Abstract
Measurement invariance, or the degree to which an instrument measures constructs consistently across subgroups, is critical for appropriate interpretations of measures. Given sex differences in the phenotypic and clinical presentation of autism spectrum disorder (ASD), it is particularly important to examine measurement invariance in autism instruments to ensure that ASD measures are not biased toward the more common male ASD phenotype. This study represents an important preliminary investigation evaluating the measurement equivalence of the Autism Impact Measure (AIM) across children and adolescents with ASD. The results indicated that the AIM demonstrated measurement invariance at the configural, metric, and scalar levels across sex in all five domains, including Repetitive Behavior, Communication, Atypical Behavior, Social Reciprocity, and Peer Interaction. These results suggest that ASD core symptoms assessed by the AIM were similar among male and female groups. In addition, the latent means for all five factors were not statistically significantly different across sex groups, revealing no systematic differences on any of the AIM subscales for males and females. Overall, this study showed that the AIM detects core ASD symptoms across all five areas equivalently in males and females and is not biased toward males with ASD.