Examining differential responses of youth with and without autism on a measure of everyday activity performance

Abstract

Purpose This study further investigated items with differential item function (DIF) in the Social/Cognitive domain of a measure of everyday activity performance, the Pediatric Evaluation of Disability Inventory-Computer Adapted Test version for Autism “PEDI-CAT (ASD),” to understand possible sources of response variation in a heterogeneous sample of youth with autism compared to the national standardization sample. Methods Cross-sectional design. A convenience sample of parents who identified they had a child between 3 and 21 years (M = 11.9 years, SD = 4.67 years) with autism (n = 365) completed an online survey that included the PEDI-CAT (ASD) and descriptive measures. For 28 items previously identified as having DIF, the PEDI-CAT (ASD) expected item score curves for the autism sample were compared to the original PEDI-CAT standardization sample. The weighted area between expected score curves (wABC) was also calculated; values >0.24 indicate significant DIF. Results All items had wABC that exceeded the criterion. Compared with peers without disabilities at the same ability level, 11 items were significantly more difficult for the youth with autism and 16 items were significantly easier. One item demonstrated non-uniform DIF. Conclusion Differential responses could indicate that: (1) children with autism have a different developmental pattern of skill acquisition for everyday activities in the Social/Cognitive domain, or (2) parents of children with autism utilize a unique appraisal process when assessing their children’s functional performance of everyday activities. Further research is required to better understand the factors leading to differential responses on the targeted items. The study illustrates the value of in-depth analysis of DIF to gain insight into the impact of a clinical condition on functional performance.

Validity, reliability, and usability of the Pediatric Evaluation of Disability Inventory‐Computer Adaptive Test for autism spectrum disorders

“…11 Our research also indicated that performance on items of the Social/Cognitive domain is different for young people with autism and for other populations; 11,12 therefore, the PEDI-CAT (ASD) software incorporates an equating technique to ensure that the criterion-based domain scaled score is a valid measure of a young person's performance, while remaining comparable to scaled scores for this domain obtained in other populations. 11,12 Additional research is needed to establish whether or not the PEDI-CAT (ASD) yields stable scores when change is not expected to occur. Evidence regarding the relationship between PEDI-CAT (ASD) scores and scores from other criterion-standard assessments is also needed to determine how assessment instruments are related to each other and which may be better suited to answering particular research questions.…”

“…This study also provided insight into the frame of reference that parents may use when reporting on their child's functional performance. 12 The PEDI-CAT (ASD) software 'speedy' option selects the next item to assess based on the relative difficulty of the previously assessed item and the parent response to that item. However, it appeared that parents often evaluated the appropriateness of each PEDI-CAT (ASD) item based on their child's age, not function.…”

Adaptive Skills in FXS: A Review of the Literature and Evaluation of the PEDI-Computer Adaptive Test (PEDI-CAT) to Measure Adaptive Skills

“…Adaptations to the PEDI-CAT may also provide an opportunity to address the issue of floor effects evidenced in both the PEDI-CAT and VABS-3 for this population. As the PEDI-CAT publishers have previously made adaptations to items tailored to specific populations, such as ASD [ 25 , 38 ], this could be considered for FXS as well; however, that would require additional validation.…”

“…The PEDI-CAT has been translated and cross-culturally adapted into several versions, including Spanish, Dutch and Brazilian-Portuguese [ 19 , 20 ]. It has been used to assess motor functioning in children with learning disorders and children with spina bifida, as well as all domains of functioning in autism spectrum disorder (ASD), Down syndrome, cerebral palsy, Angelman syndrome, infants and toddlers receiving EI services and many others [ 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 ]. Several studies show strong psychometric evidence for the PEDI-CAT.…”

“…In 2015, the PEDI-CAT was adapted for the ASD population by including additional question items and directions [ 38 ]. Kramer and colleagues [ 25 ] examined the test-retest and concurrent validity of the new version (PEDI-CAT-ASD), a speedy version in 39 children and adolescents with autism spectrum disorder (ASD). Results demonstrated a reliability for all PEDI-CAT-ASD domain scores (ICC > 0.86) and a high correlation to VABS-2 domain raw scores.…”