Date of Award


Document Type


Degree Name

Doctor of Education (Ded)


Educational and School Psychology

First Advisor

Joseph F. Kovaleski, D.Ed.

Second Advisor

Mark J. Staszkiewicz, Ed.D.

Third Advisor

Timothy J. Runge, Ph.D.

Fourth Advisor

Lisa C. Newell, Ph.D.


The number of young children with autism is ever-increasing, and school psychologists are more frequently required to identify these students. Valid screening tools are needed in order to focus school psychologists' time on those students in need of intensive evaluations in Early Intervention programs. The purpose of this study was to examine the utility of the Modified Checklist for Autism in Toddlers (M-CHAT; Robins, Fein, & Barton, 1999), an autism screening tool, for preschool-age students who have been identified as needing special education services, using both traditional scoring methods and the newly introduced Best7 scoring procedure. This study also examined which items on the M-CHAT were most associated with students' receiving an educational classification of autism, as discriminated from students with other developmental delays. In addition, this study attempted to determine whether a two-factor solution (i.e., social communication deficits and unusual behaviors) or a different model best described parent ratings on the M-CHAT. Overall, findings indicated that the M-CHAT correctly classified 62% of students in the current sample, while incorrectly classifying 38%. Sensitivity was .64 for the current sample while the specificity was .60. The positive predictive value for the M-CHAT was .61 and the negative predictive value was .64. This held true regardless of the scoring method. No differences were found with the use of traditional scoring vs. Best7 scoring methods. It was found that the failure of Question 13 (Does your child imitate you?), and Question 2 (Does your child take an interest in other children?) were the best predictors of students receiving an educational classification of autism. Results of principal components analyses indicated that the M-CHAT is composed of two components, though the emphasis of these components was different for students with and without autism. While the factors were sufficiently different to prevent combining the groups for a single factor analysis, neither group demonstrated a clear delineation between social communication and unusual behaviors. Thus, the M-CHAT does not appear to be measuring autism as per the newly recommended two-factor model of social communication and unusual behaviors.