Måleegenskaper ved den norske versjonen av Autism Spectrum Screening Questionnaire (ASSQ)

Abstract

The Autism Spectrum Screening Questionnaire (ASSQ) was first published in 1999 by Ehlers, Gillberg and Wing. It was translated to Norwegian by Anita Eriksson and Anne Mari Sund in 1996. The instrument is developed to serve as a screening tool to identify children with possible autism spectrum disorder. Adult informants who know the child well completes the questionnaire. ASSQ has one main scale and recommended cut off values (indication possible autism) for teachers and parents. The test consists of 27 items, and no formal competency demands are listed for use or interpretation of results. A literature search resulted in 121 articles: 21 Norwegian and 9 Swedish or Danish articles were included in the present review. The articles included documentation of interrater reliability, construct validity, criterion validity, discriminant validity, and norms for a limited age group. Mean values were found for different clinical groups (autism, cerebral palsy, neurological disorders, premature children, child born by mothers with drug abuse, or smoking during pregnancy). The ASSQ has excellent test-retest reliability. The test's construct validity seems in general to be satisfactory, although more research is needed to test its convergence and discriminative validity. The ASSQ seems to have good properties as a screening tool for autism, as scores above cutoff identified by teachers or parents indicate possible autism spectrum disorder in need of further examination. When used in research projects, it is uncertain how accurate it is to use scores above the clinical cut-off value as a proxy for diagnosed autism, as there will be many false positives without the test results being supplemented with a thorough diagnostic assessment. The validity of using the total score as a continuous measure of where one is placed on the autism spectrum is also uncertain.