Betrouwbaarheid en validiteit van de schalen van disruptief gedrag in de oudermodule van de DB-DOS
MetadataShow full item record
Objective: To examine the internal consistency, the interrater reliability and the validity of the Parent Module of the Disruptive Behavior Diagnostic Observation Schedule (DB-DOS), a new observational method for assessing preschool disruptive behavior. Method: A total of 107 coded DB-DOS sessions formed the sample, which were coded within three domains: problems in ‘anger modulation’, problems in ‘behavioral regulation’, and ‘competence’. The internal consistency was assessed with Cronbach’s alpha. The interrater reliability was assessed with Interclass Correlation Coefficient. Three groups have been composed on their diagnoses; a group of children with DBD, a group of children with DBD+ADHD and a control group. For assessing the validity form a categorical point of view, the group differences were tested using multivariate analyses of variance (MANOVA), followed by univariate analyses of variance (ANOVA). For assessing the validity from a dimensional point of view the scores of the DB-DOS were correlated with the outcomes of parent and teacher reports about aggressive behavior, global severity and effects on the family. Results: First, the internal consistency and inter raterreliability were high for the domains ‘anger modulation’ and ‘behavioral regulation’. For ‘competence’ the internal consistency was medium and the inter raterreliability was low. Second, there was only an effect of group on the domain ‘behavior regulation’. Following univariate analyses of variance revealed that both the DBD group and the DBD + ADHD group scored significantly higher on this domain than the control group. Third, the Pearson correlation coefficients of the domain ‘behavioral regulation’ and CBCL, TRF, CGAS, and IFS were significant. Of the domain ‘anger modulation’ only the IFS was not significant and of the domain ‘competence’ only the IFS father was significant. Conclusion: The results of this study are promising, considering that the sample is small and the results are from the parent module only. In next studies the other two contexts need to be included in the analyses and the domain ‘competence’ needs further improvement because it’s poor reliability.