Adaptation of Autism Spectrum Quotient and Camouflaging Autistic Traits-Questionnaire into Estonian: exploring possible gender effects

Abstract

Diagnosing autism in adults is difficult. Current paper tries to fill a gap in self-report screening instruments by adapting into Estonian two questionnaires: Autism Spectrum Quotient (AQ, Baron-Cohen et al., 2001) and Camouflaging Autistic Traits-Questionnaire (CAT-Q, Hull et al., 2019). Sample (N = 498, f = 409, m = 89) consisted of autistic (n = 42), self-suspected autistic (n = 105), autistic close relative (n = 95) and non-autistic (n = 256) participants. Using exploratory factor analysis, a 5-factor solution was found in AQ-24, total variance explained was 44.2% with Cronbach’s alpha of .86. CAT-Q-25, with five new items, retained the original three-factor solution, with total variance explained 56.1% (Cronbach’s alpha .95). As expected, correlations between autism questionnaires were higher than questionnaires measuring related constructs. Although correlation with anxiety measures i.e., Social Phobia and Anxiety Inventory (Turner et al., 1989) were significantly high as well. Group and sex interaction effect onto CAT-Q-25 total and subscale scores when controlling for either general or social anxiety revealed that a statistically significant interaction effect appeared only when controlling for social anxiety. It is worth considering applying item response theory approach to find the best set of questions. It is also important to explore both sex effects in autism and the effects of anxiety to autism related traits, and the possible ways how anxiety might affect autistic individuals.