Validade e confiabilidade da versão brasileira da Drive for Muscularity Scale (DMS) para jovens universitárias da cidade de Juiz de Fora - MG

Abstract

The pursuit of muscularity can be defined as the desire to achieve a muscular body as well as the concern and motivation in increasing muscle mass. There is a lack of research on this construct in women and, consequently, psychometric studies of scales that have the intention to evaluate the pursuit of muscularity in that audience. Given this knowledge absence, the aim of this study was to evaluate the validity and reliability of the Drive for Muscularity Scale (DMS) for Brazilian women. This is a cross-sectional investigation that includes the participation of 242 women for Confirmatory Factor Analysis (CFA) and convergent validity. Of these, 61 women were invited to participate in the reliability test, remaining 38 for retesting. The age of participants ranged from 18 to 35 years. Self-report instruments were applied to analyze the convergent validity of DMS, which were the following: Rosenberg SelfEsteem Scale (self-esteem), Beck Depression Inventory (depressive symptoms), Eating Attitudes Test-26 (risky behaviors for eating disorders) and Commitment Exercise Scale (psychological commitment to physical exercise). Factorial validity of DMS was evaluated by the CFA. The confidence level of the scale was determined by calculating Cronbach's alpha coefficient and in addition, by the test retest technique. For the latter, were used as a resource: the t test for repeated measures, Pearson’s correlation coefficient and the intraclass correlation coefficient (ICC). Were tested four models of DMS: 1) original model with two factors; 2) one-dimension model; 3) one-dimension model with another estimation method and 4) original model with two factors for women. However, none of these fit well, according to the values of the tests and multiple indexes calculated. DMS was associated only with self-esteem and self-acceptance (rpearson = 0.13; p <0.05) and risky behaviors for eating disorders (rpearson = 0.16; p <0.05). The value for t-test was t (40) = 2.2, p = 0.03. The ICC was equal to 0.86, p <0.001. According to the CFA, none of the tested models is suitable for application in the female audience. DMS was only correlated with self-esteem, self-acceptance and risky behaviors for eating disorders. The opposite happened with depressive symptoms and commitment to physical exercise. As for reliability, the t test showed a significant difference, indicating that the applied instrument didn’t provide sufficient reliability. The r values of Pearson’s explained 79% of the association between the two phases (test and retest). In addition, the ICC showed 73% correlation between test and retest. Finally, Cronbach's alpha coefficient demonstrated in both tests a low variance, that is, there was congruence between each item and the rest of the items of DMS. We conclude that although the instrument has shown some evidence of validity, the overall results were not satisfactory. More studies are needed to create a new instrument to measure this construct in the female audience.