Abstract
BACKGROUND: The psychometric properties of the Revised Restraint Scale (RRS) have been well established in western populations but not in Chinese adolescents. PURPOSE: This study investigated the psychometric properties of RRS and its validity in different subgroups for Hong Kong Chinese adolescents. METHOD: In 2007, 909 Hong Kong students aged 12 to 18 years (55.3% boys) completed a questionnaire including demographic items, RRS, Eating Attitudes Test (EAT-26), and Motivation for Eating Scale (MFES)-physical. Moreover, subjects' height and weight were measured. To examine the factor structure of RRS, the whole sample was randomly split into two groups (sample 1: N=454 and sample 2: N=455) for exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), respectively. Convergent and discriminant validity of RRS were investigated by correlating the RRS with EAT-26 and MFES-physical. Multigroup CFA was conducted to test the three-factor model of RRS in different sex, age, and weight status subgroups. RESULTS: Results of EFA for sample 1 revealed three strongly correlated factors for the RRS construct, and were supported by the CFA results in sample 2. Multigroup CFA further suggested that the three-factor model of RRS was stable across sex, age, and weight status subgroups. CONCLUSIONS: A new three-factor model is proposed for Hong Kong adolescents in this study. In general, RRS is a reliable and valid measure of restrained eating for adolescents, regardless of sex, age, and weight status.