Validation of a Brazilian version of the Checklist for Assessment of Gender Disadvantage (CAGED) in adolescents.

Abstract

Objective: To validate and culturally adapt a Brazilian version of the Checklist for Assessment of Gender Disadvantage (CAGED) for girls, boys, and gender-diverse adolescents, and determine the optimal cutoff point.

Methods: The original instrument, designed to assess perceptions of gender inequity among girls, was validated alongside a version for boys and gender-diverse people through seven stages: translation, synthesis, back translation, expert evaluation, pre-testing with 29 adolescents, re-evaluation, and clinical validation with 459 adolescents. To assess perceived gender discrimination, in addition to the original binary response options (yes/no), three alternative responses were tested to clarify whether "it never happened" or "it happened but does not bother me". The Content Validation Index (CVI) was calculated, with scores ≥0.80 indicating acceptable agreement. The scale's reliability was assessed using Cronbach's alpha, with a threshold of ≥0.70 indicating good internal consistency. The Self-Reporting Questionnaire (SRQ-20) was used as the gold standard to construct a receiver operating characteristic (ROC) curve and determine the optimal sensitivity, specificity, and cutoff point for the CAGED.

Results: The translation demonstrated excellent agreement, with all CVIs exceeding 0.80. The Cronbach's alpha in the pre-test was 0.97. Both response formats proved adequate, with CVIs of 0.83 for the binary and 0.79 for the three-response alternatives. The cutoff point with the highest sensitivity (72.08%) and specificity (82.61%) was 4. The area under the ROC curve was 0.823.

Conclusions: The CAGED was successfully translated and culturally adapted for Brazilian adolescents, showing high internal consistency and clarity.