Content uploaded by Renee Botta
Author content
All content in this area was uploaded by Renee Botta on Jun 21, 2016
Content may be subject to copyright.
10.1177/0093650203252409ARTICLECOMMUNICATION RESEARCH • June 2003Nathanson, Botta • Mediation and Body Image Disturbance
AMY I. NATHANSON
RENÉE A. BOTTA
Shaping the Effects of Television
on Adolescents’ Body Image
Disturbance
The Role of Parental Mediation
The purpose of this study was to explore the relationship between several
forms of parental mediation and adolescents’ television processing, emotions
while viewing,and body image disturbance. A survey of both parents and ado-
lescents (at two stages) revealed that most traditional measures of parental
mediation were not significantly related to any adolescent outcomes.However,
measures that captured the extent to which parents discussed characters’
appearance and body size were consistent predictors of the same outcomes.The
data revealed that parental mediation of this incidental content—even if it
criticized the television images—encouraged deep processing of the images
and negative emotions, which in turn led to indicators of body image distur-
bance. The implications of these findings for research on both parental media-
tion and body image are discussed.
Keywords: parental mediation; body image; television; adolescents
Research suggests that adolescent females who view more television are
more dissatisfied with their bodies (Harrison & Cantor, 1997) and are more
likely to show symptoms of eating disorders than are females who view less
television (Botta, 2000a; Stice, Schupak-Neuberg, Shaw, & Stein, 1994).
Other research indicates that the way we process body images on television is
a stronger predictor of eating-disordered thoughts and behaviors than mere
exposure (Botta, 1999a, 1999b, 2000a, 2000b; Cusumano & Thompson, 2001;
Harrison, 2000). However, a separate body of work on parental active media-
tion indicates that media effects are less likely when parents talk with their
304
COMMUNICATION RESEARCH, Vol. 30 No. 3, June 2003 304-331
DOI: 10.1177/0093650203252409
© 2003 Sage Publications
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
children about television (Nathanson, 2001a). The purpose of this study was
to explore the role of parents’ active mediation in either facilitating or miti-
gating the effects of television processing on adolescents’ attitudes and
behaviors that are symptomatic of eating disorders.
The Predictors of Body Image Disturbance
Slade (1994) defined body image as the “loose mental representation of the
body” (p.500). Body image is an overall concept consisting of related but dis-
tinct dimensions addressing feelings, thoughts, and perceptions about body
size and shape. Body Image Disturbance (BID) is when those feelings,
thoughts, and perceptions become distorted or agitated and can include
actions such as unhealthy and disordered eating. For example, disturbed
body image feelings include being dissatisfied with one’s body (body dissatis-
faction), disturbed body image perceptions include overestimating one’s body
size (perceptual distortion), disturbed body image thoughts include chronic
thoughts about weight loss and weight gain (drive to be thin), and disturbed
body image actions include eating-disordered behaviors such as bingeing and
purging foods (bulimia) and restrained eating accompanied with dangerous
weight loss (anorexia). It is important to understand the predictors of BID so
that we can develop effective interventions.
Family communication. Communication within the family plays an
important role in developing and preventing eating disorders. Researchers
have found that families with anorexic members are more likely to empha-
size obedience and conformity than other families (Bruch, 1982). Families
with bulimic members have lower quality communication (Shisslak,
McKeon, & Crago,1990),less open communication (Killian, 1994), and bursts
of overt conflict (Humphrey, 1989), which result in negative self-image and
self-directed hostility. Eating-disordered behaviors also have been predicted
by parents’ preoccupation with weight (Connors, 1996; Taylor et al., 1998)
and parental comments about weight, such as encouraging weight loss and
giving body-related compliments (Wertheim, Mee, & Paxton, 1999).
Family communication also can play a positive role in disrupting eating-
disordered behaviors. For example, Botta and Dumlao (2002) found that open
communication and skilled conflict resolution between fathers and daugh-
ters might offset anorexic symptomatology by allowing daughters more
autonomy and thus less need to control their eating. Furthermore, Botta and
Dumlao found that skilled conflict resolution between fathers and daughters
is less likely to result in bulimic symptomatology because unresolved conflict
adds to a sense of powerlessness in which the daughter may direct the
305
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
hostility toward herself in a cycle of bingeing and purging.Family communi-
cation and conflict resolution also may trigger or prevent other factors that
lead to BID (Botta & Dumlao, 2002; Frank & Jackson, 1996; McNamara &
Loveland, 1990).
Television processing. Visual media play an important role in eating disor-
ders by promoting unrealistic images of thinness and beauty. The British
Medical Association Board of Science and Education (2000) reported that
models and actresses in the 1990s had 10% to 15% body fat, whereas the aver-
age body fat for a healthy woman is 22% to 26%. Rubinstein and Caballero
(2000) found that the weight of Miss America winners has decreased by 12%
since the 1920s. Indeed, the body size and shape of Miss America winners
have declined to a point where the winners are regularly “undernourished”
according to World Health Organization standards (Rubinstein & Caballero,
2000). Adding to the problem of overly thin images is a finding by King,Touyz,
and Charles (2000) that young women who feel bad about their bodies think
that celebrities are thinner than the celebrities really are.
The thinning of women’s bodies in media during recent decades has coin-
cided with an increase in eating disorders among 15- to 24-year-old females
(Lucas, Crowson, OaposFallon, & Melton, 1999). And, Becker and Burwell
(1999) found that after satellite television was introduced in Fiji, bulimia
rates increased from 3% to 15% among Fijian teenage girls. The effects of
media exposure may be particularly strong for adolescents who lack social
support from parents (Stice, Spangler, & Agras, 2001). Despite the fact that
television becomes less important than some other media as youngsters
reach adolescence (Avery, 1979; Larson,1995), the research linking television
viewing to body image-related effects indicates that this medium remains an
important agent of socialization.
In addition, evidence coupling BID with the ways viewers process thin
media images has emerged (Botta, 1999a). First, watching thin television
characters may prompt viewers to make social comparisons between them-
selves and the characters. In this way, viewers think about how their bodies
compare to the bodies of the models and actors they see on television. This
comparison motivates viewers to think about what they might do to make
their bodies look like the ones they see on television.
Second, viewers may critically view the unrealistically thin images.Criti-
cal viewing has been defined as “an identification of the social context of the
concepts presented in the media” (Cohen, 1994, p. 98). According to Potter
(1998), critical viewing skills are among the more advanced skills required of
the media literate. The critical viewer is skeptical of media content and
actively evaluates and reflects on it (Anderson, 1983). Most media literacy
306
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
curricula teach critical viewing of television production and the story content
itself (Singer & Singer, 1998), under the assumption that critical viewers are
less vulnerable viewers.
When viewers question or challenge the way that characters look,they are
engaging in critical viewing of the incidental content. Incidental content is
peripheral to the main story line (Botta 2000a; Lowe & Durkin, 1999;
Meadowcroft & Reeves, 1989) and includes stereotypes, such as the thin
ideal. Indeed, most body-related content is presented incidentally. When
viewers engage in critical viewing of the incidental content, they question
whether characters’ appearances are realistic and become acutely aware of
the size and shape of television models and actors.
Third, viewers can engage in critical viewing of the central content. Cen-
tral content includes elements that are essential to plot, such as the story
line, content, and character development (Botta, 2000a; Collins, 1970). When
viewers focus on central content, they are less distracted by incidental con-
tent, including the size and shape of the characters’ bodies. As a result,
increased processing of central content leaves less room to process incidental
content. In this way, viewers are more likely to be affected by the central con-
tent of the programming than the incidental thin media images.
Although each form of processing requires viewers to be active consumers
of television messages, they often are related to distinct outcomes. For
instance, making social comparisons with thin characters is positively
related to anorexic behaviors, a drive to be thin, and body dissatisfaction
(Botta, 1999a) and negatively related to self-reported physical attractiveness
(Martin & Kennedy, 1993). Surprisingly, critically viewing the incidental,
body image-related content also is related to some aspects of BID (Botta,
1999a, 2000a, 2000b; Milkie, 1999). Ordinarily, we would expect that adoles-
cents who question whether thin images are realistic would resist those
images and maintain positive self-perceptions. However, it seems that the
process of devoting one’s attention to the thin images,regardless of one’s eval-
uation of the images, is sufficient to set into motion an unhealthy response to
viewing (Botta, 2000b). On the other hand,critically viewing the central con-
tent is related to fewer symptoms of BID (Botta, 2000b), perhaps because all
of one’s resources are devoted toward understanding the plot.
Emotions. The emotions viewers experience while viewing thin images
may predict BID.Negative affect is one of the distinguishing features among
individuals feeling normal and disordered levels of body dissatisfaction
(Connors, 1996).Research also has indicated that negative affect can induce
eating-disordered behaviors (e.g., Polivy, Herman, & McFarlane, 1994; Stice
et al., 1994). Moreover, negative affect can be the consequence of exposure to
307
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
thin-ideal media for some adolescents (Harrison, 2001; Stice et al.,2001). As a
result, it is possible that exposure first affects emotions,which in turn affect
BID.
It should be noted that such affect is not the same as the affective compo-
nent of BID, which includes how dissatisfied one feels about one’s body (e.g.,“I
feel fat,” and “I feel anxious about the size of my hips”). Negative emotions
that induce BID are not specific to one’s body. These feelings can reflect a gen-
eral mood or a reaction to television content. Adolescents with increased BID
have reported experiencing increased negative affect when viewing thin
media images (Botta, 2000a). This finding seems consistent with media-
effects scholars’ supposition that negative emotions experienced while view-
ing television intensify the experience and therefore the effects of the experi-
ence (Potter, 1998). Thus, negative emotions aroused during viewing may
serve as another important motivator for increased BID.
Parental Active Mediation
Research on parental mediation suggests that parents can shape the nature
of media effects on their children. Although the parental mediation construct
is multidimensional, the most attention has been devoted to understanding
the consequences of parent-child discussion of television (or active media-
tion). In some cases,active mediation enhances positive effects, such as learn-
ing from educational or prosocial television (Desmond, Singer, Singer, Calam,
& Colimore, 1985;Salomon, 1977). Active mediation also has been effective in
reducing television-induced aggression (Corder-Bolz, 1980; Hicks, 1968;
Nathanson, 1999) and discouraging children’s acceptance of stereotypical
gender roles (Nathanson, Wilson, McGee, & Sebastian, 2002).
Not all active mediation efforts are successful. Nathanson (2001b) found
that children interpreted active mediation to mean that their parents
approved of and enjoyed the mediated content. In fact, other work has shown
that active mediation can produce unexpected results. For example, Austin
and Nach-Ferguson (1995) found that active mediation was associated with
greater awareness of brands of alcohol among children. Further, Huesmann,
Eron, Klein, Brice, and Fischer (1983) found that children who received infor-
mation about violent television programs (e.g., information regarding cam-
era techniques) were more aggressive than other children.
Perhaps the reason for these discrepancies has to do with the nature of the
active mediation messages. Austin, Bolls, Fujioka, and Engelbertson (1999)
identified two types of messages. Negative mediation refers to messages that
refute or condemn the mediated content (e.g., “That’s not real,” and “That
show is wrong”). Positive mediation occurs when parents endorse or praise
308
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
the content (e.g.,“I love this show,” and “He sure is cool”). Nathanson (2001a)
suggested that parents’ comments that are neither negative nor positive (i.e.,
neutral active mediation) constitute a third type of active mediation. This cat-
egory includes messages that refer to content without relaying parents’ opin-
ions, such as providing supplemental information or asking children ques-
tions (e.g., “This show is filmed in New York,” and “What do you think will
happen next?”).
The puzzling finding that mediation can produce both desirable and unde-
sirable effects becomes more sensible when we consider this typology of
active mediation message types. Austin and Nach-Ferguson’s (1995) active
mediation scale includes elements of neutral mediation, and Nathanson’s
(2001b) measure simply taps the extent to which parents talk to their chil-
dren about violent television. In both cases, if children were thinking of
instances in which their parents made neutral or positive statements about
the content in question, then the authors’ findings are not surprising. Fur-
ther, the informational messages of Huesmann et al. (1983) were oriented
around getting the children to pay closer attention to violent television
scenes. As a result,the practice of having children devote attention to violent
images may have encouraged the children to store them in memory. These
findings suggest that the effects of active mediation may depend on its
valence, with some types enhancing effects and others decreasing them.
Although the majority of work on parental mediation has been done with
young children, some research has explored its role among adolescents.Much
of this work indicates weak or nonsignificant relationships between some
forms of parental mediation and outcomes, such as aggression (McLeod,
Atkin, & Chaffee, 1972a, 1972b; Nathanson, 2001c). However, Austin (1993)
found that active mediation was significantly related to a variety of political
attitudes and behaviors, suggesting that parents may be influential in con-
texts deemed important to adolescents. Further, certain styles of active medi-
ation may be more effective than others. Nathanson (2002) suggested that
adolescents might respond to neutral and interactive approaches to media-
tion because these styles allow adolescents to contribute to the conversation
and are not condescending in tone.
Parental Mediation, Television Processing, Emotions, and BID
The influence of parental mediation on adolescents’ BID has not yet been
explored. This study examined the relationship between a variety of parental
active mediation styles and adolescents’ television processing (both pertain-
ing to incidental and central content), emotions while viewing,and BID.Our
primary interest was to explore the effectiveness of mediation of central
309
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
content (i.e., messages about the plot or behaviors of main characters) and
mediation of incidental content (i.e., messages about the thinness-depicting
images).
Most previous mediation research focuses on the mediation of central con-
tent. In these studies,the targets of mediation (e.g.,a character’s aggressive
behavior) were both (a) part of the central story, and (b) either the same or
very similar to the effects the experimenters wished to influence (e.g., view-
ers’ aggressive behavior).
However, mediation of central content also could be related to media
effects that have little to do with the central plot. For example, adolescents
whose parents encourage them to focus on the plot of a program that features
unrealistically thin actresses should be less affected by the thin images than
other adolescents. This form of mediation may be effective because it discour-
ages adolescents from focusing on the thin images and encourages them to
process plot-relevant details instead.
For adolescents,a neutral approach in which parents do not express their
own opinions (what we call neutral mediation of central content) should be
particularly effective because it does not impose the parents’ attitudes on the
adolescent and is not condescending or threatening. Positive comments about
the central content (positive mediation of central content) also should encour-
age adolescents to attend to the plot details rather than the characters’
appearances. Adolescents who hear this form of mediation may appreciate
the central elements of the program, such as the complexity of the story. How-
ever, they should not exhibit any evidence that they have devoted much
attention or thought to the characters’ appearances.
Although negative comments about central content (negative mediation of
central content) also focus on plot elements, the tone of this approach may
produce a backlash among adolescents. Previous work shows that condemna-
tions of content can produce undesirable effects among this population
(Nathanson, 2002). It is possible that adolescents resent hearing a lecture
that expresses parents’ negative attitudes. Other work also finds that older
children may start to ignore this kind of mediation and require different
approaches (Nathanson & Yang, 2003).As a result, we may find that negative
mediation of central content leads adolescents to disregard parents’ com-
ments about the content and focus on the incidental content instead. Or, in an
effort to distinguish themselves from their parents, perhaps adolescents who
hear negative comments form more positive attitudes toward the mediated
program, including more positive attitudes toward and acceptance of charac-
ters’ appearances. In general, because previous work shows that negative
mediation of central content produces undesirable outcomes, we predicted
the same kind of relationship in this study.
310
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Hypothesis 1a: Neutral mediation of central content will be related to less
body image processing (i.e., less processing of body-related informa-
tion, fewer social comparisons), more processing of central content,
fewer negative emotions while viewing, and less BID (i.e., less body dis-
satisfaction, less drive for thinness, and fewer symptoms of anorexia
and bulimia).
Hypothesis 1b: Positive mediation of central content will be related to less
body image processing (i.e., less processing of body-related informa-
tion, fewer social comparisons), more processing of central content,
fewer negative emotions while viewing, and less BID (i.e., less body dis-
satisfaction, less drive for thinness, and fewer symptoms of anorexia
and bulimia).
Hypothesis 1c: Negative mediation of central content will be related to
more body image processing (i.e., more processing of body-related infor-
mation, more social comparisons), less processing of central content,
more negative emotions while viewing, and more BID (i.e., more body
dissatisfaction, more drive for thinness, and more symptoms of
anorexia and bulimia).
Further, we explored the possibility that body image processing and neg-
ative emotions are intervening variables in the relationship between medi-
ation and BID. That is, we examined whether the effect of mediation on BID
is indirect and channeled through the processing or emotions that are expe-
rienced during viewing. As a result, the following research question was
posed:
Research Question 1: Is the relationship between mediation of central con-
tent and adolescents’ BID indirect (i.e., mediated by adolescents’ televi-
sion processing and negative emotions)?
To date, researchers have not studied parental mediation of incidental
content. However, the body image research suggests that actions that draw
adolescents’ attention to body image issues will encourage body image pro-
cessing, negative emotions, and BID. As a result, mediation that focuses on
television characters’ bodies—even if it expresses criticism—may encourage
detrimental effects.
That is, the simple act of making comments about a character’s appear-
ance—whether good, bad, or neutral—should encourage adolescents to
attend to the very images that adolescents should disregard. Their height-
ened attention to the images may initiate social comparison processes and
negative emotions that may in turn lead to BID.This possibility, which could
introduce a considerable challenge to researchers hoping to prevent harmful
media effects, was explored in this study:
311
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Hypothesis 2a: Neutral mediation of incidental content will be related to
more body image processing (i.e., more processing of body-related infor-
mation, more social comparisons), less processing of central content,
more negative emotions while viewing, and more BID (i.e., more body
dissatisfaction, more drive for thinness, and more symptoms of
anorexia and bulimia).
Hypothesis 2b: Positive mediation of incidental content will be related to
more body image processing (i.e., more processing of body-related infor-
mation, more social comparisons), less processing of central content,
more negative emotions while viewing, and more BID (i.e., more body
dissatisfaction, more drive for thinness, and more symptoms of
anorexia and bulimia).
Hypothesis 2c: Negative mediation of incidental content will be related to
more body image processing (i.e., more processing of body-related infor-
mation, more social comparisons), less processing of central content,
more negative emotions while viewing, and more BID (i.e., more body
dissatisfaction, more drive for thinness, and more symptoms of
anorexia and bulimia).
Research Question 2: Is the relationship between mediation of incidental
content and adolescents’ BID indirect (i.e., mediated by adolescents’
television processing and negative emotions)?
Method
Participants and Procedures
Data from adolescent-parent pairs were collected. Both older adolescents
(i.e., college students) and younger adolescents (i.e.,12- to 17-year-olds) were
included. This age range was selected because eating disorders most com-
monly begin in adolescence (Smolak & Levine, 1996),with the average age of
onset for anorexia and bulimia at 15 and 18, respectively (American Psychi-
atric Association, 2000). Further, eating disorders are on the rise among 15-
to 24-year-old females,who are considered the most vulnerable to eating dis-
orders (Lucas et al., 1999).
First, 160 college students (68% female) were recruited from introductory
communication classes at two Midwestern universities in exchange for extra
course credit. Males and females completed different questionnaires,as some
measures were not relevant for each gender. On average the students were 21
years of age. The majority (82%) were Caucasian,although nearly 11% were
African American. Students could earn additional extra credit if one of their
parents and one of their 12- to 17-year-old siblings (if applicable) completed a
survey. College students who did not have a 12- to 17-year-old sibling were
given the option to recruit another parent and 12- to 17-year-old pair. In this
case, the college student did not complete a questionnaire. As a result, our
312
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
data include responses from (a) college students, their parent,and their 12- to
17-year-old sibling, (b) college students and their parent, or (c) a parent and
his or her 12- to 17-year-old child (with no necessary familial relationship
with the college student who recruited them).
After receiving a brief lesson in questionnaire administration, the college
students were asked to deliver the surveys to the other participants and then
return them to the researchers in a sealed envelope. A total of 220 parents
completed a survey. Approximately 81% were mothers whose average age
was 46. About 60% had completed some college. This population was 72%
Caucasian and 21% African American. In addition, 149 of the parents’ 12- to
17-year-olds participated (62% female). The average age of this sample was
16. Approximately 71% classified themselves as Caucasian and 24% reported
that they were African American.
Parent Survey
For the mediation questions,parents were asked to think about their “12- to
17-year-old children who are currently living at home.” If parents did not
have any 12- to 17-year-olds living at home, then they were asked to “think
back to when the college student who gave you this survey was in that age
range and was living at home.” We sought to capture the kind of mediation
style that parents used with 12- to 17-year-olds because body image issues
become very important during this time (Lucas et al., 1999; Smolak & Levine,
1996). This meant, in the case of college students, that we correlated media-
tion practices used several years ago with students’ current processing styles,
emotions, and BID. This practice is justified because (a) college students who
are living away from home are unlikely to receive much parental mediation,
and (b) college students’ processing styles, emotions, and BID are likely the
product of their socialization experiences (including receiving parental medi-
ation). This approach also allowed us to determine whether mediation
received during younger adolescence is related to outcomes experienced dur-
ing older adolescence.
Questions were developed to tap parental mediation messages targeted at
either central or incidental content that were positive, negative,or neutral in
valence. Parents were asked to think about how often they make specific
statements when their children watch television. For each question, response
options ranged from never (coded as 1) to every time (coded as 5).
All of the questions tapping mediation of central content focused on the
plot and the overall theme underlying the program. A principal components
factor analysis with oblimin rotation was performed to determine whether
the set of items reflected negative, positive, and neutral mediation. This
313
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
analysis revealed one item that did not exhibit any strong loadings on any
particular factor. After removing this item, the factor analysis revealed the
three central content factors described below (explaining 67% of variance).
The questions measuring mediation of incidental content emphasized char-
acters’ physical appearances (especially their body size) and eating behav-
iors. A principal components factor analysis with oblimin rotation was per-
formed and revealed three weak items. When the analysis was redone with
these items removed, it revealed three factors reflecting negative, positive,
and neutral mediation of incidental content (explaining 69% of variance).
Neutral mediation of central content. Four items tapped how often parents
encourage their adolescents to think about the most central aspects of pro-
grams, such as the plot and major topic (e.g., “How often do you encourage
your children to try to predict what will happen at the end of the story?” and
“How often do you discuss with your children how the story relates to their
own lives?”). This factor accounted for 42% of variance. Responses to the
questions were averaged to create a scale (α= .82, M= 2.6, SD = .84).
Positive mediation of central content. With three questions, parents were
asked how often they make positive (i.e., supportive, confirming) comments
about central story features, such as the plot or the program more generally
(e.g., “How often do you tell your children that what happens on the show is
realistic?” and “How often do you tell your children that you like the program
they are watching?”). This factor accounted for 11% of variance. Responses to
the three questions were averaged to create a scale (α= .76, M= 2.8,SD = .78).
Negative mediation of central content. With four questions, parents
reported how often they make negative (i.e., critical,condemning, contradic-
tory) comments about the central features of a story (e.g., “How often do you
tell your children that you don’t like the overall message in the program?”
and “How often do you tell your children not to copy what they see on TV?”).
This factor accounted for 14% of variance. Responses to these questions were
averaged to create a scale (α= .83, M= 3.1, SD = .85).
Neutral mediation of incidental content. With five questions, parents
reported how often they encourage their children to think about the inciden-
tal elements of a program, such as characters’ appearance (e.g., “How often do
you seek your children’s opinions about how a character’s body looks?” and
“How often do you encourage your children to think about how thin the
actresses are?”). For each question, parents were asked to consider whether
they encourage these discussions “without offering [their] own opinions.”
314
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
This factor accounted for 46% of the variance. Responses to the five questions
were averaged to create a scale (α= .91, M=1.8, SD = .77).
Positive mediation of incidental content. Three questions were used to
measure how often parents compliment, endorse,or make generally positive
remarks about a character’s appearance (e.g., “How often do you tell your
children that you like how thin an actress looks?” and “How often do you
make positive comments about a character’s physical appearance?”). This
factor accounted for 10.5% of the variance. Responses were averaged to create
a scale (α= .72, M= 2.1, SD = .71).
Negative mediation of incidental content. With four questions, parents
reported how often they criticize, condemn,or make generally negative com-
ments about a character’s appearance (e.g.,“How often do you tell your chil-
dren that the actresses on television are too thin?” and “How often do you
make negative comments about a character’s physical appearance?”). This
factor accounted for 12.5% of the variance. Responses were averaged to create
a scale (α= .79, M=2.1, SD = .82).1
Adolescent Surveys
The surveys for the older adolescents and the younger adolescents were iden-
tical within gender. Across the genders, the surveys were the same except
that the female version of the survey included questions tapping drive for
thinness, anorexia, and bulimia. Unless noted otherwise, adolescents indi-
cated their agreement on scales ranging from never (1) to always (5).
Processing of central content. With eight questions, adolescents indicated
how often they think about information that is central to a program’s story
line while viewing television (e.g., “I fill in ‘missing information’ to help me
figure out why the characters do certain things” and “I wonder what the TV
characters would do if events in the story were somewhat different”).
Responses were averaged to create a scale for the older adolescents (α= .77,
M= 3.2, SD = .53) and the younger adolescents (α= .80, M= 2.9,SD = .67).
Processing of incidental content. With six questions, adolescents reported
how often they think about or question characters’ appearances and eating
behaviors (e.g., “I notice how much the characters eat or do not eat” and “I
think about what TV producers do to make the women look so thin”).
Responses were averaged to create a scale for both the older adolescents (α=
.81,M= 2.4, SD = .71) and the younger adolescents (α= .80,M= 2.1, SD = .72).
315
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Social comparison processing. Adolescents responded to eight questions
measuring how often they compare television characters’ bodies to their own
or their friends’ bodies (e.g., “I think about how my body compares to the TV
characters’ bodies” and “I think about what I might do to have a body similar
to the characters’ bodies”). Responses were averaged to create a scale for the
older adolescents (α= .93, M= 2.7, SD = .88) and the younger adolescents (α=
.93, M= 2.2, SD = .87).
Negative emotions during viewing. Adolescents reported how much they
feel eight different emotions (e.g., guilty, sad, frustrated) when they think
about or question what TV characters look like. The participants used a 5-
point scale ranging from does not describe my feelings at all (1) to describes
my feelings very well (5). Responses were averaged to create a scale reflecting
the experience of negative affect for both the older adolescents (α= .89, M=
2.1, SD = .84) and the younger adolescents (α= .89, M= 1.8, SD = .78).
BID.Four dimensions of BID that have been validated in other work were
measured. Although research shows that measures of perceptual distortion
are less useful discriminators between controls and those clinically diag-
nosed with eating disorders (Slade, 1994), measures of body dissatisfaction,
drive for thinness, anorexia, and bulimia have been shown to be valid and
reliable indicators of BID (Garner & Olmsted, 1991; Slade, 1994; Thompson,
Heinberg, Altabe, & Tantleff-Dunn, 1999). Hence, these four measures were
adapted from Garner and Olmsted (1991) and used in the present study. The
adapted versions of these measures have been used successfully in previous
work (e.g., Botta, 1999a, 2000b).
With nine items,the participants’ body dissatisfaction was assessed.Ado-
lescents indicated how dissatisfied they are with their bodies (e.g., “I think
that my stomach is too big” and “I think my chest is too small”). Average
scores were computed for both the older adolescents (α= .88, M= 3.0, SD =
.93) and the younger adolescents (α= .87, M= 2.6, SD = .95), with high scores
indicating greater dissatisfaction.
Female adolescents’ drive for thinness was measured with 14 items tap-
ping motivation to be thin (e.g., “I am terrified of gaining weight” and “I am
preoccupied with the desire to be thinner”). Reliability analyses indicated
that 2 items should be removed from the scale for college students and 1 item
should be removed from the scale for their siblings. After removing those
items, average scores were computed to create scales for the older adoles-
cents (α= .94, M= 3.0, SD = .98) and younger adolescents (α= .91, M= 2.5,
SD = 1.1).
316
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Female adolescents completed eight items tapping anorexia (e.g., “I do not
eat for one or more days to lose weight” and “I take pills to stop myself from
eating”). However, among the college students, the combination of only six of
the items produced a reliable scale (α= .82, M= 2.0, SD = .78). In the case of
the 12- to 17-year-olds,a 7-item scale was the most reliable (α= .87, M= 1.7,
SD = .79).
Symptoms of bulimia also were assessed by having the female adolescents
complete 11 items (e.g.,“I vomit to lose weight or to keep from gaining weight”
and “I have gone on eating binges where I could not stop”). Reliability analy-
ses indicated that 1 item should be dropped from the scale for college stu-
dents. Responses were averaged to create scales for college students (α= .88,
M= 1.8, SD = .67) and the 12- to 17-year-olds (α= .83, M= 1.6, SD = .51).
Television viewing. With two questions, adolescents reported how many
hours they usually watch television on an average weekday and on an aver-
age weekend day. These estimates were converted into minutes to create two
single-item measures of weekday viewing (M= 159.70, SD = 100.24 for the
older adolescents, M= 216.39, SD = 214.13 for the younger adolescents) and
weekend viewing (M= 189.92, SD = 133.68 for the older adolescents, M=
231.12, SD = 153.29 for the younger adolescents). These measures were
taken so that we could control for the influence of mere exposure to television
and more precisely predict the influence of parental mediation of television.
Results
Hierarchical multiple regression was used to examine the hypotheses.In the
first step, the adolescents’ television viewing scores were entered so that we
could assess the influence of parental mediation of television and not simply
adolescents’ exposure to television. To test Hypothesis 1a, Hypothesis 1b,and
Hypothesis 1c, the three forms of central content mediation (i.e., positive,
negative, neutral) were simultaneously entered on the second step.This pro-
cedure was followed for each of the eight criteria (i.e., processing of central
content, processing of incidental content, social comparison processing, nega-
tive emotions, body dissatisfaction, drive for thinness, anorexia, and
bulimia). Likewise, to test Hypothesis 2a, Hypothesis 2b, and Hypothesis 2c,
the three forms of incidental mediation were entered at the same time (after
controlling for television viewing). This procedure was repeated for each of
the same criteria.
In response to Hypothesis 1a, we found that neutral mediation of central
content was positively related to anorexia symptoms (β= .26, p< .05) among
college students; however, this form of mediation was not significantly
317
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
related to any of the outcomes among the younger adolescents. Further, both
Hypothesis 1b and Hypothesis 1c were rejected, as neither positive nor nega-
tive mediation of central content was significantly related to any of the out-
comes in either sample. As a result,none of the hypotheses involving media-
tion of central content were supported.
An answer to Research Question 1 was not pursued given that mediation
of central content was significantly related to only one indicator of BID. In
general, mediation of central content information was not strongly related to
outcomes among older or younger adolescents (see Table 1).
In response to Hypothesis 2a, we found that neutral mediation was signifi-
cantly related to older adolescents’ anorexia (β= .20, p< .05). In addition, it
was significantly related to making more social comparisons (β= .19, p< .05),
engaging in more body image processing (β= .23, p< .01), having more nega-
tive emotions (β= .20, p< .05), and having a stronger drive for thinness (β=
.23, p< .05) among the 12- to 17-year-olds.
In response to Hypothesis 2b, we found that positive mediation was signif-
icantly related to experiencing negative emotions among college students
(β= .20, p< .05). There were no significant relationships between positive
mediation of incidental content and 12- to 17-year-olds’ outcomes.
In response to Hypothesis 2c, we found that negative mediation of inciden-
tal content was significantly related to making more social comparisons
among both college students (β= .19, p< .05) and the 12- to 17-year-olds (β=
.18, p< .05), more body image processing among the 12- to 17-year-olds (β=
.18, p< .05), and more negative emotions among the 12- to 17-year-olds (β=
.23, p< .01). The results addressing Hypothesis 2 appear in Table 2.
The pattern of relationships for the mediation of incidental content sug-
gests that the valence of the mediation is not particularly important. The
relationships between the outcomes and the three forms of mediation were of
similar strength and direction. In only one case (even among the
nonsignificant relationships) was the relationship between mediation and
the outcomes negative (β= –.02). As a result, to more efficiently examine the
possibility that the relationship between mediation and adolescents’ BID is
indirect (Research Question 2), all of the items reflecting mediation of inci-
dental content were combined into one scale (α= .89, M= 2.0, SD = .63).
Baron and Kenny’s (1986) method for exploring intervening variables was
used to investigate Research Question 2. According to them, intervening
variables are identified after running three separate regressions. In the first,
the dependent variable is regressed onto the independent variable. In the
second regression, the intervening variable is regressed onto the independ-
ent variable. In the final regression,the dependent variable is regressed onto
both the independent and intervening variables (i.e., to test the relationship
318
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Table 1
The Relationship Between Parental Mediation of Central Content and Adolescents’ Television Processing, Negative Emotions,
and Body Image Disturbance
Central Incidental Social Negative Body Drive for
Processing Processing Comparison Emotion Dissatisfaction Thinness Anorexia Bulimia
O/Y O/Y O/Y O/Y O/Y O/Y O/Y O/Y
Step 1:
Weekday view –.05/.20* –.09/–.11 –.10/–.05 –.07/–.11 –.07/–.04 .05/.12 .04/.13 .05/.03
Weekend view .19*/–.03 –.05/.05 .05/.10 .11/.25* –.07/–.03 .05/–.18 .14/–.27* .11/–.14
Step 2:
Neutral central .06/.09 –.02/.03 –.01/.12 .01/.15 –.06/.06 .04/.01 .26*/–.03 .09/–.02
Positive central .04/.03 .03/–.12 –.01/–.06 .00/.00 .04/–.14 –.03/.08 .18/–.18 .09/–.07
Negative central .06/.02 –.06/.09 .03/.07 .04/.08 –.04/–.05 –.15/.05 –.02/.00 –.19/–.09
Total R2.04/.04 .02/.04 .01/.05 .01/.08* .02/.05 .04/.03 .11*/.09 .09/.03
Note. The table entries reflect betas upon entry into the equation. O/Y = older adolescents/younger adolescents.
*p< .05.
319
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Table 2
The Relationship Between Parental Mediation of Incidental Content and Adolescents’ Television Processing, Negative Emotions,
and Body Image Disturbance
Central Incidental Social Negative Body Drive for
Processing Processing Comparison Emotion Dissatisfaction Thinness Anorexia Bulimia
O/Y O/Y O/Y O/Y O/Y O/Y O/Y O/Y
Step 1:
Weekday view –.05/.20* –.10/–.13 –.10/–.08 –.07/–.10 –.07/–.05 .05/.13 .04/.13 .05/.04
Weekend view .19*/–.04 –.05/.07 .04/.13 .11/.24* –.07/–.01 .05/–.17 .14/–.28* .11/–.15
Step 2:
Neutral incidental .08/.13 .08/.23** .14/.19* .11/.20* .01/.10 .12/.23* .20*/.19 .09/.19
Positive incidental .03/.07 .08/.03 .15/.04 .20*/.08 .05/–.02 .16/.09 .16/.05 .05/.10
Negative incidental .12/.05 .10/.18* .19*/.18* .15/.23** .11/.06 .13/.13 .13/.16 .03/.12
Total R2.05/.05 .03/.08* .05/.06 .05/.10** .03/.02 .04/.08 .07/.09 .03/.06
Note. The table entries reflect betas upon entry into the equation. O/Y = older adolescents/younger adolescents.
*p< .05. **p< .01.
320
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
between two predictors and the dependent variable simultaneously). The
third regression should reveal a significant relationship between the inter-
vening and the dependent variables and a nonsignificant relationship
between the independent and the dependent variables (with the intervening
variable controlled). In this study, only college students’ anorexia and 12- to
17-year-olds’ drive for thinness were used as dependent variables because
they were the only BID indicators to hold significant relationships with medi-
ation of incidental content. In each analysis, adolescents’ television viewing
was included at a first step as a control.
In the case of the older adolescents, we found evidence that both social
comparison and negative emotions were intervening variables in the rela-
tionship between mediation of incidental content and anorexia (see Figure 1).
We found that (a) mediation was significantly related to anorexia (β= .20, p<
.05), (b) mediation was significantly related to social comparison processes
(β= .20, p< .05), (c) social comparisons were related to anorexia (β= .52, p<
.001), and (d) the once-significant relationship between mediation and
anorexia became nonsignificant (β= .12, p= .18) when social comparisons
were controlled.
In the case of negative emotions, we found that (a) mediation was signifi-
cantly related to feeling negative emotions (β= .18, p< .05), (b) negative emo-
tions were significantly related to anorexia ( β= .45, p< .001), and (c) the
321
Nathanson, Botta • Mediation and Body Image Disturbance
.52*** Social Comparisons
.45*** Negative Emotions
.60*** Social Comparisons
.51*** Negative Emotions
.39*** Incidental Processing
.20* Social Comparisons
.18* Negative Emotions
.19* Social Comparisons
.22* Negative Emotions
.20* Incidental Processing
(Drive to be Thin)
Body Image Disturbance
(Anorexic Behaviors )
Teens
Incidental
Parental Mediation
.20
.20*
College
Body Image Processing
Figure 1. The Indirect Effects of Incidental Parental Mediation on Adolescents’ Body
Image Disturbance
*p< .05. ***p< .001.
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
relationship between mediation and anorexia (reported above) became
nonsignificant when negative emotions were controlled (β= .11, p= .23).
Among the younger adolescents, we found evidence that all three process-
ing variables served as intervening variables between mediation and drive
for thinness (see Figure 1). In the case of social comparisons, we found that
mediation was related to drive for thinness (β= .20, p= .06). Although this
relationship is only marginally significant, it is of the same magnitude as the
relationship between mediation and college students’ anorexia reported
above. In addition, we found that (a) mediation was significantly related to
making social comparisons (β= .19, p< .05), (b) social comparisons were
related to drive for thinness (β= .60, p< .001), and (c) controlling social com-
parisons rendered the relationship between mediation and drive for thinness
nonsignificant (β= .05, p= .58).
For negative emotions, we found that (a) mediation was related to nega-
tive emotions (β= .22, p< .01), (b) negative emotions were related to drive for
thinness (β= .51, p< .001), and (c) the relationship between mediation and
drive for thinness (reported earlier) became nonsignificant when negative
emotions were controlled (β= .03, p= .75).
In the case of body image processing, we found that (a) mediation was
related to body image processing (β= .20, p< .05), (b) body image processing
was significantly related to drive for thinness (β= .39, p< .001), and (c) con-
trolling body image processing reduced the relationship between mediation
and drive for thinness to nonsignificance (β= .12, p= .24).
Discussion
This study is the first to explore the influence of mediation on adolescents’
body image processing and BID. Our study suggests that certain forms of
well-intended mediation messages may contribute to detrimental outcomes.
Few significant relationships emerged between mediation of central con-
tent and adolescents’ outcomes. This is consistent with previous work that
has relied on a conceptualization of mediation similar to ours. It is possible
that adolescents’ emerging disinterest in television makes it more difficult
for parents to be effective in talking with their children about general content
issues.
However, we found quite a few relationships between mediation of inci-
dental content and adolescents’ outcomes. Perhaps this population is more
attentive to parents’ comments that are focused on this topic because the
importance of body image increases during adolescence. The pattern of rela-
tionship suggested little distinction among the three forms of incidental
322
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
mediation. Regardless of their valence, conversations focusing on television
characters’ bodies and appearances encouraged undesirable outcomes.
Moreover, we found that mediation of incidental content influences BID by
first encouraging certain forms of body image processing and producing neg-
ative emotions. Hence, mediation’s most immediate effect is on adolescents’
processing activities and the emotions they experience during viewing. By
critically evaluating the characters’ appearances, comparing themselves to
the televised images, and feeling bad about themselves, the adolescents are
more vulnerable to developing a strong drive for thinness or symptoms of
anorexia. On the other hand, if parents focused less on thin images when dis-
cussing television, then their adolescents also would focus less on the images
and therefore would be less vulnerable to developing BID. These results sup-
port Botta’s (2000a) recommendation that adolescent girls should not think
about body image content.
These findings are particularly interesting in the case of negative media-
tion, neutral mediation, and critical viewing. They suggest that even adoles-
cents who draw the logical conclusion that televised images are unrealistic
are susceptible to experiencing BID. This poses particular challenges to those
wishing to prevent the development of these disturbances. Simply put,
appeals to rational arguments may not be effective.
Austin and Nach-Ferguson (1995) noted that when it comes to adoles-
cents, there often is a distinction between what is known (on a rational level)
and what is desired (on an emotional level). They found that adolescents who
knew the dangers of binge drinking reported engaging in the activity because
they believed drinking was “cool.” Perhaps a similar problem occurs when
adolescents watch television that features unrealistic body types. Despite
knowing that they could never look the same way as television characters,the
adolescents believe that emulating those characters will bring them popular-
ity and acceptance.
Our findings parallel those from Irving and Berel (2001) who tested a
media literacy intervention aimed at body images. They found that although
the intervention increased college women’s skepticism of media images, it
made no difference in their BID. They posited that the intervention inter-
rupted the logical, but not the emotional, decision-making processes. The
same may have occurred with our sample. Logically, the adolescents may
have been skeptical of the images, but emotionally, their need to adhere to the
pressure to be thin remained intact.
During adolescence in particular, it also is possible that parental mes-
sages about body image issues are less compelling than the images them-
selves. Adolescents may believe that media sources are more trustworthy
than parents in providing tips for winning peer approval (Strasburger, 1995).
323
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
In fact, Taylor et al. (1998) found that wanting to emulate media images was a
stronger predictor of weight concerns than parental concerns about thinness.
Or, adolescents may believe that their parents do not have the right to dic-
tate the standards for their appearance. As Smetana and Daddis (2002)
explained, adolescents may accept parents’ rules regarding moral and con-
ventional issues (e.g.,lying, cursing,having bad manners) but may reject par-
ents’ attempts to dictate the adolescents’ personal issues, including “the state
of one’s body” (p. 564), choice of hairstyles, and media consumption habits.As
a result, adolescents may disregard parents’ comments about the appearance
of television characters and simply focus on the images their parents have
directed them toward.
It is interesting to note that mediation of incidental content was linked
with anorexia among college students but with drive for thinness among the
younger adolescents still living at home. College students may be better able
to get away with being anorexic (e.g., engaging in extreme behaviors) when
they live away from home.The 12- to 17-year-olds may not have as much con-
trol over their behaviors at home, so instead their problems are manifested in
a strong drive for thinness. It also may be that because the 12- to 17-year-olds
in this study are at the average age for onset of anorexia, they are not yet fully
engaged in anorexic behaviors. Lending credence to these arguments is the
fact that the 12- to 17-year-olds in this study engaged in less anorexic behav-
iors than the college students (M= 1.7 and 2.0, respectively, t(26) = 3.8, p<
.001).
More important from a parent’s perspective, however, is the possibility
that the results indicate a delayed onset of parental mediation effects. For the
college women, this study looked at the relationship between what parents
said to their adolescents when they were in high school and their attitudes
and behaviors now that they are away from home. For the younger adoles-
cents, however, this study looked at the relationship between what their par-
ents are currently saying and what the adolescents are currently doing. It
could be that incidental mediation predicts a stronger drive to be thin for ado-
lescents living at home, whereas it predicts the increased likelihood that the
child will become anorexic when she moves out of the house.
At the same time, the differences in findings between the college students
and the 12- to 17-year-olds could reflect a methodological difference in how
mediation of both groups was assessed. Parents engaged in retrospective
reporting of their mediation of the college students; as a result, the accuracy
of their reports depended on parents’ ability to recall their behaviors. A panel
survey is a better way to determine whether the effects of parental mediation
change as adolescents transition from high school to college.
324
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
It is important to note that the relationship between mediation and BID
may reflect the opposite causal direction: our data could reflect the effects of
having a BID on mediation. That is,parental mediation may be a response to
adolescents’ body image attitudes and behaviors. Parents may discuss thin
images they see on television because they are worried about the impact of
those images on their adolescents’ disordered attitudes and behaviors. Par-
ents may talk about thin images on television as a way of indirectly talking
about the issue with their daughters.
More work, with larger, more representative samples, would allow us to
generalize these findings to a diverse population. Although this study relied
on a convenience sample, this sample is appropriate for this topic.By target-
ing a college student population, teens, and their families, we increased our
chances of including adolescents with BID (Lucas et al., 1999; Pawluck &
Gorey, 1998; Smolak & Levine, 1996). Nevertheless, because these data were
not based on a probability sample, we must be careful to avoid making inap-
propriate generalizations.
In addition, future research also should examine mediation of other media
that promote the thin ideal. Female teenagers use magazines for information
on how to behave and what to look like, and researchers have found that mag-
azine exposure and processing are predictors of BID (e.g., Botta, 2002, in
press; Harrison & Cantor, 1997). This study focused on television because
previous work shows that viewing this medium is related to body image-
related effects among adolescents; nevertheless, other forms of media also
are important components of adolescents’ media repertoire and should be
studied.
Finally, more work should explore the actual content of parents’ mediation
of incidental content. For instance, it is possible that the surprising effects of
mediation of incidental content are due to the fact that parents who comment
on characters’ bodies also communicate their dissatisfaction with their own
bodies (e.g., “That actress is so pretty and thin. I wish I was as thin as her”).
Even if parents do not communicate their own body dissatisfaction directly,
adolescents may interpret their parents’ critical comments as jealousy. As a
result, adolescents may learn to be critical of their bodies from their parents.
Or it is possible that some of our mediation items were interpreted differ-
ently than we had intended, a situation that may have contributed to our
findings. For example, parents’ reports of how often they “make negative com-
ments about a character’s appearance” may have included instances where
they criticized an overweight character (e.g., “That actress is too fat”). It
would not be surprising if messages such as these predicted harmful body
image outcomes. Research that solicits parental mediation in an open-ended
325
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
format would help us understand the nature of these messages and how they
are delivered.
Overall, this study suggests that some forms of parental mediation may
backfire by encouraging adolescents to process images of unrealistic and
unattainable bodies. The priorities for future work in this area should include
analyzing the content of mediation and establishing the direction of causality
between mediation and BID. With more studies of this kind, we should arrive
at more satisfactory solutions to handling the effects of television on adoles-
cents’ body image processing and disturbances.
Note
1. Parents may distort their responses to parental mediation items to provide
socially desirable answers. As a result, mediation that is perceived to be beneficial for
children (e.g., negative mediation of incidental content) may be overreported relative
to mediation that is perceived to be detrimental to children (e.g., positive mediation of
incidental content).
The parents’ survey included two items derived from Paulhus’s (1991) balanced
inventory of desirable responding scale (“I sometimes tell lies if I have to,”and “I have
done things that I don’t tell other people about.”) to assess whether parents’ reporting
of mediation was significantly related to parents’ tendency to provide socially desirable
responses. Response options ranged from not true (1) to very true (7). The average of
these two items, r(220) = .38,p< .001, M= 3.7, SD = 1.7, was not significantly correlated
with any of the six parental mediation scales. As a result, there does not appear to be
any relationship between parents’ tendency to provide socially desirable responses (or
their tendency to lie) and parents’ responses to our mediation items.
References
American Psychiatric Association. (2000). Eating disorders. In Diagnostic
and statistical manual of mental disorders (4th ed., text revision). Wash-
ington, DC: Author.
Anderson, J. A. (1983). Television literacy and the critical viewer. In J. Bryant
& D. R. Anderson (Eds.), Children’s understanding of television: Research
on attention and comprehension (pp. 297-330).New York: Academic Press.
Austin, E. W. (1993).Exploring the effects of active parental mediation of tele-
vision content. Journal of Broadcasting & Electronic Media,37,147-158.
Austin, E. W., Bolls, P., Fujioka, Y., & Engelbertson, J. (1999). How and why
parents take on the tube. Journal of Broadcasting & Electronic Media,43,
175-192.
Austin, E. W., & Nach-Ferguson, B. (1995). Sources and influences of young
school-age children’s general and brand-specific knowledge about alcohol.
Health Communication,7, 1-20.
326
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Avery, R. K.(1979).Adolescents’ use of the mass media.American Behavioral
Scientist,23, 53-70.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator distinction in
social psychological research: Conceptual, strategic, and statistical con-
siderations.Journal of Personality and Social Psychology,51, 1173-1182.
Becker, A. E., & Burwell, R. A. (1999, May). Acculturation and disordered eat-
ing in Fiji. Poster session presented at the meeting of the American Psy-
chiatric Association, Washington, DC.
Botta, R. A.(1999a). Television images and adolescent girls’ body image dis-
turbance. Journal of Communication,49(2), 22-41.
Botta, R. A. (1999b, May). The role of television image exposure, perceptions
and processing in body image disturbance. Paper presented at the annual
conference of the International Communication Association, San Fran-
cisco, CA.
Botta, R. A. (2000a, June). Frustrated and guilty. Critical but not active: The
link between critical viewing and body image disturbance. Paper pre-
sented at the annual conference of the International Communication
Association, Acapulco, Mexico.
Botta, R. A. (2000b). The mirror of television: A comparison of Black and
White adolescents’ body image. Journal of Communication,50 (3), 144-
159.
Botta, R. A. (2002, April). Gender and ethnic differences in adolescents’ social
comparisons and body image disturbance. Paper presented at the meeting
of the Broadcast Educators Association, Las Vegas, NV.
Botta, R. A. (in press). For your health? The relationship between magazine
reading and adolescents’ body image and eating disturbances. Sex Roles.
Botta, R. A., & Dumlao, R. (2002). How do conflict and communication pat-
terns between fathers and daughters contribute to or offset eating disor-
ders? Health Communication,14, 199-219.
British Medical Association Board of Science and Education. (2000). Eating
disorders, body image and the media. London: British Medical
Association.
Bruch, H.(1982). Anorexia nervosa: Therapy and theory. American Journal of
Psychiatry,139, 1531-1538.
Cohen, J. R. (1994). Critical viewing and participatory democracy. Journal of
Communication,44(4), 98-114.
Collins, W. A. (1970). Learning of media content: A developmental study.
Child Development,41, 1133-1142.
Connors, M. E. (1996). Developmental vulnerabilities for eating disorders. In
L. Smolak & M. P. Levine (Eds.), The developmental psychopathology of
eating disorders: Implications for research, prevention, and treatment
(pp. 285-310). Mahwah, NJ: Lawrence Erlbaum.
Corder-Bolz, C.R. (1980). Mediation: The role of significant others. Journal of
Communication,30(3), 106-118.
Cusumano, D. L., & Thompson, K. J. (2001). Media influence and body image
in 8-11-year-old boys and girls: A preliminary report on the multidimen-
sional media influence scale. International Journal of Eating Disorders,
29, 37-44.
327
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Desmond, R. J., Singer, J. L., Singer, D. G., Calam, R., & Colimore, K. (1985).
Family mediation patterns and television viewing: Young children’s use
and grasp of the medium. Human Communication Research,11,461-480.
Frank, S., & Jackson, S. (1996). Family experiences as moderators of the rela-
tionship between eating symptomology and personality disturbances.
Journal of Youth and Adolescence,25, 55-72.
Garner, D. M., & Olmsted, M. P. (1991). Eating disorders inventory—2.
Odessa, FL: Psychological Assessment Resources, Inc.
Harrison, K. (2000).The body electric:Thin ideal media and eating disorders
in adolescents. Journal of Communication,50(3), 119-143.
Harrison, K. (2001). Ourselves, our bodies: Thin-ideal media, self-discrepancies,
and eating disorder symptomatology in adolescents. Journal of Social and
Clinical Psychology,20, 289-323.
Harrison, K., & Cantor, J. (1997). The relationship between media consump-
tion and eating disorders. Journal of Communication,47(1), 40-67.
Hicks, D. J. (1968). Effects of co-observer’s sanctions and adult presence on
imitative aggression. Child Development,39, 303-309.
Huesmann, L. R., Eron, L. D., Klein, R., Brice, P., & Fischer, P. (1983). Miti-
gating the imitation of aggressive behaviors by changing children’s atti-
tudes about media violence. Journal of Personality and Social Psychology,
44, 899-910.
Humphrey, L. (1989). Observed family interactions among subtypes of eating
disorders using structural analysis of social behavior. Journal of Con-
sulting and Clinical Psychology,57(2), 206-214.
Irving, L. M., & Berel, S. R. (2001) Comparison of media-literacy programs to
strengthen college women’s resistance to media images. Psychology of
Women Quarterly,25, 103-111.
Killian, K. (1994). Fearing fat: A literature review of family systems under-
standings and treatments of anorexia and bulimia. Family Relations,43,
311-318.
King, N., Touyz, S., & Charles, M. (2000). The effect of body dissatisfaction on
women’s perceptions of female celebrities.International Journal of Eating
Disorders,27, 341-347.
Larson, R. (1995).Secrets in the bedroom: Adolescents’ private use of media.
Journal of Youth and Adolescence,24, 535-550.
Lowe, P. J., & Durkin, K. (1999). The effect of flashback on children’s under-
standing of television crime content. Journal of Broadcasting & Electronic
Media,43, 83-97.
Lucas, A. R., Crowson, C. S., OaposFallon, W. M., & Melton, L. J. (1999). The
ups and downs of anorexia nervosa. International Journal of Eating Disor-
ders,26, 397-405.
Martin, M. C., & Kennedy, P. F. (1993). Advertising and social comparison:
Consequences for female preadolescents and adolescents. Psychology and
Marketing,10, 513-530.
McLeod, J. M., Atkin, C. K., & Chaffee, S. H. (1972a). Adolescents, parents,
and television use: Adolescent self-report measures from Maryland and
Wisconsin samples.In G. A. Comstock & E. A.Rubinstein (Eds.), Television
328
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
and social behavior: Vol. 3. Television and adolescent aggressiveness (pp. 173-
238). U.S. Government Printing Office.
McLeod, J. M., Atkin, C. K., & Chaffee, S. H. (1972b). Adolescents, parents,
and television use: Self-report and other-report measures from the Wis-
consin sample. In G. A. Comstock & E. A. Rubinstein (Eds.), Television and
social behavior: Vol. 3. Television and adolescent aggressiveness (pp. 239-
313). U.S. Government Printing Office.
McNamara, K., & Loveland, C. (1990). Differences in family functioning
among bulimics, repeat dieters, and nondieters. Journal of Clinical Psy-
chology,46, 518-523.
Meadowcroft, J. M., & Reeves, B. (1989). Influence of story schema develop-
ment on children’s attention to television. Communication Research,16,
352-375.
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(2), 190-210.
Nathanson, A. I.(1999).Identifying and explaining the relationship between
parental mediation and children’s aggression. Communication Research,
26, 124-143.
Nathanson, A. I. (2001a). Mediation of children’s television viewing: Working
toward conceptual clarity and common understanding.In W. B. Gudykunst
(Ed.), Communication Yearbook 25 (pp. 115-151). Mahwah, NJ: Lawrence
Erlbaum.
Nathanson, A. I.(2001b). Parent and child perspectives on the presence and
meaning of parental television mediation. Journal of Broadcasting &
Electronic Media,45, 210-220.
Nathanson, A. I. (2001c). Parents versus peers: Exploring the significance of
peer mediation of antisocial television. Communication Research,28, 251-
274.
Nathanson, A. I.(2002). The unintended effects of parental mediation of anti-
social television on adolescents. Media Psychology,4, 207-230.
Nathanson, A. I., Wilson, B. J., McGee, J., & Sebastian, M. (2002). Counter-
acting the effects of female stereotypes on television via active mediation.
Journal of Communication,52(4), 922-937.
Nathanson, A. I.,& Yang,M. (2003). The effects of mediation content and form
on children’s responses to violent television. Human Communication
Research,29, 111-134.
Paulhus,D. L. (1991).Measurement and control of response bias. In J. P. Robinson,
P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of personality and social
psychological attitudes (pp. 17-59). San Diego, CA: Academic Press.
Pawluck, D. E., & Gorey, K. M. (1998) Secular trends in the incidence of
anorexia nervosa: Integrative review of population-based studies. Inter-
national Journal of Eating Disorders,23, 347-352.
Polivy, J., Herman, C. P., & McFarlane, T. (1994). Effects of anxiety on eating:
Does palatability moderate distress-induced overeating in dieters? Jour-
nal of Abnormal Psychology,103, 505-510.
Potter, W. J. (1998). Media literacy. Thousand Oaks, CA: Sage.
329
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Rubinstein, S., & Caballero, B. (2000). Is Miss America an undernourished
role model? JAMA: Journal of the American Medical Association,283,
1569.
Salomon, G.(1977). Effects of encouraging Israeli mothers to co-observe “Ses-
ame Street” with their five-year-olds. Child Development,48, 1146-1151.
Shisslak, C. M., McKeon, R. T., & Crago, M. (1990). Family dysfunction in nor-
mal weight bulimic and bulimic anorexic families. Journal of Clinical Psy-
chology,46, 185-189.
Singer, D. G., & Singer, J. L. (1998). Developing critical viewing skills and
media literacy in children. The Annals of the American Academy of Pediat-
rics,557, 164-180.
Slade, P. D. (1994). What is body image? Behavioral Research Therapy,32,
497-502.
Smetana, J. G., & Daddis, C. (2002). Domain-specific antecedents of parental
psychological control and monitoring: The role of parenting beliefs and
practices. Child Development,73, 563-580.
Smolak, L., & Levine, M. P. (1996) Adolescent transitions and the develop-
ment of eating problems. In L. Smolak, M. P. Levine, & R. Striegel-Moore
(Eds.), The developmental psychopathology of eating disorders (pp. 207-
233). Mahwah, NJ: Lawrence Erlbaum.
Stice, E., Schupak-Neuberg, E., Shaw, H. E., & Stein, R. (1994). Relation of
media exposure to eating disorder symptomatology: An examination of
mediating mechanisms. Journal of Abnormal Psychology,103, 836-840.
Stice, E., Spangler, D., & Agras, W. S. (2001). Exposure to media-portrayed
thin-ideal images adversely affects vulnerable girls: A longitudinal exper-
iment. Journal of Social and Clinical Psychology,20, 270-288.
Strasburger, V. C. (1995). Adolescents and the media: Medical and psychologi-
cal impact. Thousand Oaks, CA: Sage.
Taylor, C. B., Sharpe T., Shisslak, C., Bryson, S., Estes, L. S., Gray, N., et al.
(1998). Factors associated with weight concern in adolescent girls. Inter-
national Journal of Eating Disorders,24, 31-42.
Thompson, J. K., Heinberg, L. J., Altabe, M., & Tantleff-Dunn, S. (1999).
Exacting beauty: Theory, assessment, and treatment of body image distur-
bance. Washington, DC: American Psychological Association.
Wertheim, E. H., Mee, V., & Paxton, S. J. (1999). Relationships among adoles-
cent girls’ eating behaviors and their parents’ weight-related attitudes
and behaviors. Sex Roles,41, 169-187.
Amy I. Nathanson (Ph.D., 1998, University of Wisconsin–Madison) is an
assistant professor in the School of Journalism and Communication at Ohio
State University. Her research interests include the uses and effects of media
among children, with particular attention paid to the role of parents in shap-
ing media-related effects.
330
COMMUNICATION RESEARCH • June 2003
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from
Renée A. Botta (Ph.D., 1998, University of Wisconsin–Madison) is an assis-
tant professor in the Department of Communication at Cleveland State Uni-
versity. Her research interests include theoretical frameworks for understand-
ing how adolescents use, process, interpret, and are affected by media, and the
social and health implications of communication for adolescents.
331
Nathanson, Botta • Mediation and Body Image Disturbance
at Auraria Library on June 20, 2016crx.sagepub.comDownloaded from