Prediction of body image dissatisfaction in university students by multivariate statistical methods

Résumé

Overweight and obesity are responsible for significant costs of the Brazil’s public health system and can modify the self-perception of the body image, attitudes, cognitions and thoughts. This study intends to propose predictive models to evaluate factors related to body image dissatisfaction among university students. Undergraduate students, with a mean age of 22.4 years, were interviewed in Minas Gerais, Brazil, by means of a behavioral questionnaire and evaluated the body image by scale of figures of silhouettes according to the Brazilian biotype. Contingency analysis pointed out dissatisfaction in 67.1% of the students and the time dedicated to sedentary activities presented significant influence on body image. Female undergraduate students from health science courses with recent weight gain tended to wish smaller silhouettes, while male students wished larger body sizes. Multinomial and logistic regression analyses revealed a good predictive power, but the multinomial regression was more advantageous and effective in demonstrating a satisfactory predictive model comprised of a larger number of variables to explain the body image dissatisfaction.