Utility of the Psychoeducational Profile-3 for Assessing Children with Autism Spectrum Disorders

Background: Reliable and valid estimates of cognitive and language abilities of children with autism spectrum disorders are crucial as they have implications for diagnosis, treatment, and prognosis (Delmolino, 2006; Koegel, Koegel, & Smith, 1997; Lovaas, 1987). However, it can be difficult to assess this population due to difficulties with noncompliance and attention (Koegel et al., 1997; Villa et al., 2010). The Psychoeducational Profile-3 (PEP-3) was designed to assess children with autism spectrum disorders and has undergone two revisions since it was created in 1979. Empirical support for the revised version of the measure suggests it may be a valuable assessment tool. Few studies have examined the psychometric soundness of the newest version, the PEP-3. Additionally, the authors suggest that the new version may be useful for facilitating diagnoses of autism spectrum disorders (Schopler, Lansing, Reichler, & Marcus, 2005). In this regard, it would be useful to know how the PEP-3 relates to symptoms of autism spectrum disorders.  

Objectives:  This study explored the convergent and discriminant validity of the PEP-3 for assessing the development of cognitive and language skills in clinically referred children with autism spectrum disorders with various levels of functioning. This study also examined the relationship between PEP-3 scores and autism spectrum disorder symptomology.  

Methods: Data was collected from the files of 136 children with an autism spectrum disorder who had completed the PEP-3 between 2005 and 2010. Age equivalent scores were collected from the PEP-3, the Child Development Inventory (CDI), the Vineland Adaptive Behavior Scale 2^nd Edition (VABS-2), and the Autism Diagnostic Observation Schedule (ADOS).

Results: Spearman Rho correlations indicated that the PEP-3 was significantly correlated with the CDI, VABS-2, and MPR and that similar domains tended to be more strongly correlated than dissimilar domains. Importantly, the PEP-3 cognitive and language domains were strongly correlated with the cognitive and language domains on the CDI, VABS-2, and MPR. Mean difference scores indicated some significant differences between the measures assessing cognitive and language domains. Regarding the association with autism spectrum disorder symptomology, the results indicated that the PEP-3 was significantly negatively related to the ADOS total score, r(82) = -.40, p < .001. A between subjects MANOVA with diagnosis (Autistic Disorder, Asperger’s, and Pervasive Developmental Disorder- Not Otherwise Specified) as the independent variable indicated a significant multivariate effect of diagnosis on PEP-3 scores, F(18, 148) = 2.28, p < .05.  Post hoc comparisons indicated significant differences between some of the diagnostic groups on the PEP-3 subtests. 

Conclusions: The PEP-3 was found to provide a measure of cognitive and language development that is comparable to other commonly utilized standardized measures of development for children. The findings suggest that the PEP-3 cognitive and language domains may useful for determining diagnostic categories on the autism spectrum.