Psychometric properties of a silhouette scale on body image in adults

Abstract

Body image is a multifaceted construct involving several dimensions. Itis socially defined and the way we feel about our own image influences various aspects of our lives. Dozens of scales have been developed to measure the perception of body image, however, they are made up of contoured images whichrepresent bodies from the leanest to the fattest, not contemplating silhouettes with more apparent muscle mass. Thus, the aim of this study is to validate psychometric qualities of the silhouette scale proposed by Damasceno (2004) for measuring body image in adults. The sample consisted of59 students of both genders from the Department of Physical Education in a University. Anthropometric measurements followed the recommendations of ISAK and in order tostandardize the measurements the participants were asked: "What figure best represents your body?" (actual silhouette); "What figure represents the best the body you would like to have?" (ideal silhouette). The second visit occurred in an interval of 15 days afterthe first one and the same questions were then asked to them.The instrument's validity was calculated from the correlation coefficient with second-order polynomial adjustment of SA in relation to BMI (r = 0.77, p = 0.001) and G% (r = 0.58, p = 0.001). The Intraclass Correlation Coefficient (ICC) for the test and re-test ofthe Actual Silhouette (AS)and Ideal Silhouette (IS)werecalculatedfor reliability, with values of 0.94 and 0.93, respectively. The effect size was also calculated, showing a trivial result for SA and small for SI. The presented instrument has good psychometric qualities and can generally be applied to the adult Brazilian population.

Headnote

Abstract

Body image is a multifaceted construct involving several dimensions. Itis socially defined and the way we feel about our own image influences various aspects of our lives. Dozens of scales have been developed to measure the perception of body image, however, they are made up of contoured images whichrepresent bodies from the leanest to the fattest, not contemplating silhouettes with more apparent muscle mass. Thus, the aim of this study is to validate psychometric qualities of the silhouette scale proposed by Damasceno (2004) for measuring body image in adults. The sample consisted of59 students of both genders from the Department of Physical Education in a University. Anthropometric measurements followed the recommendations of ISAK and in order tostandardize the measurements the participants were asked: "What figure best represents your body?" (actual silhouette); "What figure represents the best the body you would like to have?" (ideal silhouette). The second visit occurred in an interval of 15 days afterthe first one and the same questions were then asked to them.The instrument's validity was calculated from the correlation coefficient with second-order polynomial adjustment of SA in relation to BMI (r = 0.77, p = 0.001) and G% (r = 0.58, p = 0.001). The Intraclass Correlation Coefficient (ICC) for the test and re-test ofthe Actual Silhouette (AS)and Ideal Silhouette (IS)werecalculatedfor reliability, with values of 0.94 and 0.93, respectively. The effect size was also calculated, showing a trivial result for SA and small for SI. The presented instrument has good psychometric qualities and can generally be applied to the adult Brazilian population.

Keywords: -Reliability; Body ideal; Validity and reliability; validity of studies.

10.7752/jpes.2020.03180

Introduction

Body image is a multifaceted construct characterized by its complex definition due to subjective aspects such as cognition, affective responses, behavioral traits, and perception (Cash & Smolak, 2011; Slade, 1994). These characteristics constantly interact among each other, restructuring the body image of the individual itself. Therefore, it is a continuous process whichlasts for one's entire life and is developed by sociocultural aspects, interpersonal relationships, values, attitudes inherent to culture, and media influence (Banfield & McCabe, 2002; Cash & Smolak, 2011; Slade, 1994).

In addition, the way anindividual perceives themselfinfluences behavior, relationships, and how we comprehend the world (Barros, 2005). It is common to find people who are unsatisfied with their body image, and this issue might be considered a risk factor since it is related toeating disorders, depression, and suicide(Adami et al., 2012; Tritschler & Barrow, 2000). Thus, body image perception and body dissatisfaction are measured by body size estimate techniques; one such technique is the silhouette scale which ais simple, efficient, and feasible instrument (Cash & Smolak, 2011; Tritschler & Barrow, 2000). It is composed offrontal images of silhouettes which are ordered horizontally in a crescent way of silhouettes whichvary from thin to fat or strong bodies (Figure1). The instrument asks the individual which of thesilhouettes represent theiractual body,and which body the individual would like to have. The calculus of the difference between the actual body and the desired one is used to define body image dissatisfaction level (Ahmadi, Heyrani, & Yoosefy, 2018; Cash & Smolak, 2011; R. Gardner & Brown, 2010; R. M. Gardner, Jappe, & Gardner, 2009).

In fact, dozens of scales for different populations havecreated and several were tested(Castro, Damasceno, Miranda, Lima, & Vianna, 2011; R. Gardner & Brown, 2010; R. M. Gardner et al., 2009; Hildebrandt, Langenbucher, & Schlundt, 2004; Kakeshita, 2008; Li, Hu, Ma, Wu, & Ma, 2005; Lima, Orlando, Teixeira, Castro, & Damasceno, 2008; Lønnebotn et al., 2018; Peterson, Ellenberg, & Crossan, 2003; Stunkard, Sorensen, & Schulsinger, 1983; Yepes, Viswanathan, Bovet, & Maurer, 2015). R. Gardner and Brown (2010)assessed many of themin a review about perceptive scales. In Brazil, Kakeshita (2008)proposed a silhouette scale for children and adults, which is the only one that has been validated for the Brazilian population. Nevertheless, the scale does not include muscular bodies, which are highly appreciated nowadays. Moreover, it seems that athletic bodies are preferred, being characterized by voluminous muscle mass (Damasceno, 2004; Damasceno et al., 2011; Lima et al., 2008; Pope, Olivardia, Gruber, & Borowiecki, 1999) mainly motivated by media with toys and advertisements selling the desire for sculptured bodies (Pope et al., 1999). Additionally, Voracek and Fisher (2002) analyzed anthropometric characteristics of women photographed inPlayboy Magazine fromits first publication to nowadays, showing a visible decrease in body mass index and waist-hip ratio.

In fact, according to the desired perfect body, Benkeser, Biritwum, and Hill (2012); Damasceno (2004); Damasceno et al. (2011); Damasceno et al. (2008)verified that females are the most affected. A thin body characterized by low weight is the most pursued by women, while large volumes of muscles and low body fat isthe most wanted among men (Damasceno, 2004; Damasceno et al., 2011). Therefore, it is likely both genderswant a more muscular body when compared to the ones they have (Damasceno et al., 2011). However, instruments whichmeasure body image throughout several types of silhouettes including muscular, thin, and fat bodies are lacking in the literature(Hildebrandt et al., 2004). Scales whichare concerned about muscle mass are included in studies byDamasceno et al. (2008); Lima et al. (2008), Castro et al. (2011), Gyory, Brandäo, Agresta, and Rebustini (2003) and Pedretti, Neves, Silva, Toledo, and Lima (2003); however, none of them werevalidated. Thus, the aim of this study is to validate the psychometric qualities of the silhouette scale proposed by Damasceno (2004)and adapted by Pedretti et al. (2003) for measuring body image in adults (Figure 1).

Material & methods

Study design

This is a transversal study performed in two visits.The study was approved by the ethical committee and all participants signed the informed consent form. The study protocol was approved by the Human Research Ethics Committee (no.1.140.803, CAAE - 1.140.803), Federal University of Pernambuco, Brazil. All the volunteers agreedto participate in the study, and parents/guardians also signed the written informed consent. Participants

The sample consisted in 59 students of the Department of Physical Education in a University, 38 of them are men and 21 women (Table 1).

Procedure

In the first visit, all participants were oriented about the aim and procedures of the study and signed the consent term.Anthropometric measurements were then performed. In addition, in order tostandardize the measurements the participants were asked: "What figure best represents your body?" (actual silhouette); "What figure best represents the body you would like to have?" (ideal silhouette). Next, the second visit occurred afteran interval of 15 days fromthe first one and the same questions were asked to them.

Anthropometric measurements

Anthropometric measurements followed the recommendations of the International Society for Advancement of Kinanthropometry, International Conference (International Society for Advancement of Kinanthropometry. International Conference (8th: 2002: Manchester England), Reilly, & Marfell-Jones, 2003). Thus, weight, height, and skinfolds (triceps, medial gastrocnemius) were quantified. Materials used for the assessments were: electronic scale (Filizola, Brazil), measuring tape (Sanny, Brazil), and adipometer (Lange, USA). In addition, body mass index (BMI) was obtained by the formula weight/height2. Criterion validity was determined as a result of the correlation between anthropometric characteristics and the silhouette scale.

Body image measurements

The silhouette scale created by Damasceno (2004) and adapted by Pedretti et al. (2003) is composed of15 images. Its values vary from -7, whichrepresents a silhouette with high levels of fat tissue, to +7 representing high levels of muscular mass. The scale also includes different bodies for men and women. The central silhouette, represented by the value of 0, denotes a thin body with lower muscle mass and body fat. Statistical analysis

All analyses are presented in mean and standard deviation. Validity was measured by the correlation coefficient. Reliability was measured using intraclass correlation coefficient (ICC), effect size, and Wilcoxon test. Alpha level was estimated in 5%. Analyses were doneby SPSS for Windows v. 20.

Results

Scale validity was calculated by correlation coefficient with polynomial adjustment between actual silhouette, BMI, and body fat% altogether and stratified by gender(Table 2).

The intraclass correlation coefficient (ICC) of the actual and ideal silhouette presented values of 0.94 and 0.93, respectively. Moreover, the effect size of the responses in the first visit related to the second was calculated. The results for the actual silhouette were considered trivial, whereas the desired silhouette was weak (Table 3).

In addition, body image measurements were compared by the Wilcoxon test. The analyses showed no differences between the first and second moments for actual silhouette (z=-0.459; p=0.659) and ideal silhouette (z=-0.819; p=0.413) (Table 4).

Discussion

The results showed a correlation between anthropometric measurements with the silhouette scale proposed by Damasceno (2004), indicating good construct validity. Therefore, the instrument interprets how well the participants perceive themselves; thus, the silhouette scale is a valid instrument for measuring body image in the adult population.

A correlation coefficient with quadratic adjustment was used to report the results due to the scale presenting a "U" correlation with variables such as BMI. This occurred because BMI increases on the extremities of the scale sinceone extremity is related to increased body fat index and the opposite to augmented lean mass index; therefore, linear regression is inappropriate since the slope would be insufficient to demonstrate tendency. Another studybyour research group found the same behavior of a correlation between BMI and AS (Damasceno et al., 2011).

The scale presented reasonable resultsaccording to validity. Even though body image is a dynamic variable, participants maintained the same or really close perceptions of actual and desired silhouette within 15 ± 5 days of interval between measurements. Furthermore, in a review conductedby R. Gardner and Brown (2010), scale validity for the adult population presented lower values when compared to this study,however with only one exception for the scale of Stunkard et al. (1983)whichyielded a correlation of 0.99. However, regarding reliability, the scale presented values of 0.81 for actual and 0.92 for desired silhouette. The onlyscale whichpresented higher reliability values thanthe present study was created by Tucker et al.; nevertheless, the scale is only for men, and presented values of 0.94 and 0.96 for actual and desired silhouette, respectively. In addition, Kakeshita (2008) validated a body image scale in the Brazilian population using the Pearson correlation (r=0.93. p<0.05) in which the reliability of the scale was also measured (r=-0.79). Their findings show good psychometrics results; however, it does not consider muscular bodies.

Thus, in considering the values presented, the scale proposed by Damasceno (2004) reports similar values when compared with those related in the literature. However, it is important to emphasize that the proposed scale includes both bodies with high adiposity levels as well as those with increased muscle mass. Therefore, the scale attends to the ideal body present in society nowadays, especially in the young population. The instrument is also feasible and might be easily used in gyms and other locations where body image is important to measure.

Conclusions

The body image scale yielded good psychometric qualities and couldbe applied to the adult population. The scale also not only enables measurementsfor bodies with high levels of fat, but also forincreased muscle mass.

Conflicts of interest -The authors do not have any conflicts of interest to declare.

Sidebar

Published online: May 30, 2020

(Accepted for publication: May 18, 2020)

References

References

Adami, F., Schlickmann Frainer, D. E., de Souza Almeida, F., de Abreu, L. C., Valenti, V. E., Piva Demarzo, M. M., . . . de Oliveira, F. R. (2012). Construct validity of a figure rating scale for Brazilian adolescents. Nutr J, 11, 24. doi:10.1186/1475-2891-11-24

Ahmadi, S., Heyrani, A., & Yoosefy, B. (2018). Prevalence of body shape dissatisfaction and body weight dissatisfaction between female and male students. Journal of Physical Education and Sport, 18(4), 2264 - 2271.

Banfield, S. S., & McCabe, M. P. (2002). An evaluation of the construct of body image. Adolescence, 37(146), 373-393. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/12144166

Barros, D. D. (2005). Imagem corporal: a descoberta de si mesmo %J História, Ciências, Saúde-Manguinhos.12, 547-554. Retrieved from http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104- 59702005000200020&nrm=iso

Benkeser, R. M., Biritwum, R., & Hill, A. G. (2012). Prevalence of overweight and obesity and perception of healthy and desirable body size in urban, Ghanaian women. Ghana Med J, 46(2), 66-75.

Cash, T. F., & Smolak, L. (2011). Body image : a handbook of science, practice, and prevention (2nd ed.). New York, NY: Guilford Press.

Castro, A. P. A., Damasceno, V. O., Miranda, J. A., Lima, J. R. P., & Vianna, J. M. (2011). Photo Silhouettes for Assessment of Body Image of Bodybuilders. Revista brasileira de medicina do esporte, 17, 250-253. Retrieved from http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1517- 86922011000400007&nrm=iso

Damasceno, V. O. (2004). Correlation between the level of dissatisfaction with a body image and anthropometric profile of practitioners of gym.(Master Science). Castelo Branco University, Rio de Janeiro.

Damasceno, V. O., Vianna, J. M., Novaes, J. S., Lima, J. R. P., Fernandes, H. M., & Reis, V. M. (2011). Relationship between anthropometric variables and body image dissatisfaction among fitness center users. Revista de Psicología del deporte, 20(2), 367 - 382.

Damasceno, V. O., Vianna, V. R. A., Vianna, J. M., Lacio, M. L., Lima, J. R. P., & Novaes, J. S. (2008). Body image and ideal body. Revista brasileira de ciência e movimento, 14(2), 81-94.

Gardner, R., & Brown, D. (2010). Body image assessment: A review of figural drawing scales. Personality and Individual Differences, 48, 107-111. doi:10.1016/j.paid.2009.08.017

Gardner, R. M., Jappe, L. M., & Gardner, L. (2009). Development and validation of a new figural drawing scale for body-image assessment: the BIAS-BD. J Clin Psychol, 65(1), 113-122. doi:10.1002/jclp.20526

Gyory, R., Brandão, M. A., Agresta, M. C., & Rebustini, F. (2003). Dismorfia muscular em fisiculturistas do sexo feminino. Paper presented at the Anais do XXVI simpósio Internacional de Ciências do Esporte, São Paulo.

Hildebrandt, T., Langenbucher, J., & Schlundt, D. G. (2004). Muscularity concerns among men: development of attitudinal and perceptual measures. Body Image, 1(2), 169-181. doi:10.1016/j.bodyim.2004.01.001

International Society for Advancement of Kinanthropometry. International Conference (8th : 2002 : Manchester England), Reilly, T., & Marfell-Jones, M. (2003). Kinanthropometry VIII : proceedings of the 8th International Conference of the International Society for the Advancement of Kinanthropometry (ISAK). London ; New York: Routledge.

Kakeshita, I. S. (2008). Adaptation and validation of Silhouette Scales for Brazilian children and adults. (PhD in Psychobiology). University of Sao Paulo, Ribeirão Preto.

Li, Y., Hu, X., Ma, W., Wu, J., & Ma, G. (2005). Body image perceptions among Chinese children and adolescents. Body Image, 2(2), 91-103. doi:10.1016/j.bodyim.2005.04.001

Lima, J., Orlando, F., Teixeira, M., Castro, A., & Damasceno, V. (2008). Silhouettes to assess body image of bodybuilders. Cardernos Arquivos Sanny, 1(1), 26-30.

Lønnebotn, M., Svanes, C., Igland, J., Franklin, K. A., Accordini, S., Benediktsdóttir, B., . . . Gómez Real, F. (2018). Body silhouettes as a tool to reflect obesity in the past. PLoS One, 13(4), e0195697. doi:10.1371/journal.pone.0195697

Pedretti, A., Neves, M. V., Silva, E. R. A., Toledo, S. A., & Lima, J. R. P. (2003). Validation of silhouettes to assess the body image of women with an interest in weight loss and hypertrophy. Paper presented at the Proceedings of the XXVI International Symposium on Sport Sciences, São Paulo.

Peterson, M., Ellenberg, D., & Crossan, S. (2003). Body-image perceptions: reliability of a BMI-based Silhouette Matching Test. Am J Health Behav, 27(4), 355-363. doi:10.5993/ajhb.27.4.7 Pope, H. G., Jr., Olivardia, R., Gruber, A., & Borowiecki, J. (1999). Evolving ideals of male body image as seen through action toys. Int J Eat Disord, 26(1), 65-72. doi:10.1002/(sici)1098-108x(199907)26:1<65:aideat8> 3.0.co;2-d

Slade, P. D. (1994). What is body image? Behaviour Research and Therapy, 32(5), 497-502. doi:10.1016/0005- 7967(94)90136-8

Stunkard, A. J., Sorensen, T., & Schulsinger, F. (1983). Use of the Danish Adoption Register for the study of obesity and thinness. Res Publ Assoc Res Nerv Ment Dis, 60, 115-120.

Tritschler, K. A., & Barrow, H. M. (2000). Barrow & McGee's practical measurement and assessment (5th ed.). Philadelphia: Lippincott Williams & Wilkins.

Voracek, M., & Fisher, M. L. (2002). Shapely centrefolds? Temporal change in body measures: trend analysis. Bmj, 325(7378), 1447-1448. doi:10.1136/bmj.325.7378.1447

Yepes, M., Viswanathan, B., Bovet, P., & Maurer, J. (2015). Validity of silhouette showcards as a measure of body size and obesity in a population in the African region: A practical research tool for generalpurpose surveys. Popul Health Metr, 13, 35. doi:10.1186/s12963-015-0069-6