Content uploaded by Kristen Harrison
Author content
All content in this area was uploaded by Kristen Harrison on May 05, 2020
Content may be subject to copyright.
http://crx.sagepub.com/
Communication Research
http://crx.sagepub.com/content/39/3/338
The online version of this article can be found at:
DOI: 10.1177/0093650211401376
2012 39: 338 originally published online 16 March 2011Communication Research
Nicole Martins and Kristen Harrison
Television Use and Self-Esteem : A Longitudinal Panel Study
Racial and Gender Differences in the Relationship Between Children's
Published by:
http://www.sagepublications.com
can be found at:Communication ResearchAdditional services and information for
http://crx.sagepub.com/cgi/alertsEmail Alerts:
http://crx.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://crx.sagepub.com/content/39/3/338.refs.htmlCitations:
What is This?
- Mar 16, 2011 OnlineFirst Version of Record
- Apr 24, 2012Version of Record >>
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Communication Research
39(3) 338 –357
© The Author(s) 2012
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0093650211401376
http://crx.sagepub.com
Racial and Gender
Differences in
the Relationship
Between Children’s
Television Use and
Self-Esteem: A
Longitudinal Panel Study
Nicole Martins1 and Kristen Harrison2
Abstract
A longitudinal panel survey of 396 White and Black preadolescent boys and girls was
conducted to assess the long-term effects of television consumption on global self-esteem.
The results revealed television exposure, after controlling for age, body satisfaction,
and baseline self-esteem, was significantly related to children’s self-esteem. Specifically,
television exposure predicted a decrease in self-esteem for White and Black girls and
Black boys, and an increase in self-esteem among White boys. The findings are discussed
in terms of cultivation theory and social identity theory.
Keywords
children, television, self-esteem, race, social identity, cultivation
Global self-esteem, generally defined as a feeling of self-worth and fundamental respect
for oneself (Rosenberg, 1972), is recognized for the significant role it plays in mental
health and behavioral disorders (McGee & Williams, 2000; U.S. Department of Health and
Human Services, 2009). Known predictors of self-esteem include peer influences (e.g., peer
acceptance, popularity) and family relationships (see Harter, 1999 for review). Research-
ers also have begun to examine the contemporary influences of the media on self-esteem
because of established relationships between media exposure and other psychological variables
1Indiana University, Bloomington
2University of Illinois at Urbana-Champaign
Corresponding Author:
Nicole Martins, Indiana University, 1229 E. Seventh St., Bloomington, IN 47405
Email: nicomart@indiana.edu
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 339
such as attitudes and beliefs. For example, media exposure is correlated with the endorse-
ment of gender stereotypes (Herrett-Skjellum & Allen, 1996), real-world perceptions of
racial and ethnic minorities (Mastro & Kopacz, 2006), and body dissatisfaction (Dohnt &
Tiggemann, 2006). Given these relationships, it is widely believed that the media play an
important role in shaping self-conceptions (see Sahlstein & Allen, 2002). Yet this assump-
tion has rarely been tested with global self-esteem, particularly in samples of preadolescent
children when self-esteem is just beginning to develop. Therefore, the purpose of this
investigation is to examine whether media consumption is related to children’s self-esteem
by means of a longitudinal panel survey. Furthermore, because media may not affect all chil-
dren in the same way, differences by gender and racial/ethnic background are examined.
The Importance of Self-Esteem
Self-esteem is an important construct with significant behavioral and emotional ramifications,
particularly for children. Research has shown that high self-esteem is correlated with student
motivation, persistence, and academic achievement (for review see Baumeister, Campbell,
Krueger, & Vohs, 2003). Research also indicates that high self-esteem among elementary
school children is related to the health of their peer relationships (Berndt, 2002). In con-
trast, youth with low self-esteem are at risk for adopting health-compromising behaviors,
such as alcohol (Scheirer, Botvin, Griffin, & Diaz, 2000) and drug abuse (Howard &
Jenson, 1999). Low self-esteem has also been linked to other health-risk behaviors such as
disordered eating and suicidal ideation (McGee & Williams, 2000; Wilburn & Smith,
2005). Researchers have argued that youth with low self-esteem are at risk for adopting
such behaviors because these actions are regarded as the only means available for coping
with feelings of low self-worth (Baumeister, 1999) or because these youth are more likely
to succumb to peer pressure (McGee & Williams, 2000). In accord, many advocacy groups
have created programs to increase self-esteem in students, under the assumption that grade
point averages would increase and delinquency would decrease, thereby leading to happier
and more successful lives (National Association for Self-Esteem, 2005).
Media Exposure and Self-Esteem
According to a recent national survey of over 2,000 children aged 8 to 18 years, the typical
American child spends an average of 7 hours per day with entertainment media (Rideout,
Foehr, & Roberts, 2010). Given the sheer amount of time that children and adolescents
spend with the media, researchers are interested in the role that the media play in children’s
development and mental health. Recent evidence indicates that television viewing dis-
places one-on-one communication with parents and caregivers (Christakis et al., 2009).
For example, one small laboratory-based study found that parents interacted less with their
children when the television set was turned on (Tanimura, Okuma, & Kyoshima, 2007).
Researchers argue that not only are these interactions crucial to children’s language devel-
opment, but screen time is associated with less opportunities for children to engage in
activities that have been shown to nurture self-esteem such as reading, athletics, and hob-
bies (Centers for Disease Control and Prevention, 2009).
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
340 Communication Research 39(3)
The extant research that has focused on the impact of the media on children’s health has
focused on body image and eating disorders. Indeed, content analyses demonstrate that
females in the media are growing thinner over time (e.g., Spitzer, Henderson, & Zivian,
1999) and are underweight compared to females in the U.S. population (Greenberg, Eastin,
Hofschire, Lachlan, & Brownell, 2003). Research also supports that girl’s exposure to these
images results in lower body satisfaction in the long term (Harrison & Hefner, 2006). More-
over, boys are not immune to these effects. Like female bodies, content analyses have shown
that males in the media are underweight compared to the U.S. population (Greenberg, et al.,
2003) and that exposure to media featuring the muscular male body can result in an increase
in depression and muscle dissatisfaction among boys (Harrison & Bond, 2007).
Yet the focus on body satisfaction, or appearance esteem, ignores the notion that self-esteem
is a multidimensional construct. Global self-esteem includes cognitive, social, and physical
appearance dimensions that combine to create an overall self-evaluation (Sahlstein & Allen,
2002). Given the media’s effect on physical appearance, it seems reasonable to presume
that the media influence these other dimensions of self-esteem as well. Undeniably, media per-
sonalities promote more than unrealistic body ideals. The characters featured on television
are also successful and powerful (Signorielli, 1993). For example, Signorielli and Kahlenberg
(2001) content analyzed 11 one-week samples of prime-time television dramas between
fall 1990 and spring 1998 for characters’ marital status, race, and occupations. The results
revealed that White collar jobs occupied by men were overrepresented on television com-
pared to actual U.S. labor force statistics (35% vs. 17%), leading the authors to conclude that
television programs place a large emphasis on prestigious and glamorous occupations.
Prestigious and glamorous occupations are also associated with affluent lifestyles.
Research supports that adolescents’ exposure to programs that place a large emphasis on
prestige and affluence is related to their tendency to overestimate other people’s affluence.
For example, Potter (1991) found that television viewing was positively correlated with
adolescents’ agreement with perceptions of the normality of affluence such as “most peo-
ple have enough money to live a very good life” in his sample of youth in Grades 8 through
12. The relationship remained significant even when controlling for demographics, infor-
mation processing ability, and information-seeking variables. Signorielli (1993) found that
television exposure was related to adolescents’ wanting high status jobs that would earn
them money but also be relatively easy and allow them to take long vacations. These images,
then, may result in unrealistic assumptions about the real world. When these expectations
about the real world are not realized, the evaluation may result in unhappiness and dissat-
isfaction with one’s life (Buijzen & Valkenburg, 2003).
The amount of time children spend with the media has led researchers to call for further
investigation into the relationship between media exposure and global self-esteem (Polce-
Lynch, Myers, Kliewer, & Kilmartin, 2001). Despite this call for research, little work has
directly tested this relationship, particularly in the long run and with ethnically diverse
samples of children. The scant research that has linked media exposure to children’s self-
esteem has done so in studies where body-change strategies (e.g., lose weight, increase
muscle mass) are the outcomes of interest. For instance, Ricciardelli and McCabe (2001)
examined the moderating role of self-esteem in the relationship between sociocultural
influences (e.g., media exposure, peer teasing, parent comments) and drive for muscularity
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 341
in a sample of 587 boys aged 11 to 15. Results revealed a significant interaction, such that
the relationship between media exposure and drive for muscularity was stronger for boys
with low rather than high self-esteem.
In studies where the relationship between media exposure and self-esteem has been
directly tested, few have done so with samples of children. Wiggins (1987) found no rela-
tionship between self-esteem and television use in his sample of 490 children in Grades 4
through 12. One possible explanation for this null finding is that Wiggins asked children to
report the amount of television they watched during the week, but not on the weekend
when they have more time to devote to television viewing. Thus, artificially low weekly
viewing scores could have reduced the variance necessary to find a significant relationship.
Stroman (1986) examined the relationship between television viewing and self-concepts in
Black children in Grades 3 through 6. She found a positive relationship between television
viewing and self-concept for girls but not for boys. However, it is challenging to compare
the findings of these two studies because Stroman studied self-esteem in Black children.
Thus, the limited collection of research on media exposure and children’s self-esteem has
used different racial groups making generalization difficult.
Theoretical Approaches Regarding
Self-Esteem and Media Exposure
Assuming self-esteem is in fact related to media exposure, what are some theoretical
explanations for this relationship? One possible mechanism is cultivation theory, which
posits that people’s view of social reality is modeled by the media (Gerbner, Gross,
Signorielli, Morgan, & Shanahan, 2002). One of the assumptions of cultivation theory is
that certain common features of the television landscape pervade all forms of program-
ming. Cultivation theory offers an explanation for how white collar jobs, the thin ideal,
power, and wealth may come to be perceived as commonplace and easily achievable. In
support of this theoretical model, Cheung and Chan (1996) found that extensive television
viewing was positively associated with the endorsement of materialistic values in their
large sample of adolescents in Hong Kong. The authors reasoned that this finding was due
to television’s emphasis on fine clothes, extravagant cars, and luxurious homes.
Cultivation analysis has been criticized for assuming that the television landscape is
relatively homogenous (Hawkins & Pingree, 1981). Indeed, there are studies that have
found that exposure to certain genres, but not television viewing overall, significantly
predicted the outcome variable of interest (e.g., Shrum, 1996; Wilson, Martins, & Marske,
2005). Although this is an important point to make, we would be remiss to deem that
overall time spent with the media is not important. Harrison (2006) addressed this very
idea with her scope of self-theory. She argues that heavy television viewers allow televi-
sion to displace the real-life experiences that might build self-complexity and resilience.
That is, television’s messages may encourage a narrow-minded view of the self, because
the time demands of a heavy television diet limit adolescents’ opportunities to take part
in the real-world experiences that would otherwise broaden and add complexity to their
self-definition.
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
342 Communication Research 39(3)
If media messages are cultivating unrealistic expectations for the real world that result
in self-esteem being affected beyond body esteem, then an adequate test of these pro-
cesses should rule out alternative explanations for effects. One such explanation is that
individuals who exhibit low self-esteem also suffer from lower body esteem. That is, any
relationship we may find between television exposure and self-esteem may be explained
by the shared variance between body esteem and self-esteem. We argue that body esteem
may explain some portion of the relationship between television exposure and self-esteem
but not all of it. Even when controlling for body esteem, lower self-esteem should be more
pronounced among children reporting higher television exposure. Thus, we predict:
Hypothesis 1: Television viewing will significantly predict lower self-esteem, inde-
pendent of body esteem.
A second theoretical approach that may explain a relationship between self-esteem
and media exposure is social identity theory. Tajfel’s social identity theory proposes
that people are motivated to maintain a positive self-concept by comparing favorable
characteristics of their in-group with parallel unfavorable characteristics of the relevant
outgroup (Taijfel, 1978; Turner, 1982). Individuals perceived to be similar on salient
dimensions, such as gender or race/ethnicity, are identified as in-group members and
those considered dissimilar are identified as outgroup members. Once these classifications
have been made, comparison strategies are used to help protect the self-concept by evalua-
ting those who are similar to themselves as better than those in the pertinent outgroup
(Abrams & Hogg, 1990).
Research demonstrates that not only are young children able to make simple in-group/
outgroup distinctions based on gender and race, but children also attach favorable evalua-
tions to in-group members presumably to maintain their self-concept (Corenblum & Annis,
1993). In their classic study, Kuhn and colleagues (1978) found that 2-and-3-year-olds
tended to attribute positive characteristics to their own sex and negative characteristics to
the other. In terms of race, Kowalski (2003), assessed in-group and outgroup attitudes among
a diverse sample of preschool children. As predicted, preschoolers in the study displayed a
positive bias (e.g. smart, friendly, nice, clean) toward their own racial or ethnic group. Yet
the sources for comparison may not be based in reality. Instead, these comparisons may be
derived from the mass media (Hogg, Terry, & White, 1995). Given the ubiquitous nature
of media in the lives of children, the potential is high for media content to influence chil-
dren’s self-esteem because these messages may provide a comparative basis for positive
in-group evaluations. Thus, the impact these comparisons may have on self-esteem depend
on what media messages are made available.
Gender Differences
The representation of males and females on television has been the subject of many studies
over the past half century. Several studies have demonstrated that on television, men appear
twice as frequently as women (Signorielli, 1997; Signorielli, McLoud, & Healy, 1994). This
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 343
underrepresentation is exacerbated in children’s television programming, with male char-
acters appearing up to four times more frequently in children’s Saturday morning programs
(Calvert, 1999). When females are featured, they are typically portrayed in a stereotypical
way. For example, Thompson and Zerbinos (1997) showed that male characters were more
likely than female characters to be portrayed as independent and assertive, whereas female
characters were more likely than male characters to be portrayed as emotional, warm, and
affectionate.
What kind of effect, then, will exposure to these types of portrayals have on boys’ and
girls’ self-esteem? Surprisingly, there is little research that has explored the role of the
media in gender differences in children’s self-esteem. However, both cultivation theory
and social identity theory would predict that exposure to these images may serve to
increase the self-esteem among boys. Cultivation research contends that children (espe-
cially heavy viewers) will likely adopt the belief that boys are dominant, assertive, and
powerful because these are the perspectives most frequently seen on television. Social
identity theory would predict that boys would use these characteristics, then, as a basis of
comparison to maintain their self-concept. In contrast, we should expect that self-esteem
will decrease among girls. Heavy-viewing television viewers may be more likely to
believe that girls are passive, weak, and insignificant. These stereotypic media portrayals
provide comparative elements that may result in a negative self-evaluation among girls.
Based on cultivation and social identity theory, we predicted:
Hypothesis 2: Television viewing will significantly predict higher boys’ self-esteem,
but lower girls’ self-esteem.
Racial Differences
The case can be made that Black youth are particularly at risk for developing low self-
esteem. Cultivation theory predicts that the consequences of television exposure are most
pronounced for its most frequent viewers (Gerbner et al., 2002); and in fact, African American
children have been found to consume more media than Whites or other minority groups
(Roberts et. al., 2010). Social identity theory predicts that efforts to maintain or bolster
one’s self-concept are dependent on the types of comparison that are made available (Fry-
berg, 2003). Unfortunately for African American children, the majority of comparisons
available in the television world consist of characters who are unprofessional and provoca-
tively dressed compared to White counterparts (Mastro & Greenberg, 2000). In addition,
children who watch television news will witness a world where Blacks are more likely to
be shown as perpetrators of crime than Whites (Dixon & Linz, 2000). Thus, frequent exposure
to stereotypical images that portray Blacks as unprofessional and criminal may lead view-
ers to believe that these are true characteristics of Blacks in the real world. In this way, it is
believed that frequent media use may negatively influence the evaluations among African
American youth, thereby leading to lower self-esteem.
Despite this concern, evidence to support low self-esteem among African Americans
is limited. Although African American youth watch more television, Black youth report
higher self-esteem than non-Black youth. In their meta-analytic synthesis of 261 studies,
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
344 Communication Research 39(3)
Gray-Little and Hafdahl (2000) found higher self-esteem scores for Black children, teens,
and young adults than for White participants. An explanation for this finding could be that
African Americans do not turn to the larger White society, and mainstream media, as their
reference points, but turn instead to the Black community as their source of support
(Gray-Little & Hafdahl, 2000). By drawing mostly in-group comparisons and by attribut-
ing negative feedback to prejudice against the group (not the self), African Americans
may buffer their self-esteem from the media’s often one-dimensional images. Yet this
explanation does not account for the mix of positive and null relationships reported earlier
(Stroman, 1986). In accord, our first research question asks:
Research Question 1: Is there a racial difference in the relationship between televi-
sion exposure and subsequent self-esteem?
Sex, Race, and Television Exposure?
Hypothesis 2 and Research Question 1 concern whether gender and race moderate the
relationship between television exposure and self-esteem. Considering these two variables,
we extend our inquiry by investigating whether sex and race of the child modifies the
extent to which television exposure influences self-esteem. Reviewing previous studies
that have examined sex and race differences in self-esteem reveals that they did not look at
diverse samples together, calling into question the need for the present study. Thus, the
final research question asks:
Research Question 2: Will gender and race moderate the relationship between tele-
vision exposure and subsequent self-esteem?
In summary, available evidence suggests that a relationship between television
exposure and self-esteem may exist among young children of both genders. What is
still unclear is the causal direction of this relationship and whether the same
relationship exists among African Americans. Although a longitudinal design does
not provide an airtight demonstration of causality, it goes much further than cross-
sectional designs toward demonstrating causality by fulfilling the temporal-order
requirement of causal order (Davis, 1985). A longitudinal design also allows for the
control of key variables that could put research participants at risk for lowered self-
esteem independent of their television exposure, including baseline self-esteem and
baseline body satisfaction.
Method
Participants
A sample of elementary school children participated in a multiyear panel study in schools
situated in three lower-middle to upper middle-class communities in the Midwestern
United States. Because one of the goals of the study was to test race as a moderator, schools
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 345
were preselected on the basis of racial diversity. The first two waves of data collection,
which took place 1 year apart, were the focus of analysis. Thirty-four participants were
unable or unwilling to report their race/ethnicity; of those who did report their race/ethnicity,
the majority identified themselves as Black or African American (N = 235), followed by
White or European American (N = 194); due to the small number of participants reporting
a race other than White or Black, we chose to focus only on White and Black participants
in our analyses (N = 429). The number of girls who participated in both waves was 223 and
the number of boys who participated in both waves was 206. At baseline, 136(28%) were
second graders, 162 (37%) were third graders, and 131 (31%) were fourth graders. Ages
ranged from 7 to 12 (M = 8.72, SD = 1.02) at baseline to 8 to 13 (M = 9.75, SD = 1.03) at
follow-up.
Procedure
All participants obtained parental consent and signed their own assent forms at the start of
testing. Trained male and female undergraduate research assistants administered the ques-
tionnaires to groups of two to five boys and girls (with female assistants working with the
girls and male assistants working with the boys), separated by gender and separated from
one another by barriers so their responses would remain private. The research assistant sat
opposite each group of children and read each item out loud, pointing to the questionnaire
to indicate the appropriate item along with the response options. To reduce primacy and
recency effects, the order of response options was alternated across questionnaires, and
t tests were used to compare responses across the different questionnaire versions. There
were no effects of order. Participants received self-esteem and body satisfaction measures
first, followed by television exposure measures, followed by demographic measures. When
they finished completing their questionnaires, children received a novelty pencil in thanks
for their participation and were escorted back to their classrooms. The same procedure was
followed for both baseline and follow-up questionnaires.
Measures
Television exposure. We measured television exposure according to the procedure out-
lined by Harrison (2000). Participants were asked to report how many hours of television
they watched the preceding weekday (a) in the morning before school, (b) during school,
(c) after school before dinner, and (d) after dinner before bed. Participants were then asked
how many hours of television they watched on a typical Saturday and Sunday (a) in the
morning before lunch, (b) in the afternoon before dinner, and (c) after dinner before bed.
For each item, the response options ranged from 0 to 5 hours. A weekly television viewing
index was created by multiplying the number of hours reported for the preceding weekday
by 5 and adding it to the number of Saturday and Sunday hours. A histogram of the televi-
sion viewing variables revealed that the distributions were normal until the 85-hour point
(approximately 12 hours of viewing per day), after which it was positively skewed, with
3 participants reporting viewing an average of more than 15 hours per day. Due to the strong
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
346 Communication Research 39(3)
possibility that their responses to the rest of the measures would be unreliable, the partici-
pants who reported viewing an average of more than 85 hours of television per week at
either wave of data collection were dropped from further analyses, which reduced the
sample to N = 396.
Self-esteem. Participants completed the 11-item Lawrence Self-Esteem Questionnaire
(LSEQ, Lawrence, 1981) as a measure of overall feeling of self-worth (i.e., global self-
esteem). This scale is based on the definition that self-esteem is the “the child’s affective
evaluation of the sum total of his or her characteristics both mental and physical” (Lawrence,
1981, p. 246). This scale has been validated for children in this age group (Hart, 1985) and
has been included in several large-scale studies with children (Viner & Cole, 2006; Williams
& Currie, 2000). This is an ideal measure to use for two reasons. First, the instrument
includes several items to assess the behavior of the child and the child’s perceptions of
others concerning their behavior which are important indicators of self-esteem (Cooper-
smith, 1967; Harter, 1999). Moreover, the measure does not contain items that relate to body
shape or body satisfaction, allowing us to examine the relationship between television
viewing and self-esteem independently of body esteem. Second, when measures are devel-
oped for young children, a number of difficulties are encountered, such as their cognitive
ability, language development, and short attention spans (Hughes, 1984). Thus, the LSEQ
is ideal because the questions and the response items are appropriate for the developmental
level of the children in this study. Sample items include “When you want to tell a teacher
something, do you usually feel foolish?” and “Are there lots of things about yourself you
would like to change?” Response options are: 0 (never), 1 (not sure), and 2 (always). Responses
to these questions were reverse coded such that higher scores were indicative of higher
self-esteem. Scale reliability estimates were α = .71 at baseline and α = .79 at follow-up.
The reliability estimates are consistent with other samples of the same age (α = .76, Williams
& Currie, 2000) and ethnicity (α = .70, Martins, 2008).
Race/ethnicity and gender. On the last page of the questionnaire, participants were asked
to report their race/ethnicity and their gender. The race/ethnicity question was open ended.
Response options were recorded by the research assistant and later coded according to U.S.
census categories. In cases where multiple ethnicities were listed (e.g., “I am Black and
White”) the non-White category was given preference to avoid undercounting participants
of color.
Control variables. The age of the child was used as a covariate in these analyses. We chose
to control for age because the distribution of ages across the sample was relatively narrow,
making meaningful developmental predictions with regard to age differences difficult.
Recall that global self-esteem, not just body esteem, is of concern here. Because one
could argue that any relationship we may find between television exposure and self-
esteem may be explained by the shared variance between body esteem and self-esteem,
body esteem must be controlled to assess whether television exposure predicts global
self-esteem above and beyond body perceptions. Therefore, satisfaction with perceived
body shape was measured and used as a control variable. Perceived body shape was
measured by administering a pictorial scale featuring line drawings of prepubescent chil-
dren, with boys viewing a scale of male figures and girls viewing a scale with female
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 347
figures (Tiggemann & Wilson-Barrett, 1998). Each scale featured seven bodies ranging
from markedly thin to markedly fat. We measured participants’ perceived body shape by
showing them the child figure drawings and directing them to “Circle the picture that
shows how YOU look.” Next, we asked participants to “Circle the picture that shows
how you WANT to look.” Satisfaction with perceived body shape was calculated as the
ideal minus the current figure rating. Thus, negative scores indicated a desire to be thin-
ner. Good test-retest reliability has been found for such figure-rating scales with young
children (Collins, 1991).
Results
Preliminary Analyses
Preliminary analyses tested for attrition biases that might have compromised the general-
izibility of the findings. The attrition rate of White and Black children in this study was
15% and was primarily due to students moving out of the school district or absenteeism.
Participants who provided complete data did not differ significantly from those who did
not on any of the variables assessed in Wave 1.
Descriptive Statistics
The descriptive statistics for each major control, predictor, and criterion variables are pre-
sented in Table 1. In particular, the data revealed that higher scores on the self-esteem
measure were significantly correlated with hours of television watched per week for all
children in both waves. At baseline, self-esteem was significantly correlated with body
satisfaction, but this finding was not significant at follow-up.
More differences emerge when the sample is divided into groups based on gender and
race/ethnicity. Black girls reported significantly more overall television exposure at
follow-up (Table 2). However, the mean body satisfaction scores and the mean self-esteem
scores for both White and Black girls at baseline and follow-up were not significantly dif-
ferent from one another. Several differences also emerged for males (Table 3). Black boys
reported significantly higher overall television exposure at both baseline and follow-up.
There were no race differences among the boys for body satisfaction or self-esteem at
either wave.
Hypothesis and Research Questions
To test our hypotheses and answer our research questions, we conducted a hierarchical
multiple regression analysis. Self-esteem assessed at the baseline was entered on Step 1.
Control variables were entered on Step 2: age, baseline body esteem, and Time 2 TV view-
ing; Time 1 TV viewing, race, and gender were entered on Step 3; interaction terms for
Time 1 TV viewing by race, Time 1 TV viewing by gender, and race by gender were
entered on Step 4; and the three-way interaction term representing the Time 1 TV viewing
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
348 Communication Research 39(3)
by race by gender was entered on the last step. Our outcome of interest was children’s self-
esteem at follow-up. To reduce problems with multicollinearity, continuous predictor
variables were mean centered prior to creating the polynomial term (Aiken & West, 1991).
The results of the analysis revealed that Time 1 self-esteem was a marginal predictor of
Time 2 self-esteem scores. Baseline self-esteem accounted for 9% of the variance in chil-
dren’s self-esteem at follow-up, Fchange(1, 394) = 3.43, p = .06. The control variables, as
a set, did not significantly account for additional variance in children’s self-esteem,
Fchange(1, 391) = .19, p = ns. Baseline television viewing failed to predict self-esteem at
follow-up (β = .00, t = 1.30, p = ns). The models containing the interaction terms also
failed to account for additional variance in children’s self-esteem at follow-up.
Cohen, Cohen, Aiken, and West (2003) acknowledge that the analysis proposed above
is often the “simplest” model for examining the relationship between a given X and Y in
two-wave longitudinal data sets. This model, however, raises the question as to what the
optimal interval should be for examining causal effects of one variable on another. As
McArdle and Woodcock (1997) point out, consideration of the timing of effects is a theo-
retical task rarely addressed in the social sciences. Consequently, if the effects of baseline
Table 1. Baseline and Follow-Up Means, Standard Deviations, and Zero-Order Correlations
for Key Variables (N = 396)
Variable M SD 1 2 3 4 5
Baseline
Age 8.74 1.01
Gender — — 0.03
Race/ethnicity — — 0.01 -0.07
Body satisfaction -0.56 1.35 -0.08 0.22** 0.07
Hours of TV/week 40.53 23.01 -0.01 0.01 0.13** -0.08
Global self-esteem 1.16 0.42 0.07 0.19** -0.03 0.14** -0.15**
Follow-up
Age 9.76 1.05
Gender — — 0.03
Race/ethnicity — — -0.01 0.05
Body satisfaction -0.60 1.27 -0.04 0.17** 0.11
Hours of TV/week 39.91 22.94 -0.09 0.10 0.31** -0.03
Global self-esteem 1.08 0.51 -0.05 -0.11 -0.03 -0.10 -0.12*
Note: Body satisfaction was calculated as the ideal minus the current figure rating, with -6 being the
smallest and 4 being the largest. Hours of television viewed per week ranged from 0 to 85. Self-esteem
ranged from 0 to 2. Codes assigned to the sex variable were 0 = girls and 1 = boys. Codes assigned to
the race variable were 0 = White and 1 = Black. Means and standard deviations are not provided for
categorical variables; readers interested in zero-order correlations broken down by gender and race/
ethnicity should contact the first author.
*p < .05. **p < .01. ***p<.001.
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 349
Table 2. Baseline and Follow-Up Means and Standard Deviations for Key Variables for Girls
(N = 213)
White Black
Variable Min. Max. M SD M SD t test
Baseline
Body satisfaction -6 3 -0.89 1.52 0.80 1.40 -0.47
Hours of TV/week 0 85 38.76 23.24 41.78 23.28 -0.93
Global self-esteem 0.09 2.0 -1.08 0.45 1.08 0.39 -0.07
Follow-up
Body satisfaction -6 2 -0.85 1.38 -0.61 1.44 0.49
Hours of TV/week 0.50 80 29.64 18.81 41.46 21.42 -4.21***
Global self-esteem 0.18 2 1.08 0.45 1.09 0.46 0.95
Note: The minimum and maximum values for variables at baseline represent the minimum and maximum
values of the scales for both waves except for body satisfaction. For body satisfaction, -6 = actual
minimum and 6 = actual maximum.
***p < .001.
Table 3. Baseline and Follow-Up Means and Standard Deviations for Key Variables for Boys
(N = 183)
White Black
Variable Min. Max. M SD M SD t test
Baseline
Body satisfaction -4 4 -0.40 1.29 -0.090 0.91 -1.82
Hours of TV/week 0 85 35.94 22.73 45.06 22.03 -2.76**
Global self-esteem 0 2 1.28 0.42 1.22 0.40 1.01
Follow-up
Body satisfaction -5 2 -0.41 1.27 -0.22 0.93 -1.26
Hours of TV/week 0 85 32.18 22.33 43.34 22.22 -3.641**
Global self-esteem 0.18 2 1.01 0.54 1.03 0.53 -0.23
Note: The minimum and maximum values for all variables except body satisfaction (baseline and
follow-up) represent the minimum and maximum values of the scales. For body satisfaction, -6 = actual
minimum and 6 = actual maximum.
**p < .005.
television viewing on self-esteem take place over a much shorter period of time than is
represented by the interval between study waves, then Cohen et al. (2003) caution that
employing X1, (baseline television viewing in this case) may “lead to a serious underesti-
mate of its impact” (p. 573). Under these circumstances, then, using X2 (Time 2 TV viewing)
in predicting self-esteem at follow-up may provide a truer estimate of the longitudinal
impact of television viewing on self-esteem.
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
350 Communication Research 39(3)
Given the null findings reported above, it appears as though we may have such a cir-
cumstance. In accord with Cohen and colleagues’ (2003) recommendation, we decided to
employ a “quick acting effects approach” (p. 573) where Time 2 television viewing was
used to predict Time 2 self-esteem.1 Thus, the second hierarchical regression analysis
included self-esteem at the baseline on Step 1; age, baseline body esteem, and baseline TV
exposure were entered as covariates on Step 2; Time 2 TV viewing, race, and gender were
entered on Step 3; interaction terms for Time 2 TV viewing by race, Time 2 TV viewing
by gender, and race by gender were entered on Step 4; and the three-way interaction term
representing Time 2 TV viewing by race by gender was entered on the fifth step.
The results of the regression analysis are reported in Table 4. As can be seen, Time 1
self-esteem was a marginal predictor of Time 2 self-esteem scores. Baseline self-esteem
accounted for 9% of the variance in children’s self-esteem at follow-up, Fchange(1, 394) =
3.43, p = .06. The covariates (Step 2), as a set, did not significantly predict children’s self-
esteem. The predictor variables entered on Step 3 collectively accounted for an additional
2% of the variance in children’s self- esteem, Fchange(1, 388) = 2.81, p = .04. In particular,
Time 2 television viewing was negatively related to children’s self-esteem. The two-way
interactions between Time 2 TV viewing by race, Time 2 TV viewing by gender, and gen-
der by race approached significance, Fchange(1, 385) = 2.20, p = .06. The contribution of the
three-way interaction between television viewing, sex, and race (Step 5) was statistically
Table 4. Summary of Hierarchical Regression Analysis for Variables Predicting Self- Esteem
(N = 396)
B SE βt∆R2
Step 1
Time 1 self-esteem .11 .06 0.09** 1.85 .01**
Step 2
Age -.02 .03 -0.03 -0.64
Baseline Body esteem .00 .02 0.00 0.06
Time 1 TV viewing .00 .00 0.06 1.19 .00
Step 3
Time 2 TV viewing -.02 .01 -0.94** -2.57
Race -.08 .07 -0.08 -1.15
Gender -.13 .07 -0.13** -1.97 .02**
Step 4
Time 2 TV × race .01 .01 0.66 1.98**
Time 2 TV × gender .02 .01 1.11 2.95**
Race × gender .05 .03 0.14 1.53 .02**
Step 5
Time 2 TV × race × gender -.01 .00 -0.95 -2.44*** .02***
Note: All coefficients are from the full model.
**p < .05. ***p <.01.
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 351
significant, Fchange(1, 384) = 5.97, p = .01 explaining an additional 2% of the variance. The
entire model explained 25% of the variance in children’s follow-up self-esteem scores.
Hypothesis 2 predicted that television exposure would significantly predict higher self-
esteem among boys, but lower self-esteem among girls. In order to fully verify the form of
the significant two-way interaction between Time 2 TV viewing and gender as reported
above, we computed a simple slopes analysis (Aiken & West, 1991). The analysis revealed
a disordinal interaction, such that Time 2 television viewing was negatively related to
girls’ self-esteem (β = -.20, p = .01), but not significantly related to boys’ self-esteem
scores (β = .00, p = ns).
Our first research question asked if there was a racial difference in the relationship between
television exposure and subsequent self-esteem. Another simple slopes analysis was com-
puted to examine the form of the significant two-way interaction between Time 2 television
viewing and race. The analysis revealed another disordinal interaction, such that Time 2
television viewing was negatively related to Black children’s self-esteem (β = -.23, p = .00)
but not significantly related to White children’s self-esteem scores (β = .10, p = ns).
Our last research question asked if gender and race would moderate the relationship
between television exposure and children’s self-esteem. A simple slopes analysis was con-
ducted to examine the form of the significant three-way interaction as reported in Table 4.
The analysis revealed that Time 2 television viewing significantly predicted decreased
self-esteem in White girls (β = -.30, p = .05), Black girls (β = -.35, p = .01) and Black boys
(β = -.26, p = .04). In contrast, Time 2 television viewing predicted increased self-esteem
in White boys (β = .38, p = .01). Thus, our analyses reveal that television viewing signifi-
cantly predicts children’s self-esteem (Hypothesis 1), but this relationship is dependent
upon children’s gender (Hypothesis 2) and race (Research Questions 2 and 3).
Discussion
We sought to determine whether changes in television use would be associated with self-
esteem measured 1 year later in a sample of White and Black boys and girls. We found that
television exposure predicted a decrease in self-esteem for all children except White boys.
A look at the effect size, as estimated by the change in R2, points to the fact that an televi-
sion viewing is a weak-to-moderate predictor of lowered self-esteem for White girls, Black
girls, and Black boys, and a weak-to-moderate predictor of self-esteem in White Boys.
Why did television exposure predict a decrease in self-esteem among all children except
White boys? There are at least three potential explanations for this pattern. First, the major-
ity of television content serves to reinforce both gender-role and racial stereotypes
(Herrett-Skjellum & Allen, 1995). Indeed, males are portrayed as powerful, strong, and
rational whereas females are portrayed as frail, emotional, and sensitive (Thompson &
Zerbinos, 1997). Black male characters are disproportionately shown as buffoons, or as
menacing and unruly youths, and Black female characters are typically shown as exotic
and sexually available (Marable, 1996). The media comparisons made available for White
boys, therefore, are quite positive in nature. Social identity theory would predict that expo-
sure to these messages serve to bolster the self-concept of White boys because their basis for
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
352 Communication Research 39(3)
comparison reveals that prestige, power, and wealth are commonplace and easily achiev-
able for White males on television. Indeed, research demonstrates that upward compari-
sons can actually be beneficial to people when they are led to believe that attainment of the
depicted achievements is possible (see Suls, Martin, & Wheeler, 2002).
A second possible explanation concerns the nature of the portrayal of White female
characters and African American characters on television. Returning to the idea that televi-
sion serves to reinforce gender and racial stereotypes, social identity theory would predict
that White girls and Black children are using these messages as a basis for self-evaluation
that, in turn, negatively impact their self-esteem. Given that media messages about White
girls and Black children are often negative, the next question to ask is why would children
use these messages as a basis of comparison? Milkie (1999) argues that viewers struggle to
avoid self-evaluations with media messages because the mass media alter societal ideas
about what is normative. If children believe that others (e.g., peers, family) use such mes-
sages to evaluate them, White girls and Black children cannot simply ignore mass media
messages as a comparative referent. Future research should not only continue to examine
the kinds of messages children are extracting from the mass media, but also examine what
role peer and family networks share in mitigating or exacerbating comparisons to these
messages and subsequent impact these comparisons have on self-esteem.
A third explanation why White girls and Black children in this sample reported lower
self-esteem could be due to the fact that television viewing is displacing real-life experi-
ences that might build self-esteem. In support of this idea, Harrison (2006) found that
children who watched more than 20 hours of television per week reported significantly
fewer unique self-descriptors than children who watched less than 20 hours. She argued
that heavy television viewing predicted a drop in self-complexity because there is less vari-
ance in the content of available media messages, and television does not depict the diver-
sity found in real life. If we relate these findings to self-esteem, then, children who are not
doing other things besides watching television cannot help but compare themselves to
what they see on the screen. However, children who watch less and interact more, are able
to discover other things about themselves (e.g., fit, funny) that they might not have discov-
ered if they were watching television. Of course, this theory does not explain the increase
in self-esteem for White boys. Clearly, the relationship between television exposure and
children’s self-complexity is an area ripe for future research.
Limitations, Strengths, and Future Directions
Overall, our findings support television consumption has an impact on children’s self-
esteem, at least in the short term. Yet an alternative explanation for the findings reported
here could be due to the specific content to which the children were exposed. In particular,
the Black participants in this sample could be watching programming which exacerbates
negative portrayals of African Americans. For example, Ward (2004) examined the impact
of multiple forms of media use (e.g., exposure to music videos, sports, Black-oriented
television) on self-esteem and racial self-esteem among 156 African American high school
students. She found that exposure to music videos, but not exposure to prime-time
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 353
television overall, predicted lower self-esteem and racial self-esteem among both genders.
Music videos often rely on shortcuts and cultural stereotypes because they are a story-
telling format with little time to devote to deep characterizations; thus, it is no surprise that
Ward found that music videos were a predictor of low self-esteem. A limitation to this
study, then, is that genre distinctions were not assessed. Nonetheless, future research
should continue to measure overall time spent with television to assess whether television
viewing is displacing key activities and interactions that would serve to nurture children’s
self-esteem.
In spite of this key limitation, our study does possess the strength of a longitudinal
design. The data suggest that television exposure positively predicts children’s self-esteem,
independent of age, initial self-esteem, and body satisfaction. Because longitudinal panel
studies are not experiments, however, causal claims must be considered with caution. It
could be argued, for instance, that individuals with low self-esteem seek out media con-
sumption in an effort to feel better about themselves or to take their minds off of their
problems. If that were the case here, self-esteem assessed in Wave 2 should predict televi-
sion viewing during the same time period. We conducted regression analyses to test this
possibility, and found that controlling for age, body satisfaction, and baseline television
exposure, self-esteem assessed in Wave 2 did not predict television viewing. The fact that
self-esteem did not predict television viewing for the race/gender subgroups does not
negate the possibility of a reciprocal causal relationship; it merely suggests that for these
children, reciprocation between television exposure and self-esteem does not yet play a key
role. We may not see a reciprocal relationship for all children until they mature and form
their identities.
Despite its limitations this study makes several unique contributions to the literature on
media exposure and self-esteem among children. First, the use of a racially diverse sample
allowed for investigation of potential racial differences. Second, the use of a global self-
esteem measure allowed us to assess the impact of the media on other dimensions of self-
esteem (e.g., cognitive, social) independent of body esteem. Third, the use of a longitudinal
design extends existing cross-sectional research linking media exposure with self-esteem
in children. In particular, our initial analysis assumed that optimal interval for examining
the effect of television viewing on children’s self-esteem would be 1 year. However, our
results show that this time span may be inappropriate. Given elementary school children’s
rate of development, messages and habits from a year ago may have exerted their impact
on self-perceptions, and those effects have gone. Thus, future research should continue to
examine the dynamic nature of self-esteem in young children using time frames shorter
than 1 year.
Authors’ Note
We thank Amy Marske and Veronica Hefner for overseeing data collection.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interests with respect to the authorship and/or
publication of this article.
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
354 Communication Research 39(3)
Funding
This research was funded by a Scholar Faculty award to the second author from the William T.
Grant Foundation.
Note
1. It should also be noted that this model eliminates the need to use a raw change scores (in this
case, change in television viewing and change in self-esteem from Time 1 to Time 2) which
is considered to produce misleading findings (for further discussion, see Cohen et al., 2003;
Edwards, 2001).
References
Abrams, D., & Hogg, M. (1990). An introduction to the social identity approach. In D. Abrams &
M. Hogg (Eds.), Social identity theory: Constructive and critical advances (pp. 1-9). Hertfordshire, UK:
Harvester Wheatsheaf.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury
Park, CA: SAGE.
Baumeister, R. F. (1999). The self in social psychology. Philadelphia, PA: Taylor and Francis.
Baumeister, R. F., Campbell, J. D., Krueger, J. I., & Vohs, K. D. (2003). Does high self-esteem cause bet-
ter performance, interpersonal success, happiness, or healthier lifestyles? Psychological Science in the
Public Interest, 4, 1-44.
Berndt, T. J. (2002). Friendship quality and social development. Current Directions in Psychological Sci-
ence, 11, 7-10.
Buijzen, M., & Valkenburg, P. M. (2003). The effects of television advertising on materialism, parent-child
conflict, and unhappiness: A review of research. Applied Developmental Psychology, 24, 437-456.
Calvert, S. L. (1999). Children’s journeys through the information age. Boston, MA: McGraw-Hill.
Centers for Disease Control & Prevention. (2009). Parents get in the game: Parents can play an active role
in children’s long-term health. Retrieved from http:// www.cdc.gov/youthcampaign
Cheung, C. K., & Chan, C. F. (1996).Television viewing and mean world value in Hong Kong’s adoles-
cents. Social Behavior and Personality, 24, 351-364.
Christakis, D. A., Gilkerson, J., Richards, J. A., Zimmerman, F. J., Garrison, M. M., Xu, D., et al. (2009).
Audible television and decreased adults words, infant vocalizations, and conversational turns. Archives
of Pediatric Adolescent Medicine, 163(6), 554 -558.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis
for the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum.
Collins, M. E. (1991). Body figure perceptions and preferences among preadolescent children. Interna-
tional Journal of Eating Disorders, 10, 199-208.
Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco, CA: Freeman.
Corenblum, B., & Anis, R. (1993). Development of racial identity in minority and majority children: An
affect discrepancy model. Canadian Journal of Behavioral Science, 25, 499-521.
Davis, J. A. (1985). The logic of causal order. Series: Quantitative applications in the social sciences
(No. 55). Newbury Park, CA: SAGE.
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 355
Dixon, T., & Linz, D. (2006). Overrepresentation and underrepresentation of African Americans and Latinos
as lawbreakers on television news. Journal of Communication, 50, 131-154.
Dohnt, H., & Tiggemann, M. (2006). The contribution of peer and media influences to the development of
body satisfaction and self-esteem in young girls: A prospective study. Developmental Psychology, 42,
929-936.
Edwards, J. R. (2001). Ten difference score myths. Organizational Research Methods, 4, 265-287.
Fryberg, S. (2003). Really? You don’t look like an American Indian: Social representations and social
group identities. Dissertation Abstracts International,64(1549), 3B.
Gerbner, G., Gross, L., Morgan, M., Signorelli, N., & Shanahan, J. (2002). Growing up with television:
Cultivation processes. In J. Bryant, & Zillmann (Eds.), Media effects: Advances in theory and research
(2nd ed., pp. 43-67). Mahwah, NJ: Lawrence Erlbaum.
Gray-Little, B., & Hafdahl, A. R. (2000). Factors influencing racial comparisons of self-esteem: A quantita-
tive review. Psychological Bulletin, 126, 26-54.
Greenberg, B. S., Eastin, M., Hofschire, L., Lachlan, K., & Brownell, K. D. (2003). Portrayals of overweight
and obese individuals on commercial television. American Journal of Public Health, 93, 1342-1348.
Harrison, K. (2000). Television viewing, fat stereotyping, body shape standards, and eating disorder symp-
tomology in grade school children. Communication Research, 27, 617-640.
Harrison, K. (2006). Scope of self: Toward a model of television’s effects on self-complexity in adoles-
cence. Communication Theory, 16, 251-279.
Harrison, K., & Bond, B. (2007). Gaming magazines and the drive for muscularity in preadolescent boys:
A longitudinal examination. Body Image, 4, 269-277.
Harrison, K., & Hefner, V. (2006). Media exposure, current and future body ideals, and disordered eating
among preadolescent girls: A longitudinal panel study. Journal of Youth and Adolescence, 35, 153-163.
Hart, J. G. (1985). LAWSEQ: Its relation to other measures of self-esteem and academic ability. British
Journal of Educational Psychology, 55, 167-169.
Harter, S. (1999). The construction of the self: A developmental perspective. New York, NY: Guilford.
Hawkins, R. P., & Pingree, S. (1981). Using television to construct social reality. Journal of Broadcasting,
25, 347-364.
Herrett-Skjellum, J., & Allen, M. (1996). Television programming and sex stereotyping: A meta-analysis.
Communication Yearbook, 19, 157-185.
Hogg, M., Terry, D., & White, K. (1995). A tale of two theories: A critical comparison of identity theory
with social identity theory. Social Psychology Quarterly, 58, 255-269.
Howard, M. O., & Jensen, J. M. (1999). Inhalant use among antisocial youth; prevalence and correlates.
Addictive Behaviors, 24, 59-74.
Hughes, H. (1984). Measures of self-concept and self-esteem for children ages 3-12 years: A review and
recommendations. Clinical Psychology Review, 4, 657-692.
Kowalski, R. M. (2003). The emergence of ethnic and racial attitudes in preschool-aged children. The Journal
of Social Psychology, 143, 677-690
Kuhn, D., Nash, S. C., & Brucken, L. (1978). Sex role concepts of two- and three-year-olds. Child Develop-
ment, 49, 445-451.
Lawrence, D. (1981). Development of a self-esteem questionnaire. British Journal of Educational Psychol-
ogy, 51, 245-251.
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
356 Communication Research 39(3)
Marable, M. (1996). Reconciling race and reality. In E. Dennis & E. Pease (Eds.), The media in black and
white (pp. 11-17). New Brunswick, NJ: Transaction.
Martins, N. (2008). Social aggression on television and its relationship to children’s aggression in the class-
room. Unpublished doctoral dissertation. University of Illinois at Urbana-Champaign.
Mastro, D. E., & Greenberg, B. (2000). The portrayal of racial minorities in prime-time television. Journal
of Broadcasting & Electronic Media, 44, 690-703.
Mastro, D. E. & Kopacz, M. A. (2006). Media representations of race, prototypicality, and policy reason-
ing: An application of self-categorization theory. Journal of Broadcasting & Electronic Media, 50,
305-322.
McArdle, J. J., & Woodcock, R. W. (1997). Expanding test-retest designs to include developmental time-
lag components. Psychological Methods, 2, 403-435.
McGee, R., & Williams, S. (2000). Does low self-esteem predict health compromising behaviors among
adolescents? Journal of Adolescence, 23, 569-582.
Milkie, M. A. (1999). Social comparisons, reflected appraisals, and mass media: The impact of pervasive
beauty images on Black and White girls’ self-concepts. Social Psychology Quarterly, 62, 190-210.
National Association for Self-Esteem. (2005). Review of research. Fulton, MD: Author.
Polyce-Lynch, M., Myers, B. J., Kliewer, W., & Kilmartin, C. (2001). Adolescent self-esteem and gender:
Exploring relations to sexual harassment, body image, media influence, and emotional expression. Journal
of Youth and Adolescence, 30, 225-244.
Potter, W. J. (1991). Examining cultivation from a psychological perspective. Communication Research,
18, 92-113.
Ricciardelli, L. A., & McCabe, M. P. (2001c). Self-esteem and negative affect as moderators of sociocultural
influences on body dissatisfaction, strategies to decrease weight, and strategies to increase muscle tone
among adolescent boys and girls. Sex Roles, 44, 89-207.
Rideout, V., Foehr, U. G., & Roberts, D. F. (2010). Generation M2: Media in the Lives of 8-18 year-olds.
Washington DC: Kaiser Family Foundation.
Rosenberg, M. (1972). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.
Sahlstein, E., & Allen, M. (2002). Sex differences in self-esteem: A meta-analytic assessment. In M. Allen,
R. Preiss, B. Gayle, & N. Burrell (Eds.), Interpersonal communication research: Advances through
meta-analysis (pp. 59-72). Mahwah, NJ: Erlbaum.
Scheier, L. M., Botvin, G. J., Griffin, K. W., & Diaz, T. (2000). Dynamic growth models of self-esteem and
adolescent alcohol use. Journal of Early Adolescence, 20, 178-209.
Shrum, L. J. (1996). Psychological processes underlying cultivation effects. Human Communication
Research, 22, 482-509.
Signorielli, N. (1993). Prime-time violence 1993-2001: Has the picture really changed? Journal of Broadcast-
ing & Electronic Media, 47, 36-57.
Signorielli, N. (1997). Reflections of girls in the media: A content analysis. Menlo Park, CA: Kaiser Family
Foundation.
Signorielli, N., & Kahlenberg, S. (2001). Television’s world of work in the nineties. Journal of Broadcast-
ing & Electronic Media, 45, 4-22.
Signorielli, N., McLeod, D., & Healy, E. (1994). Gender stereotypes in MTV commercials: The beat goes
on. Journal of Broadcasting & Electronic Media, 38, 91-101.
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
Martins and Harrison 357
Spitzer, B. L., Henderson, K.,A., & Zivian, M.,T. (1999). Gender differences in population versus media
body sizes: A comparison over four decades. Sex Roles, 40, 545-565.
Stroman, C. A. (1986). Television viewing and self-concept among Black children. Journal of Broadcast-
ing & Electronic Media, 30, 87-93.
Suls, J., Martin, R., & Wheeler, L. (2002). Social comparison: Why, with whom, and with what effect?
Current Directions in Psychological Science, 11, 159-163.
Taijfel, H. (1982). Differentiation between social groups: Studies in the social psychology of intergroup
relations. London, UK: Academic.
Tanimura, M., Okuma, K., & Kyoshima, K. (2007). Television viewing, reduced parental utterance, and
delayed speech development in infants and young children. Archives of Pediatric and Adolescent Medi-
cine, 161, 618-619.
Thompson, T. L., & Zerbinos, E. (1997). Television cartoons: Do children notice it’s a boy’s world? Sex
Roles, 37, 415-432.
Tiggemann, M., & Wilson-Barrett, E. (1998). Children’s figure ratings: Relationship to self-esteem and
negative stereotyping. International Journal of Eating Disorders, 23, 83-88.
Tuner, J. (1982). Towards a cognitive redefinition of the social group. In H. Tajfel (Ed.), Social identity and
intergroup relations. Cambridge, UK: Cambridge University Press.
U.S. Department of Health and Human Services. (2009). Building self-esteem: A self-help guide. Retrieved
from http://www.mentalhealth.samhsa.gov
Viner, R. M., & Cole, T. J. (2006). Adult socioeconomic, educational, social, and psychological outcomes
of childhood obesity: a national birth cohort study. British Medical Journal, 330, 1354 -1358.
Ward, L. M. (2004). Wading through the stereotypes: Positive and negative effects of media use on Black
adolescents’ conceptions of self. Developmental Psychology, 40, 284-294.
Wiggins, J. D. (1987). Self-esteem, earned grades, and television viewing habits of students. School Counselor,
35, 128-133.
Wilburn, V. R., & Smith, D. E. (2005). Stress, self-esteem, and suicidal ideation in late adolescence. Adoles-
cence, 40, 33-45.
Williams, J. M., & Currie, C. (2000). Self-esteem and physical development in early adolescence: Pubertal
timing and body image. Journal of Early Adolescence 20, 129-149.
Wilson, B. J., Martins, N., & Marske, A. (2005). Children’s and parents’ fright reactions to kidnapping
stories in the news. Communication Monographs, 72, 46-70.
Bios
Nicole Martins (PhD, University of Illinois at Urbana-Champaign, nicomart@indiana.edu) is
an assistant professor in the Department of Telecommunications at Indiana University.
Kristen Harrison (PhD, University of Wisconsin-Madison, krishar@illinois.edu) is an associate
professor in the Department of Communication at the University of Illinois at Urbana-Champaign.
at UNIVERSITY OF MICHIGAN on April 24, 2012crx.sagepub.comDownloaded from
