
media (ˇ D .46, p < .05). The significant interaction supersedes the significant main
effect for realism/believability, providing support for the hypothesis. The significant
interaction is shown in Figure 2.
Table 3
Realism, Believability and Valence of American Media
Realism and Believability of American Media Mean
Std.
Deviation
Realism of American media 3.89 1.38
Believability of American media 3.83 1.31
Realism and believability combined 3.85 1.30
Valence of African-Americans in Am. print media 3.31 1.30
Valence of African-Americans in Am. visual media 3.34 1.37
Valence of African-Americans in Am. media in general 3.38 1.24
Tan et al./AFRICAN-AMERICAN STEREOTYPES IN SOUTH KOREA 581
Figure 2
Stereotypes of African-Americans: Factor 1 (Work Ethic)
As Figure 2 shows, perceived positive media portrayals led to positive stereotypes
(hard-working, honest, good morals, polite, generous) when these portrayals were
perceived to be real/believable, and perceived negative portrayals led to negative
stereotypes (lazy, dishonest, poor morals, rude, selfish) when these portrayals were
perceived to be real/believable. The results for Factor 4 (attractiveness) are similar
to the results for Factor 1. Factor 4 (R-square D .12, p < .001) was significantly
predicted by interest in American people and culture (ˇ D .16, p < .01), use of
American media (ˇ D .47, p < .05), and the interaction between American media
valence and realism/believability (ˇ D .46, p D .05). These results show that interest
in Americans and frequent use of American media led to positive stereotyping
of African-Americans. The significant interaction between valence and realism/
believability supports the hypothesis. As Figure 3 shows, perceived positive media
portrayals of African-Americans led to positive stereotyping of African-Americans
as beautiful, neat and intelligent when these media portrayals were perceived to be
real/believable; perceived negative portrayals led to negative stereotyping of African-
Americans as not beautiful, sloppy and not intelligent when media portrayals were
believable. Real/believable and positive portrayals led to significantly more positive
stereotypes (estimated mean D .20, p < .01) compared with real/believable and
negative portrayals (estimated mean D �.3). The difference in stereotypes between
positive and negative media portrayals was not significant (p D .12) when these
portrayals were not perceived to be real/believable (see Figure 3).
Discussion
The results provide some support for a cognitive processing model of media
effects on stereotyping. This model suggests that stereotyping may result from the
582 Journal of Broadcasting & Electronic Media/December 2010
Figure 3
Stereotypes of African-Americans: Factor 4 (Attractiveness)
evaluation of information by receivers. This study focused on valence of media
portrayals and perceived realism/believability of the media. Consistent with this
model, it was found that stereotypes were the result of information evaluation rather
than frequency of media use. The valence of stereotypes depended on an interaction
between perceived valence and perceived realism/believability of media portrayals.
Positive media portrayals that were perceived to be real and believable resulted
in some positive stereotyping; negative media portrayals that were perceived to be
real and believable resulted in some negative stereotyping. Positive and negative
portrayals had no effect on stereotyping when they were not perceived to be real
and believable.
As such, the major finding is that an interaction between perceived information
valence and realism/believability influenced some stereotypes. Not all stereotypes
were affected by the predicted interaction.
Descriptions of African-Americans in Factor 1 (work ethic: hardworking, honest,
good morals, polite, generous) and Factor 4 (attractiveness: intelligent, beautiful,
sloppy) were significantly predicted by the interaction between valence and realism/
believability. These descriptions were also generally agreed upon by respondents
(SDs, respectively, D .98 and .93). On the other hand, there was less agreement
among respondents in their description of African-Americans in Factor 2 (aggression:
arrogant, violent, aggressive; SD D 1.05) and Factor 3 (likability: not prejudiced,
humorous, hedonic; SD D 1.05), which were not predicted by the interaction. Thus,
one explanation for the differential effects of the interaction is that the influence
of information valence and realism/believability on stereotypes was moderated by
the degree of agreement among receivers on the stereotypes: the more agreement
with the stereotype, the more likely an interaction effect. Based on this logic, it
Tan et al./AFRICAN-AMERICAN STEREOTYPES IN SOUTH KOREA 583
may be speculated that the more likely the respondents endorsed stereotypical
traits of African-Americans, the more likely the valence and realism/believability
of African-American media portrayals influenced their stereotypes. This speculation
draws upon the selective exposure hypothesis (Hart et al., 2009), which suggests
that audiences, in an attempt to reduce cognitive dissonance, are more likely to
attune to, and be affected by, media messages that are congruent with their existing
predispositions. If greater levels of stereotype endorsement could in fact influence
the interaction of information valence and realism/believability on stereotyping, this
warrants further investigation.
Additionally, the results suggest that the predicted interaction effect between
media valence and media realism/believability on stereotypes is moderated by
the valence of stereotypes. Although regression analyses show that the interactions
were significant and in the predicted directions for stereotype factors 1 and 4, real/
believable and positive portrayals did not lead to significantly more positive stereo-
types in Factor 1 compared with real/believable negative portrayals. For Factor 4,
real/believable and positive portrayals led to significantly more positive stereotypes
compared with real/believable negative portrayals. Factor 1 included the following
traits: hardworking, honest, good morals, polite, generous. As Table 1 shows, re-
spondents rated African-Americans positively on these traits with a mean of 4.38
on a 7-point scale (7 D most positive rating). Factor 4 included the following traits:
beautiful, neat and intelligent. Respondents rated African-Americans negatively on
these traits with a mean of 3.29. Thus, a ceiling effect may have been observed
for Factor 1, suggesting that media valence and media realism/believability may
not influence positive stereotype traits as much as negative stereotype traits. These
results suggest that negative stereotypes are more susceptible to information and me-
dia influence. This supports previous research that has shown negative information
can be more influential than positive information (e.g., Fiske, 1980), and the media
can form and reinforce negative stereotypes of African-Americans (e.g., Appiah,
2002; Dixon, 2006; Oliver, 1999; Rada, 2000). Taking this into consideration, and
because prior studies revealed that negative stereotypes of African-Americans may
influence opinions on race-coded issues (Mastro & Kopacz, 2007; Tan et al., 2000)
and prime racial attitudes during evaluations of a U.S. president (Valentino, 1999),
future studies might consider investigating whether foreign residents’ stereotypes,
and their exposure to negative portrayals of African-Americans could influence
their opinions of the foreign policies of President Barrack Obama (the first African-
American president of the United States) and his administration.
In summary, the findings suggest that the relationships between stereotyping and
media portrayals are complex, cannot be adequately explained by main effects, and
depend on the valence of traits used to describe the stereotyped group and receiver
agreement on the stereotypes. This study revealed evidence that an interaction effect
between perceived valence and perceived realism/believability of portrayals could
influence Korean high school students’ stereotyping of African-Americans. As such,
it is recommended that future research on the media’s influence on stereotyping
consider examining the direct and indirect effects of these two variables on for-
584 Journal of Broadcasting & Electronic Media/December 2010
eigners’ stereotyping of other American racial minority groups, notably, Latinos and
Asian-Americas.
Limitations
Some limitations should be noted when interpreting the results of this study. First,
the preliminary test of the model does not adequately capture the wide range of in-
formation sources that the respondents might use to learn about African-Americans.
Personal contact and use of the Internet, for example, were not measured, and the
study’s measures of traditional media use did not include use of specific media
genres (e.g., news and entertainment). These limitations underscore the importance
of examining the effects of non-media sources of information, such as parents,
teachers, peers, and personal contact with Americans in the United States, and the
effects of Internet sources such as blogs and social media Web sites (e.g., Facebook).
It is important to examine whether evaluations of the perceived valence and realism/
believability of fictional versus non-fictional portrayals could have different effects
on stereotyping.
Second, the sample was purposive, rather than random. To the extent possible,
a more heterogeneous group of respondents, including students from more diverse
high schools and individuals from the general population, is suggested for surveys.
Third, post-hoc explanations were used for differential support of the hypothesis,
rather than data specifically generated to test for alternative explanations. It is recom-
mended that researchers develop a set of competing hypothetical path models that
integrate the effects of perceived valence and realism/believability on stereotyping,
and test the fit of these competing models with more robust analytical techniques
such as structural equation modeling (e.g., Mastro & Kopacz, 2007).
Fourth, the respondents may have been sensitized by the stereotype measures. It is
possible that the study may be underestimating the degree to which they stereotyped
African-Americans negatively. Future studies on the media’s effects on stereotypes
ought to account for the influence of social desirability (see Crowne & Marlowe,
1960), and use alternative and multiple measures of stereotyping, to include less
intrusive measures (e.g., Steele & Aronson, 1995).
Finally, regression analysis of survey results does not establish causal direction.
Therefore, this study’s conclusion that media use leads to stereotyping should be
tested in more rigorous designs, such as in experiments.
References
Appiah, O. (2002). Black and White viewers’ perception and recall of occupational characterson television. Journal of Communication, 52(4), 776–793.
Ashmore, R. D., & Del Boca, F. K. (1981). Conceptual approaches to stereotypes and stereo-typing. In D. L. Hamilton (Ed.), Cognitive processes in stereotyping and intergroup behavior(pp. 1–35). Hillsdale, NJ: Lawrence Erlbaum Associates.
Tan et al./AFRICAN-AMERICAN STEREOTYPES IN SOUTH KOREA 585
Austin, E. W., & Freeman, C. (1997). Effects of media, parents and peers on African-Americanadolescents’ efficacy toward the media and the future. Howard Journal of Communications,8(3), 275–290.
Bandura, A. (2002). Social cognitive theory of mass communications. In J. Bryant & D. Zillman(Eds.), Media effects: Advances in theory and research (2nd ed., pp. 121–153). Hillsdale,NJ: Lawrence Erlbaum Associates.
Busselle, R. W. (2001). Television exposure, perceived realism, and exemplar accessibility inthe social judgment process. Media Psychology, 3(1), 43–67.
Cappella, J. N., Lerman, C., Romantan, A., & Baruh, L. (2005). News about genetics andsmoking: Priming, family smoking history, and news story believability on inferences ofgenetic susceptibility to tobacco addiction. Communication Research, 32(4), 478–502.
Cha, J. (1992). Korean adults and college students’ attitudes toward foreigners. Report ofresearch granted by the Ministry of Education of Korea in 1986.
Cha, J., & Choi, I. (1992). College students’ attitude toward foreigners. Psychological Science,1(1), 1–23.
Cohen, P., Cohen, J., West, S. G., & Aiken, L. S. (2002). Applied multiple regression/correlationfor the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum Associates.
Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent ofpsychopathology. Journal of Consulting Psychology, 24(4), 349–354.
Dalisay, F., & Tan, A. (2009). Assimilation and contrast effects in the priming of AsianAmerican and African American stereotypes through TV exposure. Journalism and MassCommunication Quarterly, 86(1), 7–22.
Dates, J., & Stroman, C. (2001). Portrayals of families of color on television. In J. Bryantand J. A. Bryant (Eds.), Television and the American family (pp. 207–225). Mahwah, NJ:Lawrence Erlbaum Associates.
DeFleur, M. L., & DeFleur, M. H. (2003). Learning to hate Americans: How US media shapenegative attitudes among teenagers in twelve countries. Spokane, WA: Marquette.
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(2), 162–187.
Dixon, T. L., & Linz, D. (2000). Overrepresentation and underrepresentation of African Amer-icans and Latinos as lawbreakers on television news. Journal of Communication, 50(2),131–154.
Dominick, J. R., & Greenberg, B. S. (1972). Attitude towards violence: The interaction of televi-sion exposure, family attitudes, and social class. In G. A. Comstock & E. A. Rubinstein (Eds.),Television and social behavior (Vol. 3). Government Printing Office, Washington, DC.
Elasmar, M. G. (2003). An alternative paradigm for conceptualizing and labeling the processof influence of imported television programs. In M. E. Elasmar (Ed.), The impact of in-ternational television: A paradigm shift (pp. 157–169). Mahwah, NJ: Lawrence ErlbaumAssociates.
Entman, R. M. (1994). Representation and reality in the portrayal of blacks on networktelevision news. Journalism Quarterly, 71(3), 509–520.
Fiske, S. T. (1980). Attention and weight in person perception: The impact of negative andextreme behavior: Impression and recall order effects in behavior-based impression forma-tion. Journal of Personality and Social Psychology, 37(10), 1758–1768.
Friedman, T. L. (2006). The world is flat: A brief history of twenty-first century. New York:Farrar, Straus & Giroux.
Fujioka, Y. (1999). Television portrayals and African-American stereotypes: Examination oftelevision effects when direct contact is lacking. Journalism and Mass CommunicationQuarterly, 76(1), 52–74.
Fyock, J., & Stangor, C. (1994). The role of memory biases in stereotype maintenance. BritishJournal of Social Psychology, 33, 331–344.
Gerbner, G., Gross, L., Morgan, M., Signorielli, N., & Shanahan, J. (2002). Growing up withtelevision: Cultivation processes. In J. Bryant and D. Zillman (Eds.), Media effects: Advancesin theory and research (2nd ed., pp. 43–67). Mahwah, NJ: Lawrence Erlbaum Associates.
586 Journal of Broadcasting & Electronic Media/December 2010
Gerdes, L. I. (Ed.) (2006). Globalization. San Diego, California: Greenhaven Press.Gibbons, J. A., Lukowski, A. F., & Walker, W. R. (2005). Exposure increases the believability of
unbelievable news headlines via elaborate cognitive processing. Media Psychology, 7(3),273–300.
Gilbert, D. T., & Hixon, J. G. (1991). The trouble of thinking: Activation and application ofstereotypic beliefs. Journal of Personality and Social Psychology, 60(4), 509–517.
Greenberg, B. S., Mastro, D., & Brand, J. E. (2002). Minorities and the mass media: Televisioninto the 21st century. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theoryand research (2nd ed., pp. 333–351). Mahwah, NJ: Lawrence Erlbaum Associates.
Ha, J. W., & Yang, E. K. (2002). The regionalization of television and Korean cultural wave inEast Asia. Journal of Broadcasting Research, winter, 67–103.
Han, H. (2007). Social issues and changes of employment policy in the response to the increaseof foreign workers in Korea. Conference paper presented to the International Conferenceon Government Innovation and Human Resources Development, Seoul, Korea.
Hart, W., Albarracin, D., Eagly, A., Brechan, I., Lindberg, M., & Merrill, L. (2009). Feelingvalidated versus being correct: A meta-analysis of selective exposure to information. Psy-chological Bulletin, 135(4), 555–588.
Hwang, J., Kim, Y., Lee, M., Choi, H., & Lee, D. (2007). The examination of multi-ethnicity,multiculturalism in Korean society. Research Reports, 19, 1–23.
Internet World Stats (2008). Retrieved Oct. 30, 2008, from http://www.internetworldstats.com/Jang, T. (2001). Korean college students’ preferences of particular race, and ethnicity. Con-
temporary Critical Studies, 14(1), 99–113.Kim, H., & Oh, S. (1999). The stereotypes, affects, and attitudes of college students and citizens
toward North Koreans: A comparison with those toward other foreigners. PsychologicalScience, 8(1), 1–22.
Lang, A. (2000). The limited capacity model of mediated message processing. Journal ofCommunication, 50(1), 46–70.
Lang, A., & Friestad, M. (1993). Emotion, hemispheric specialization and visual and verbalmemory for television messages. Communication Research, 20(5), 647–670.
Madon, S., Guyll, M., Aboufadel, K., Montiel, E., Smith, A., Palumbo, P., & Jussim, J. (2001).Ethnic and national stereotypes: The Princeton trilogy revisited and revised. Personality andSocial Psychology Bulletin, 27(8), 996–1010.
Mastro, D. E., & Kopacz, M. A. (2007). Media representations of race, prototypicality, andpolicy reasoning: An application of self-categorization theory. Journal of Broadcasting andElectronic Media, 50(2), 305–322.
Messaris, P., & Woo, J. (1991). Image vs. reality in Korean-Americans’ responses to massmediated depictions of the United States. Critical Studies in Mass Communication, 8(1),74–90.
Oliver, M. B. (1999). Caucasian viewers’ memory of Black and White criminal suspects in thenews. Journal of Communication, 49(3), 46–60.
Pettigrew, T. F. (1998). Intergroup contact theory. Annual Review of Psychology, 49(1), 65–85.
Pew. (2009). Retrieved July 23, 2009, from http://pewglobal.org/reports/display.php?ReportIDD264
Rada, J. A. (2000). A new piece of the puzzle: Examining effects of television portrayals ofAfrican Americans. Journal of Broadcasting & Electronic Media, 44(4), 704–715.
Rokeach, M. (1968). Beliefs, attitudes, and values. San Francisco: Jossey-Bass.Rubin, A. M., & Perse, E. M. (1987). Audience activity and soap opera involvement: A uses
and effects investigation. Human Communication Research, 14(2), 246–268.Schiappa, E., Gregg, P., & Hewes, D. (2005). The parasocial contact hypothesis. Communica-
tion Monographs, 72(1), 92–115.Schneider, D. J. (2004). The psychology of stereotyping. New York: The Guilford Press.Shim, D. (2006). Hybridity and the rise of Korean popular culture in Asia. Media, Culture, and
Society, 28(1), 25–44.Smith, T. W. (1991). Ethnic images in the United States. The Polling Report, 7(1), 1–5.
Tan et al./AFRICAN-AMERICAN STEREOTYPES IN SOUTH KOREA 587
Snyder, M., & Miene, P. (1994). On the function of stereotypes and prejudices. In M. P. Zanna& J. M. Olson (Eds.), The psychology of prejudice: The Ontario Symposium (pp. 33–54).Hillsdale, NJ: Lawrence Erlbaum Associates.
Stangor, C., & Schaller, M. (1996). Stereotypes as individual and collective representations. InC. N. Macrae, C. Stangor, & M. Hewstone (Eds.), Stereotypes and stereotyping (pp. 3–37).New York: The Guilford Press.
Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance ofAfrican Americans. Journal of Personality and Social Psychology, 69(5), 797–811.
Tajfel, H. (1981). Social stereotypes and social groups. In J. C. Turner & H. Giles (Eds.),Intergroup behavior (pp. 144–167). Chicago: The University of Chicago Press.
Tan, A., Fujioka, Y., & Tan, G. (2000). Television use, stereotypes of African Americans andopinions on affirmative action: An affective model of policy reasoning. CommunicationMonographs, 67(4), 362–371.
Tan, A., Zhang, L., Zhang, Y., & Dalisay, F. (2007). Stereotypes of Americans and use ofAmerican and Chinese information sources by Chinese high school students. Presented atMidwest Association for Public Opinion Research (MAPOR), Chicago, IL, November 2007.
Tan, A., Zhang, Y., Zhang, L., & Dalisay, F. (2009). Stereotypes of African-Americans andmedia use among Chinese high school students. Howard Journal of Communications, 20(3),260–275.
The Ministry of Justice (2008). Retrieved July 31, 2009 from http://www.immigration.go.kr/Valentino, N. A. (1999). Crime news and the priming of racial attitudes during evaluations of
the president. Public Opinion Quarterly, 63(3), 293–320.van de Vijver, F. J. R., & Leung, K. (1997). Methods and data analysis for cross-cultural research.
Newbury Park, CA: Sage.World Bank (2008). Retrieved July 1, 2009, from http://siteresources.worldbank.org/DATA
STATISTICS/Resources/GDP.pdfYim, H. (2002). Cultural identity and cultural policy in South Korea. The International Journal
of Cultural Policy, 8(1), 37–48.Yoon, T.-J. (1997). Mass media and the reproduction of the international order: Presentation
of American culture by American television programs aired in Korea, 1970–1989. Unpub-lished thesis, the University of Minnesota.
Copyright of Journal of Broadcasting & Electronic Media is the property of Broadcast Education Association
and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for individual use.