The Sociocultural Attitudes Toward Appearance Questionnaire (SATAQ-3): reliability and normative comparisons of eating disordered patients

Examination of the Sociocultural Attitudes Towards Appearance Questionnaire-3 in a mixed-gender young-adolescent sample.

“…Thompson et al (2004) found internal reliability scores to range from 0.96 EXAMINATION OF THE SATAQ-3 9 (Internalization -General and Information), to 0.89 (Internalization -Athlete), whereas internal consistencies were slightly lower in the other non-clinical, adult female sample ranging from 0.90 (Internalization -General) to 0.85 (Internalization -Athlete and Pressures). Similar results were found in a disordered eating inpatient sample (Calogero et al, 2004) with only the Internalization -Athlete scale producing Cronbach's  below 0.90, ranging from 0.77 to 0.84 depending on age and diagnosis. Cronbach's  in the Jordanian sample (Madanat et al, 2006) were marginally lower ranging from 0.82 (InternalizationAthlete) to 0.89 (Information), while internal consistency of the scales in the male sample (Karazsia & Crowther, 2008) was high ranging from 0.95 (Information) to 0.85…”

“…The convergent validity of the SATAQ-3 has been previously assessed (Calogero et al, 2004;Thompson et al, 2004) In their two samples, Thompson et al (2004) found correlations between the SATAQ- Mean item scores from previous research (Thompson et al, 2004;Calogero et al, 2004) showed that disordered eating adolescents generally had higher mean SATAQ-3 scores than disordered eating adults, most notably on the Internalization -General scale, with this difference reported as significant in the original study (Calogero et al). By diagnostic categories, participants with bulimia nervosa generally had higher internalization scores than participants with other diagnoses.…”

“…Thompson et al (2004) designed the SATAQ-3 with the four-factor solution being confirmed in both of their samples of college-aged women, with an identical factor structure found in the Jordanian sample (Madanat et al, 2006), providing some evidence for the crosscultural utility of the measure. This solution was largely replicated in the clinical sample (Calogero et al, 2004) with all original items loading highest on the original factor, apart from one Internalization -Athlete item (I compare my body to that of people in "good shape"), which loaded higher on the Internalization -General factor. In their confirmatory factor analyses of the SATAQ-3, Markland and Oliver (2008) found the same item to again be problematic and suggested this item be omitted, however their findings overall also supported the 4-factor solution of the SATAQ-3.…”

“…To our knowledge of published studies, the SATAQ-3 has been evaluated in adult female community samples (Madanat, Hawks, & Brown, 2006;Markland & Oliver, 2008;Thompson et al, 2004), a clinical, predominantly adult sample (and a minority of adolescents : Calogero, Davis, & Thompson, 2004), and one study has examined its properties with males (a young-adult community sample: Karazsia & Crowther, 2008). The current study is the first published investigation of the psychometric properties of the SATAQ-3 in a community sample of adolescent females and males (younger than 18 years of age), who are most typically the population involved in universal prevention programs.…”

“…This rotation method allows for pre-existing associations between factors, which has been previously found for the SATAQ-3 scales (Calogero et al, 2004;Thompson et al, 2004). Items where factor loadings were greater than .30 were retained for further analyses.…”