See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/225802798

Body Image and Self-Esteem Across Age and

Gender: A Short-Term Longitudinal Study

Article in Sex Roles · November 2010

DOI: 10.1007/s11199-010-9813-3

CITATIONS READS

47 1,609

4 authors, including:

Matthew Fuller-Tyszkiewicz Lina Ricciardelli

Deakin University Deakin University
72 PUBLICATIONS 705 CITATIONS 185 PUBLICATIONS 7,528 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

The Alcohol Intervention Training Program View project

All content following this page was uploaded by Matthew Fuller-Tyszkiewicz on 20 May 2014.

The user has requested enhancement of the downloaded file.

Sex Roles (2010) 63:672–681
DOI 10.1007/s11199-010-9813-3

ORIGINAL ARTICLE

Body Image and Self-Esteem Across Age

and Gender: A Short-Term Longitudinal Study
David Mellor & Matthew Fuller-Tyszkiewicz &
Marita P. McCabe & Lina A. Ricciardelli

Published online: 15 July 2010

# Springer Science+Business Media, LLC 2010

Abstract This study examined the relationship between theory (Fredrickson and Roberts 1997) was used to shape
body image and self-esteem across time and age for men the design of the current study, which examined this
and women. Participants were 150 men and 239 women association across the adult lifespan for both women and
aged between 20 and 86 years (M=59.73), recruited from men in Australia. Body dissatisfaction has been linked
the general population in Australia. They completed a with a range of adverse psychosocial consequences,
questionnaire assessing self-esteem, body dissatisfaction, including poor self-esteem, depression, eating disorders
body importance, and a range of other domains of life and obesity (Darby et al. 2007; Stice 2002; Wiederman
satisfaction at baseline and 2 years later. Cross-sectionally, and Pryor 2000). Therefore, we also aimed to evaluate,
we found that higher self-esteem was associated with lower using a short-term longitudinal design, whether the
body dissatisfaction, but there was no consistent relation- direction of the relationship between body dissatisfac-
ship between these variables across time for age and tion and global evaluations of self-worth declines with
gender. Women were more dissatisfied with their bodies age for either or both genders, after accounting for
than men; yet, men placed greater importance on their other areas of life satisfaction. Since findings from re-
appearance than women, and also reported high levels of search into body image and related issues in Australia
body dissatisfaction. are generally consistent with those found in other
Western contexts, the findings of the current study are
Keywords Body dissatisfaction . Self-esteem . Gender . likely to be generalizable to similar countries in the Western
Age . Longitudinal design world.
As in other Western countries, in Australia dissatisfac-
tion with physical appearance is a common experience for
Introduction many female adolescents and women, and more recently,
male adolescents and men (McCabe and Ricciardelli
There have been limited theoretically-based longitudinal 2004a; Tiggemann and Lynch 2001). Indeed, hospital-
studies on body image and its relationship with self- isation for body dissatisfaction and eating disorders, and
esteem among adult men and women. Objectification the need for continued long term psychological and
physical care for eating disordered patients (who are
predominantly female) is a huge financial burden on the
D. Mellor : M. Fuller-Tyszkiewicz : M. P. McCabe : public health system (Australian Institute of Health and
L. A. Ricciardelli
Welfare 2007; Pratt and Woolfenden 2002). Consequently,
School of Psychology, Deakin University,
Melbourne, Australia there is a need to further understand aspects of body
dissatisfaction across both gender and age, and the
D. Mellor (*) Australian Federal Government has made body dissatis-
School of Psychology, Deakin University,
faction and related problems a priority area for research
221 Burwood Highway,
Burwood, Victoria 3125, Australia (Ellis 2009; Roxon 2008). It is within this context that the
e-mail: [email protected] current project was undertaken.
Sex Roles (2010) 63:672–681 673

Objectification Theory, Gender and Body Dissatisfaction of reproductive potential (aged 15 to 40 years) and, on that
basis, we may expect gender differences in body image
According to objectification theory, sociocultural norms importance and body dissatisfaction to be greatest in this
relevant to physical appearance influence the way individuals period. However, the majority of studies examining body
view their own bodies and the impact that such evaluations image concerns have been from westernized countries
have on their sense of self-worth. Fredrickson and Roberts (predominantly USA, England, Australia and Canada) and
(1997) argue that in Westernized cultures these norms, have sampled adolescents and young adults (aged 18 to
which are communicated via mass media, friends, family, 25 years; Grogan 2008), a peak period for eating disorders
work colleagues, and even strangers encountered in every- and excessive dieting (van Hoeken et al. 2005). However,
day life, are more pervasive and emphasize more strongly both men and women continue to experience a range of
the importance of appearance for women than for men. changes to their physical appearance from around age forty
Consistent with this, Tiggemann and Pennington (1990) and beyond, including an increase in body size and weight,
found in an Australian study that gender differences in levels the appearance of wrinkles and loss of skin elasticity, as
of body dissatisfaction begin to appear around the age of 13 well as graying, thinning and (particularly in males) loss of
and they are still apparent beyond middle adulthood. hair (Andres 1989). Fredrickson and Roberts (1997) argue
Indeed there is evidence to suggest that this gender that the impact that the physical aging process has on body
difference is robust across a range of Western cultures other dissatisfaction and mental health will depend on the extent
than Australia, including the USA (Chang and Christakis to which an individual is able to relinquish or modify the
2003) and France (Parnot et al. 2006), as well as non- unrealistic appearance standards that s/he internalized
Westernized cultures, such as China (Chen and Jackson throughout youth and early adulthood. For individuals
2008), and India (Iqbal et al. 2006). Studies investigating who fail to modify their appearance-related beliefs, we
gender differences in body image importance have found may anticipate a worsening of body dissatisfaction and its
similarly consistent results; namely, that across a wide negative consequences. Conversely, for individuals who
range of contexts women report greater concern about successfully step out of the objectification limelight, body
physical appearance than men (e.g., Canada, Morry and image concerns may exert less impact on their wellbeing
Staska 2001; United States, Noll and Fredrickson 1998). and feelings of self-worth. In this way, age-related changes
However, although women exhibit greater body dissatis- in body dissatisfaction may be mediated by changes in
faction than men, men in these studies also exhibit body body image importance.
image concerns (e.g., the USA, Forbes et al. 2001;
Australia, McCabe and Ricciardelli 2004a). Thus, as Forbes Body Image and Self-Esteem
et al. argued, to focus only on women implies that body
dissatisfaction is in some way a female problem or a Some researchers (e.g., Webster and Tiggemann 2003) have
problem that has different causes in women compared to pointed to the decline in the importance one attributes to
men. Forbes and colleagues suggested that while body appearance over the adult life span, arguing that this may
dissatisfaction may be more likely to have severe con- serve as a buffer against the ill-effects of appearance
sequences in women (e.g., clinical eating disorders), male concerns. Such an explanation may also help to account
body dissatisfaction is a major factor in the widespread use for the decline in the prevalence of eating disorders with
of anabolic steroids. Along similar lines, Tiggemann and age (Heatherton et al. 1997; Tiggemann and Lynch 2001).
colleagues have argued that elements of objectification On the other hand, there is evidence to suggest that the
theory may generalize to male populations. Tiggemann and influence of body dissatisfaction on global evaluations of
Kuring (2004) demonstrated in a sample of Australian self-worth remains stable or actually strengthens in middle
university students aged 17 to 45 years that men who age. For instance, Tiggemann and Stevens (1999) compared
engage in appearance-related self-monitoring reported the strength of association between self-esteem and body
heightened levels of appearance anxiety, body shame and dissatisfaction across different age groups in Australia and
depressed mood. In another Australian study, Martins et al. found that the relationship was stronger in women aged 30
(2007) have also shown a link between self-objectification, to 49 years than in younger women. In another Australian
appearance-related self-monitoring and body dissatisfaction study, Webster and Tiggemann (2003) found that the
in both gay and heterosexual men. relationship between self-esteem and body dissatisfaction
was stronger for women aged 35 to 49 (the ‘middle age’
Objectification Theory, Age and Body Image group) than for women aged 20 to 34 (‘young adulthood’)
or 50 to 65 (‘older adults’). Furthermore, they showed that
Fredrickson and Roberts (1997) argued that the emphasis self-esteem and body dissatisfaction were comparable for
on female appearance is most pronounced during the period women in the young adulthood and middle age groups, and
674 Sex Roles (2010) 63:672–681

that this relationship was non-significant in the older adult considerably more concerned with their physical appearance
group. Wilcox (1997) examined levels of body dissatisfac- than the general population.
tion, body image importance and self-esteem in men and A more stringent test of the age-related tenets of
women in the US, each grouped in age by decades (20–29, objectification theory would entail evaluation of whether
30–39, etc. up to the age of 79). She found that level of the role of body image importance for evaluations of self-
body dissatisfaction and magnitude of the relationship worth generalize to populations characterized by normative
between body attitudes and self-esteem did not vary as a levels of body dissatisfaction and which provide a larger
function of age or gender. Interestingly, Wilcox also found that representation of middle age and elderly participants.
body image importance in this sample did not vary across the Moreover, the majority of studies described in the preced-
adult lifespan, suggesting that body image importance may ing sections failed to evaluate the relative contribution of
not necessarily decline with age. These findings are consistent appearance concerns for self-esteem within the context of
with Fredrickson and Roberts’ (1997) assertion that age- other, noted determinants of self-esteem. According to
related changes in body dissatisfaction (and its influence on Harter (1999) individuals make domain-specific judgments
mental health) are dependent on the extent to which body as well as global judgments of their self-worth, and
image importance changes across adulthood. typically the evaluations they make vary across domains.
Although the majority of studies that have evaluated the Further, domain-specific evaluations contribute to global
link between self-esteem and body image concerns have self-evaluations differentially. Although self-evaluations of
been cross-sectional, several researchers have implemented physical appearance have been shown consistently to
longitudinal designs in order to investigate the directional contribute strongly to global self-esteem, the omission of
nature of this relationship. Tiggemann (2005) evaluated other domains in previous studies is an important oversight
levels of body dissatisfaction and self-esteem in Australian given that prior research outside the field of body image
female high school students at two time points, separated (e.g., Cummins et al. 2003) has shown that the priority of
by 2 years. She found that body dissatisfaction at Time 1 other domains of life satisfaction (e.g., satisfaction with
predicted change in self-esteem, but that Time 1 self-esteem health, achievements, relationships, community engagement,
was unrelated to change in body dissatisfaction. Findings accumulation of material possessions, feelings of safety, and
from Paxton et al.'s (2006) study, which was also within an future security) changes across adulthood and may become
Australian context, suggest that the contribution of body more important predictors of wellbeing as we age. Failure to
dissatisfaction to self-esteem may change across the include these other domains of (dis)satisfaction may result in
different phases of life. Despite a consistent negative over-estimates of the importance of appearance concerns for
correlation between self-esteem and body dissatisfaction evaluations of self-worth. Thus, a more thorough investigation
across gender and age group, Paxton and colleagues found of the contribution of body dissatisfaction for self-esteem
that initial body dissatisfaction was predictive of subse- across the adult lifespan is warranted.
quent low self-esteem 5 years later in early-adolescent girls The purpose of the present study was to determine the
and mid-adolescent boys, but not in mid-adolescent girls or extent to which age- and gender-related differences in body
early-adolescent boys. dissatisfaction—and the nature of the relationship between
Other findings highlight the possibility that it is body body image concerns and self-esteem—as previously observed
image importance, moreso than body dissatisfaction, which in the literature are attributable to changes in importance of
impacts self-esteem. Cooper and Fairburn (1993) tracked appearance across adulthood. On the basis of objectification
body dissatisfaction, importance of appearance, self-esteem theory (Fredrickson and Roberts 1997), it was anticipated that
and mood in a sample of individuals in the UK who were there would be gender- and age-related differences in
being treated for bulimia nervosa. They found that change perceptions of body image, such that body image dissatis-
in body dissatisfaction over the course of 10 to 12 weeks of faction would be higher among women than men (Hypothesis
treatment was closely associated with change in mood, 1), and decrease with age (Hypothesis 2). Given the greater
whereas change in the importance of appearance was emphasis in Westernized cultures placed on the importance of
associated with changes in self-esteem. Similarly, in a physical appearance for young women (Fredrickson and
sample of obese individuals in the US who were recovering Roberts 1997), it was further anticipated that age-and gender-
from gastric bypass surgery, Masheb et al. (2006) found related effects on body dissatisfaction would be mediated by
that body image importance was associated with change in individual differences in body image importance (Hypotheses
self-esteem, while body dissatisfaction was associated with 3 and 4, respectively). To test the temporal effects of body
changes in both self-esteem and mood. While the studies by dissatisfaction on self-esteem, participants were tested at two
Cooper and Fairburn (1993) and Masheb et al. (2006) time points, separated by 2 years. It was hypothesized that in
implicate body image importance in evaluations of self- regression analyses, body dissatisfaction at Time 1 would
worth, individuals in these studies are likely to be predict self-esteem at Time 2 for both men (Hypothesis 5) and
Sex Roles (2010) 63:672–681 675

women (Hypothesis 6). It was anticipated that the magnitude ten items. Five items are positively-worded and five
of the relationship between body dissatisfaction and self- negatively-worded. Item examples are “I feel that I have a
esteem would change across the adult lifespan in accordance number of good qualities” and “At times I feel that I am no
with accompanying changes in the relevance of various good at all.” To allow for greater distinction in levels of
domains of self-worth. Specifically, a moderating effect self-esteem, Rosenberg’s (1965) original 4-point response
between body dissatisfaction and age would predict T2 self- scale was replaced with an 11-point response scale, with 0=
esteem scores for both men (Hypothesis 7) and women Strongly disagree, 5=Neutral, and 10=Strongly agree.
(Hypothesis 8), such that the influence of body dissatisfaction Self-esteem scores were obtained by adding all item
on self-esteem would be expected to decline with age. responses, after the negatively-worded item responses had
been reverse-coded. High scores indicated a strong sense of
self-esteem. The scale has sound psychometric properties,
Method with test-retest correlations over a period of 2 weeks
typically being in the range of .82–.88 and Cronbach’s
Participants alpha for various adolescent and adult samples being in the
range of .77–.88 (Rosenberg 1986). In this study, alpha was
The participants were initially drawn from the Australian Unity .86 at both T1 and T2.
Wellbeing project. This project, which commenced in 2001, Body importance and body dissatisfaction were assessed
monitors the subjective wellbeing of the Australian population. by two subscales of the Body Image and Body Change
Each survey involves a sample of 2,000 people drawn Inventory (McCabe and Ricciardelli 2004b). The Body
according to census data to be representative of the geographic Importance Scale is made up of seven items. The first four
distribution of the Australian population. The participants are items, which assess level of importance placed on weight,
new to the project, selected to be 18 years or over and fluent in shape and muscles (e.g., “How important to you is your
English, with an even gender composition. All reports and raw weight compared to other aspects of your life?”), are given to
data from these surveys are available from http://www.deakin. all participants regardless of gender. The remaining three
edu.au/research/acqol/index_wellbeing/index.htm. items are gender-specific: men are asked to rate the
At the end of the initial telephone interview, participants importance of chest, shoulders and arms, whereas women
are asked whether they would be willing to be involved are asked to rate the importance of hips, thighs and legs. This
with the project on a future occasion. Those who reply division into gender-specific items is based on prior research
‘yes’ are followed-up with a paper questionnaire on an which has shown that the former set is more relevant for men
annual basis thereafter. and the latter for women (McCabe and Ricciardelli 2004a, b).
All participants in the current study were members of a Participants respond on a 5-point scale ranging from 0 (very
cross-sectional study who had volunteered for further important) to 4 (not important at all), and a total score is
contact and were enrolled in our longitudinal study. Of obtained by summing the responses.
the 779 questionnaires mailed out to these volunteers at The Body Image Dissatisfaction Scale consists of seven
Time 1 (T1), 546 completed questionnaires were returned items which assess the level of dissatisfaction that is
(70.1% response rate). These 546 respondents were associated with each aspect of the body, or each of the
surveyed again at Time 2 (T2), 2 years later. Of these, body parts described above (e.g., “How satisfied are you
389 responded (71.2%; 49.9% of the T1 initial sample; 150 with your abdominal region/stomach?”). Participants respond
men and 239 women). Their ages ranged from 20 to on a 5-point scale ranging from 0 (very happy) to 4 (very
86 years, with a mean age of 59.73 years (SD=14.24). The unhappy). Scores are derived in a similar manner as for body
drop-outs were significantly younger (M=53.42 years, SD= image importance.
15.65) than the respondents, t(544)=4.66, p<.001. Gender Scores on the above scales have been shown to have
difference in drop-out rates was non-significant, χ2 (df=1)= high levels of internal consistency reliability with adolescent
0.52, p>.05. Of the remaining 389 participants, gender males and females (Cronbach’s α>.92), and they have
differences in age and BMI were non-significant; t(1, 387)= demonstrated concurrent validity with other key measures
0.93, p>.05, and t(1, 387)=−0.37, p>.05, respectively. of body concerns, the Stunkard’s Figure Body Drawings and
the Body Dissatisfaction subscale of the Eating Disorders
Measures Inventory (Ricciardelli and McCabe 2002). In the current
study, Cronbach’s alphas at T1 and T2, respectively, were.91
Participants completed all measures at T1 and at T2. Weight and.89 for men’s body dissatisfaction, .92 and .90 for men’s
and height were recorded on a demographic information sheet. body image importance, .92 and .90 for women’s body
Self-esteem was measured with the Rosenberg Self- dissatisfaction, and .92 and .91 for women’s body image
Esteem Scale (Rosenberg 1965). This scale is made up of importance.
676 Sex Roles (2010) 63:672–681

Domains of satisfaction were measured using the items of and Fidell 2007). On that basis, 18 multivariate outliers
the Personal Wellbeing Index (PWI; International Wellbeing were identified and removed, resulting in a final sample of
Group [IWG] 2006). The PWI consists of seven-items, each 389 cases being used in the analyses reported below.
prefaced with “How satisfied are you with...” The stems are:
“your standard of living,” “your health,” “your achievements Descriptive Statistics
in life,” “your personal relationships,” “how safe you feel,”
“feeling part of your community,” “your future security.” Preliminary analyses were conducted to evaluate the signifi-
Participants respond on an 11-point scale ranging from very cance of mean differences observed across time and gender
dissatisfied (0) to very satisfied (10), with neutral (5) as a for the domains of satisfaction (including body dissatisfac-
midpoint. Each item represents a separate domain as prior tion), self-esteem and body image importance (as shown in
research has demonstrated their unique predictive value for Table 1). Repeated measures MANCOVA with gender and
key outcome variables, including overall wellbeing (IWG time as independent variables and BMI as a covariate
2006). A total wellbeing score was not required in this study; indicated main effects for time, F(1, 386)=16.21, p<.01;
each item was used as a separate domain measure. and gender, F(1, 386)=7.19, p<.01. The interaction effect of
time by gender was non-significant, F(1, 386)=3.11, p>.05.
Procedure Further inspection of the main effect for gender revealed
that women scored significantly higher than men on the
Approval to conduct the study was obtained from the Deakin following domains of satisfaction: body dissatisfaction,
University Human Research Ethics Committee. At both T1 F(1, 386)=9.75, p<.01, satisfaction with living standards,
and T2, surveys with a cover letter were posted to the F(1, 386),=5.18, p<.05, satisfaction with achievements, F(1,
participants and when completed, they were returned in a 386)=11.11, p<.01, and satisfaction with level of commu-
reply-paid envelope. The questionnaires, designed to take nity connectedness, F(1, 386)=7.67, p<.01. Surprisingly,
approximately 20 min to complete, were completed anony- men scored higher than women on body image importance,
mously and matched at T1 and T2 by the use of a code. F(1, 386)=9.01, p<.01. Despite the gender difference in
body dissatisfaction, it should be noted that body dissatis-
faction was relatively common for men and women: roughly
Results one-third of men (34.5% at T1 and 35.3% at T2) and one-
half of women (46.8% at T1 and 48.8% at T2) in the present
Data Cleaning and Screening sample had a score on the dissatisfied end of the spectrum
(i.e., dissatisfaction score >14).
Prior to the main analysis, data were screened to ensure they Inspection of univariate effects across time revealed that
met the assumptions of multivariate analysis (Tabachnick scores on self-esteem varied over time, F(1, 387)=9.77,

Table 1 Means and standard deviations for key variables across time and gender.

Men (n=150) Women (n=239)

T1 T2 T1 T2

Self esteem 73.44 (15.64) 76.82 (16.02) 76.05 (15.99) 76.29 (16.33)
Body Dissatisfactiona,b 11.40 (6.24) 11.78 (6.17) 13.58 (6.66) 13.86 (6.62)
Body Importancea,b 11.90 (6.15) 11.42 (5.99) 10.11 (6.17) 9.73 (5.85)
Living standarda,b 7.48 (1.58) 7.48 (1.85) 7.83 (1.73) 7.86 (1.78)
Health 6.94 (1.96) 6.96 (2.15) 7.30 (1.83) 7.16 (1.94)
Achievementsa,b 6.53 (1.93) 6.53 (2.01) 7.19 (1.88) 7.04 (2.09)
Relationships 7.30 (2.12) 7.35 (2.18) 7.61 (2.19) 7.65 (2.09)
Safetyb 7.66 (1.75) 7.55 (1.95) 7.75 (1.72) 7.91 (1.53)
Communitya,b 6.83 (1.92) 6.76 (2.02) 7.32 (1.98) 7.25 (1.87)
Security 7.03 (1.89) 6.81 (2.12) 7.16 (1.98) 7.04 (2.09)

Possible scores:
Self esteem 0–100 (higher scores reflect greater self-esteem), Body Dissatisfaction 0–28 (higher scores reflect greater dissatisfaction with appearance),
Body image importance 0–28 (higher scores reflect greater satisfaction), Personal wellbeing domains (living standard, health, achievements,
relationships. safety, community, security) 0–10 (higher scores reflect greater satisfaction). a and b denote significant differences (p<.05) between men and
women at T1 and T2, respectively
Sex Roles (2010) 63:672–681 677

p<.01, due primarily to changes in self-esteem for men. As mediate the relationships between age and body dissatis-
a result, the interaction between gender and time was also faction (p<.05 for the indirect effect) and gender and body
significant, F(1, 387)=7.28, p<.01. dissatisfaction (p<.05 for the indirect effect), although it
While the absolute difference in self-esteem values for should be noted that body image importance served as a
both men and women across T1 and T2 is quite small (in fact, partial mediator as there remained a significant amount of
for women the difference was smaller than the standard error co-variation in the direct pathways from both age and
of the mean at T1), the standard deviation around this mean gender to body dissatisfaction. In combination, these
suggests that there was considerable intra-individual vari- predictor variables accounted for roughly 10% variance in
ability in self-esteem across time, with both male and female body dissatisfaction.
participants changing their self-esteem rating (on average)
by over ten points (see Table 1). Such a finding is not Relationship between Self-Esteem and the Domains
consistent with interpretations based purely on fluctuations of Satisfaction (H5–H8)
in self-esteem due to sampling error. Furthermore, bivariate
correlations revealed that change in self-esteem scores did Hierarchical regression analyses were conducted to evaluate:
not vary as a function of age for men (r=.07, p>.01) or (a) the contribution of body dissatisfaction at T1 to self-esteem
women (r=.01, p>.01). at T2 for men (Hypothesis 5) and women (Hypothesis 6), and
(b) the moderating influence of age on the relationship
Path Model for Relationship between Gender, Age, Body between body dissatisfaction and self-esteem (Hypotheses 7
Image Importance and Body Dissatisfaction (Hypotheses 3 and 8).
and 4) Tables 2 and 3 display the results of hierarchical
regressions conducted to evaluate the contributions of the
A path analysis was conducted through AMOS v17 to T1 domains of satisfaction scores on T2 self-esteem
evaluate: (a) gender- and age-related differences in body separately for women and men, respectively. For brevity,
dissatisfaction (H1 and H2, respectively), and (b) the only the significant predictors at each step of the model are
mediational influence of body image importance for the age- shown in these tables. T1 self-esteem scores were entered
body dissatisfaction (H3) and gender-body dissatisfaction into the model at Step 1 in order to provide an estimate of
(H4) relationships. BMI was added as a covariate to control the residual change in this variable over time. Residual
for differences in body dissatisfaction attributable to body change was chosen in preference to absolute change scores
mass. Bias corrected bootstrap estimates were used in as the former method is less susceptible to biased regression
preference for Sobel’s z-test as the former method more coefficients due to the effects of regression to the mean
accurately calculates indirect effects (Shrout and Bolger 2002). (Twisk 2003). T1 BMI was also entered at Step 1 to control
As shown in Fig. 1, age, gender and BMI all made for the effect that individual differences in body mass may
significant direct contributions to body dissatisfaction. The otherwise have on the relationship between body dissatisfac-
direction of these coefficients indicates that (on average) tion and self-esteem. Domains of satisfaction (including body
body dissatisfaction was higher among women than men, dissatisfaction) and age were entered at Step 2, followed by
younger participants, and/or those with a high BMI. Age the interaction terms for domains by age at Step 3.
and gender were also significantly related to body image As shown in Table 2, T1 self-esteem scores were strongly
importance, such that physical appearance was deemed related to T2 self-esteem scores for female participants. T1
more important by men and/or those who were younger. BMI was also a significant contributor to self-esteem for
Finally, body image importance was found to significantly women (β=−.10, p<.05). While the inclusion of the domain

Fig. 1 Path model testing the

mediational influence of body
image importance for gender- and
age-related differences in body
dissatisfaction. BI body image
importance, BD body dissatisfac-
tion, and BMI body mass index.
All paths (with the exception of
BMI➔BI) are significant (p<.05).
Values associated with each path
are standardized regression
coefficients. For visual clarity,
error terms have been omitted.
678 Sex Roles (2010) 63:672–681

Table 2 Significant predictors

of T2 self-esteem for female Step Sig IVs R2 ΔR2 B se β t sr
participants (n=239).
1 Self esteem T1 .58** .76** .04 .74 17.86 .74
BMI T1 −.10* .04 −.10 −2.35 −.10
2 Self esteem T1 .61** .03 .62** .06 .61 11.03 .45
BMI T1 −.10* .04 −.10 −2.36 −.10
Ach 1.41** .51 .16 2.76 .11
3 Self esteem T1 .61** .01 .63** .52 .16 2.66 .45
T1 Time 1, T2 Time 2, BD Body BMI T1 −.10* .04 −.10 −2.36 −.10
Dissatisfaction, BMI Body Mass Ach 1.39** .52 .16 2.66 .11
Index, Ach satisfaction with life BD × Age −.02* .01 −.10 −2.00 −.08
achievements

scores at Step 2 and interaction terms at Step 3 failed to cultures and, second, that the influence of body dissatisfac-
significantly improve the models for female participants tion on global estimates of self-worth would vary as
(Step 2: Fchange (9, 238)=1.80, p>.05; Step 3: Fchange (8, functions of age, gender and the importance one attributes
230)=.69, p>.05), it is important to note that several to physical appearance. Unlike previous studies (e.g.,
variables made unique contributions to prediction of T2 Wilcox 1997) which have simply explored the correlation
self-esteem scores. A main effect for the achievement between body image importance and self-esteem, the
domain was found at Step 2 (β=.16, p<.01) and remained present study included a variety of domains of satisfaction
significant at Step 3 (β=.16, p<.01). The interaction between to determine the relative importance of body image for self-
age and body dissatisfaction was also a significant predictor esteem across the adult lifespan.
of T2 self-esteem after controlling for main and interaction
effects of the various domains, T1 BMI, and T1 self-esteem Gender Effects
(β=−.10, p<.05). Using Preacher et al.'s (2006) online
program for post-hoc probing of moderation effects, it was Present findings suggest that the relationship between
revealed that body dissatisfaction was a significant predictor gender and perceptions of body image is perhaps more
of self-esteem at low values for age (specifically, age complex than previously thought (e.g., Fredrickson and
<31.62 years). Roberts 1997). Fredrickson and Roberts argued that in
T1 self-esteem scores were strongly related to T2 self- Westernized cultures sociocultural norms relevant to phys-
esteem scores in male participants also (see Table 3). However, ical appearance are more pervasive and emphasize more
inclusion of domains of satisfaction as main effects at Step 2 strongly the importance of appearance for women than for
and interaction terms with age at Step 3 failed to significantly men. While women in the present study did report greater
improve the model; Fchange (9, 127)=.58, p>.05, and Fchange body dissatisfaction than men, a considerable proportion of
(8, 119)=.40, p>.05, respectively. Moreover, none of the men (approximately 35%) and women (approximately
other individual predictors uniquely contributed to T2 self- 45%) scored at the dissatisfied end of the scale, providing
esteem after controlling for T1 levels of self-esteem. further support for Forbes et al.’s (2001) notion that
appearance concerns are becoming a normative experience
for both genders. Thus, our findings suggest that objecti-
Discussion fication theory could apply to both men and women.
Moreover, men placed more importance on their appear-
The present study evaluated two important and related ance than women, a finding that is contrary to those of
tenets of objectification theory (Fredrickson and Roberts previous researchers (Morry and Staska 2001; Noll and
1997): first, that age- and gender-related differences in body Fredrickson 1998) who have drawn on sociocultural
dissatisfaction result from the greater importance placed on explanations emphasizing the role of societal values and
physical appearance for young women in Westernized media messages in perpetuating the greater importance of

Table 3 Significant predictors

of T2 Self-esteem for male Step Sig IVs R2 ΔR2 B se β t sr
participants (n=150).
1 Self esteem T1 .64** .81** .05 .79 15.16 .78
2 Self esteem T1 .65** .01 .78** .07 .76 11.35 .60
3 Self esteem T1 .66** .01 .77** .07 .75 10.59 .57
T1 Time 1, T2 Time 2
Sex Roles (2010) 63:672–681 679

appearance for women than for men (e.g., Fredrickson and Age Effects
Roberts 1997).
These findings could be explained in two ways. Firstly, it The current study is one of only a limited number of studies
may be that the men in our sample are placing more that have examined body image issues across the adult
importance on appearance and are more dissatisfied with lifespan. Webster and Tiggemann (2003) have suggested
their bodies than men in previous studies. Indeed, several that people attribute less importance to their physical
authors have raised concern about the recent trends toward appearance as they move into middle and late adulthood,
increased body image disturbances among males (Forbes et and that this reduced emphasis on appearance may protect
al. 2001; McCabe and Ricciardelli 2004a; Thompson and against negative self-evaluations associated with body
Cafri 2007). In fact, other studies have found that body image. While this proposal makes intuitive sense, it is not
image importance is higher among adolescent boys than supported by the current study, which instead showed a
among adolescent girls (e.g., Ricciardelli and McCabe decline in appearance dissatisfaction across age despite an
2003). We have also seen an increase in marketing towards increase in body importance across the adult life span. A
the attainment of a male-oriented beauty ideal, popularly possible explanation of the current findings is that body
coined by the media as the ‘metro-sexual male’. The image importance has a different meaning as we age.
emergence of a beauty market for men, including special- Tiggemann (2004) argues that while the human body is still
ized skin and hair care products as well as cosmetic important in later life, it is satisfaction with the functional
procedures which have traditionally been the domain of aspects of the body which take precedence rather than
women (Bamberger 1988; Cottle 1988; Grogan 2008), may satisfaction with appearance. If this interpretation is correct,
result in increased pressure among men to conform to a then evaluations of body functioning, rather than appearance
particular standard of physical appearance. A second (as measured in this study) may be a more appropriate measure
possible explanation for these discrepant findings relates of age-related changes in body satisfaction, particularly for
to the measures of body dissatisfaction and body image older populations.
importance used in the present study. Unlike many prior
studies, we used gender-specific measures of body image Relationship between Self-Esteem and Body Dissatisfaction
dissatisfaction and importance, which ask about salient
body parts for males and females, respectively. As such, As expected, the current study showed that self-esteem had
these measures may be better equipped to detect male body a moderate, negative correlation with body dissatisfaction
image issues than prior measures developed, normed and for both men and women, and the strength of this
validated on samples of women. relationship remained constant over a 2-year period and
Even if this latter explanation is correct, it is interesting across adulthood. This finding adds support to Wilcox
that males report less body dissatisfaction while placing (1997), who also found that the relationship between self-
more importance on appearance than that reported by their esteem and attitudes towards one’s body does not vary as a
female counterparts. Perhaps men have more realistic function of age or gender. The stability of this relationship
expectations about whether or not they will achieve the indicates that, regardless of age or gender, people with
male body image ideal, or it may be that the criterion they lower self-esteem are more likely to report dissatisfaction
compare themselves against is less unrealistic or more with their body shape or size. Given that body image tends
variable than the thin ideal against which women compare to represent a large portion of an individuals’ self-concept,
themselves. Indeed, in contrast to female depictions it is not surprising that self-esteem and body dissatisfaction
whereby the broadcast of thinness is consistent and are related; however, few studies have examined the
overwhelming (Fouts and Burggraf 2000; Livingstone direction of this relationship over time, particularly in
1996), Fawkner and McMurray (2002) found that the middle age and elderly samples. Overall, the results
majority of 34 men in their focus groups noted that there demonstrated that the other domains of life satisfaction
is an image diversity with respect to media presentations of did not predict self-esteem over time, with the only
men. While it may be true that today’s men are exposed to significant exception being satisfaction with achievement
more idealized male bodies than ever before, it is also likely among female participants at T1 predicting T2 levels of
that many images of “less than perfect” or “alternative” self-esteem.
men are also in circulation. This suggestion warrants further The results of previous longitudinal studies examining
investigation. Nevertheless, these findings suggest that men body dissatisfaction and self-esteem have demonstrated
may be focusing more on their bodies in recent years and that that, for adolescents, dissatisfaction with one’s body
they are dissatisfied with different aspects of their bodies predicted lower self-esteem several years later (Paxton et
than females. These gender differences need to be considered al. 2006; Tiggemann 2005) but that self-esteem did not
in any health promotion or clinical interventions. predict levels of body dissatisfaction over the same time
680 Sex Roles (2010) 63:672–681

period (Tiggemann 2005). In light of this, we hypothesized psychological and sociocultural factors. Interestingly, this
that body dissatisfaction at T1 would predict self-esteem study found that men placed greater importance on their
2 years later. The results did not support this hypothesis. appearance than women, and that many reported high levels of
Body dissatisfaction did predict changes in self-esteem body dissatisfaction. This finding is not consistent with
2 years later among women aged less than 31 years, but not previous research and the general consensus that women are
for men nor for women above this age. For many young more concerned with their appearance and experience greater
women, the high societal value attached to physical body dissatisfaction. Due to the association between body
appearance perpetrated by a variety of sociocultural dissatisfaction and low self-esteem, as well as the risk of
influences (e.g., peers, family, and the media) influences developing body image disturbances and eating disorders,
dissatisfaction with physical appearance. Therefore, dissat- these findings may suggest that body image disturbances are
isfaction with physical appearance is likely to predict self- of concern among Australian men.
esteem over time among young adult women. On the other
hand, for older women, self-esteem may play a protective
role in relation to how they feel about their appearance. References
There are limitations to the generalizability of findings
from the current study. Firstly, the final sample of 389 Andres, R. (1989). Does the ‘best’ body weight change with age? In
participants constitutes one half of the initial sample. A A. J. Stunkard & A. Baum (Eds.), Perspectives in behavioural
larger number of men and women, and respondents from medicine: Eating, sleeping and sex (pp. 99–107). Hillsdale:
Erlbaum.
the younger adult age group need to be included in future Australian Institute of Health & Welfare. (2007). Young Australians:
studies. It is possible that the low response rate from these Their health and wellbeing. Canberra: Author.
two groups led to a biased sample, and so the results need Bamberger, M. (1988, April 20). Performance enhancers. Sports
to be treated with caution. BMI data were self-reported, and Illustrated, 59–65.
Chang, V. W., & Christakis, N. A. (2003). Self-perception of weight
so future studies need to determine if the above relationships appropriateness in the United States. American Journal of Preven-
still exist when objective measures of height and weight are tive Medicine, 24, 332–339. doi:10.1016/S0749-3797(03)00020-5.
employed. Chen, H., & Jackson, T. (2008). Prevalence and sociodemographic
correlates of eating disorder endorsements among adolescents
and young adults from China. European Eating Disorders
Implications of Present Findings Review, 16, 375–385. doi:10.1002/erv.837.
Cooper, P. J., & Fairburn, C. G. (1993). Confusion over the core
The findings of this study raise some interesting issues for psychopathology of bulimia nervosa. International Journal of
researchers and health policy makers. Primarily, the finding Eating Disorders, 13, 385–389. doi:10.1002/1098-108X(199305)
13:4<385::AID-EAT2260130406>3.0.CO;2-W.
that men in our study placed more importance on their Cottle, M. (1988, May). Turning boys into girls. The Washington
appearance than women, yet were less dissatisfied, suggests Monthly, 32–37.
that different health promotion strategies may be required to Cummins, R. A., Eckersley, R., Lo, S. K., Okerstrom, E., Hunter, B.,
target men and women in relation to their self-care. The & Davern, M. (2003). Australian Unity Wellbeing Index:
Cumulative psychometric record. Retrieved from Australian
finding that the relationship between self-esteem and body Centre on Quality of Life website: http://www.deakin.edu.au/
dissatisfaction was strongest for women aged 31 or research/acqol/instruments/wellbeing_index.htm
younger, with people with lower self-esteem being more Darby, A., Hay, P., Mond, J., Rodgers, B., & Owen, C. (2007). Disordered
likely to be dissatisfied with their body shape or size eating behaviours and cognitions in young women with obesity:
Relationship with psychological status. International Journal of
suggests that interventions should be targeted towards Obesity, 31, 876–882. doi:10.1038/sj.ijo.0803501.
females during an early developmental period. Ellis, K. (2009). New advisory group to tackle negative body image
[media release]. Retrieved from: http://www.deewr.gov.au/Ministers/
Conclusion Ellis/Media/Releases/Pages/Article_090304_121922.aspx
Fawkner, H. J., & McMurray, N. E. (2002). Body image in men: Self
reported thoughts, feelings, and behaviors in response to media
The findings from the current study confirm the existence images. International Journal of Men’s Health, 1, 137–150.
of a relationship between self-esteem and body image doi:10.3149/jmh.0102.137.
among adult men and women, whereby higher self-esteem Forbes, G. B., Adams-Curtis, L. E., Rade, B., & Jaberg, P. (2001).
Body dissatisfaction in women and men: The role of gender-
was related to lower body dissatisfaction. However, the typing and self-esteem. Sex Roles, 44, 461–484. doi:10.1023/
findings are preliminary, as the relationship was not A:1011982112899.
consistently demonstrated in the longitudinal data. Previous Fouts, G., & Burggraf, K. (2000). Television situation comedies:
studies have also fallen short of demonstrating a predictive Female weight, male negative comments and audience reactions.
Sex Roles: A Journal of Research, 42, 925–932. doi:10.1023/
relationship between self-esteem and body image, suggesting A:1007054618340.
that the relationship is more complex than originally concep- Fredrickson, B. L., & Roberts, T. (1997). Objectification theory:
tualized and that it may be moderated by a range of biological, Toward understanding women’s lived experiences and mental
Sex Roles (2010) 63:672–681 681

health risks. Psychology of Women Quarterly, 21, 173–206. Ricciardelli, L. A., & McCabe, M. P. (2002). Psychometric evaluation
doi:10.1111/j.1471-6402.1997.tb00108.x. of the Body Change Inventory: An assessment instrument for
Grogan, S. (2008). Body image. New York: Routledge. adolescent boys and girls. Eating Behaviors, 3, 45–49.
Harter, S. (1999). The construction of the self. A development doi:10.1016/S1471-0153(01)00044-7.
perspective. New York: Guilford. Ricciardelli, L. A., & McCabe, M. P. (2003). A longitudinal analysis
Heatherton, T. F., Mahamedi, F., Striepe, M., Field, A. E., & Keel, P. of the role of psychosocial factors in predicting body change
(1997). A 10-year longitudinal study of body weight, dieting, and strategies among adolescent boys. Sex Roles, 45, 349–360.
eating disorder symptoms. Journal of Abnormal Psychology, 106, doi:10.1023/A:1022942614727.
117–125. doi:10.1037/0021-843X.106.1.117. Rosenberg, M. (1965). Society and the adolescent self-image. Prince-
International Wellbeing Group. (2006). Personal Wellbeing Index. Re- ton: Princeton University Press.
trieved from Australian Centre on Quality of Life website: http:// Rosenberg, M. (1986). Self-concept from middle childhood through
www.deakin.edu.au/research/acqol/instruments/wellbeing_index.htm adolescence. In J. Suls (Ed.), Psychological perspectives on the
Iqbal, N., Shahnawaz, M. G., & Alam, A. (2006). Educational and self (Vol. 3, pp. 107–136). Hillsdale: Erlbaum.
gender differences in body image and depression among Roxon, N. (2008). Assisting people with eating disorders [media
students. Journal of the Indian Academy of Applied Psychology, release]. Retrieved from http://www.health.gov.au/internet/ministers/
32, 365–368. publishing.nsf/Content/mr-yr08-nr-nr135.htm?OpenDocument
Livingstone, S. (1996). On the continuing problems of media effects. &yr=2008&mth=10
In J. Currant & M. Gurevitch (Eds.), Mass media and society (pp. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and
305–324). London: Arnold. nonexperimental studies: new procedures and recommendations.
Martins, M., Tiggemann, M., & Kirkbridge, A. (2007). Those speedos Psychological Methods, 7, 422–445. doi:10.1037/1082-
become them: The role of self-objectification in gay and 989X.7.4.422.
heterosexual men’s body image. Personality and Social Psychol- Stice, E. (2002). Risk and maintenance factors for eating pathology: A
ogy Bulletin, 33, 634–647. doi:10.1177/0146167206297403. meta-analytic review. Psychological Bulletin, 128, 825–848.
Masheb, R. M., Grilo, C. M., Burke-Martindale, C. H., & Rothschild, doi:10.1037//0033-2909.128.5.825.
B. S. (2006). Evaluating oneself by shape and weight is not the Tabachnick & Fidell. (2007). Using multivariate statistics. Boston:
same as being dissatisfied about shape and weight: A longitudi- Pearson/Allyn & Bacon.
nal examination in severely obese gastric bypass patients. Thompson, J. K., & Cafri, G. (2007). The muscular ideal. Washington
International Journal of Eating Disorders, 39, 712–720. DC: American Psychological Association. doi:10.1037/11581-000.
doi:10.1002/eat.20311. Tiggemann, M. (2004). Body image across the adult life span:
McCabe, M. P., & Ricciardelli, L. A. (2004a). Body image Stability and change. Body Image, 1, 29–41. doi:10.1016/
dissatisfaction among males across the lifespan: A review of S1740-1445(03)00002-0.
past literature. Journal of Psychosomatic Research, 56, 675–685. Tiggemann, M. (2005). Body dissatisfaction and adolescent self-
doi:10.1016/S0022-3999(03)00129-6. esteem: Prospective findings. Body Image, 2, 129–135.
McCabe, M. P., & Ricciardelli, L. A. (2004b). A longitudinal study of doi:10.1016/j.bodyim.2005.03.006.
pubertal timing and extreme body change behaviors among Tiggemann, M., & Pennington, B. (1990). The development of gender
adolescent boys and girls. Adolescence, 39, 145–166. differences in body size dissatisfaction. Australian Psychologist,
Morry, M. M., & Staska, S. L. (2001). Magazine exposure: Internal- 25, 306–313. doi:10.1080/00050069008260025.
ization, self-objectification, eating attitudes, and body satisfaction Tiggemann, M., & Stevens, C. (1999). Weight concern across the
in male and female university students. Canadian Journal of life-span: Relationship to self-esteem and feminist identity.
Behavioural Science, 33, 269–279. doi:10.1037/h0087148. International Journal of Eating Disorders, 26, 103–106.
Noll, S. M., & Fredrickson, B. L. (1998). A meditational model doi:10.1002/(SICI)1098-108X(199907)26:1<103::AID-
linking self-objectification, body shame, and disordered eating. EAT14>3.0.CO;2-0.
Psychology of Women Quarterly, 22, 623–636. doi:10.1111/ Tiggemann, M., & Lynch, J. E. (2001). Body image across the life span in
j.1471-6402.1998.tb00181.x. adult women: The role of self-objectification. Developmental
Parnot, L., Rousseau, A., Benazet, M., Faure, B., Lenhoret, E., Sanchez, Psychology, 37, 243–253. doi:10.1037/0012-1649.37.2.243.
A., et al. (2006). Comparaison du vécu corporel en fonction du Tiggemann, M., & Kuring, J. K. (2004). The role of body
genre chez l’adolescent et le jeune adulte (Comparison of body objectification in disordered eating and depressed mood. British
dissatisfaction as function of gender in the adolescent and young Journal of Clinical Psychology, 43, 299–311.
adult). Journal de Thérapie Comportementale et Cognitive, 16 Twisk, J. W. R. (2003). Applied longitudinal data analysis for epidemi-
(2), 45–48. doi:10.1016/S1155-1704(06)70197-5. ology: A practical guide. New York: Cambridge University Press.
Paxton, S. J., Neumark-Sztainer, D., Hannan, P. J., & Eisenberg, M. E. Van Hoeken, D., Seidell, J., & Hoek, H. W. (2005). Epidemiology. In
(2006). Body dissatisfaction prospectively predicts depressive J. Treasure, U. Schmidt, & E. van Furth (Eds.), The essential
mood and low self-esteem in adolescent girls and boys. Journal handbook of eating disorders (pp. 11–34). West Sussex: Wiley.
of Clinical Child & Adolescent Psychology, 35, 539–549. Webster, J., & Tiggemann, M. (2003). The relationship between
doi:10.1207/s15374424jccp3504_5. women’s body satisfaction and self-image across the life span:
Pratt, B. M., & Woolfenden, S. (2002). Interventions for preventing The role of cognitive control. The Journal of Genetic Psychology,
eating disorders in children and adolescents. Cochrane Database 164, 241–252. doi:10.1080/00221320309597980.
of Systematic Reviews, Issue 2. Art. No.: CD002891. doi: Wiederman, M. W., & Pryor, T. L. (2000). Body dissatisfaction,
10.1002/14651858.CD002891. bulimia, and depression among women: The mediating role of
Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools drive for thinness. International Journal of Eating Disorders, 27,
for probing interaction effects in multiple linear regression, 90–95.
multilevel modelling and latent curve analysis. Journal of Educa- Wilcox, S. (1997). Age and gender in relation to body attitudes: Is there
tional and Behavioral Statistics, 31, 437–448. doi:10.3102/ a double standard of aging? Psychology of Women Quarterly, 21,
10769986031004437. 549–565. doi:10.1111/j.1471-6402.1997.tb00130.x.

View publication stats