The assessment of individual differences between young children with a pervasive developmental disorder by means of behaviour scales which are derived from direct observation

J Child Psychol Psychiatry. 1994 Feb;35(2):333-49. doi: 10.1111/j.1469-7610.1994.tb01166.x.

Abstract

Data obtained by direct observation of 112 3-6-year-old normal children and 31 children with a pervasive developmental disorder aged 3-6 were used to construct behaviour scales by means of simultaneous component analysis. This is a technique for finding behaviour clusters (components) common to different groups by weighting the variables such that the resulting components maximize variance accounted for when summed over the groups (Milsap & Meredith, 1988, Psychometrika, 53, 123-134; Berge & Kiers, 1990, Nederlands Tijdschrift voor de Psychologie, 45, 221-226). An evaluation of the component structure that was found is given for both groups. Results show uncorrelated components for the normal group, while some of the same components are intercorrelated in the clinical group. Scales were constructed which are shown to have discriminative value with respect to subgroups within the group of patients.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Autistic Disorder / classification
  • Autistic Disorder / diagnosis
  • Autistic Disorder / psychology
  • Child
  • Child Behavior Disorders / classification
  • Child Behavior Disorders / diagnosis
  • Child Behavior Disorders / psychology
  • Child Development Disorders, Pervasive / classification
  • Child Development Disorders, Pervasive / diagnosis*
  • Child Development Disorders, Pervasive / psychology
  • Child, Preschool
  • Female
  • Humans
  • Individuality*
  • Intelligence
  • Male
  • Psychiatric Status Rating Scales / statistics & numerical data
  • Psychometrics
  • Reproducibility of Results
  • Stereotyped Behavior