Cluster analysis of children and adolescents with brain damage and learning disabilities using neuropsychological,...

The purpose of the study was to employ psychoeducational, neuropsychological, and sociobehavioral (Conners Rating Scale) variables in determining if definable subtypes exist within a diverse population of subjects with learning disabilities (LD) and documented brain damage. The sample of 95 subjects (27% female and 73% male) had been referred for neuropsychological assessment at a large, Midwestern medical center. Mean age was 10.6 years. Brain damage (BD) was documented for 45% of the sample. The first cluster analysis employed neuropsychological, psychoeducational, and sociobehavioral data and revealed four interpretable clusters. A second cluster analysis excluded sociobehavioral data and yielded two interpretable clusters. In neither analysis did a cluster consist exclusively of BD or LD subjects. Results were interpreted as supporting the importance of the sociobehavioral component in LD subtyping, as well as supporting the contention that parallels may exist in cerebral function and/or structure between the LD and BD classifications.