Alejandro Tapia-De-Jesús, Irais Castillo Rangel, Yazmín González-Alvarado, Miriam Barajas Márquez, Alicia Parra Carriedo, KayLoni Olson, Samantha Plezia, Claudia Unikel-Santoncini
EFA supported a five-factor structure (peers, media, muscular, family, and thin) explaining 69.36% of the variance. CFA showed excellent model fit (CFI = 0.99; RMSEA = 0.034; SRMR = 0.049). Internal consistency was high (subscales α = 0.86-0.96; total-scale α = 0.88-0.90), while test-retest reliability ranged from moderate to strong (r = 0.56-0.90). Convergent validity was supported by significant correlations with EDE-Q7 and BSQ-8D. Strict invariance was supported, and structural models showed adequate fit indices, although some latent parameters may differ between groups; scores were higher in the self-report clinical group.
OBJECTIVE: This study aimed to validate the Sociocultural Attitudes Towards Appearance Questionnaire-4 (SATAQ-4) in Mexican women, evaluating its factor structure, internal consistency, temporal stability, convergent validity, and measurement invariance in women with and without an eating disorder (ED) diagnosis.
METHOD: The sample included 1220 women (123 with a self-reported clinical ED diagnosis and 1097 without one), aged 18-73. Participants completed the SATAQ-4, the seven-item Eating Disorder Examination Questionnaire (EDE-Q7), and the Body Shape Questionnaire (BSQ-8D). Analyses included exploratory factor analysis (EFA), confirmatory factor analysis (CFA), reliability testing (Cronbach's alpha, McDonald's omega, and test-retest), and measurement invariance analyses.
RESULTS: EFA supported a five-factor structure (peers, media, muscular, family, and thin) explaining 69.36% of the variance. CFA showed excellent model fit (CFI = 0.99; RMSEA = 0.034; SRMR = 0.049). Internal consistency was high (subscales α = 0.86-0.96; total-scale α = 0.88-0.90), while test-retest reliability ranged from moderate to strong (r = 0.56-0.90). Convergent validity was supported by significant correlations with EDE-Q7 and BSQ-8D. Strict invariance was supported, and structural models showed adequate fit indices, although some latent parameters may differ between groups; scores were higher in the self-report clinical group.
DISCUSSION: The SATAQ-4 demonstrates strong psychometric properties in Mexican women. Its invariance supports meaningful comparisons between self-report clinical and non-clinical groups, making it a valuable tool for identifying sociocultural risk factors and informing prevention and intervention strategies.