Date of Graduation

Spring 2020

Degree

Master of Science in Applied Behavior Analysis

Department

Psychology

Committee Chair

Jordan Belisle

Keywords

PEAK Relational Training System, PEAK Comprehensive Assessment, Construct Validity, Internal Consistency, Principle Component Analysis, Construct, Language, Cognition

Subject Categories

Applied Behavior Analysis | Experimental Analysis of Behavior

Abstract

The purpose of the current study was to conduct a pilot investigation of the internal construct validity of the four modules of the PEAK Comprehensive Assessment (PCA). The PCA has been developed through robust research over the past five years (Dixon et al. 2017) and is designed to evaluate language and cognitive skills of individuals with developmental disabilities, including neurodevelopmental disorders such as Autism Spectrum Disorder. Although the PCA contains four modules exemplifying four distinct learning processes (Direct Training, Generalization, Equivalence, and Relational Learning), these four processes may represent one singular learning construct, described loosely as “executive functioning” or “cognitive ability” in domains outside of applied behavior analysis. Within applied behavior analytic models, the common feature among these modules is that all are operant learning accounts. I evaluated the construct validity of the PCA using a principle component analysis in a sample of 55 participants with disabilities collected from multiple clinical sites throughout the United States. Results supported a one-factor model, suggesting that although scores in each module may differentially direct programming decisions, they are representative of a single underlying construct. Implications of these results are discussed.

Copyright

© Nicole Rae McDonald

Open Access

Share

COinS