Evaluation and Norms of the German Version of the Aberrant Behavior Checklist for Children and Adolescents

Background:

The aberrant behavior checklist (ABC) was originally developed to measure therapy success in people with mental retardation [1]. Today it is a well-established and worldwide used measure in effectiveness research especially for medication studies. Its validity and reliability has been replicated in different studies which could show its suitability for different age groups and disorders e.g. autism spectrum disorders [e.g. 2,3,4]. It is so widely used that most reviews trying to aggregate autism research can only base their meta-analyses on this questionnaire. Nevertheless the validation is not sufficiently done for every translation. The German version of the ABC has only been validated for adults with mental retardation so far [5] but not for children and adolescents with autism spectrum disorders where it is also widely used and necessary to ensure international comparability of therapy outcomes.

Objectives:

The aim of the present study is to (1) validate the German ABC in children and adolescents with autism spectrum disorder including the factor structure of the subscales, (2) examine its reliability and (3) develop norms.

Methods:

1003 parents of children aged between 4 and 17 years (M: 11.11, SD: 4.09, ♂: 569) were recruited at the Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy of the University Hospital in Frankfurt and filled in the questionnaires. Four different samples were thereby assessed: autism spectrum disorder (n = 225), healthy controls (n = 470), mentally disabled children (n = 128) and a clinical subsample with other mental disorders (n = 180). We examined the validity and reliability of the ABC under consideration of age and gender and developed German norms for this age group. A confirmatory factor analysis was used to confirm the factor structure.

Results:

Item selectivity and Cronbach's alpha were excellent, with the alpha if one of the items is deleted ranging between .963 and .969. The original factor structure of the ABC could be replicated. The retest reliability showed a value of r = 0.824 for all items and ranged from 0.796 to 0.864 in case of the five different subscales. The interrater reliability between mother and father ratings was ICC = 0.791 (CI: 0.44-0.92).

Conclusions:

Overall the psychometric properties of the German ABC for children and adolescents could be considered as good to excellent and the factor structure could be replicated. Considering this good results the German ABC can be recommended as a measure of therapy success. This is an important contribution to the international comparability of effectiveness research.