RESUMEN
BACKGROUND: Research using the IPSCAN Child Abuse Screening Tool for Children (ICAST-C), has provided ample evidence of the magnitude of violence against children. Knowledge about its psychometric characteristics and validity is limited. Hence, our objective was to translate and culturally adapt the ICAST-C in adolescents from Mexico City and determine its psychometric properties. PARTICIPANTS AND SETTING: To determine the psychometric properties of the instrument 723 adolescents between 11 and 18 years of age from 9 public secondary schools in Mexico City participated. METHODS: The study was carried out in two phases: 1) translation and adaptation of the instrument (in 5 steps) and 2) pilot evaluation of the psychometric properties. Total and factor reliabilities were determined, Pearson correlation was used for temporal stability while construct validity was determined by Confirmatory Factor Analysis (CFA), and final adequacy of the items eliminated by the CFA. RESULTS: We developed the culturally relevant Mexican Spanish version of the ICAST-C. The CFA confirmed the six-factor structure hypothesis. To improve the original model we eliminated ten items, the final model showed good global fit indices (χ2(1310) = 2207.68, p < .01, χ2/df = 1.68; CFI =0.95; RMSEA = 0.02 [CI95% 0.02-0.03]; SRMR = 0.08). Total and factor reliabilities were adequate (Alpha = 0.79-0.92, r = 0.52-0.75), except for the non-violent discipline factor (Alpha = 0.59, r = 0.38). CONCLUSIONS: While these data suggest that this version of the ICAST-C is valid and reliable for adolescents in Mexico City public secondary schools, further research should evaluate the psychometric properties in a national sample.
