Abstract
[Abstract]

Abstract

Abstract

Abstract
Our aim was to determine prototypical patterns of eating behaviour among Portuguese higher education students, and to relate these patterns with BMI. Data from 280 higher education students (63.2% females) aged between 18 and 27 years were analysed. Several eating behaviour dimensions (emotional and external eating, flexible and rigid restraint, binge eating, and eating self-efficacy) were assessed, and eating styles were derived through cluster analysis. BMI for current, desired and maximum self-reported weights and the differences between desired and current BMI and between maximum and current BMI were calculated. Women scored higher in emotional eating and restraint, whereas men showed higher eating self-efficacy. Men had higher current, desired and maximum BMI. Cluster analysis showed three eating styles in both male and female subsamples: "Overeating", "High self-efficacy" and "High restraint". High self-efficacy women showed lower BMI values than the others, and restrictive women had higher lost BMI. High self-efficacy men showed lower desired BMI than overeaters, and lower maximum and lost BMI than highly restrictive ones. Restrictive women and men differ on important eating behaviour features, which may be the cause of differences in the associations with BMI. Eating self-efficacy seems to be a central variable influencing the relationships between other eating behaviour dimensions and BMI.

Abstract

Abstract
Self-efficacy refers to the beliefs in the ability to organise and implement the action plans needed to achieve a certain result and the feeling of control over the behaviours and environment. It determines the initiation, maintenance and cessation of strategies or behaviours, being a good predictor of eating behaviour. Despite its potential in terms of research, prognosis and evaluation of interventions, we are unaware of instruments to assess global features of food self-efficacy validated for the Portuguese population. The aim of this work is to develop and validate a scale to assess general features of food self-efficacy ("Escala de Auto-Eficácia Alimentar Global"). We evaluated a sample of 276 higher education students, and the analysis was performed separately for the subsamples of females and males. The items were created through the adaptation of the six items of the self-efficacy subscale of the "Inventário Clínico de Auto-Conceito" (Serra, 1986). The analysis led to the maintenance of five of the six items initially considered. The scale shows an unifactorial structure, and the proportion of total variance explained by the principal components extracted is higher than 64%. It also shows good internal consistency, with values of Cronbach's alpha above 0.85. The study of relations with other measures showed evidence of its convergent and discriminant validities. The "Escala de Auto-Eficácia Alimentar Global" may prove to be an useful tool for the prognosis and the evaluation of interventions related to the change in eating habits.