Screening for Emotional and Behavioral Problems in Children Aged Between 8 and 11 Years Old : Transcultural Validation of the Dominic Interactive Among Aboriginal Innu Children Based on Vallerand's Methodology

Introduction: Aboriginal youths present in a large proportion emotional and behavioral problems (EBP) (Beals et al., 2005; Lemstra et al., 2013; Whitbeck et al., 2008). However, few psychological test has been validated for this population (Williamson et al., 2014). In a prevention perspective, valid psychological screening test would allow to better identify EBP among aboriginal children. The main purpose of this study is to validate a screening test for EBP among this population. The Dominic Interactive (Valla, 2000) was chosen for its good psychometric properties demonstrated in various populations (Bergeron et al., 2013; Kuijpers et al., 2014; Shojaei et al., 2009).

Methods: Based on Vallerand’s transcultural validation methodology (1989), psychometric properties of the Dominic Interactive were evaluated among Innu Aboriginal children aged between 8 and 11 years old (N = 97; Girls = 51.5%; Age : M = 9.4 (1.1)) from a rural area in Quebec. The children were met in their community school to complete the Dominic Interactive. Thirty randomly selected out of the 97 completed it a second time for test-retest reliability. Teachers completed the Teacher report form of the ASEBA for 93 of the children. For each psychometric property, statistical analyses were performed on each of the seven scales of the Dominic Interactive : Specific phobias (SP), Separation anxiety (SA), Generalized anxiety (GA), Depression (DEP), Opposition (OPP), Conduct problems (CP) and ADHD.

Results: All scales have satisfactory to excellent test-retest reliability with intra-class correlations between .70 for SP to .91 for ADHD. Internal consistency is satisfactory for 5 scales (α = .79 for GA to α = .82 for ADHD) but SP (α = .58) and SA (α = .68) scales show poor internal consistency. Some low but significant correlations (r = .21 to r = .29) are observed between the Dominic Interactive and the ASEBA scales: results suggest promising convergence for GA, DEP, CP and ADHD. About construct validity, factor analysis confirm the model for SA (p = .42; CFI = .99; RMSEA = .02), GA (p = .26; CFI = .96; RMSEA = .03) and OPP (p = .10; CFI = .96; RMSEA = .06). SP scale do not fit the model. For DEP, CP and ADHD, some indices seem to fit the model. However, small sample size compared to the larger number of items for those scales may limit statistical power. Model fit is not confirmed. The clinical standards seem appropriate for four scales (DEP, OPP, CP and ADHD) while for the three scales of anxiety the Innu children are overrepresented.

Conclusion : Besides these clinical standards that could be revised and the SP scale that shows poorer psychometric properties, the Dominic Interactive seems appropriate to screen for EBP among Aboriginal Innu children. Further research would allow to verify the generalisation of these results among other aboriginal nations and to verify the clinical utility of the tool.