Using the cut-off score of 29 has a false positive rate of 1%, however will fail to identify 14.4% of people who actually do meet the diagnostic criteria for DSM-5 defined Autism Spectrum Disorder. Higher scores indicate responses that are more consistent with autism traits.Ī total score of 29 or more is indicative of clinically significant autism traits (Broadbent, et al., 2013). Items are summed to obtain an overall total score and scores for each of the five subscales. Men with autism have an average score of 35.1 (SD = 6.9), while the mean score for women with autism is 3.81 (SD = 4.4). This analysis suggest the optimal cutoff score is 29+.ĭata for gender related norms are provided for the sample of adults with autism and the community sample (Baron-Cohen et al., 2001). Subsequent analysis by Broadbent and colleagues (2013) found that a cutoff score of 29+ had a 14.4% false negative rate, and a less than 1% false positive rate. Internal consistency was moderate to high.Ī cutoff score of 32+ was suggested in the original Baron-Cohen and colleagues (2001) article, and was found to distinguish the Autism Sample from the control Community Sample. Psychometric analysis indicated adequate inter-rater and test-retest reliability. They administered the scale to adults with DSM-IV classified Asperger syndrome (AS) or high functioning autism (HFA) (Autism sample, N = 58) and to a control sample of randomly selected individuals from the community (N = 174). The AQ was developed by Baron-Cohen and colleagues (2001).

Autism spectrum quotient aq test