The objective of this study was to refine the Children's Social Behavior Questionnaire (CSBQ), to reduce its length, and to verify its psychometric properties. The CSBQ is a questionnaire for parents or caregivers of children with PDD. The items describe a broad range of features that are typical of PDD, particularly in its milder forms. Based on conceptual judgment and factor analyses, the number of items was reduced from 96 to 49. Six subscales were constructed to allow a differentiated description of PDD problems. Estimates for internal, test-retest, and inter-rater reliability, and for convergent and divergent validity were good. Different clinical and control groups showed the hypothesized patterns in nature and degree of their problems.

The results implicate that the CSBQ not only has specific value as a measure of subtle social skills to identify pervasive developmental disorders, but that the instrument also has a specific contribution to differentiating between the two levels of ID. Furthermore, our outcomes imply a slight difference between limitations in subtle social skills as mentioned by the AAMR (American Association on Mental Retardation 2002) and limitations in subtle social skills as seen in milder forms of pervasive developmental disorders. Clinical and theoretical implications will be discussed.

The Children's Social Behavior Questionnaire (CSBQ) contains items referring to behavior problems seen in children with milder variants of PDD. Data of large samples of children diagnosed as having high-functioning autism, PDDNOS, ADHD, and other child-psychiatric disorders were gathered. Besides the CSBQ, parents completed the Autism Behavior Checklist (ABC) and the Child Behavior Checklist (CBCL). The data provided the basis for scale construction of the CSBQ, a comparison of the CSBQ scales with other instruments and a comparison of groups on scores on the CSBQ. The 5 scales obtained referred to Acting-out behaviors, Social Contact problems, Social Insight problems, Anxious/Rigid behaviors and Stereotypical behaviors. Results show that the CSBQ has good psychometric qualities with respect to both reliability and validity. A comparison of the different groups showed that significant group differences were found on all scales. In general, the autism group received the highest scores, followed by the PDDNOS group and the ADHD group. Exceptions were on the Acting-out scale, where the ADHD group scored highest and on the Social Insight scale, where no significant difference was found between the PDDNOS group and the ADHD group. Implications of the results and suggestions for further research are discussed.

No abstract

No abstract