The validity of self-report measures for young children
Saye, Kimberley Blaker
MetadataShow full item record
Questions of reliability, validity, and the ability of children to respond to child self-report measures of emotions and perceptions have historically made self-report measures inconsequential for research and clinical purposes. However, there have been many improvements in reliability and validity of these measures, and some developmental investigations indicate that children can report better than once thought. In order to add new information to this literature the current investigation assessed the validity of child self-report of emotions and behavior using the Behavior Assessment System for Children (BASC) Self-Report of Personality (SRP-C), in comparison to Teacher Rating Scales (TRS-C), and Parent Rating Scales (PRS-C).. The purpose of this study was to assess the external validity of the BASC SRP-C using a sample of 147 clinic-referred children ages 8-11 years and another sample of 36 school children ages 8-11 years. Predictive validity was assessed by correlating BASC scores with outcomes of referral for intervention and number of disciplinary reports. Child ratings were found to significantly correlate with discipline reports, while teacher and parent ratings were significantly correlated with both discipline reports and referral for intervention. Parent and child scores were correlated in the clinic sample and, overall, were found to agree at a higher-than-expected level. Predictive Discriminant Analysis was used to assess the utility of multiple informant data for clinical diagnosis. Though child reports were found to be poor predictors of diagnostic categories when examined independently, the information obtained contributed uniquely to the results of a comprehensive assessment. Positive Predictive Power and Negative Predictive Power were also computed indicating that the BASC SRP-C scales did not have utility as specific inclusionary or exclusionary criteria for the diagnosis of a disruptive behavior problem. The results are discussed in terms of their additive value for clarifying the validity of child self-reports for specific research and clinical purposes.