Many developmental disorders, including autism spectrum disorder (ASD), exhibit language impairments. Expressive language sampling (ELS) procedures yield measures reflecting real-world performance, capture language impairments common to many developmental disorders and yield useful descriptors of individual difference and development change¹,². ELS has yet to be adequately validated for individuals with ASD, however. Leonard Abbeduto and his colleagues at the University of California, Davis plan to examine the utility of ELS as an outcome measure for ASD.

In ELS, samples of spoken language are collected in structured naturalistic interactions. The samples are then analyzed to derive outcome measures reflecting important dimensions of language skill and behavior. Abbeduto and his team plan to examine various ELS measures in 108 6- to 23-year-olds with ASD at three sites (University of California, Davis, University of Minnesota and University of Washington) to validate the utility of ELS as an outcome measure for ASD, and to assess whether ELS can be implemented appropriately across multiple sites.

Assessments of ELS measures will include basic psychometric properties, including their consistency over repeated administrations, their consistency across different contexts for sampling and their relationship to other accepted measures of language skill. The extent to which the psychometric properties of ELS procedures differ with variations in participant age, gender, autism symptom severity and intelligence quotient will also be examined. Abbeduto and his colleagues will also compare the psychometric properties of three ELS contexts that have been proposed as outcome measures for ASD: conversation, narration and the Autism Diagnostic Observation Schedule – Second Edition (ADOS-2).  They will also compare the psychometric properties of ELS in ASD relative to comparable data being collected on same-age individuals with fragile X syndrome and Down syndrome in a National Institutes of Health-funded project (R01HD074346). Test-retest consistency will be assessed at 4 weeks (+/-1 week) using alternate versions of ELS materials. Internal consistency will be assessed by computing alpha coefficients within and across sampling contexts. Norm-referenced standardized tests and informant reports will provide indicators of validity.

Successful completion of this research project will provide speech-language professionals and ASD families with procedures to appropriately evaluate and measure expressive language skills in individuals with ASD. The greater understanding gained from ELS procedures will allow families and their treatment teams to more appropriately address language deficits present in ASD individuals.