Examination of the reliability and factor structure of the Autism Spectrum Quotient (AQ) in a non-clinical sample

Abstract

A self-report screening measure for high functioning autism spectrum disorders is needed for diagnostic screening and research purposes. The Autism Spectrum Quotient (AQ) has been developed for these reasons, although a comprehensive assessment of the psychometric properties of the AQ has not been completed. The purpose of the current study was to assess the distribution, internal consistency, and factor structure of the AQ in a non-clinical sample (n = 1005). The current findings demonstrate the normal distribution of autistic traits and support a three-factor structure of the AQ. Additionally, a three-factor version of the AQ yielded somewhat improved internal consistency. Implications of these findings and suggestions for further development of the AQ as a measure of the autism spectrum are offered.

The use of the Autism-spectrum Quotient in differentiating high-functioning adults with autism, adults with schizophrenia and a neurotypical adult control group

“…In this respect, it is important to consider studies in which the factor analytic structure of the AQ has been examined. Two studies suggested three underlying factors: Social skill, Details/patterns and Communication/mindreading (Austin, 2005;Hurst, Mitchell, Kimbrel, Kwapil, & Nelson-Gray, 2007) and one study suggested a broad Social interaction subscale and a second subscale Attention to detail (Hoekstra et al, 2008). In the three existing factor analytic studies, the subscale Imagination was never established as a separate subscale.…”

Modeling the impact of social problem-solving deficits on depressive vulnerability in the broader autism phenotype

“…While the psychometric properties of the overall scores produced by the AQ have consistently been shown to be reasonably good (Baron-Cohen et al, 2001;Hoekstra, Bartels, Cath, & Boomsma, 2008;Kurita and Koyoma, 2006;Wakabayahi, Baron-Cohen, Wheelwright, and Tojo, 2006), a number of studies have raised concerns regarding the internal consistencies of the five domains detailed in the initial model by Baron-Cohen et al Based on the results of subsequent factor analyses, it has been suggested that, alternatively, the AQ could be more accurately defined by a four- (Stewart & Austin, 2009), three- (Hurst, Mitchell, Kimbrel, Kwapil, & Nelson-Gray, 2007), or two-domain model (Hoekstra et al, 2008). This study utilized the Hoekstra et al (2008) two-domain model comprised of the original 'Attention to Details' factor, and a higher-order factor of 'Social Interaction', which encompasses the Social Skills, Communication, Attention Switching, and Imagination domains from the original Baron-Cohen et al (2001) model.…”

Broad Autism Phenotypic Traits and the Relationship to Sexual Orientation and Sexual Behavior

“…The traits associated with the BAP have been described as following a roughly normal distribution within the TD population (Best et al 2008;Hoekstra et al 2007;Hurst et al 2007;Pisula et al 2015). Young adults with higher levels of the BAP are similar to those with ASD in that they have been shown to struggle more with loneliness, creating and maintaining friendships (Jobe and White 2007), depression, anxiety, and being victims of bullying (Kunihira et al 2006).…”

“…It has also been used to assess these traits in TD populations (e.g., Hurst et al 2007;Jobe and White 2007;Kunihira et al 2006). Example items on the AQ include: "I find it difficult to work out people's intentions" and "I am fascinated by numbers."…”