Internal structure and measurement invariance of the Dominic Interactive among Indigenous children in Quebec

This study aims to examine the reliability and validity of the French version of the Dominic Interactive screening tool (Valla, 2008) among Indigenous children in Quebec. The Dominic Interactive is a computerized screening tool, which assesses prevalent emotional and behavioral problems in children. Participants in this study were 195 Innu Nation children aged between 8 and 11 years. Statistical analyses were performed on each of the 7 scales of the Dominic Interactive to assess reliability, factor structure, and measurement invariance across boys and girls. Results show satisfactory reliability (ranging from α_tet = .83 to .94 and from ω = .84 to .95) for 5 out of the 7 scales scores. Separation Anxiety and Specific Phobias scales failed to show adequate reliability. Confirmatory factor analyses confirm the 1-factor structure for Opposition and Conduct Problems scales (root mean square error of approximation, RMSEA ≤ .05; comparative fit index, CFI ≥ .95). Within an exploratory framework, confirmatory factor analyses also show good fit indices of relaxed models for Inattention/Hyperactivity/Impulsivity, Depression, and Specific phobias, admitting some error correlations. Generalized anxiety had poorer model fits; factor structure is not confirmed for this scale. The Separation anxiety construct appears to be better described by a 2-factor structure than by the postulated 1-factor structure. Measurement invariance between boys and girls was sufficiently supported for most of the scales, except for Specific Phobias. Therefore, results demonstrate promising reliability and validity for scales evaluating behavioral problems and depressive symptoms, but further research is still needed to determine the generalizability of these exploratory results in Indigenous populations. (PsycINFO Database Record (c) 2020 APA, all rights reserved).