Doctor of Philosophy (PhD)



Document Type



The early diagnosis of autism spectrum disorder (ASD) is tied to receiving the earliest possible intervention services that can influence a child’s development (Filipek et al., 1999). There are different measurement tools utilized to aid in the diagnostic assessment process, including broadband behavioral rating scales. The Childhood Behavior Checklist (CBCL; Achenbach & Rescorla, 2001) is a parent/caregiver completed questionnaire about one’s child (typically or atypically developing). The CBCL is one of the most utilized broadband measures at the initial stages of a diagnostic assessment or screening to aid in understanding the symptom presentation (Hyman et al., 2020). Currently, there is no established and widely accepted ASD subscale for the version that is used with children and adolescents from six to eighteen years. Data was used from the National Database for Autism Research (NDAR) for the ASD sample and an existing database through the Louisiana State University Psychological Services Center was utilized for the internalizing disorder (INT), externalizing disorder (EXT), and no diagnosis (NODX) comparison groups. Divergent from previous research (Camodeca, 2018), both the 9-item ASD subscale (Ooi et al., 2011) and 10-item ASD subscale (So et al., 2013) discriminated ASD from the three comparison groups with optimized sensitivity and specificity. Additionally, the items of the CBCL/6-18 were examined to create a new ASD subscale. Contrary to hypotheses, the results yielded a 6-item ASD subscale with poor internal consistency compared to the 9-item and 10-item ASD subscales. Nevertheless, the 6-item ASD subscale was better at discriminating ASD from the INT group and was only better than the 10-item ASD subscale for discrimination with the NODX group; the subscale did not differentiate children with ASD from those in the EXT group. This research provides preliminary data supporting an additional ASD subscale in addition to the 9-item and 10-item ASD subscales to classify individuals with ASD from those without a diagnosis. Although the 6-item ASD subscale demonstrated some promising results (e.g., optimizing sensitivity and specificity better than the other two ASD subscales), more research needs to be conducted to elucidate which CBCL items most accurately can be used to differentiate symptoms of ASD.



Committee Chair

Matson, Johnny

Available for download on Wednesday, June 19, 2024