- Jennifer L Stevenson: Department of Psychology, Ursinus College, 601 East Main Street, Collegeville, PA, 19426, USA. jstevenson@ursinus.edu. ORCID
- Kari R Hart: Department of Mathematics and Computer, The Hotchkiss School, Lakeville, CT, USA.
The current study systematically investigated the effects of scoring and categorization methods on the psychometric properties of the Autism-Spectrum Quotient. Four hundred and three college students completed the Autism-Spectrum Quotient at least once. Total scores on the Autism-Spectrum Quotient had acceptable internal consistency and test-retest reliability using a binary or Likert scoring method, but the results were more varied for the subscales. Overall, Likert scoring yielded higher internal consistency and test-retest reliability than binary scoring. However, agreement in categorization of low and high autistic traits was poor over time (except for a median split on Likert scores). The results support using Likert scoring and administering the Autism-Spectrum Quotient at the same time as the task of interest with neurotypical participants.