In the present study, teacher’s ratings, mother’s ratings, and father’s ratings on the Child Problematic Trait Inventory (CPTI) were obtained on two independent samples (N = 381 and N = 406, respectively) of Italian school age children (age range: 6–12 years). Cronbach alpha values greater than 0.80 were observed for all CPTI scale scores based on teacher’s ratings, mother’s ratings, and father’s ratings respectively. Although intraclass correlation values suggested poor parent-teacher agreement, robust weighted least square Confirmatory Factor Analyses (WLSMV CFAs) suggested that the three-factor model of the CPTI was consistently replicated as the best fitting model across all informants. Logistic regression analysis conducted in one of the samples showed that teacher’s ratings and mother’s ratings of the CPTI, but not CPTI father’s ratings, significantly predicted child’s problem behavior at school. Multiple regression analyses showed that CPTI scores were significantly predicted by selected temperament dimensions, but with different relationships for different sources of information. Overall, our findings suggest that the CPTI is a reliable and valid measure of problematic traits with a three-factor structure, although this study indicates that multiple sources of information may prove necessary when assessing psychopathy in school-age children with the CPTI.
