Dimensionality, hierarchical structure, age generalizability, and criterion validity of the GAIN's Behavioral Complexity Scale

Kendon J. Conrad*, Karen M. Conrad, Jessica Mazza, Barth B. Riley, Rod Funk, Mark A. Stein, Michael L. Dennis

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

19 Scopus citations

Abstract

This study used Rasch measurement model criteria and traditional psychometric strategies to examine key psychometric properties of the Behavioral Complexity Scale (BCS), a widely used measure of externalizing disorders that focuses on attention deficit, hyperactivity, and conduct disorders. With a sample of 7,435 persons being screened for substance use disorders, the BCS was found to (a) be unidimensional, (b) have a hierarchical severity structure, (c) be generalizable to both youths and adults, and (d) meet hypothesized correlations with criterion variables. The BCS performed well as a unidimensional measure. The Rasch severity hierarchy of attention deficit to hyperactivity to conduct disorders provided a perspective that suggested that a dimensional measure could be used as an alternative and, in some ways, as an improvement to categorical diagnosis and common dimensional approaches. The finding of 3 low-severity conduct disorder items also supported a revision of categorical criteria, especially in substance use disorders.

Original languageEnglish (US)
Pages (from-to)913-924
Number of pages12
JournalPsychological assessment
Volume24
Issue number4
DOIs
StatePublished - Dec 2012

Keywords

  • Age generalizability
  • Attention-deficit/hyperactivity disorders
  • Conduct disorders
  • Criterion validity
  • Externalizing disorders
  • Global Appraisal of Individual Needs
  • Hierarchical structure
  • Rasch model

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Clinical Psychology

Fingerprint

Dive into the research topics of 'Dimensionality, hierarchical structure, age generalizability, and criterion validity of the GAIN's Behavioral Complexity Scale'. Together they form a unique fingerprint.

Cite this