The Juumul uf Ps.vcho1ofi.v. 2002, /36(6).

581-596

Body Image Dissatisfaction:

Gender Differences in Eating Attitudes,
Self-Esteem, and Reasons for Exercise
ADRIAN FURNHAM
NICOLA BADMIN
IAN SNEADE
Department of Psychology
University College bndotr

ABSTRACT. l b o hundrcd and thirty-five adolescents completed a questionnaire on the

subject of eating attitudes. self-esteem, reasons for exercise, and their ideal versus current
body size and shape. As predicted. boys were as likely to want to be heavier as lighter,
whereas very few girls desired to be heavier. Only girls associated body dissatisfaction
with the concept of self-csteem. Male self-esteem was not affected by hody dissatisfac-
tion. Specific reasons for exercise were found to correlate with low self-esteem and dis-
ordered eating. regardless of sex. The results are discussed in relation to burgeoning pub-
lished research in this area.
Key words: body image, eating attitudes. exercise. self-esteem

AN ENORMOUS INTEREST in the topic of body image perception, distortion,

and satisfaction has developed during the past 20 to 30 years (Altabe & Thomp-
son, 1996 Borchert & Heinberg, 1996; Mandelson, White, & Mandelson, 1996;
Monteath & McCabe, 1997). Much of this work has been done with students
(Kenny & Adams, 1994; Tiggernann. Winefield. Winefield, & Goldney, 1994).
paying particular attention to women (Alley & Scully, 1994; Henderson & Vick-
ers, 1995). although more recent studies have concentrated on children (Kolody
& Sallis. 1995; Wood, Becker, & Thompson, 1996).
In this study we have focused on adolescents, who have been extensively
studied in this area because many eating disorders appear to start soon after
puberty and persist through secondary school years (Fox, Page, Peters, Arm-
strong. & Kirby, 1994; Mueller et al., 1995; Tiggemann & Dyer, 1995). This
appears to be a universal issue, and there have been numerous cross-cultural

Address correspondence to Adrian Furnham. Department of Psychology University Col-

lege London, 26 Bedford Way, London WCI OAP; ucjtsaf@ucl.ac.uk (e-tail).

58 I
582 The Juurnul of Psychuluxy

studies, for example, in Hong Kong (Davis & Katzman, 1997; Lee, Leung, Lee,
Yu, & Leung, 1996). Norway (Wichstrom. 1995). and Poland (Wlodarczyk-Bis-
aga & Dolan. 1996). as well as among minority groups in Western countries
(Ahmad, Waller. & Verduyn, 1994).
Dissatisfaction with one’s body image is often seen as a correlate of eating
disorders. This has been demonstrated by the high prevalence of eating disorders
in groups in which there is an increased emphasis on maintaining a thin, “ideal”
body (e.g., ballet dancers and models; Garner & Garfinkel, 1980) as well as ath-
letes (Sundgot-Borgen, 1993).
The link between body image dissatisfaction and eating disorders is not
clearly apparent in men. However, there is tentative evidence to suggest that the
incidence of eating disorders in men may be increasing and that current preva-
lence rates for men may be underestimated (Carlot 8c Camargo, 1991). The con-
siderable amount of research and popular writing focusing on women and eating
disorders may discourage men from admitting to what they classify as a female
disorder. There have also been gender differences noted in the labeling of behav-
ior; for example, men do not label the ingestion of a large quantity of food as
bingeing (Franco, Taniburrino, Carroll, & Bernal. 1988). Men appear to be more
interested in shape than weight. although the two are clearly related (Anderson
& Di Domenico. 1992).
This difference between the desire for a shape change in men, as opposed to
weight loss through dieting in women, may be a function of the different male
and female ideals. The male ideal is a V-shaped figure with an emphasis placed
on large biceps. chest, and shoulders, whereas the female ideal is to be extreme-
ly thin, with the emphasis placed on slim hips, bottom, and thighs. Men’s desire
for weight gain would tit with the desire to achieve the male ideal V-shaped fig-
ure and to gain additional muscle.
Women are more likely than men to describe themselves as tit. to weigh
themselves often, and to diet frequently. They are also generally more dissatis-
tied with their physical appearance than are men (Cooper & Fairburn. 1983;
Furnham & Calnan. 1998). The most marked difference in body-image percep-
tions between the sexes is dissatisfaction with weight and, to a lesser extent, with
shape, particularly the hips (Berscheid. Walster, 8c Bohrnstedt. 1973). Franco et
al. (1988). and Miller, Coffman, and Linke ( 1980) have shown that many men are
dissatisfied with their weight and shape, although somewhat less so than women;
yet some researchers have indicated that men are as dissatisfied as women with
their body shape and weight (Drewnowski & Yee, 1987; Silberstein, Striegel-
Moore, Timko, & Rodin. 1988).
The nature of body weight dissatisfaction is. however, slightly different in
men and women. Women are more likely to judge themselves overweight when
by objective standards they are not, whereas men are more likely to perceive
themselves as underweight with respect to objective standards (Furnham & Cal-
nan, 1998). These perceptions suggest that both genders misprceive their weight
 Furnham. Badmin. & Sneade S83

in comparison with others of their gender, or they make judgments about their
weight using an unhealthy standard. “Underweight”appears to have a different
meaning for men and women. Men see being underweight as bad; women see it
as good (Cash, Winstead, & Janda, 1986).
Researcherscomparing men’s and women’s dissatisfaction with their weight
need to take into account the direction of the dissatisfaction. Dissatisfaction with
body image in women is normally shown by their desire to lose weight, whereas
as many men want to gain weight as lose it. Silberstein et al. (1988) found that
only 4.4% of the women they studied wanted to become bigger compared with
46.8% of the men. In fact, men seemed to be approximately equally divided
between those desiring weight gain and those desiring weight loss. Drewnowski
and Yee (1987) agreed with these findings; they showed that 18-year-old boys
were more or less evenly split between those who wished to lose and those who
wished to gain body weight.
The greater body dissatisfaction (discrepancy between current and ideal fig-
ures) displayed by adult females has since been replicated in adolescents (Tigge-
man & Pennington, 1990) and in women who displayed particular eating abnor-
malities (Zellner, Harner. & Adler. 1989) such as bulimia (Williamson, Davis,
Goteczny, & Blouin, 1989). These measures of weight dissatisfaction-includ-
ing desired versus perceived body size (Fallon & Rozin. 1985) and desired ver-
sus actual weight (Silberstein et al.. I988)-showed no difference between men
and women on the absolute values of their weight dissatisfaction scores but
revealed gender differences in the direction of weight dissatisfaction. Twice as
many women as men desired to be thinner, whereas only two women, as opposed
to half of the men, wanted to be heavier. In the present study, we examined the
relationships among these different measures of weight dissatisfaction and self-
esteem in young adolescent boys and girls.
Research on the relationship between body satisfaction and self-esteem in
both genders has produced conflicting results, although a relationship between
body satisfaction and self-esteem has been documented for both (Lerner,
Karabenick, & Stuart, 1973; Lerner, Orlos. & Knapp, 1976). Research on the
relationship between self-esteem and the weight concern factor of the Body
Esteem Scale ((BES) in women has produced inconsistent results (Franzoi &
Henog. 1986; Franzoi & Shields, 1984). Silberstein et al. (1988) found that
men’s self-esteem was also affected by the degree of body dissatisfaction, regard-
less of the direction of the dissatisfaction (i.e., under- or overweight).
When perceived ideal discrepancies exist concerning bodily aspects that are
perceived as malleable (e.g.. weight), dissatisfaction often provokes efforts to
close the perceived ideal gap. In Western society, dieting and exercise are the
primary strategies for altering one’s body. There is evidence that dieting is a nor-
mative behavior for young women but not for young men (Garner, Rockert, Olm-
sted. Johnson, & Coscina, 1985; Rodin, Silberstein. & Striegel-Moore, 1985)and
that in general, men and women diet for different reasons and that these reasons
584 The Juurtwl u/’Ps.vrhuluxv

tend to change with age (Davis & Cowles, 1991). Some researchers have sug-
gested that it is actually the dieting behavior that is an important factor in the
development and perpetuation of eating disorders (Polivy & Herman, 1985).
Alternatively. exercise has been thought of as merely a means of quickening
weight loss or as a way of denying the unpleasant effects of dieting.
Sundgot-Borgen(1993) found growing evidence of a link between exercise
and eating disorders. Katz (1986) hypothesized that there may be a number of
reasons for this relationship, including the idea that exercise creates the initial
weight loss through a truly diminished appetite, an increased narcissistic invest-
ment in the body, or an elevated production of endorphins. Franco et at. (1988)
found that weight trainers and body builders are a subpopulationat increased risk
for developing eating disorders. Gamer et at. (1985) found that women, in con-
trast to men, exercise primarily in order to lose weight. Silberstein et at. (1988)
found that exercising for weight control reasons was associated with disordered
eating and that those who exercise for appearance rather than for health reasons
may also be more at risk for the development of eating disorders.
In this sense one may talk about positive and negative reasons for exercise,
only the latter of which are associated with disordered eating patterns. McDonald
and Thompson (ly92)replicated the Silbersteinet al. findings, although they used
minimum exercise criteria. They found that exercising for weight, tone, and to a
lesser degree for attractivenessis positively connected with eating disturbanceand
body image dissatisfaction for both men and women, whereas exercising for tit-
ness reasons for men is negatively connected with eating disturbances.Therefore,
there are strong indications that certain motivations for exercise are accompanied
by increases in disturbances, particularly body dissatisfaction.Alternatively, there
are also other motivations such as health and fitness that are associated with less
disturbance and greater self-esteem. especially for men.
In the present study we investigated the following hypotheses: (a) Women
wish to be thinner, but men are equally divided between those who wish to be
thinner and those who wish to be heavier; (b) those who are dissatisfied with
their bodies have lower self-esteem, regardless of the direction of body image
dissatisfaction; (c) negative reasons for exercise (weight control. attractiveness.
and tone) are associated with disordered eating and low self-esteem. whereas
other reasons for exercise (mood. health, enjoyment, and fitness) are not related
to disordered eating and are, in fact. related to high self-esteem.

Method
Participants
The participants (N= 235; I I I boys, 124 girls) in the present study attended
the sixth form (years 12 and 13)of four different schools in England-two gram-
mar schools and two comprehensive schools. All were native Britons. although
 Furnham. Badmin. & Snade 585

approximately 4% were of African and 8% of Asian origin. The socio-economic

classes of the participants were reasonably diverse, although middle-classchildren
we= more prevalent. The age of the participants ranged from 16 to 18 years ( I
participant was 15); the mean age was 16.8 years (SD= 0.72). The mean body
weight of the boys was 68.89 kg (SD = 10.06) and of the girls 57.07 kg (SD =
8.86). The majority of participant.were within the normal height and weight band
assessed using tables published by the British Heart Foundation (1991).

Questionnaires
All the participants completed a questionnaire that contained the following
measures:
The Eating Attitudes Test (EAT; Garner & Garfinkel, 1979) is a psycholog-
ical measure of anorexichulimic-like attitudes and beliefs. We used the &item
self-report version, in which the frequencies of attitudes and beliefs are rated
using a 6-point scale. The EAT has been validated to show a high level of con-
current validity and a consistent predictive validity across independent samples
and controls; the test also demonstrates a high degree of internal reliability (Gar-
ner & Garfinkel. 1979). In this sample, internal reliability was high (a= .91). Of
the 40 questions, all were completed except that Question 23, “I have regular
menstrual periods,” was omitted for the boys. We added four questions related to
body dissatisfaction and desire for shape change that males tend to be most dis-
satisfied with (“I would like to increase my upper body size, e.g., chest, biceps,
shoulders” and “I would like to decrease my lower body size, e.g., thighs, bot-
tom, hips”). The desire to become bigger and to gain weight were also ques-
tioned. With the inclusion of these extra four questions. the internal reliability
remained high (a= 9 5 ) .
We used the Self-Esteem Scale (Rosenberg, 1965) to assess self-esteem. The
Self-Esteem Scale is a widely used measure with acceptable reliability and valid-
ity. This sample showed a high internal reliability (a= .96). Participants rated
each item on a 4-point scale; high scores indicated high self-esteem.
Reasons for Exercise (Silberstein et al., 1988) consists of 7 subscales: exer-
cising for weight control, for fitness, for health, for improving body tone, for
improving overall physical attractiveness, for improving mood, and for enjoy-
ment. It is a 24-item questionnaire. and responses are rated on a 7-point scale
ranging from not at all important ( I ) to extremely important (7).Participants who
never exercised were asked to skip this section. Silberstein et al. (1988) reported
the alpha coefficients for all the subscales to be above .67 in this inventory (a=
-88 for the overall scale). It has been suggested that these different reasons for
exercise can be broken down into positive (a= .94) and negative (a= .84) rea-
sons. Positive reasons are to improve mood, health, enjoyment, and body fitness;
negative reasons are for weight control, body toning, and attractiveness. This
classification remains open to debate.
586 The Journul of Psychology

We used the Contour Drawing Rating Scale (CDRS; Thompson & Gray,
1991) to assess body dissatisfaction. The CDRS consists of nine drawings of a
female figure (for female participants) or a male figure (for male participants).
Each drawing increases in size from extremely thin ( I ) to very obese (9). Partic-
ipants are asked to rate their ideal figure (what they ideally wanted to look like)
and their current size (perceived figure). The discrepancy between the ideal and
current size scores is an index of body size dissatisfaction.
We gathered background information on a confidential information sheet
that included sex, age, height, present weight, ideal weight, highest past weight,
lowest past weight, parents' occupation, and current school. The information
given on weight provided another measure of body dissatisfaction: the difference
between the students' current weight and ideal weight indicated the number of
kilograms they desired to lose or gain. Thus weight dissatisfaction was a second
measure of body image dissatisfaction.

Procedure

We requested and obtained permission from the four schools before distrib-
uting the questionnaires to the students. This involved having senior leaders
examine the questionnaire for its appropriateness and acceptability. We sent the
sixth form head teachers a cover letter that explained the reasons for the study
and its aims as well as details of the measures we used. We asked the teachers to
introduce the details of the experiment to the participating students, who were
then asked to complete the questionnaire during school time. All of the partici-
pants were sixth form pupils (12th grade). We also asked the teachers to remind
the participants that all responses would be strictly confidential and could not be
identified with an individual in any way. We did this to encourage the students to
be honest when answering the questions, but it also fulfilled the informed con-
sent requirements. All participants were told that participation was not mandato-
ry, but fewer than 2% withdrew. leaving a 98% response rate.

Results
Table I shows the means and standard deviation values for all the variables
used in the analysis. We first performed a between-gender multivariate analysis
of variance (MANOVA) over the four scales, the results of which were signifi-
cant, F(3, 188) = 4.41. p < .Ol; we subsequently calculated various t tests. We
next performed a between-genderMANOVA over the eight Reasons for Exercise
subscales. the results of which were also significant, F(7.205) = 14.21. p < .Ol.
and we then subsequently calculated eight t tests on the subscale score.
These results showed clear patterns, with seven significant differences,
although when corrected for reliability (Bonferroni) the differences were not
extremely significant 0,c .Ol). Compared with the boys, the girls expressed a
 TABLE 1
Gender Differences on All Measures

Boys Girls
Variable (or scale) N a reliability M SD M SD I

CDRS discrepancy 23 I .95 -0.02 I I .33 -1.23 1.266 5.98**

Weight dissatisfaction 189 .46 0.08 7.47 4.55 7.34 4.14
EAT 225 .95 95.99 34. I 121.78 35.78 5.53**
Self-Esteem Scale 234 .96 23.77 3.17 22.90 4.33 1.72
Reason for Exercise
Weight 216 .65 9.93 4.13 13.65 3.8 I 6.90**
Body fitness 216 .87 20.97 5.39 18.71 5.5 3.05*
Mood 216 .82 13.77 5.80 16.78 5.7 3.85*
Health 214 .79 20.25 5.00 20.38 5.47 0.18
Attractiveness 214 .87 14.43 4.56 14.78 5. I 0.53
Enjoyment 215 .62 13.23 7.48 13.62 4.95 0.46
Positive reasons 214 .87 68.2 17.1 1 69.53 16.19 0.58
Negative reasons 214 .88 35.83 11.1 I 42.95 11.49 4.59**

Nure. CDRS = Contour Drawing Rating Scale. EAT = Eating Attitudes Tea.
*p < .05. * p c .01.
greater discrepancy between ideal and actual body shape, expressed greater
weight dissatisfaction, and had higher rates of abnormal eating attitudes and
behaviors. The girls also tended to exercise more often for weight loss than the
boys, whose reasons for exercising were for body fitness. Thus there was an over-
all difference; girls exercised for negative reasons more often than the boys did,
as we had predicted.
We examined the figures that the participants selected from the CDRS as
their current and their ideal figures. Boys and girls did diff'er in their selected cur-
rent and ideal figures. The percentages of boys and girls choosing each tigure as
their current and ideal are contained in Table 2. The differing percentages of boys
(2 I. I%) and girls ( 18.3%) indicate how closely the current figure matched their
ideal figure. The results show that the girls were more dissatisfied with their body
image than were their male counterparts. The majority of both male (79.9%)and
female (8I .7%) participants selected an ideal figure different from their current
figure. When the direction of the current ideal discrepancy was examined, a strik-
ing gender difference was shown. The boys were as likely to want to be heavier
(36.1%) as to be thinner (42.8%). whereas only 9 girls expressed a desire to be
heavier (8.1%) and a very large portion (74.6%) chose an ideal figure that was
thinner than their current figure.
To see the extent of body image dissatisfaction. we calculated the percent-
ages for those who selected an ideal that wds two or more figures apart from their
current figure on the CDRS continuum. Our calculation showed that three times
as many girls ( 14.5%)as boys (4.4%) had a large discrepancy of two or more tig-
ures apart, a gender difference that was statistically significant, f = 2.68, p = .001.

TABLE 2
Gender Differences in Contour Drawing Rating h i e (CDRS) Scores

Difference hetween Number ol' responses %

current and ideal ligures Boys Girls Boys Girls

> -3 2 4 I .9 3.3
-3 I 12 0.9 10.0
-2 12 31 11.2 25.6
-I 30 47 28.8 39. I
0 22 16 21.1 18.3
I 32 8 30.7 6.5
2 4 I 3.5 0.8
3 2 I I .9 0.8
~~ ~~

Note. Negative values in the first column represent a desire to be heavier (i.e.. that the ideal figure
was biggerheavier than the judged current figure).
 Fumham, Badmin, & Sneadc 589

The discrepancy between current and ideal figure for the boys was not signifi-
cant, r = I .23, p = .221 (the difference between means for current and ideal fig-
ure choice was a very small 0.17). but the discrepancy between current and ideal
figures for the girls was very significant, r = 7.94, p = .OOl (the difference
batween means for current and ideal figure choice was a big difference of 1.2).
As Table I indicates, boys and girls did differ significantly in weight dissat-
isfaction scores, which represent the discrepancy between actual and desired
weight, f = 4.14. p = .001; 29.4% of the boys wanted to gain weight compared
with 10.2% of the girls, whereas 35% of the boys wanted to lose weight com-
pared with a vast 69.0% of the girls. Those who were satisfied with their current
weight constituted 35.2% of the boys and 2 I .7% of the girls. Thus, combining
the CDRS discrepancy and weight dissatisfaction results, our hypotheses that
boys and girls are dissatisfied with their bodies and that girls wish to be thinner
and boys are equally divided between those who wish to be lighter and those who
wish to be heavier were supported.
Girls obtained significantly higher EAT scores than boys. Two girls, but no
boys, scored above the clinical cut-off of 30. The scores on the Self-Esteem Scale
showed a marginally significant difference between boys and girls, f = 1.72, p =
.OH.The girls had lower mean scores.
There were significant gender differences on exercising for weight control,
body fitness, mood, and tone. Girls exercised for weight control. mood, and tone
reasons significantly more than boys, who reported that they exercised more for fit-
ness reasons. There was no significantgender difference in those who chose exer-
cising for health, attractiveness,and enjoyment (all positive reasons for exercising).
We calculated correlations (Pearson's) for both genders separately and per-
formed a partial correlation matrix (partialing out gender) on the total group. In
considering the relationship between the CDRS (discrepancybetween the current
and ideal figure) and the weight dissatisfaction measure (discrepancy between
present and ideal weight), we used the absolute values of the scores to account
for direction as well as degree of body dissatisfaction. Overall, there were few
significant correlations separately between the EAT and self-esteem and the
seven reasons for exercising among the girls. EAT scores were significantly cor-
related with attractiveness. indicating that disordered eating was associated with
girls who exercised primarily to be more attractive. Self-esteem was correlated
with weight control, indicating that those with lower self-esteem exercised main-
ly to control their weight.
On the other hand, many of the correlations between the EAT and self-
esteem were significant for boys. Exercise for weight control and body tone was
highly correlated with the EAT, as was exercising for body fitness, health, and
attractiveness. Thus, for the boys, exercise for both positive and negative reasons
was associated with disordered eating attitudes. Self-esteem was also signifi-
cantly correlated with exercising for body fitness and tone. We investigated the
relationship between the CDRS scores and self-esteem using separate correla-
 TABLE 3
Correlations Between the Variables

Variable 1 2 3 4 5 6 7 8 9 10 11 12 13

1. EAT -.31 .04 .02 -.05 -.07 .21 -.09 .I3 -.I9 .26 .05 -.09
2. Self-esteem -.63 .I9 .04 .03 -.02 .03 -.04 .04 -.I7 .I0 -.03 -.I1
3. Weight .29 -.M .06 .02 .08 SO .12 .56 -52 .25 .09 .03
4. Body fitness .23 -.24 .30 SO .55 .I4 .25 .I4 .04 .02 -.I4 -.I4
5. Mood .I0 -.IN .23 .26 .04 .33 -07 .40 -.09 .23 .11 .01
6. Health .22 -.I7 .33 .70 .39 .08 .28 .I8 .oO .04 -.I4 -.I6
7. Attractiveness .28 -.I8 50 .58 .I7 .54 .I I .69 -.a .33 .08 -.06
8. Enjoyment -.09 .10 .I1 .I9 .39 .27 .26 .a .04 -07 .09 .I1
9. Body tone .33 -.19 .57 .55 .25 .49 .59 .I6 -.48 .a .22 .05
10. CDRS -.I7 .04 -.29 -.I0 .04 -.03 -.I2 -.07 .20 -.63 .SO .03
1 1. Weight distribution .I0 .OI .45 -.01 .oO -.09 .09 .07 .I8 -.55 -.25 .03
12. Body mass -.I0 .07 .20 .01 .21 .01 .15 .I3 .I I -.I2 .26 .87
13. Ideal body mass -.I6 .09 .01 .01 .21 .07 .I2 .I7 .oO .I4 -.I5 .90

Note. The upper triangle is for girls (n = 124):the lower triangle is for boys (n = 108). Because of missing data, the RS for correlations vary as. therefore, do
significance levels. Generally where Correlations differ .30 or more between boys and girls that difference is most often significant. CDRS = Contour Drawing
Rating Scale. EAT = Eating Attitudes Test.
 Furnham. Badmin, & Sneade 59 I

TABLE 4
Regression Analysis

Dissatisfaction
EAT with figure
Predictor variable P I P t
~~

Sex 0.13 1.94** .09 I .4m

Self-esteem -.5 1 8.33*** .02 0.29
Positive reasons -.06 0.94 .05 0.92
Negative reasons .I9 2.66** -.25 3.62***
Body mass .30 2.00** -.63 7.16***
Ideal body mass -.37 2.47*** .58 6.59***
Dissatisfaction with figure .oo 0.99

Note. FortheEAT.R2=.41,F(7, I S ) = 17.59.p<.01. For Dissatisfactionwith6gure,RL=.43.F(7.

LSS) = 1 8 . 9 4 . ~< .Wl.
**p<.01.***p<.001.
tions for boys and girls. We found a nonsignificant result for the boys ( r = .W),
whereas the correlation for females was significant ( r = -.17, p = .01).
The correlationsbetween the weight dissatisfaction measure and self-esteem
showed no significant correlation for the boys ( r = .01) but a modest positive cor-
relation for the girls ( r = .lo, p < .05). This result supports the CDRS discrepan-
oy and self-esteem correlation results. We also examined the relationship
between the self-esteem scores and the EAT separately for the two genders.
Boys' self-esteem scores were significantly correlated with their EAT scores ( r =
6 3 , p = .OOI). Girls' scores also showed a positive significant correlation with
their EAT scores ( r = .3 I,p = .001). Table 3 contains the statistical details. Final-
ly, we calculated a number of multiple regressions to determine the relative
power of predictor variables. The dependent variables were scores on the Total
EAT and the CDRS (Table 4).
The results for the two regression analyses were fairly similar. Self-esteem,
negative reasons for exercise, ideal body mass, actual body mass, and gender
were significant predictors of EAT scores. There were three significant predictors
of dissatisfaction with bodylfigurelshape,which accounted for over 40% of the
variance: negative reasons for exercise, actual body mass, and ideal body mass.
Overall, these results confirm the results of previous studies as well as our
hypotheses for this study.

Discussion
These results were in agreement with previous research results (e.g., Fallon
& Rozin, 1985; Furnham & Calnan, 1998; Silberstein et al., 1988). all of which
592 The Juurnul id Psvcholux.v

were investigations of slightly older participants. Furnham and Calnan found

men and women to be generally dissatisfied with their bodies, 80% of whom
selected different ideal figures. Men and women showed comparable degrees of
body dissatisfaction but in very different and predicted directions (Furnham &
Calnan). The present findings showed that approximatelyequal numbers of boys
wanted to be heavier (37%) as wanted to be thinner (43%). whereas only 8.0%
of the girls wanted to increase their weight. Thus our first hypothesis was again
supported.
Drewnowski and Yee (1987) found a highly similar pattern of body dissatis-
faction in their sample of undergraduate men and women. Our results report a
comparable trend and reconfirm the hypothesis that body dissatisfaction is not a
characteristic unique to women. The measure of weight dissatisfaction support-
ed the body dissatisfaction findings (CDRS scores), with 29.4% of the boys
wishing to gain weight (compared with 10.2% of the girls) and 35% of the boys
wishing to lose weight (compared with a staggering 69% of girls).
The statement by Cash et al. (1986) that underweight men seem unhappy
with their bodies, whereas underweight women appear satisfied, is lent some
support by our present results, highlighting the finding that the nature of male
body dissatisfaction is different from that of females.
It is interesting to note the pattern of differences on individual items of the
questionnaires, which strongly suggests that the sources of body dissatisfaction
may differ for the two genders. The most marked of these was found on the addi-
tional items to the EAT, which we added to take into account previous research
showing that body dissatisfaction stems from two different areas for the two gen-
ders. On these additional items, 73% of the boys wanted to increase their upper
body size (compared with 22% of the girls) and 41% of the boys wanted to
decrease their lower body size (compared with 63% of the girls).
These statistics support the notion that the desire to gain weight by men is
associated with the wish to become more muscular and to achieve the male ideal
of a V-shaped figure. Body perfectionism is no longer a phenomenon exclusive
to women (if indeed it ever was). However, these are anatomical factors that can
be enhanced and changed primarily by rigorous exercise or weight training and
not solely by means of dieting. This may suggest why, although there was an
obvious desire for shape change in the boys in our present study, it was not
reflected by disordered eating patterns or negative reasons for exercise.
In contrast, the female ideal of being extremely thin, as presented through
advertising by the diet industry, can be seen as a breeding ground for disordered
eating and negative reasons for exercise (Champion & Furnham. 1999). The
notion of these ideals has been supported by the results of the present study and
would thus suggest that neither men nor women can escape the sociocultural
pressure to achieve the ideal body shape.
As we expected, the girls scored higher than the boys on a measure assess-
ing behaviors and attitudes associated with disordered eating. However, readers
 Furnham. Badmin. & Sneade 593

should remember that our participants came from the socioeconomic middle
class, which means that our results may not generalize to working-class adoles-
cents. Furthermore, the girls, more than the boys, reported that they exercised for
weight control, mood, health, and tone reasons, and for the girls these reasons for
exercise were highly correlated with scores from the EAT. According to R d i n et
al. (1985). the pervasive incidence of dieting and weight concerns among women
can be considered normative discontent.
The findings suggest that, for many women, exercising may be part of the
complex of attitudes and behaviors that make up normative discontent. although
it could be argued that some of these reasons for exercise (e.g., weight control,
m d regulation) are strongly advocated by the medical profession. There is lit-
tle doubt that Western societies encourage and reward the pursuit of the perfect
body because it is an ideal that symbolizes the attainment of numerous personal
virtues and achievements. Unfortunately, the body image standards to which
women currently aspire are far beyond what can be achieved with healthy or sen-
sible levels of dieting or exercise (Brownell. 1991).
In the current study we found that dissatisfaction with body image and
weight was not significantly correlated with self-esteem for boys but was signif-
icant for the girls. The hypothesis that those who are dissatistied with their bod-
ies have lower self-esteem. regardless of the direction of body image dissatisfac-
tion, was supported; we found that body image dissatisfaction is more closely
related to low self-esteem for girls than for boys. This result confirms research
on the relationship between body satisfaction and self-esteem showing that
female body image satisfaction is highly correlated with self-esteem (Lerner et
al., 1973). However, it does not replicate the findings of Silberstein et al. (1988)
who found that there was no relationship between female body dissatisfaction
and self-esteem, which Silberstein et al. reasoned was because of the normative
discontent mechanism described earlier.
However, as reported by Davis and Cowles ( 1991). dieting is a more com-
mon behavior among women who report that the way they feel about their bod-
ies is important to the way they feel about themselves in general; this relation-
ship was not found for the men in their study. Results of our present study would
support this view. There is considerable evidence that men are more visually
interested in the bodies of the opposite sex than are women (see Mazur, 1986, for
a Ireview). As a result, women have traditionally displayed a greater commitment
to the pursuit of cultural standards of beauty than have men. In an era that cher-
ishes a very slender body for women, those for whom body satisfaction is a cen-
tral aspect for self-esteem are more likely to be preoccupied with weight and
prone to negative eating attitudes.
It could also be that rather than body dissatisfaction causing low self-esteem
in women the reverse could be true (low self-esteem may cause individuals to be
dissatisfied with their bodies). Individuals with a negative sense of self, situated
in a culture obsessed with the pursuit of physical perfection at almost any cost.
594 The Juuml uf Psychuluxy

would reflect these feelings of worthlessness in how they perceive themselves

(thus providing a distorted body image).
It is interesting to note that in the further correlations of self-esteem with the
EAT and in both positive and negative reasons for exercise, there were few gen-
der differences. Positive reasons for exercise were not significantly correlated
with the EAT or the self-esteem scores for either gender. This result only partly
supports the hypothesis that exercising for weight control, attractiveness, and tone
would be associated with disordered eating and low self-esteem, whereas exercis-
ing for mood. health. enjoyment. and fitness would not be related to disordered
eating and would instead be related to high self-esteem, regardless of gender.
The majority of previous studies on body image dissatisfaction have con-
centrated on women and the desire to lose weight. This study's results provide
evidence that men are not satisfied with their bodies either but that this body
image dissatisfaction also operates in the direction of weight gain. These findings
also suggest there may be important sex differences in the etiology of eating dis-
orders, and thus further research in this area can be used to develop effective psy-
chological interventions for these disorders.

REFERENCES
Ahmad, S.. Walter. G., & Verduyn, C. (1994). Eating attitudes and body satisfaction
among Asian and Caucasian adolescents. Journal of Adolescence, 17.461-470.
Alley, T.,& Scully, K..(1994). Thc impact of actual and perceived changes in body weight
on women's physical allracliveness. Basic and Applied Social Psychology, 15.
535-542.
Altabe. M.,& Thompson, J. (1996). Body image: A cognitive self-schema construct. Cog-
nition Therapy and Research, 20, I7 I - 193.
Anderson, A. E..& Di Domenico. L. (1992). Diet vs. shape content of popular male and
female magazines: A dose response relationship to the incidence of eating disorders.
International Journal of Eating Disorders. I I, 283-287.
Berscheid. E.. Walster. E.. & Bohmsledl, G. (1973). The happy American body: A survey
report. Psychology Today, 7(6). 1 19- I 3 I.
Borchert, J.. & Heinberg, L. (1996). Gender schema and gender role discrepancy as cor-
relates of body image. The Journal of Psychology, 130, 547-559.
British Heart Foundation. (1991). Food and your hean Heart information series. (Whole
No. 7). London: Author.
Brownell. K. D.(1991). Dieting and the search for the perfect body: Where physiology
and culture collide. Behuviour Therapy, 22, 1-12.
Carlot. D.. & Camargo. C. (1991). Review of bulimia nervosa in males. American Jour-
nal of Psychiatr): 148. 831-843.
Cash, T. F., Winstead. B. A.. & Janda. L. H. (1986). The great American shape-up. Psy-
chology Today, 30, 37.
Champion, H.. & Fumham. A. (1999). The effect of the media on body dissatisfaction in
adolescent girls. Eimpean Earing Disonlers Review, I. 2 13-228.
Cooper. l? J.. & Fairbum. C. G.(1983). Binge eating and self-induced vomiting in thc
community: A preliminary study. British Journal of Psychiatry, 142. 139-144.
Davis, C..& Cowles. M. P.( I 991). Body image and exercise: A study of relationships and
comparisons between physically active men and women. Sex Roles, 25, 33-44.
 Furnham. Badmin. 8c Snrade 595

Davis, C., & Katzman, M. (1997). Charting new territory. Sex Roles, 36*449-459.
Drewnowski, A., & Yee. D. K. (1987). Men and body image: Are males satisfied with their
body weight? Psychosamatic Medicine, 49. 626-634.
Fallon, A. E., & Rozin. P. (1985). Sex differences in perceptions of body’shape. Journal
Of Abnormal Psyrhology, 94. 102- 105.
Pox. K. A., Page, A.. Peters, M.. Armstrong. N., & Kirby, B.(1994). Dietary restraint and
fatness in early adolescent girls and boys. Journal ofAdolescence. 17. 149-161.
Pranco. S. N.,Tambumno. M. B.. Carroll. B. T.. & Bemal. G. A. A. (1988). Eating atti-
tudes in college males. International Joutnal nf Eating Disorders, 7, 285-288.
Franzoi, S. L.. & Henog. M. E. (1986). The Body Esteem Scale: A convergent and dis-
criminate validity study. Journal of Personality Assessment, 50. 24-31.
Franzoi. S.L., & Shields, S. A. (1984). The Body Esteem Scale: Multi-dimensionalstruc-
ture and sex differences in a college population. Journal of Personality Assessment. 48.
173-178.
Fumham. A.. & Calnan, A. (1998). Eating disturbances. self-esteem, reasons for exercis-
ing and body weight dissatisfaction in adolescent males. Eumpean Eating Disorders
Review, 6.58-72.
Gamer, D.M.. & Garfinkel, P. E.(1979). The Eating AttitudesTest: An index of the symp-
toms of anorexia nervosa. Psychological Medicine, 9, 213-279.
Gamer. 0. M., & Garfinkel. P. E. (1980). Socio-cultural factors in the development of
anorexia nervosa Psychological Medicine, 10, 647-656.
Gamer, D. M.. Rockert, W., Olmstead. M. P.,Johnson, C.. & Coscina, D. V. (1985). Psy-
cho-educational principles in the treatment of bulimia and anorexia nervosa. In D. M.
Gamer & P.E.Garfinkel (Eds.). Handbook of psychotherapy for anorexia nervosa and
bulimia. New Yurk: Guilford.
Henderson. L.. & Vickers. P. (1995). Health and beauty? Health Education Journal, 54.
61-73.
Katz. J. L. (1986). Long-distance running, anorexia nervosa and bulimia: A report of two
cases. Comprehensive Psychiatv. 27, 74-78.
Kenny. D.. & Adams. R. (1994). The relationship between eating attitudes. body mass
index. age, and gender in Australian university students. Australian Psychologist. 29.
128- 134.
Kolody. B.. & Sallis. J. (1995). A prospective study of ponderosity, body image, self-con-
cept and psychological variables in children. Journal of Developmental and Behav-
ioural Paediatrics. 16. 1-5.
Lee. S..Leung. T.. Lee, A.. Yu. H..& Leung. C. (1996). Body dissatisfaction among Chi-
nese undergraduates and its implications for eating disorders in Hong Kong. Interna-
tional Journal of Enting Disorders. 20, 77-89.
Lerner, R. M.. Karabenick, S.A., & Stuart. J. L. (1973). Relations among physical attrac-
tiveness. body attitudes and selfconcept in male and female college students. The Jour-
nal of Psychology, 85, 1 19-129.
Lemer, R. M.. Orlos, J. B., & Knapp, J. R. (1976). Physical attractiveness, physical effec-
tiveness and self-concept in late adolescents. Adolescence, I I . 3 13-326.
Mandelson. B.. White, D.. & Mandelson. M. (1996). Self-esteem and body esteem. Jour-
MI of Applied Developmental Psychology, 13. 32 1-346.
Mazur. A. (1986). U.S. trends in feminine beauty and over-adpation. The Journal of Sex
Research, 22. 28 1-303.
McDonald, K., & Thompson. J. K. (1992). Eating disturbance, body image dissatisfaction
and reasons for exercising: Gender differences and correlation findings. International
Journal of Eating Disorders, 11, 289-292.
Miller. T. M., Coffman. J. G..& Linke, R. A. (1980). Survey on body image, weight and
S96 The Juurnctl uf Psvchologv

diet of collegc studcnts. Joirrnnl of Atiierican Dietetic Association. 77, 844-846.

Monteath, S.. & McCabe. M. (1997). The inlluencc of societal factors on female body
image. Thr Journal of Social Psychology, 137. 708-727.
Mueller. C., Field, T., Yando. R., Harding, J.. Gonzalez. K.. Lasko. D.. & Bendell. D.
( 1995). Undereating and overeating conccrns among adolescents. Journal of Child Psy-
rhology and Psychiatty, 36. 1019-1025.
Polivy. J.. & Herman. C. P. (1985). Dieting and bingcing: A causal analysis. Atnericati
Psychologist. 40. I93-20 I.
Rodin. J.. Silherstcin. L.. & Striegel-Moore, R. (1985). Womcn and weight: A normative
discontcnt. In T. B. Sondercggcr (Ed.). P.tycho1og.v and gender: 1984 Nebrash Sym-
posium Motiiwtion (pp. 267-308). Lincoln: University of Nebraska Press.
Rasenberg, M. (1965,s).Society arid thc adolrscrnt self-image. Princeton. NJ: Princeton
University Prcss.
Silberstein. L. R.. Stricgel-Mom. R. H.. Timko, C.,& Rodin. J. ( 1988). Behavioural and
psychological implications of body dissatisfaction: Do men and women differ? Se.r
Roles, 19. 219-232.
Sundpt-Borgen. J. ( 1993). Prevalence of eating disorders in elitc female athletes. Inter-
nationalJ o u m l of Sport Nutrition. 3, 2940.
Thompson, M. A.. & Gray. J. J. (1991). Dcvelopmenl and validation o f a New Body Image
Assessment Scale. Journal cfPersonuli/y Assessment, 64,258-269.
Tiggemann, M.. & Dyer, G . (1995). ldcal body shape preferences and cating disorder
scores in rrdolcscent women. Psyrhology a d Health. 10. 345-347.
Tipgemann. M.. & Pcnnington. B. ( 1990). Thc development of gender differences in
body-size dissalisfaclion. Australian Psychcdogist. 25. 306-3 I 3.
Tiggcmann, M., Winelield, H.. Winelield, A.. & Goldney, R. (1994). Gender differences
in the psychological corrclates of body weight in young adults. Psychology and Health.
9.39545 I.
Wichstrom. L. ( 1995). Social. psychological and physical correlates of cating problems.
Psvrhologiral Medicine. 25. 567-579.
Williamson. D. A.. Davis, C. J.. Goreczny, A. J.. & Blouin. D. C. (1989). Body-image dis-
turbances in bulimia nervosa: lnllucncus on actual body sizc. JournulofAbnomtal Psv-
chologx 98.97-99.
Wlodarczyk-Bisaga. K.,& Dolan. B. (1996). A two-stage epidemiologicalstudy of abnor-
mal eating attiludcs and their prospective risk factors in Polish schoolgirls. Psycholog-
icul Medicine. 26. 1-12.
Wood. K.. Baker. J.. & Thompson. 1. (1996). Body image dissatisfaction in pre-adoles-
cent children. Joirrnctl of Applied Developmental Psyrhology, 17. 85- 100.
Zellner, D. A., Harncr. D. E.. & Adler, R. L.(I989). Effects of eating abnormalities and
gender on pcrceptions of desirablc body shape. Journal of Abnormal Psychology Y8.
93-96.

Origittal iitatiusrript received March 12. 2001

Final revision accepted Febrrcaty IS, 2002