Abstract
This study examined the structural and item-level psychometric properties of the Emotion Regulation Questionnaire for Children and Adolescents (ERQ-CA) in 525 Brazilian youth aged 8–17 years (M = 13.84; SD = 3.18), of both sexes (50.87% girls), enrolled in public schools. Participants completed the ERQ-CA and mental health and well-being measures. We evaluated the proposed two-factor structure, measurement invariance across sex and age, item discrimination and threshold parameters using Item Response Theory (IRT), Differential Item Functioning (DIF), and associations with external indicators. The results provided partial support for two-factor model, with heterogeneous global fit indices and evidence of localized residual dependence. Measurement invariance analyses supported equality of factor loadings and thresholds across sex and age. Sensitivity analyses indicated that age-related loading invariance was somewhat dependent on how the age groups were defined. DIF was detected in two Suppression items, but its score-level impact was minimal for sex and small but detectable for age. Associations with external variables were predominantly small to moderate in magnitude and showed distinct patterns for Reappraisal and Suppression. Overall, the findings support the interpretability of the ERQ-CA while also identifying localized model limitations and group-related item functioning that should be considered when interpreting scores.