The MacArthur three-city outcome study: Evaluating multi-informant measures of young children's symptomatology

Objective: Three sites collaborated to evaluate the reliability and validit y of 2 measures, developed in tandem to assess symptomatology and impairmen t in 4- to 8-year-old children: the Berkeley Puppet Interview Symptomatolog y Scales (BPI-S) and the Health and Behavior Questionnaire (HBQ). Method: I n this case-control study, mothers, teachers, and children reported on mult iple dimensions of children's mental health for 120 children (67 community and 53 clinic-referred children). Results: The BPI-S and the parent and tea cher versions of the HBQ demonstrated strong test-retest reliability and di scriminant validity on a majority of symptom scales. Medium to strong effec t sizes (Cohen d) indicated that children in the clinic-referred group were viewed by all 3 informants as experiencing significantly higher levels of symptomatology than nonreferred, community children. Conclusion: The availa bility of a set of multi-informant instruments that are psychometrically so und, developed in tandem, and developmentally appropriate for young childre n will enhance researchers' ability to investigate and understand symptomat ology or the emergence of symptomatology in middle childhood.