(PS3-B43) Factor Structure of the Autism Spectrum Adaptive Scale Questionnaire (ASAS-Q)

Intro: Autism spectrum disorder (ASD), now commonly referred to as autism spectrum condition(s) (ASC), is a neurodevelopmental disorder that is believed to affect approximately 1 in 44 individuals. As there is limited research on the etiological basis of ASC, diagnoses are given based on an individual’s behavior. Although, the process of receiving a ‘traditional diagnosis’ can be time consuming, expensive, and difficult to procure. Having an accurate self-report scale for diagnosing ASC in adults may be an efficient way to offer useful clinical information during assessments. However, the current self-report scales for detecting ASC have limitations, causing controversy over their effectiveness. A few examples of these limitations include the AQ-50 and SQ-40 not utilizing modern diagnostic content and containing validity issues, the RAADS-R and RAADS-14 not matching its subscales with the DSM and ICD’s diagnostic criterion, and the RBQ-2A-R using too small of a sample to conduct necessary analyses to ensure the scale’s accuracy. As an attempt to overcome the limitations of current self-report measures, we developed a new self-report measure for detecting ASC in adults called the Autism Spectrum Adaptive Scale Questionnaire (ASAS-Q). This study aims to assess the factor structure of the ASAS-Q. 

Method: 994 adult volunteers with and without history of ASC were recruited via Research Match (391 with autism, 594 without autism). To determine the factor structure of the ASAS-Q, we first ran an Exploratory Factor Analysis (EFA), which suggested three factors. Based on these initial findings, we then reran the analysis, limiting the number of factors to 3. Using a loading value cutoff of above .4, items that either did not load onto any factor or cross loaded onto more than one factor were deleted. The final analysis converged in six iterations. Finally, we ran a reliability analysis to test the internal consistency. 

Results: In its current experimental form, ASAS-Q has 38 items. Bartlett’s tests of sphericity, which tests overall significance of the correlations within the correlation matrix, was significant, (χ² (780) = 24860.18, p = .000). Likewise, the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy indicated a high relationship among variables (KMO = .97). These tests indicated that a factor analytic model is acceptable for this dataset. Upon evaluating the items loading on each factor, we determined that the three factors include social-emotional reciprocity, communication, and self-regulation. Cronbach’s Alpha indicated high internal consistency among each of the three subscales: social-emotional reciprocity (α = .87), communication (α = .90), and self-regulation (α = .94), Further, Cronbach's Alpha indicated that overall scale has excellent internal consistency, (α = .97).

Conclusions: These findings indicate that the ASAS-Q may serve as a useful new self-report scale for diagnosing ASC in adults. The factors on the ASAS-Q are similar to the main components of ASC outlined in DSM-V, with an emphasis on self-regulatory behaviors and adaptive items. Future studies will determine the reliability and validity of this new self-report measure. Clinical implications will be discussed.