Body Dissatisfaction Prospectively Predicts Depressive Mood and Low Self-Esteem in Adolescent Girls and Boys

ArticleinJournal of Clinical Child & Adolescent Psychology 35(4):539-49 · December 2006with1828 Reads
Impact Factor: 3.31 · DOI: 10.1207/s15374424jccp3504_5 · Source: PubMed
Abstract

This research examined whether body dissatisfaction prospectively predicted depressive mood and low self-esteem in adolescent girls and boys 5 years later. Participants were early-adolescent girls (n = 440, Time 1 M age = 12.7 years) and boys (n = 366, Time 1 M age = 12.8 years) and midadolescent girls (n = 946, Time 1 M age = 15.8 years) and boys (n = 764, Time 1 M age = 15.9 years). After controlling for Time 1 of the relevant dependent variable, ethnicity, socioeconomic status, and body mass index, Time 1 body dissatisfaction was a unique predictor of Time 2 depressive mood and low self-esteem in early-adolescent girls (depressive mood: F = 4.80, p < .05; self-esteem: F = 9.64, p < .01) and midadolescent boys (depressive mood: F = 12.27, p < .001; self-esteem: F = 9.38, p < .01) but not in early-adolescent boys or midadolescent girls. These findings are consistent with the hypothesis that body dissatisfaction is a risk factor for depressive mood and low self-esteem in both girls and boys but in different phases of adolescence.

Full-text

Available from: Susan J Paxton
Body Dissatisfaction Prospectively Predicts Depressive Mood
and Low Self-Esteem in Adolescent Girls and Boys
Susan J. Paxton
School of Psychological Science, La Trobe University, Melbourne
Dianne Neumark-Sztainer, Peter J. Hannan, and Marla E. Eisenberg
School of Public Health, University of Minnesota
This research examined whether body dissatisfaction prospectively predicted depres
-
sive mood and low self-esteem in adolescent girls and boys 5 years later. Participants
were early-adolescent girls (n = 440, Time 1 M age = 12.7 years) and boys (n = 366,
Time 1 M age = 12.8 years) and midadolescent girls (n = 946, Time 1 M age = 15.8
years) and boys (n = 764, Time 1 M age = 15.9 years). After controlling for Time 1 of
the relevant dependent variable, ethnicity, socioeconomic status, and body mass in
-
dex, Time 1 body dissatisfaction was a unique predictor of Time 2 depressive mood
and low self-esteem in early-adolescent girls (depressive mood: F = 4.80, p < .05;
self-esteem: F = 9.64, p < .01) and midadolescent boys (depressive mood: F = 12.27,
p < .001; self-esteem: F = 9.38, p < .01) but not in early-adolescent boys or
midadolescent girls. These findings are consistent with the hypothesis that body dis-
satisfaction is a risk factor for depressive mood and low self-esteem in both girls and
boys but in different phases of adolescence.
Body dissatisfaction is widespread among adoles-
cents in Western cultures (Neumark-Sztainer, Story,
Hannan, Perry, & Irving, 2002; Ricciardelli & McCabe,
2001). The negative impact of body dissatisfaction in
adolescent girls and boys has been gauged in terms of
the immediate distress, low self-esteem, and depres-
sive mood with which it is consistently associated
(Allgood-Merten, Lewinsohn, & Hops, 1990; Cash,
2002; Keery, van den Berg, & Thompson, 2004; Wert
-
heim, Koerner, & Paxton, 2001; Wichstrom, 1999). In
addition, the negative impact of body dissatisfaction
has been demonstrated in longitudinal studies in which
it has been identified as a reliable predictor of the in
-
creases in use of dieting and disordered eating symp
-
toms and clinical eating disorders (e.g., Killen et al.,
1996; Stice, 2001, 2002). However, less research atten
-
tion has been directed toward the role of body dis
-
satisfaction as a prospective predictor of other signifi
-
cant psychological problems, most notably depressive
mood and low self-esteem.
A theoretical rationale for a role for body dissatis
-
faction in predicting an increase in depression in girls
has been articulated by Stice and Bearman (2001).
They proposed that puberty moves girls away from the
current thin beauty ideal and this change precipitates
body dissatisfaction. As appearance is a critical eval
-
uative dimension for girls in Western culture, body dis-
satisfaction directly contributes to increase in depres-
sive mood. Although this rationale was specifically di-
rected toward a relation between body dissatisfaction
and depressive mood in girls, boys are under increas-
ing pressure to meet an unrealistic, lean, muscular
body ideal (Cafri et al., 2005; Cohane & Pope, 2001).
It has been argued that this pressure could have similar
consequences in adolescent boys (Holsen, Kraft, &
Roysamb, 2001).
In adolescent girls, a number of structural equation
modeling studies that used cross-sectional data have
strongly implicated body dissatisfaction as a predictor
of low self-esteem and depression (e.g., Keery et al.,
2004; van den Berg, Wertheim, Thompson, & Paxton,
2002; Wichstrom, 1999). Providing firmer support for
body dissatisfaction preceding depression and self-es
-
teem, a number of studies of adolescent girls have re
-
ported that body dissatisfaction prospectively predicts
depressive mood (Holsen et al., 2001; Johnson & War
-
dle, 2005; Rierdan, Koff, & Stubbs, 1989; Stice &
Bearman, 2001), and one study observed that body dis
-
satisfaction predicts an increase in low self-esteem
(Johnson & Wardle, 2005). Holsen and colleagues ob
-
served that body dissatisfaction predicted depressive
mood within a single sample of girls 13 to 15 years old
but not in girls 15 to 18 years old They did not report
whether body dissatisfaction at first assessment (13
years) predicted depression at final assessment (18
years). Holsen and colleagues proposed that early ado
-
Journal of Clinical Child and Adolescent Psychology
2006, Vol. 35, No. 4, 539–549
Copyright © 2006 by
Lawrence Erlbaum Associates, Inc.
539
Correspondence should be addressed to Susan Paxton, La Trobe
University, School of Psychological Science, Bundoora, Melbourne,
VIC 3086, Australia. E-mail: susan.paxton@latrobe.edu.au
Page 1
lescence is the most challenging period for girls, as it is
associated with entry into puberty, change in body
shape, and greater capacity for self-reflection. Stice
and Bearman observed that body dissatisfaction was
directly related to increases in depressive mood and
also indirectly related, with the effect mediated by diet
-
ing and bulimic symptoms.
The research in girls to date is indicative of a risk
factor role of body dissatisfaction for depressive mood
and low self-esteem over a relatively short time period
(1 to 2 years). However, it is unclear whether the effect
of body dissatisfaction is over 1 to 2 years or whether it
predicts depression or self-esteem over a longer time
period (5 years). In addition, the observation of greater
impact of body dissatisfaction on depression in early
rather than midadolescence made by Holsen and col
-
leagues (2001) has yet to be confirmed in an independ
-
ent investigation.
Although attention has focused on girls in the exam
-
ination of the relation between body dissatisfaction and
depressive mood, Holsen et al. (2001) also examined
the nature of the relation in teenage boys. They found
body dissatisfaction did not predict depressive mood in
a single sample of boys ages 13 to 15 years but did pre-
dict depressed mood from age 15 to 18 years. They
suggested that the underlying psychological processes
behind the body image–depressed mood relation may
be similar in boys and girls but that the age difference
may reflect typically later timing of puberty and ap-
pearance-related pressures. Holsen et al. did not exam-
ine whether body dissatisfaction prospectively pre-
dicted depressive mood over an extended follow-up
period in the boys and did not examine whether body
dissatisfaction assessed in boys in middle adolescence
predicted an increase in depression in the early adult
years. As far as we are aware, such prospective predic
-
tors of self-esteem in boys have yet to be explored. It
would be valuable to understand whether body dissat
-
isfaction in boys predicts depressive mood and low
self-esteem at this later stage as the consequences of
body dissatisfaction are often ignored in boys and this
information could help guide early intervention for
these possible outcomes.
This study aimed to extend existing research exam
-
ining the relation between body dissatisfaction and
psychological outcomes in a number of ways. We ex
-
plored whether body dissatisfaction is a prospective
risk factor for boys’ and girls’ psychological outcomes
over a relatively long (5-year) follow-up period. We
aimed to extend Holsen et al.s (2001) exploration of
developmental differences in outcomes of body dissat
-
isfaction by exploring two longer developmental
phases (12 to 17 years and 15 to 21 years) and using
both self-esteem and depressive mood as outcome vari
-
ables. We explored these questions in two cohorts of
girls and boys involved in the Eating Among Teens
Project (Project EAT). The first cohort was initially as
-
sessed in early adolescence, whereas the second cohort
was initially assessed in midadolescence (Time 1).
Both were reassessed 5 years later (Time 2). It was hy
-
pothesized that body dissatisfaction at Time 1 would
prospectively predict depressive mood and self-esteem
at Time 2, after controlling for baseline levels of the
relevant variable, body mass index (BMI), and demo
-
graphic variables in early-adolescent girls and mid
-
adolescent boys but in not early-adolescent boys or
midadolescent girls.
Method
Participants
Project EAT–I is a large epidemiological study of
socioenvironmental, personal, and behavioral determi
-
nants of dietary intake and weight status among ado
-
lescent girls and boys from diverse ethnic and so
-
cioeconomic backgrounds (Neumark-Sztainer, Story,
Hannan, & Croll, 2002). In-class surveys and height
and weight measurements were completed by students
from junior and senior high school in 31 Minnesota
schools (N = 4,746). Project EAT–II aimed to resurvey
as many of the original participants as could be con-
tacted 5 years later (2003–2004). Surveys were mailed
to the address provided by participants during EAT–I.
If the survey was not returned, a reminder postcard,
then a survey, then a second reminder postcard, and
then a final survey were sent by a commercial mail car-
rier to encourage response. If mail was returned due to
an incorrect address, Internet tracking services were
used to identify a current address where possible, and
the mailing procedure recommenced.
The University’s Institutional Review Board Hu
-
man Subjects Committee approved all protocols. At
Time 1, letters were sent home to parents in advance of
survey administration, describing the study and re
-
questing active or passive consent, in accordance with
requirements of each school. On the day of survey ad
-
ministration, project staff described study procedures,
the confidentiality of the data, and the voluntary nature
of participation. They also responded to questions as
needed to obtain assent from those students with pa
-
rental consent. At Time 2, initial letters were sent to all
participants in the older cohort and parents of all par
-
ticipants in the younger cohort, explaining the study
and inviting them and their children to participate in
EAT–II. If no refusal was received after a delay of 3
weeks, survey materials, including detailed consent
and assent forms, were sent. Completion and return of
surveys implied consent to participate in the study.
In this manner, 3,672 of the original participants
were contacted by mail and 2,516 completed surveys,
representing 52.0% of the original cohort and 68.4% of
participants who were contactable for EAT–II. The fi
-
540
PAXTON, NEUMARK-SZTAINER, HANNAN, EISENBERG
Page 2
nal sample who completed valid surveys in both EAT–I
(Time 1) and EAT–II (Time 2) consisted of 440 girls in
the early-adolescent cohort (Time 1 M age = 12.7
years, SD =.74; Time 2 M age = 17.2 years, SD = .62);
366 boys in the younger cohort (Time 1 M age = 12.8
years, SD = .76; Time 2 M age = 17.2 years, SD = .78);
946 girls in the older cohort (Time 1 M age = 15.8
years, SD = .81; Time 2 M age = 20.4 years, SD = .82);
and 764 boys in the older cohort (Time 1 M age = 15.9
years, SD = .78; Time 2 M age = 20.5 years, SD = .82).
The ethnic and racial background of the sample was
61.9% White, 11.1% African American, 4.5% His
-
panic, 17.8% Asian, 1.9% Native American, and 2.7%
of mixed or other ethnicity. There was also diversity in
socioeconomic status (SES): 13% low, 17% mid
-
dle-low, 25% middle, 28% middle-high, and 17% high.
Measures
Body dissatisfaction and BMI. Body dissatisfac
-
tion was assessed with a modification of the Body
Shape Satisfaction Scale (Pingitore, Spring, & Gar
-
field, 1997). Each respondent rated his or her satisfac
-
tion with 10 aspects of his or her body shape and parts
(e.g., height, weight, body shape, waist, body build,
and shoulders). The original scale included features of
the body that could potentially be influenced by weight
change. The modified scale used in this study also in-
cluded items that could not be readily changed (e.g.,
height and body build) as these could also be sources of
dissatisfaction and were likely to be relevant for boys.
Items were rated on a 5-point Likert scale from 1 (very
satisfied)to5(very dissatisfied) and item responses
were summed (higher scores reflecting higher dissat
-
isfaction). Principal components analyses, conducted
separately by gender, confirmed the unitary structure
of the body dissatisfaction construct for both girls and
boys. The items were approximately equally weighted,
making the sum a suitable summary. The single factor
explained 59% and 61% of the variance in girls and
boys, respectively; all other eigenvalues were less than
1. Cronbach’s alpha was .92 and .91 in the early- and
midadolescent girls, respectively, and .93 in both co
-
horts of boys. At Time 1, weight and height were mea
-
sured using standardized equipment and procedures,
and BMI was calculated (kg/m
2
).
Demographic variables. Gender, age, ethnicity,
and SES were based on Time 1 youth self-report.
Ethnicity was categorized as White, African Ameri
-
can, Asian American, Hispanic, Native American, or
mixed. SES was categorized as low, low-middle, mid
-
dle, upper-middle, or upper SES. The prime determi
-
nant of SES category was parental education level, de
-
fined by the highest level of educational attainment of
either parent. However, other variables were taken into
account, including family eligibility for public assis
-
tance, eligibility for free or reduced-cost school meals,
and parental employment status (Neumark-Sztainer,
Story, Hannan, & Croll, 2002).
Depressive mood. Depressive mood was as
-
sessed with the six-item Depressive Mood Scale devel
-
oped by Kandel and Davies (1982). Items such as
“During the past 12 months, how often have you been
bothered or troubled by feeling unhappy, sad, or de
-
pressed” were rated on a 3-point scale from 1 (not
at all)to3(very much), and item responses were
summed. Higher scores indicated higher depressed
mood. Kandel and Davies validated the Depressive
Mood Scale in adolescents who presented at two psy
-
chiatric clinics. They observed a significant corre
-
lation between the Depressive Mood Scale and the
depressive mood subscale of the Symptom Check
List–90 (Derogatis, 1977) and higher scores on the De
-
pressive Mood Scale in adolescents diagnosed with
major depressive illness compared to those who were
not. Supporting the validity of Kandel and Davies’
measure, in this study at Time 2 strong significant cor
-
relations existed between scores on the Depressive
Mood Scale and suicidal ideation, assessed on a
3-point scale of 1 (yes, during the last year),2(yes,
more than a year ago), and 3 (never): early-adolescent
girls r = .46, p < .001; early-adolescent boys r = .26, p <
.001; midadolescent girls r = .35, p < .001; and mid-
adolescent boys r = .38, p < .001. In our research, inter-
nal reliability of the Depressive Mood Scale was sup-
ported by Cronbach’s alpha of .78 in both cohorts of
girls and .78 and .80 in early and midadolescent boys,
respectively.
Self-esteem. Self-esteem was assessed with a
shortened version of the Rosenberg Self-Esteem Inven
-
tory (Rosenberg,1965). The original scale has 10 items,
suchas “Onthe whole, I am quite satisfiedwith myself,
rated on 4-point scale from 1 (strongly disagree)to4
(strongly agree). Item scores are summed, and higher
scores are indicative of higher self-esteem. Validity of
the scale is supported by a significant positive correla
-
tion with peer ratings of self-esteem (Demo, 1985), and,
in a review of the convergent and discriminant validity
of the Rosenberg Self-Esteem Inventory, Byrne (1983)
concluded that this measure was one of the most valid
measuresof global self-esteem. Wylie (1989) reported a
satisfactory 7-month test–retest correlation of r = .73 in
high school students and a 2-week test–retest reliability
of r = .85 in college students. Internal reliability is sup
-
ported by high Cronbach’s alpha of .80 (Paxton, Norris,
Wertheim, Durkin, & Anderson, 2005). The original
10-item scale was shortened to 6 items to minimize the
length of the survey instrument by omitting twopositive
and two negative items. The internal consistency of the
scale remained satisfactory, with a Cronbach’salpha of:
.75 and .79 for early- and midadolescent girls, respec
-
541
BODY DISSATISFACTION PREDICTS DEPRESSION AND LOW SELF-ESTEEM
Page 3
tively, and .74 and .79 for early- and midadolescent
boys, respectively.
Data Analysis
Distributions of the scores for depressive mood
and self-esteem were confirmed to be symmetric and
approximately normal, and thus the assumptions of
multiple regression analysis of independence, constant
variance and linearity, were met. There was a very
small amount of missing item data (accounting for mi
-
nor differences in sample sizes in analyses), but this
was of such a minor scale as to introduce negligible
bias. To examine whether body dissatisfaction pro
-
spectively predicted depressive mood after controlling
for initial depressive mood, demographic variables and
BMI, multiple regression analyses were conducted for
each gender and cohort group separately in which
Time 2 depressive mood was the dependent variable.
The independent variables (Time 1 depressive mood,
SES category, ethnic category, Time 1 BMI, and Time
1 body dissatisfaction) were entered into the regression
simultaneously. To ensure that any failure to observe a
predictive effect of body dissatisfaction was not due
merely to shared variance with BMI, multiple regres-
sion analyses were conducted in which Time 2 depres-
sive mood was the dependent variable and Time 1 de-
pressive mood, SES category and ethnic category, but
not Time 1 BMI, were entered simultaneously as inde-
pendent variables. A similar strategy was used to ex-
amine whether body dissatisfaction at Time 1 prospec-
tively predicted low self-esteem 5 years later.
To test the differences postulated a priori between
the genders and the cohorts for each of the outcomes,
an analysis of covariance using all data and including
the interactions of gender and cohort with body dis
-
satisfaction was conducted (Jaccard, Turrisi, & Wan,
1990). Each analysis yielded a single degree-of-free
-
dom F test of any differences in slopes of body dissatis
-
faction on the outcome, and targeted contrasts were ex
-
amined between the gender and cohort groups.
Comparisons of demographic characteristics of
Project EAT–I participants who did and did not re
-
spond to Project EAT–II indicated several small but
significant differences. Consequently, in all analyses,
data were weighted to adjust for differential response
rates using the response propensity method (Little,
1986) in which the inverse of the estimated probability
that an individual responded at Time 2 was used as the
weight. Thus, all Time 2 estimates are generalizable to
the population represented by the original Project
EAT–I sample. The weighted ethnic and SES propor
-
tions of the study population were 48.3% White,
18.9% African American, 5.8% Hispanic, 19.6%
Asian, 3.6% Native American, and 3.8% mixed or
other race; and 17.8% low, 18.9% middle-low, 26.7%
middle, 23.3% middle-high, and 13.3% high SES.
Analyses were conducted with SAS 8.2, using propen
-
sity weights to account for nonresponse bias.
Results
Characteristics of Sample
and Univariate Correlates
of Time 2 Body Dissatisfaction
Table 1 shows mean scores for Time 1 and Time 2
depressive mood and self-esteem and mean scores for
Time 1 body dissatisfaction and BMI for each cohort of
girls and boys. Univariate correlations between Time 1
and Time 2, body dissatisfaction, depressive mood,
and self-esteem are shown in Table 2. Notably, Time 1
body dissatisfaction was positively correlated with
Time 2 depressive mood and inversely correlated with
Time 2 self-esteem in early- and midadolescent girls
and midadolescent boys. However, these correlations
were not observed in early-adolescent boys. Time 1
BMI was positively correlated with Time 2 depressive
mood and inversely correlated with Time 2 self-esteem
in early- and midadolescent girls. Time 1 BMI was also
positively associated with Time 2 depressive mood in
the early-adolescent boys but not midadolescent boys.
542
PAXTON, NEUMARK-SZTAINER, HANNAN, EISENBERG
Table 1. Means and Standard Deviations of Time 1 and Time 2 Depressive Mood, Self-Esteem, and Body Dissatisfaction
and Time 1 Body Mass Index
Scale
Range
Early Adolescence Midadolescence
Girls
a
Boys
b
Girls
c
Boys
d
M SD M SD M SD M SD
Time 1 Depressive Mood 6–18 10.5 2.7 9.4 2.6 11.3 2.7 9.8 2.7
Time 2 Depressive Mood 6–18 11.6 3.0 10.3 2.9 11.7 2.9 10.3 2.9
Time 1 Self-Esteem 6–24 17.6 3.4 18.4 3.6 17.2 3.4 18.8 3.4
Time 2 Self-Esteem 6–24 17.5 3.6 18.8 3.5 17.9 3.4 19.1 3.3
Time 1 Body Dissatisfaction 10–50 26.9 9.9 23.0 9.4 28.9 9.3 23.1 8.4
Time 2 Body Dissatisfaction 10–50 28.5 9.9 24.9 9.7 28.8 9.1 23.5 8.6
Time 1 Body Mass Index 21.7 4.4 21.3 5.2 22.6 4.5 23.0 4.2
Note: Within each group, the sample size for each analysis differs to a small degree due to differences in missing data.
a
n = 394–437.
b
n = 337–358.
c
n = 887–937.
d
n = 735–751.
Page 4
The Relation Between Time 1 Body
Dissatisfaction and Time 2 Depressive
Mood and Self-Esteem
Early-adolescent girls. Table 3 presents sum
-
mary statistics for multiple regression analyses in
which we examined whether Time 1 body dissatisfac
-
tion predicted Time 2 depressive mood and low self-es
-
teem in early-adolescent girls. As hypothesized, after
controlling for Time 1 depressive mood, demographic
variables, and BMI, Time 1 body dissatisfaction was a
unique predictor of depressive mood 5 years later. Sim
-
ilarly, Time 1 body dissatisfaction was a unique inverse
predictor of Time 2 self-esteem. Whereas Time 1 BMI
was a significant univariate correlate of Time 2 depres
-
sive mood and self-esteem, when entered into the
regression with body dissatisfaction, it did not make a
unique contribution to variance in Time 2 depressive
mood or self-esteem. When the regression analyses
were conducted without controlling for BMI, the effect
of Time 1 body dissatisfaction was similar as when
BMI was controlled (depressive mood: estimate = .05,
SE = .02, F = 7.06, p < .008; self-esteem: estimate =
–.08, SE = .02, F = 16.64, p < .0001).
Early adolescent boys. Table 4 presents sum
-
mary statistics for multiple regression analyses in which
543
BODY DISSATISFACTION PREDICTS DEPRESSION AND LOW SELF-ESTEEM
Table 2. Pearson’s Correlations Between Time 2 Depressive Mood and Self-Esteem and Time 1 Depressive Mood,
Self-Esteem, Body Dissatisfaction, Body Mass Index, and SES
Girls Boys
BD 1 BD 2 Dep 1 Dep 2 SE 1 SE 2 BD 1 BD 2 Dep 1 Dep 2 SE 1 SE 2
Early Adolescence
a
BD 2 .35 .23
Dep 1 .33 .23 .14 .11
Dep 2 .22 .31 .32 .09
ns
.28 .32
SE 1 –.47 –.32 –.52 –.25 –.45 –.21 –.37 –.18
SE 2 –.32 –.57 –.31 –.52 .37 –.10
ns
–.43 –.30 –.55 .26
BMI 1 .32 .26 .02
ns
.12 –.08
ns
–.18 .42 .17 .15 .11 –.26 –.07
ns
SES –.13 –.11 –.23 –.09
ns
.21 .05
ns
–.13 –.22 –.11 –.10
ns
.21 .10
ns
Mid-Adolescence
b
BD 2 .48 .42
Dep 1 .31 .15 .33 .21
Dep 2 .18 .34 .39 .24 .39 .40
SE 1 –.52 –.29 –.50 –.28 –.42 –.22 –.47 –.25
SE 2 –.28 –.52 –.27 –.55 .47 –.27 –.45 –.30 –.58 .39
BMI 1 .33 .32 .02
ns
.12 –.12 –.11 .31 .30 .07
ns
.07
ns
–.09 –.06
ns
SES –.12 –.15 –.01
ns
–.03
ns
.06
ns
.05
ns
–.11 –.12 –.05
ns
–.01
ns
.13 .04
ns
Note: With 300 observations correlations of 0.11, 0.15, and 0.19 are statistically significant at p < .05, .01, and .001 levels, and with 700 observa
-
tions the corresponding critical correlations are 0.073, 0.10, and 0.13. BD1=Time1body dissatisfaction; BD2=Time2body dissatisfaction;
Dep1=Time1depressive mood; Dep2=Time2depressive mood; SE 1 = Time 1 self-esteem; SE 2 = Time 2 self-esteem; BMI 1 = Time 1 body
mass index; SES = socioeconomic status; ns = not significant at p < .05.
a
Girls: n = 376–435; boys: n = 324–354.
b
Girls: n = 857–929; boys: n = 717–747.
Table 3. Summary of Multiple Regression Analyses Predicting Time 2 Depressive Mood and Time 2
Self-Esteem in Early Adolescent Girls
df Estimate SE F p
Time 2 Depressive Mood
a
Ethnicity 5 0.42 .832
SES 4 0.06 .993
Time 1 Depressive Mood 1 .32 .06 24.66 <.001
Time 1 Body Mass Index 1 .05 .04 2.13 .146
Time 1 Body Dissatisfaction 1 .04 .02 5.32 .022
Time 2 Self-Esteem
b
Ethnicity 5 0.44 .822
SES 4 0.26 .906
Time1 Self-Esteem 1 .28 .06 23.79 <.001
Time 1 Body Mass Index 1 –.07 .04 2.7 .102
Time 1 Body Dissatisfaction 1 –.07 .02 9.64 .002
Note: Sample sizes differ slightly due to differences in missing data. SES = socioeconomic status.
a
n = 356, R
2
= .14.
b
n = 350, R
2
= .21.
Page 5
we examined whether Time 1 body dissatisfaction pre
-
dicted Time 2 depressive mood and low self-esteem in
early-adolescent boys. When controlling for BMI, as
hypothesized and as one might expect from the ab
-
sence of a univariate association, Time 1 body dissatis
-
faction was not a unique predictor of Time 2 depressive
mood or self-esteem at 5-year follow-up. Similarly,
when BMI was not controlled, body dissatisfaction did
not significantly contribute to the variance in Time 2
depressive mood or self-esteem (depressive mood: es
-
timate = .01, SE = .02, F = 0.50, p < .48; self-esteem:
estimate = –.01, SE = .02, F = .27, p < .60).
Midadolescent girls. Table 5 presents summary
statistics for multiple regression analyses in which we
examined whether Time 1 body dissatisfaction pre
-
dicted Time 2 depressive mood and low self-esteem in
midadolescent girls. Although Time 1 body dissatisfac
-
tion was a univariate correlate of Time 2 depressive
mood and self-esteem, in the multiple regression analy
-
ses, after controlling for Time 1 of the dependent vari
-
able,demographicvariables,andBMI,Time1bodydis
-
satisfactiondidnotuniquely contributeto the prediction
of the outcome variables. It was of interest that Time 1
BMI did significantly contribute to the prediction of
Time 2 depressive mood and self-esteem. However, the
failureofTime1bodydissatisfactiontopredicttheTime
2variableswasnot aresult of shared variancewith Time
1 BMI because, when the multiple regression analyses
that did not control for BMI were examined, Time 1
body dissatisfaction still was not a significant predictor
of depressive mood or self-esteem 5 years later (depres
-
sive mood: estimate = .01, SE = .01, F = 1.92, p < .17;
self-esteem:estimate=.02,SE=.01,F=1.36,p<.24).
Midadolescent boys. Finally, Table 6 presents
summary statistics for multiple regression analyses in
which we examined whether Time 1 body dissatisfac
-
tion predicted Time 2 depressive mood and low self-es
-
teem in midadolescent boys. As hypothesized, after en
-
tering control variables, Time 1 body dissatisfaction
made significant unique contributions to the prediction
544
PAXTON, NEUMARK-SZTAINER, HANNAN, EISENBERG
Table 4. Summary of Multiple Regression Analyses Predicting Time 2 Depressive Mood and Time 2
Self-Esteem in Early Adolescent Boys
df Estimate SE F p
Time 2 Depressive Mood
a
Ethnicity 5 2.36
b
.041
SES 4 0.56 .690
Time 1 Depressive Mood 1 .33 .06 31.12 <.001
Time 1 Body Mass Index 1 .01 .03 0.04 .838
Time 1 Body Dissatisfaction 1 .02 .02 0.75 .386
Time 2 Self-Esteem
c
Ethnicity 5 3.19
d
.008
SES 4 1.68 .155
Time 1 Self-Esteem 1 .27 .06 20.95 <.001
Time 1 Body Mass Index 1 –.01 .04 0.11 .745
Time 1 Body Dissatisfaction 1 .00 .02 0.00 .979
Note: Sample sizes differ slightly due to differences in missing data.
a
n = 297, R
2
=.18.
b
Hispanic ethnicity predicted higher depressive mood; estimate = 1.36, SE = 0.66.
c
n = 291, R
2
= .17.
d
African American ethnicity predicted higher self-esteem; estimate = 1.46, SE = 0.54.
Table 5. Summary of Multiple Regression Analyses Predicting Time 2 Depressive Mood and Time 2
Self-Esteem in Midadolescent Girls
df Estimate SE F p
Time 2 Depressive Mood
a
Ethnicity 5 0.50 .775
SES 4 0.73 .574
Time 1 Depressive Mood 1 .39 .04 113.62 <.001
Time 1 Body Mass Index 1 .06 .02 5.99 .015
Time 1 Body Dissatisfaction 1 .01 .01 0.27 .602
Time 2 Self-Esteem
b
Ethnicity 5 0.94 .455
SES 4 0.36 .834
Time 1 Self-Esteem 1 .44 .04 147.98 <.001
Time 1 Body Mass Index 1 –.07 .03 6.76 .009
Time 1 Body Dissatisfaction 1 .00 .01 0.10 .754
Note: Sample sizes differ due to differences in missing data.
a
n = 843, R
2
= .15.
b
n = 815, R
2
= .23.
Page 6
of Time 2 depressive mood and self-esteem (Table 6).
Similar findings were observed when BMI was not
controlled (depressive mood: estimate = .05, SE = .01,
F = 14.95, p < .0001; self-esteem: estimate = –.05, SE =
.01, F = 12.36, p < .0005).
Comparison of Gender
and Cohort Effects
Confirming the hypothesis that Time 1 body dissat
-
isfaction would predict Time 2 depressive mood and
self-esteem in early- but not midadolescent girls and
mid- but not early-adolescent boys, in the analyses of
covariance the 1 degree-of-freedom F tests of interac
-
tion were significant for the differences in slopes of
body dissatisfaction on Time 2 depressive mood, F(1,
2181) = 4.05, p = .044, and self-esteem, F(1, 2181) =
10.6, p = .001. Tables 3 through 6 show the stratum
specific effects of differences in body dissatisfaction
on depressive mood and self-esteem measured 5 years
later, controlling for ethnicity, SES, Time 1 BMI, and
Time 1 depressive mood or self-esteem, respectively.
Table 7 shows the effect of the interaction of body dis
-
satisfaction with gender and cohort, confirming the
stratification analyses by showing significant effects in
the expected direction for each outcome only in early-
adolescent girls and midadolescent boys. Figure 1
presents the stratified results graphically, showing the
maximum predicted difference on the Depressive
Mood scale at Time 2 between 2 participants who dif
-
fered at Time 1 by the maximum on the body dissatis
-
faction scale (50 compared to 10). Smaller differences
in body dissatisfaction would lead to proportionally
smaller differences on the Depressive Mood at Time 2.
Figure 2 can be interpreted similarly but for effects on
self-esteem. Note in Figure 1 and 2 the similarity of
slopes in early-adolescent girls and midadolescent
boys and the essentially null slopes in the other two
gender and cohort groups.
Discussion
This study found body dissatisfaction to be a pro
-
spective risk factor for depressive mood and low self-
esteem over a sustained period of time (5 years) in both
545
BODY DISSATISFACTION PREDICTS DEPRESSION AND LOW SELF-ESTEEM
Table 6. Summary of Multiple Regression Analyses Predicting Time 2 Depressive Mood and Time 2
Self-Esteem in Midadolescent Boys
df Estimate SE F p
Time 2 Depressive Mood
a
Ethnicity 5 4.93
b
<.001
SES 4 0.43 .788
Time 1 Depressive Mood 1 .36 .04 82.46 <.001
Time 1 Body Mass Index 1 .01 .02 0.08 .773
Time 1 Body Dissatisfaction 1 .05 .01 12.45 <.001
Time 2 Self-Esteem
c
Ethnicity 5 3.78
d
.002
SES 4 1.87 .115
Time 1 Self-Esteem 1 .33 .04 74.69 <.001
Time 1 Body Mass Index 1 –.01 .03 0.08 .784
Time 1 Body Dissatisfaction 1 –.05 .02 9.38 .002
Note: Sample sizes differ due to differences in missing data.
a
n = 705, R
2
= .19.
b
African American ethnicity predicted lower depression; estimate = –1.24, SE = 0.32.
c
n = 695, R
2
= .20.
d
African American ethnicity predicted higher self-esteem; estimate = 1.43, SE = 0.37.
Table 7. Slopes of Depressive Mood and of Self-Esteem at Time 2 Regressed on Body
Dissatisfaction at Time 1 in Analyses of Covariance Examining the Three-Way Interaction
Between Gender, Cohort, and Body Dissatisfaction
Estimate SE t p
Depressive Mood .045
Early Adolescent Girls .037 .015 2.44 .015
Midadolescent Girls .009 .010 0.89 .372
Early Adolescent Boys .015 .018 0.83 .404
Midadolescent Boys .042 .012 3.40 .001
Self-Esteem .002
Early Adolescent Girls –.055 .017 –3.18 .001
Midadolescent Girls –.017 .012 –1.35 .176
Early Adolescent Boys .018 .021 0.86 .389
Midadolescent Boys –.043 .014 –3.02 .003
Page 7
girls and boys. However, the findings demonstrated
that body dissatisfaction predicted depressive mood
and low self-esteem in early- but not midadolescent
girls and mid- but not early-adolescent boys. Although
several interpretations are possible, our prospective re
-
search is consistent with the proposal that body dissat
-
isfaction is a risk factor for increases in depressive
mood and low self-esteem. Given the high value placed
on achieving the current body ideal (Stice & Bearman,
2001), perceived failure to live up to these standards is
likely to increase negative self-evaluation. In addition,
in Western societies, which tend to view characteristics
such as body weight, muscularity, and leanness as un
-
der individual control, an adolescent who is dissatis
-
fied with his or her body is likely to perceive this to be
the result of personal inadequacy. Such negative beliefs
are likely to have the long-term effect of increasing de
-
pressive mood and low self-esteem.
Body dissatisfaction and depressive mood and low
self-esteem may also have a spiral relation. Recent re
-
search suggests self-esteem is a prospective risk factor
for body dissatisfaction in girls across adolescence
(Paxton, Eisenberg, & Neumark-Sztainer, in press) and
depressive mood is a risk factor for body dissatisfaction
in mid- to late-adolescent boys (Paxton et al., in press;
Presnell,Bearman,& Stice, 2004).Bodydissatisfaction
maycontributetodepressivemoodand lowself-esteem,
which may in turn increase body dissatisfaction. Possi
-
bly, in some adolescents the nature of the relation be
-
tween body dissatisfaction and negative affect is in one
direction, whereas in others it is in the opposite.
Body dissatisfaction predicted depressive mood and
low self-esteem at different developmental stages in
girls compared to boys, consistent with the study over a
shorter time period by Holsen and colleagues (2001). It
is a time of physical changes in girls that frequently
move them away from the thin social ideal. As illus
-
trated in Table 1, this was a period of increase in body
dissatisfaction in our sample. In addition, it is a devel
-
opmental stage in which there are strong pressures to
546
PAXTON, NEUMARK-SZTAINER, HANNAN, EISENBERG
Figure 1. Differences in depressive mood (Time 2) for different levels of Time 1 depressive mood in early- and midadolescent girls and
boys, adjusted for Time 1 depressive mood, BMI, ethnicity, and SES in gender and cohort stratified analyses.
Figure 2. Differences in self-esteem (Time 2) for different levels of Time 1 body dissatisfaction in early- and midadolescent girls and
boys, adjusted for Time 1 self-esteem, BMI, ethnicity, and SES in gender and cohort stratified analyses.
Page 8
conform and be accepted by friends and peers (Berndt
& Hestenes, 1996). Under these conditions, poor body
image during this period could lead to feelings of
unattractiveness and uncertainty about acceptability,
contributing to increases in depressive mood and low
self-esteem. In late-adolescent girls, body dissatisfac
-
tion was not a unique predictor of increases in psy
-
chological symptoms. Notably, there was little mean
change in body dissatisfaction during this period. Girls
who are vulnerable to a negative psychological impact
of body dissatisfaction may already have been affected
earlier in teenage years and during the latter period re
-
main stable but high on body dissatisfaction and de
-
pressive mood and low on self-esteem. During this pe
-
riod, other factors may become more pertinent triggers
to increased depression and low self-esteem, such as
failed dieting attempts and related self-criticism or re
-
lationship problems.
It has been suggested that boys become aware of ex
-
ternalsocialpressurestoconformtoanappearanceideal
laterthangirls(Holsenetal.,2001).However,ifthiswas
the explanation of the differences between cohorts, one
would expect a substantial increase in body dissatisfac
-
tion in the later developmental phase in boys, which is
not the case. Rather, there was an increase in body dis-
satisfaction across the assessment period in the
early-adolescent boys. Although body dissatisfaction
itself may not intensify, the importance attached to
physicalappearance may become greaterduringthisde-
velopmental period compared to the earlier phase, and
the interaction between these factors may contribute to
increases in depressive mood and low self-esteem.
Except in the case of the midadolescent girls, BMI
was not a unique predictor of negative outcomes, con
-
sistent with previous research over a shorter time pe
-
riod (Needham & Crosnoe, 2005; Stice & Bearman,
2001). In our research, the pattern was somewhat dif
-
ferent in girls from midadolescence to young adult
-
hood. In this group, Time 1 BMI but not body dissatis
-
faction was a unique predictor of negative affect. It is
notable, however, that Time 1 body dissatisfaction was
correlated with Time 2 depressive mood and low self-
esteem. Our results suggest that body dissatisfaction
and BMI are closely related in midadolescent girls and
share a substantial amount of variance in the prediction
of depressive mood and low self-esteem. It is unlikely
that the observed relations with depressive mood and
low self-esteem were due to BMI per se. It is more
likely that these relations were mediated by negative
consequence of failed dieting experiences (Stice &
Bearman, 2001) or the negative evaluation, weight
teasing, and social discrimination frequently experi
-
enced by larger young women (e.g., Crandall, 1994;
Rothblum, Brand, Miller, & Oetjen, 1989; Wertheim et
al., 2001).
Although differences in the pattern of prospective
risk factors are suggested by our study, our design lim
-
its conclusions that can be drawn regarding timing of
the relation between Time 1 body dissatisfaction and
Time 2 depressive mood and low self-esteem. As par
-
ticipants were only assessed at 5-year intervals, it is
possible that body dissatisfaction did prospectively
predict psychological outcomes at some point during
the time period but not precisely when assessed. For
example, in girls, body dissatisfaction at 15.5 years
may predict low self-esteem at 17.5 years but not at
20.5 years. In addition, our focus in this research was
on the potential role of body dissatisfaction as a predic
-
tor of psychological outcomes. It also should be noted
that the overall variance accounted for by the regres
-
sion analyses was not very high (early-adolescent girls:
depressive mood R
2
= .14, self-esteem R
2
= .21; early-
adolescent boys: depressive mood R
2
= .18, self-esteem
R
2
= .17; midadolescent girls: depressive mood R
2
=
.15, self-esteem R
2
= .23; early-adolescent boys: de
-
pressive mood R
2
= .19, self-esteem R
2
= .20). A wide
range of other variables is also likely to contribute to
the prediction of increases in depressive mood and low
self-esteem in adolescents over this time period, such
as family functioning, peer interactions, and stressful
life events. The impact of these variables may also vary
according to developmental stage.
This study has a number of strengths that include the
large and diverse sample of boys and girls, assessed at
two developmental stages; the 5-year follow-up period
that facilitated examination of body dissatisfaction as a
long-term risk factor for mental health outcomes; as-
sessment instruments that were sound for both girls and
boys; and measured height and weight. This study also
hasa number of limitations.Attritionfromthefirsttothe
follow-up assessment was substantial (47%). To over
-
comepotential biasresulting fromthis, a weighting pro
-
cedure was used to make findings more generalizable to
the population represented in the EAT–I sample. It is
also not clear the extentto which the results from this re
-
search are generalizable to other populations. The con
-
sistency of our findings with those of Holsen and col
-
leagues (2001), who conducted their research in
Norway, provides support for the generalizability of the
findings, but further research is required in other popu
-
lations to confirm it. An additional limitation is that in
research of this kind it is possible that the observed rela
-
tions may be the result of a variable not assessed that in
-
fluencedbothbodydissatisfactionanddepressivemood
and low self-esteem. For example, sexual or physical
abusepriortoTime1measurementoccasionmayleadto
both Time 1 body dissatisfaction and, especially if con
-
tinuing, to Time 2 depressive mood and self-esteem and
thusbe the common underlying basis for the association
wefound between Time1bodydissatisfactionandTime
2 outcomes. Thus, further research is required to con
-
firm causal relations indicated by this research. In addi
-
tion, having only two time points in longitudinal re
-
search limits interpretations that can be drawn. Having
547
BODY DISSATISFACTION PREDICTS DEPRESSION AND LOW SELF-ESTEEM
Page 9
more frequent assessments would assist in elucidating
mechanisms of action and mediators of the relations ob
-
served.
A number of areas for future research emerge from
this study. Although body dissatisfaction has been iden
-
tified as a prospective risk factor for depressive mood
and self esteem, this study did not explore whether par
-
ticular patterns of persistence of body dissatisfaction
were more strongly associated with the development of
depressive mood and self-esteem. It would be valuable
to examine whether relations between Time 1 body
dissatisfaction and psychological mood are stronger
when body dissatisfaction is stable throughout the fol
-
low-up period rather than present at first assessment
but diminished over time. In this research it was also
noted that ethnic background predicted depressive
mood and self-esteem in boys but not girls, a finding
that requires further exploration. In particular, being
African American was psychologically protective in
boys in the sample, consistent with previous research
(e.g., Gray-Little & Hafdahl, 2000). Recent research,
however, indicates that depression in African Ameri
-
can teens is higher if they live among non-Hispanic
White Americans than African Americans (Wight,
Aneshensel, Botticello, & Sepulveda, 2005). Thus, it
will be important to consider the generalizability of
this observation in other social environments.
References
Allgood-Merten, B., Lewinsohn, P. M., & Hops, H. (1990). Sex dif-
ferences and adolescent depression. Journal of Abnormal Psy
-
chology, 99, 55–63.
Berndt, T. J., & Hestenes, S. L. (1996). The developmental course of
social support: Family and peers. In L. Smolak, M. P. Levine, &
R. Striegel-Moore (Eds.), The developmental psychopathology
of eating disorders: Implications for research, prevention and
treatment (pp. 77–106). Hillsdale, NJ: Lawrence Erlbaum As
-
sociates, Inc.
Byrne, B. M. (1983). Investigating measures of self-concept. Mea
-
surement and Evaluation in Guidance, 16, 115–126.
Cafri, G., Thompson, J. K., Ricciardelli, L., McCabe, M., Smolak,
L., & Yesalis, C. E. (2005). Pursuit of the muscular ideal: Physi
-
cal and psychological consequences and putative risk factors.
Clinical Psychology Review, 25, 215–239.
Cash, T. F. (2002). Cognitive behavioral perspectives on body image.
In T. F. Cash & T. Pruzinski (Eds.), Body image: A handbook of
theory, research and clinical practice (pp. 38–46). New York:
Guilford.
Cohane, G. H., & Pope, H. G., Jr. (2001). Body image in boys: A re
-
view of the literature. International Journal of Eating Disor
-
ders, 29, 373–379.
Crandall, C. S. (1994). Prejudice against fat people: Ideology and
self-interest. Journal of Personality & Social Psychology, 66,
882–894.
Demo, D. H. (1985). The measurement of self-esteem: Refining
our methods. Journal of Personality & Social Psychology, 48,
1450–1502.
Derogatis, L. R. (1977). SCL–90 administration, scoring and proce
-
dures manual–1. Baltimore: Johns Hopkins University School
of Medicine.
Gray-Little, B., & Hafdahl, A. R. (2000). Factors influencing racial
comparisons of self-esteem: A quantitative review. Psychologi
-
cal Bulletin, 126, 26–54.
Holsen, I., Kraft, P., & Roysamb, E. (2001). The relationship be
-
tween body image and depressed mood in adolescence: A
5-year longitudinal panel study. Journal of Health Psychology,
6, 613–627.
Jaccard, J., Turrisi, R., & Wan, C. K. (1990). Interaction effects in
multiple regression. Newbury Park, CA: Sage.
Johnson, F., & Wardle, J. (2005). Dietary restraint, body dissatisfac
-
tion and psychological distress: A prospect analysis. Journal of
Abnormal Psychology, 114, 119–125.
Kandel, D. B., & Davies, M. (1982). Epidemiology of depressive
mood in adolescents: An empirical study. Archives of General
Psychiatry, 39, 1205–1212.
Keery, H., van den Berg, P., & Thompson, J. K. (2004). An evalua
-
tion of the Tripartite Influence Model of body dissatisfaction
and eating disturbance with adolescent girls. Body Image, 1,
237–251.
Killen, J. D., Taylor, C. B., Hayward, C., Haydel, K. F., Wilson, D.
M., Hammer, L., et al. (1996). Weight concerns influence the
development of eating disorders: A 4-year prospective study.
Journal of Consulting & Clinical Psychology, 64, 936–940.
Little, R. J. A. (1986). Survey nonresponse adjustments for estimates
of means. International Statistical Review, 54, 137–139.
Needham, B. L., & Crosnoe, R. (2005). Overweight status and de
-
pressive symptoms during adolescence. Journal of Adolescent
Health, 36, 48–55.
Neumark-Sztainer, D., Story, M., Hannan, P. J., & Croll, J. (2002).
Overweight status and eating patterns among adolescents:
Where do youth stand in comparison to the Healthy People
2010 Objectives? American Journal of Public Health, 92,
844–851.
Neumark-Sztainer, D., Story, M., Hannan, P. J., Perry, C., & Irving,
L. M. (2002). Weight-related concerns and behaviors among
overweight and non-overweight adolescents: Implications for
preventing weight-related disorders. Archives of Pediatrics and
Adolescent Medicine, 156, 171–178.
Paxton, S. J., Eisenberg, M. E., & Neumark-Sztainer, D. (in press).
Prospective predictors of body dissatisfaction in adolescent
girls and boys: A five-year longitudinal study. Developmental
Psychology.
Paxton, S. J., Norris, M., Wertheim, E. H., Durkin, S. J., & Anderson,
J. (2005). Body dissatisfaction, dating, and importance of thin
-
ness to attractiveness in adolescent girls. Sex Roles: A Journal
of Research, 53, 663–675.
Pingitore, R., Spring, B., & Garfield, D. (1997). Gender differences
in body satisfaction. Obesity Research, 5, 402–409.
Presnell, K., Bearman, S. K., & Stice, E. (2004). Risk factors for
body dissatisfaction in adolescent boys and girls: A prospec
-
tive study. International Journal of Eating Disorders, 36,
389–401.
Ricciardelli, L. A., & McCabe, M. P. (2001). Dietary restraint and
negative affect as mediators of body dissatisfaction and bulimic
behavior in adolescent girls and boys. Behaviour Research &
Therapy, 39, 1317–1328.
Rierdan, J., Koff, E., & Stubbs, M. L. (1989). A longitudinal analysis
of body image as a predictor of the onset and persistence of
adolescent girls’ depression. Journal of Early Adolescence, 9,
454–466.
Rosenberg, M. (1965). Society and adolescent self-image. Princeton,
NJ: Princeton University Press.
Rothblum, E. D., Brand, P., Miller, C., & Oetjen, H. (1989). Results
of the NAAFA survey on employment discrimination: Part II.
NAAFA Newsletter, 17, 4–6.
Stice, E. (2001). A prospective test of the dual pathway model of
bulimic pathology: Mediating effects of dieting and negative af
-
fect. Journal of Abnormal Psychology, 110, 124–135.
548
PAXTON, NEUMARK-SZTAINER, HANNAN, EISENBERG
Page 10
Stice, E. (2002). Risk and maintenance factors for eating pathology:
A meta-analytic review. Psychological Bulletin, 128, 825–848.
Stice, E., & Bearman, S. K. (2001). Body image and eating distur
-
bances prospectively predict increases in depressive symptoms
in adolescent girls: A growth curve analysis. Developmental
Psychology, 37, 597–607.
van den Berg, P., Wertheim E. H., Thompson, J. K., & Paxton, S. J.
(2002). Development of body image, eating disturbance, and
general psychological functioning in adolescent females: A
replication using covariance structure modeling in an Austra
-
lian sample. International Journal of Eating Disorders, 31,
46–51.
Wertheim, E. H., Koerner, J., & Paxton, S. J. (2001). Longitudinal
predictors restrictive eating and bulimic behavior in adolescent
girls. Journal of Youth and Adolescence, 30, 69–81.
Wichstrom, L. (1999). The emergence of gender difference in de
-
pressed mood during adolescence: The role of intensified gen
-
der socialization. Developmental Psychology, 35, 232–245.
Wight, R. G., Aneshensel, C. S., Botticello, A. L., & Sepulveda, J. E.
(2005). A multilevel analysis of ethnic variation in depressive
symptoms among adolescents in the United States. Social Sci
-
ence & Medicine, 60, 2073–2084.
Wylie, R. C. (1989). Measures of self-concept. Lincoln: University
of Nebraska Press.
Received June 27, 2005
Accepted April 13, 2006
549
BODY DISSATISFACTION PREDICTS DEPRESSION AND LOW SELF-ESTEEM
Page 11
    • "For example, interpersonal psychotherapy may help to improve psychological well-being by assisting individuals to develop their interpersonal skills, manage potentially difficult relationship transitions, and build their confidence in interacting with others (Klerman, Weissman, Rounsaville, & Chevron, 1984). Body dissatisfaction has been linked to a number of indicators of poor psychological well-being (Davison & McCabe, 2006; Jones et al., 2016; Paxton et al., 2006; Stice & Shaw, 2002) and, as expected, predicted greater psychopathology in the transgender group. Cross-sex hormone treatment and SRS are intended to better align the physical body with the person's experienced gender, though this is achieved to varying degrees between individuals . "
    [Show abstract] [Hide abstract] ABSTRACT: Research has yet to identify specific predictors of poor psychological well-being and quality of life in transgender people. This study aimed first to explore the predictive value of five factors known to be associated with poor psychological well-being in cis- and transgender people; age, self-esteem, victimisation, interpersonal problems, and body dissatisfaction. Second, to investigate the mediatory role of self-esteem and social support. Two hundred and eight participants (104 transgender and 104 cisgender controls), matched by age and gender, completed measures of these predictor variables, along with general psychopathology and functional quality of life. The results indicate that in the transgender group, greater psychopathology and greater depression were predicted by younger age (psychopathology only), lower self-esteem, greater body dissatisfaction, and greater interpersonal problems. In the cisgender group, only lower self-esteem and greater interpersonal problems were significant predictors of these factors. For quality of life, lower self-esteem and greater interpersonal problems were significant predictors of low quality of life in both groups. Self-esteem but not social support mediated the above relationships. Overall, self-esteem and interpersonal problems appear to be crucial factors that influence well-being. Those providing treatment to transgender people should pay more attention to these areas.
    Full-text · Article · May 2016 · Sexual and Relationship Therapy
    0Comments 0Citations
    • "Longitudinal studies revealed that body dissatisfaction was predicted by low self-esteem (Gilbert and Meyer, 2005). Conversely, a 5-year prospective study found body dissatisfaction responsible for low self-esteem, casting doubt on the directionality of these factors (Paxton et al., 2006). In order to understand better how body image is predicted, Phan and Tylka (2006) tested a model of disordered eating with Asian Americans to expand prior findings on the connections between self-esteem and body dissatisfaction . "
    [Show abstract] [Hide abstract] ABSTRACT: In this study, a survey of Sri Lankan adolescents explored relationships between thin-ideal internalization and body dissatisfaction-issues that have received greater research attention in East Asia and the West. Girls scored lower in thin-ideal internalization and awareness and higher in self-esteem compared to boys; however, unlike in previous studies, body dissatisfaction did not differ significantly by gender. Hierarchical linear multiple regression analyses revealed that self-esteem moderated the relationship between thin-ideal internalization and body dissatisfaction among girls, but this was not observed for boys. Self-esteem as a protective factor against body dissatisfaction among Sri Lankan female adolescents was confirmed.
    Preview · Article · Mar 2016 · Journal of Health Psychology
    0Comments 0Citations
    • "The effect of body dissatisfaction on anxiety and depression did not vary as a function of age or gender. These results tie together the research showing that body dissatisfaction predicts lower levels of self-esteem over the longer-term (Johnson and Wardle, 2005; Paxton et al., 2006; Wichstrøm and von Soest, 2016) and the research showing that self-esteem is associated with symptoms of depression and anxiety (Dooley et al., 2015; Nima et al., 2013; Orth et al., 2008; Wichstrøm and von Soest, 2016). Thus, in addition to replicating associations already found in the literature (e.g., the association between body dissatisfaction and psychological distress), the results of the present study help to refine our understanding of body dissatisfaction by more clearly identifying the process that explains the symptoms. "
    [Show abstract] [Hide abstract] ABSTRACT: This brief report tests the mediating effect of self-esteem in the relationship between body dissatisfaction and symptoms of depression and anxiety. A sample of 409 adolescents (females = 58.4%) aged between 14 and 18 years completed the Rosenberg Self-Esteem Scale, the Contour Drawing Rating Scale, the Multidimensional Anxiety Scale for Children, and the Center for Epidemiologic Studies Depression Scale. Overall, results for the indirect effects analysis were significant for both anxiety and depression, which confirmed the mediating role of self-esteem. Thus, a negative perception of one's body image has the effect of lowering self-esteem, which in turn increases psychological distress.
    Full-text · Article · Feb 2016 · Journal of Health Psychology
    0Comments 0Citations
Show more