Basic and Applied Social Psychology

ISSN: 0197-3533 (Print) 1532-4834 (Online) Journal homepage: http://www.tandfonline.com/loi/hbas20

Ideal Body Image Assessment Among Japanese

Women

Shinichi Saito & Satoko Izumi Barton

To cite this article: Shinichi Saito & Satoko Izumi Barton (2017): Ideal Body Image
Assessment Among Japanese Women, Basic and Applied Social Psychology, DOI:
10.1080/01973533.2017.1383909

To link to this article: https://doi.org/10.1080/01973533.2017.1383909

Published online: 10 Nov 2017.

Submit your article to this journal

Article views: 11

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=hbas20

Download by: [University of Florida] Date: 03 January 2018, At: 01:53

 BASIC AND APPLIED SOCIAL PSYCHOLOGY
https://doi.org/10.1080/01973533.2017.1383909

none defined

Ideal Body Image Assessment Among Japanese Women

Shinichi Saitoa and Satoko Izumi Bartonb
a
Tokyo Woman’s Christian University; bThe University of Tokyo

ABSTRACT
Using the Contour Drawing Rating Scale, we examined the ideal body sizes of young Japanese
women who resided in Tokyo. The results indicated that the relationship among the respondents'
current, ideal, perceived male-peer, and perceived female-peer ideal body sizes were not
necessarily the same across different body mass index levels. For example, the results revealed that
only the heavier respondents seemed to consider their male-peer ideal body sizes as their own
ideal body sizes. The findings also showed that although most participants wished to be slimmer,
they believed that their same-sex peers wanted to be even thinner than themselves.
Downloaded by [University of Florida] at 01:53 03 January 2018

Research has indicated that thin-ideal images in the ideal, and perceived female-peer ideal body sizes. Thus,
media directly or indirectly impact women’s distorted the aim of the present study is to further examine
body image and body dissatisfaction, which may be women’s ideal body images as measured by the CDRS,
among the risk factors for developing an eating disorder focusing on the comparison of respondents’ current,
(e.g., Grabe, Ward, & Hyde, 2008; Lopez-Guimera, ideal, perceived male-peer ideal, and perceived female-
Levine, Sanchez-Carracedo, & Fauquet, 2010; J. K. peer ideal body sizes according to BMI. Our research
Thompson & Heinberg, 1999; Tiggemann & Pickering, questions were as follows: What is the relationship
1996; Tiggemann & Polivy, 2010). The American between respondents’ perceived current, ideal, perceived
Psychological Association (2007) identified body male-peer ideal, and perceived female-peer ideal body
dissatisfaction as an important issue among women sizes? Is the relationship dependent on their BMI?
and teenage girls. A growing body of research has also
found that many young Japanese women are discontent
with their own bodies (e.g., Baba & Sugawara, 2000; Methods
Kowner, 2002).
Procedures
In past studies, body image was often assessed
using figural rating scales (e.g., Fallon & Rozin, 1985; The participants in this study were selected from
Stunkard, Sorenson, & Schulsinger, 1983; M. A. women 18–22 years of age who live in Tokyo. We
Thompson & Gray, 1995). Among them, M. A. conducted a web-based survey using a marketing
Thompson and Gray’s Contour Drawing Rating company’s (Macromill, Inc.) panel that has more than
Scale (CDRS) was one of the most frequently used 1 million registered individuals, which is one of the
(e.g., Furnham, Badmin, & Sneade, 2002; Swami et al., largest in Japan and frequently used by Japanese
2010). Previous studies using the CDRS revealed social psychologists. E-mails were sent on February 14,
that the ideal body sizes of young women were 2017, asking them to complete the web-based question-
significantly smaller than their current body sizes naire. The survey continued until February 22, and
(Furnham et al., 2002; Swami et al., 2010). 638 respondents (M age ¼ 20.55 years, SD ¼ 1.22)
To our knowledge, however, previous studies using completed the questionnaires anonymously; of these,
the CDRS did not examine if the findings would hold 67.1% were college students, and the remaining 32.9%
true regardless of different body mass index (BMI) were nonstudents. This proportion virtually corre-
levels, although some of the studies controlled for sponded with the official data on female college
BMI as a covariate in their analyses. In addition, studies students in Tokyo as reported by the Ministry of
have not sufficiently clarified the relationship between Education, Culture, Sports, Science and Technology in
women’s perceived current, ideal, perceived male-peer 2016.

CONTACT Shinichi Saito ssaito@lab.twcu.ac.jp Tokyo Woman’s Christian University, 3-1-29-302 Shimo-renjaku Mitaka-shi, Tokyo 181-0013, Japan.
© 2017 Taylor & Francis
 2 S. SAITO AND S. IZUMI BARTON

The participants were asked to provide information (BMI ¼ 21.00–22.99; n ¼ 123) and upper healthy–obese
regarding their height (M ¼ 157.9 cm, SD ¼ 5.5), actual (BMI ¼ 23.00–45.00; n ¼ 83; hereinafter referred to as
weight (M ¼ 50.8 kg, SD ¼ 7.9), and ideal weight for BMI groups).
their current height (M ¼ 46.1 kg, SD ¼ 4.6). According
to the official data of the Ministry of Internal Affairs
Measures
and Communications, the average height and weight
of young Japanese women (20–24 years) in 2015 were Contour drawing rating scale
158.5 cm (SD ¼ 5.0) and 50.8 kg (SD ¼ 5.9) respectively. M. A. Thompson and Gray’s (1995) CDRS consists of
Thus, the sample used for this study had a height and nine line drawings of women’s bodies arranged and
weight close to the average for young Japanese women. numbered from 1 (smallest) to 9 (largest). The CDRS’s
BMI was calculated as the ratio of body mass to figures are arranged “with relatively fine graduations
height squared (kg/m2). Respondents’ mean BMI was from one figure to the next that were designed to
20.35 (SD ¼ 2.88), ranging from 14.57 to 44.44. Their increase at a standard rate” (Wertheim, Paxton, &
ideal BMI was 17.84 (SD ¼ 1.73), ranging from 12.46 Tilgner, 2004, p. 200). The reported reliability and
to 24.44. validity of the CDRS were also acceptable (e.g., Gardner
As a criterion for judging BMI values, the World & Brown, 2010; Wertheim et al., 2004). Gardner and
Health Organization’s (WHO’s) classification is perhaps Brown (2010) pointed out that “potential shortcoming
Downloaded by [University of Florida] at 01:53 03 January 2018

the most frequently used: underweight (BMI < 18.50), of many existing scales arises when they are drawn with
normal range (18.50–24.99), overweight (25.00–29.99), facial and body features that reflect obvious Caucasian
and obese (BMI ≥ 30.00). Using the WHO’s classi- ethnicity” (p. 108). Thus, we employed a modified
fication, however, many young Japanese women are version of the CDRS, which obscured facial features
categorized into “normal range,” which is not appropri- with opaque boxes “to minimize the effects of facial
ate for this study’s purpose. Thus, we utilized a different features, hair style, and perceived ethnicity”
criterion in this study. (Swami et al., 2010, p. 314). As Wertheim et al. (2004)
Frederick, Forbes, Grigorian, and Jarcho (2007) recommended, we also offered rating points between
suggested that BMI scores could be divided into seven the figures. We asked the participants to select the line
categories: underweight (14.50–18.49), lower healthy drawing that (a) most closely approximated their
(18.50–20.99), middle healthy (21.00–22.99), upper perceived current body size (hereafter, current body
healthy (23.00–24.99), lower overweight (25.00–27.49), size), (b) they would like to possess the most (hereafter,
upper overweight (27.50–29.99), and obese (30.00– ideal body size), (c) they perceived as the most physically
45.00). Although Frederick et al.’s categorization was attractive to men of their own age (hereafter, male-peer
not created with Japanese women in mind, it may be ideal body size), and (d) they believed would be
useful for a Japanese sample because they divide the considered ideal by women of their own age (hereafter,
WHO’s “normal range” into three separate categories. female-peer ideal body size). Thus, this study measured
Using this categorization, 23.4% of the respondents the participants’ current, ideal, male-peer ideal, and
in this study (n ¼ 149) were considered underweight female-peer ideal body sizes (hereafter, choice category).
(BMI < 18.5). A survey conducted by the Japanese Following the previous studies (e.g., Furnham et al.,
Ministry of Health, Labour and Welfare in 2015 2002; Swami et al., 2010), we also calculated scores by
revealed that 22.3% of Japanese women in their 20s were subtracting ideal line drawing ratings from current line
underweight (BMI < 18.5). Thus, the proportion of drawing ratings as an index of body dissatisfaction.
underweight respondents in this study was close to that Positive scores indicate that participants were
of the official data. In addition, 44.4% (n ¼ 283) were in dissatisfied with their current body size and wanted
the lower healthy group, 19.3% (n ¼ 123) were in the to be thinner. Zero indicates satisfaction, whereas
middle healthy group, 7.5% (n ¼ 48) were in the upper negative scores indicate that they were dissatisfied
healthy group, 3.1% (n ¼ 20) were in the lower over- with their current body and wanted to be heavier. The
weight group, 1.4% (n ¼ 9) were in the upper over- scores ranged from −3 to 8 (M ¼ 1.96, SD ¼ 1.76).
weight group, and 0.9% (n ¼ 6) were in the obese
group. Because the participants categorized from
Results
upper healthy to obese were few, these four categories
were combined for subsequent analyses. Thus, the Table 1 shows the results of preliminary analyses
respondents were categorized into the following four indicating mean BMIs (standard deviations) and
groups: underweight (BMI < 18.5; n ¼ 149), lower educational status for the respondents who chose each
healthy (BMI ¼ 18.5–20.99; n ¼ 283), middle healthy line drawing. The effect size for the difference in mean
 BASIC AND APPLIED SOCIAL PSYCHOLOGY 3

Table 1. BMIs of participants selecting each line drawing and percentages of college students and nonstudents selecting each line
drawing.
CDRS Figures (Perceived Current Body Size)
1 2 3 4 5 6 7 8 9
% % % % % % % % %
Students 2.3 6.3 7.5 23.8 24.8 17.5 10.0 6.1 1.6 M ¼ 4.95, SD ¼ 1.69
n 10 27 32 102 106 75 43 26 7 d ¼ 0.09
Nonstudents 1.0 7.6 11.4 20.5 20.0 15.2 11.9 6.2 6.2 M ¼ 5.12, SD ¼ 1.92
n 2 16 24 43 42 32 25 13 13
M BMI 16.3 17.4 18.1 19.2 20.2 20.9 22.4 23.6 28.7 r ¼ .74
SD 1.04 1.39 1.05 1.28 1.39 1.83 1.77 2.64 5.50
n 12 43 56 145 148 107 68 39 20
Note. BMI ¼ body mass indexes; CDRS ¼ Contour Drawing Rating Scale.

ratings of current body size between college students Next, we conducted a multivariate analysis of
and nonstudents (d ¼ 0.09) was found to be smaller variance with choice category (e.g., current body size,
than Cohen’s (1992) convention for a small effect size ideal body size, female-peer ideal body size, and
(d ¼ 0.20). We also compared line drawing ratings for male-peer ideal body size) as the independent variable
ideal body size, female-peer ideal body size, and and the ratings of line-drawings as the dependent
Downloaded by [University of Florida] at 01:53 03 January 2018

male-peer ideal body size between college students variable. The analysis revealed a main effect of choice
and nonstudents. The results indicated that differences category in the underweight (g2p ¼ :20), lower healthy
between college students and nonstudents for ideal (g2p ¼ :58), middle healthy (g2p ¼ :74), and upper
body size (M ¼ 3.12, SD ¼ .89 vs. M ¼ 2.90, SD ¼ .99, healthy–obese groups (g2p ¼ :84). Then, we conducted
d ¼ 0.23), female-peer ideal (M ¼ 2.58, SD ¼ .90 vs. multiple comparisons. Figure 1 and Table 2 summarize
M ¼ 2.44, SD ¼ .99, d ¼ 0.15), and male-peer ideal body the results.
sizes (M ¼ 3.42, SD ¼ 1.03 vs. M ¼ 3.42, SD ¼ 1.17, First, with regards to the underweight group, the
d ¼ 0.00) were all small. Therefore, in the following sec- difference between the participants’ current body sizes
tion, we report the results of the analyses using the com- and their ideal body sizes (M ¼ 3.21, SD ¼ 1.26 vs.
bined data of college students and non-students. As M ¼ 2.93, SD ¼ .91, d ¼ 0.25) was small. That is, when
shown in Table 1, the respondents who chose larger defining body dissatisfaction as the discrepancy between
line-drawings had larger BMIs (r ¼ .74), which seems ideal line drawing ratings from current line drawing
to indicate the validity of the CDRS. ratings, the underweight women at the aggregate level

Figure 1. Means (with standard errors) of the ratings on the Contour Drawing Rating Scale according to body mass index groups.
Note. A smaller mean indicates a thinner body size. Only small effect sizes are shown in the figure.
 4 S. SAITO AND S. IZUMI BARTON

Table 2. Summary of descriptive statistics (means and standard deviations) and effect sizes (Cohen’s d) between four
category choices.
Underweight Lower Healthy Middle Healthy Upper Healthy—Obese
Current body size M ¼ 3.21, SD ¼ 1.26 M ¼ 4.80, SD ¼ 1.19 M ¼ 6.00, SD ¼ 1.06 M ¼ 7.47, SD ¼ 1.17
Ideal body size M ¼ 2.93, SD ¼ 0.91 M ¼ 2.86, SD ¼ 0.86 M ¼ 3.28, SD ¼ 0.91 M ¼ 3.57, SD ¼ 0.95
Female-peer ideal body size M ¼ 2.74, SD ¼ 0.95 M ¼ 2.46, SD ¼ 0.84 M ¼ 2.50, SD ¼ 1.00 M ¼ 2.47, SD ¼ 1.04
Male-peer ideal body size M ¼ 3.81, SD ¼ 1.04 M ¼ 3.25, SD ¼ 1.05 M ¼ 3.40, SD ¼ 1.02 M ¼ 3.33, SD ¼ 1.12
Current vs. ideal d ¼ 0.25 d ¼ 1.87 d ¼ 2.75 d ¼ 3.66
Current vs. female-peer ideal d ¼ 0.42 d ¼ 2.27 d ¼ 3.40 d ¼ 4.52
Current vs. male-peer ideal d ¼ 0.52 d ¼ 1.38 d ¼ 2.50 d ¼ 3.61
Ideal vs. female-peer ideal d ¼ 0.20 d ¼ 0.47 d ¼ 0.82 d ¼ 1.10
Ideal vs. male-peer ideal d ¼ 0.90 d ¼ 0.41 d ¼ 0.12 d ¼ 0.23
Female-peer vs. male-peer ideal d ¼ 1.07 d ¼ 0.83 d ¼ 0.89 d ¼ 0.80
Note. For example, Cohen’s d for “current vs. ideal” refers to the effect size calculated using the means and standard deviations of current body size and ideal
body size.

did not show body dissatisfaction. The analysis also the participants perceived that their same-sex peers
revealed that the difference between female-peer ideal wished to be even thinner than they were themselves.
body size and ideal body size (M ¼ 2.93, SD ¼ .91 vs. The results also showed that, regardless of their BMI
M ¼ 2.74, SD ¼ .95, d ¼ 0.20) was small. By contrast, group, participants’ perception of their female peers’
Downloaded by [University of Florida] at 01:53 03 January 2018

the participants’ perception of their male-peer ideal ideal body size was thinner than their male peers’ ideal
body size (M ¼ 3.81, SD ¼ 1.04) was larger than their body sizes (ds ¼ 0.80 ∼ 1.07).
own ideal body sizes (d ¼ 0.90), as well as their percep-
tion of female-peer ideal sizes (d ¼ 1.07), and current
Discussion
body sizes (d ¼ 0.52).
As for the relationship between the participants’ The results indicated that the relationships among
current body sizes and their ideal body sizes, the young women’s current, ideal, perceived male-peer,
analysis revealed that, except in the underweight group, and perceived female-peer ideal body sizes were not
the participants’ ideal body sizes were much thinner necessarily the same across different BMI groups, which
than their current body sizes (d ¼ 1.87 for the lower signify the importance of considering BMI levels in the
healthy group, d ¼ 2.75 for the middle healthy group, data analysis. For example, the findings revealed that the
and d ¼ 3.66 for the upper healthy–obese group). underweight respondents at the aggregate level did not
Next, as with the underweight group, the lower show body dissatisfaction defined as the discrepancy
healthy group’s perception of male-peer ideal body sizes between their current and ideal body sizes. Such
(M ¼ 3.25, SD ¼ 1.05) was slightly larger than their own findings were not reported in the previous studies.
ideal body sizes (M ¼ 2.86, SD ¼ .86, d ¼ 0.41). Thus, The findings of the present study showed that
the analysis revealed that the underweight and lower although most participants wished to be slimmer, they
healthy respondents’ perceptions of their male-peer believed that their same-sex peers wanted to be even
ideal body sizes were heavier than their own ideal body thinner than they were themselves. This result may be
sizes. explained by the third-person perception (Choi,
As shown in Figure 1 and Table 2, however, among Leshner, & Choi, 2008; Davidson, 1983). Davidson
the heavier respondents (i.e., the middle healthy and (1983) contended that people tend to overestimate the
upper healthy–obese groups), their own ideal and impact of media content on others compared to its
male-peer ideal body sizes did not differ. As for the impact on themselves (third-person perception) and
middle healthy group, there was little difference that this perception may lead to behavioral conse-
between their ideal body sizes and their perception quences (third-person effect). Thus, it is possible that
of male-peer ideal body sizes (M ¼ 3.28, SD ¼ 0.91 vs. young women consider their same-sex peers to be more
M ¼ 3.40, SD ¼ 1.02, d ¼ 0.12). Similarly, for upper susceptible to media content than they are themselves,
healthy–obese group, the difference between their ideal and therefore they estimate their peers’ ideal body shape
body sizes and their’ perception of male-peer ideal to be thinner than their own. However, further research
body sizes was small (M ¼ 3.57, SD ¼ 0.95 vs. M ¼ 3.33, is needed on this issue.
SD ¼ 1.12, d ¼ 0.23). The results also revealed that, except in the under-
Furthermore, the analysis revealed that, except in the weight group, perceived male-peer ideal body sizes were
underweight group, participants’ perception of their much smaller than their current body sizes, which may
female-peer ideal body sizes was thinner than their be due to their beliefs about men’s preferences of female
own ideal body sizes (ds ¼ 0.47 ∼ 1.10), indicating that thinness. The results also indicated that the heavier
 BASIC AND APPLIED SOCIAL PSYCHOLOGY 5

respondents (BMI ≥ 21.00) seemed to consider their and Individual Differences, 48, 107–111. doi:10.1016/
male-peer ideal body sizes as their own ideal body sizes. j.paid.2009.08.017
Thus, the contention that young women partially Grabe, S., Ward, L. M., & Hyde, J. S. (2008). The role of the
media in body image concerns among women: A meta-
accept the male gaze (Fallon & Rozin, 1985; Milkie, analysis of experimental and correlational studies. Psycho-
1999; Thomsen, 2002) applied to the relatively heavier logical Bulletin, 134, 460–476. doi:10.1037/0033-2909.
women. The findings, however, suggested that this is 134.3.460
not the case with thinner young women. Future Hayashi, F., Takimoto, H., Yoshita, K., & Yoshiike, N. (2006).
research should examine the possible reasons why Perceived body size and desire for thinness of young
Japanese women: A population-based survey. British
slimmer young women perceive their male-peer ideal
Journal of Nutrition, 96, 1154–1162. doi:10.1017/
body size as heavier than their own ideal body sizes. bjn20061921
One of the limitations of the present study is that the Kowner, R. (2002). Japanese body image: Structure and
sample was restricted to young women living in Tokyo. esteem scores in a cross-cultural perspective. International
Past research indicated that Japanese women who live in Journal of Psychology, 37(3), 149–159. doi:10.1080/
metropolitan areas tend to have a greater drive for thinness 00207590143000298
Lopez-Guimera, G., Levine, M. P., Sanchez-Carracedo, D., &
than those who live in smaller towns (Hayashi, Takimoto, Fauquet, J. (2010). Influence of mass media on body image
Yoshita, & Yoshiike, 2006). Thus, we should be careful and eating disordered attitudes and behaviors in females:
about the generalizability of these findings. Further A review of effects and processes. Media Psychology, 13,
Downloaded by [University of Florida] at 01:53 03 January 2018

research involving women living in local areas is needed. 387–416. doi:10.1080/15213269.2010.525737

In addition, although the underweight respondents Milkie, M. A. (1999). Social comparisons, reflected appraisals,
and mass media: The impact of pervasive beauty images on
in this study did not show body dissatisfaction, it is
Black and White girls’ self-concepts. Social Psychology
the result of an aggregate level of analysis. However, Quarterly, 62, 190–210. doi:10.2307/2695857
individual results indicated possible body dissatisfaction Stunkard, A., Sorenson, T., & Schulsinger, F. (1983). Use of
for some of the underweight respondents. Future Danish adoption register for the study of obesity and
research needs to investigate what types of underweight thinness. In S. Kety, L. P. Rowland, R. L. Sidman, &
women are dissatisfied with their body size. S. W. Matthysse (Eds.), The genetics of neurological and
psychiatric disorders (pp. 115–120). New York, NY: Raven.
Swami, V., Frederick, D. A., Aavik, T., Alcalay, L., Allik J.,
References Anderson, D., …, Zivcic-Becirevic, I. (2010). The attractive
female body weight and female body dissatisfaction in 26
American Psychological Association (2007). Report of the countries across 10 world regions: Results of the inter-
APA task force on the sexualization of girls. Washington, national body project I. Personality and Social Psychology
DC: Author. Bulletin, 36(3), 309–325. doi:10.1177/0146167209359702
Baba, A., & Sugawara, K. (2000). Drive for thinness in Thompson, J. K., & Heinberg, L. J. (1999). The media’s
adolescent women. Japanese Journal of Educational influence on body image disturbance and eating disorders:
Psychology, 48(3), 267–274. doi:10.5926/jjep1953.48.3_267 We’ve reviled them, now can we rehabilitate them?
Choi, Y., Leshner, G., & Choi, J. (2008). Third-person effects Journal of Social Issues, 55(2), 339–353. doi:10.1111/0022-
of idealized body image in magazine advertisements. 4537.00119
American Behavioral Scientist, 52, 147–164. doi:10.1177/ Thompson, M. A., & Gray, J. J (1995). Development and
0002764208321348 validation of a new body-image assessment scale. Journal
Cohen, J. (1992). A power primer. Psychological Bulletin, of Personality Assessment, 64, 258–269. doi:10.1207/
112(1), 155–159. doi:10.1037//0033-2909.112.1.155 s15327752jpa6402_6
Davidson, W. P. (1983). The third-person effect in communi- Thomsen, S. (2002). Health and beauty magazine reading and
cation. Public Opinion Quarterly, 47, 1–15. body shape concerns among a group of college
Fallon, A. E., & Rozin, P. (1985). Sex differences in women. Journalism and Mass Communication Quarterly,
perceptions of desirable body shape. Journal of Abnormal 79, 988–1007. doi:10.1177/107769900207900413
Psychology, 94(1), 102–105. doi:10.1037//0021-843x. Tiggemann, M., & Pickering, A. S. (1996). Role of television in
94.1.102 adolescent women’s body dissatisfaction and drive for
Frederick, D., Forbes, G. B., Grigorian, K. E., & Jarcho, J. M. thinness. International Journal of Eating Disorders, 20(2),
(2007). The UCLA body project I: Gender and ethnic 199–203. doi:10.1002/(sici)1098-108x(199609)20:2<199::
differences in self-objectification and body satisfaction aid-eat11>3.0.co;2-z
among 2,206 undergraduates. Sex Roles, 57, 317–327. Tiggemann, M., & Polivy, J. (2010). Upward and downward:
doi:10.1007/s11199-007-9251-z Social comparison processing of thin idealized media
Furnham, A., Badmin, N., & Sneade, I. (2002). Body image images. Psychology of Women Quarterly, 34, 356–364.
dissatisfaction: Gender differences in eating attitudes, self- doi:10.1111/j.1471-6402.2010.01581.x
esteem, and reasons for exercise. The Journal of Psychology, Wertheim, E. H., Paxton, S. J., & Tilgner, L. (2004). Test–retest
136(6), 581–596. doi:10.1080/00223980209604820 reliability and construct validity of the Contour Drawing
Gardner, R. M. & Brown, D. L. (2010). Body image Rating Scale in a sample of early adolescent girls. Body
assessment: A review of figural drawing scales. Personality Image, 1, 199–205. doi:10.1016/s1740-1445(03)00024-x