The Autism Spectrum Quotient (AQ) is a self-administered questionnaire used to measure autistic traits in adults (age 16+). An IQ of at least low average (IQ >=80) is required to take the test adequately.
Basic information | |
|---|---|
| Statements: | 50 |
| Duration: | 5–10 minutes |
| Type: | screening tool |
| Authors: | Simon Baron-Cohen |
| Publishing year: | 2001 |
| Seminal Paper: | The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism, Males and Females, Scientists and Mathematicians (Baron-Cohen et al., 2001) |
Take the test here:
The AQ
Who the test is designed for
- Adults(age 16+) of average or higher intelligence.[1]The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism, Males and Females, Scientists and Mathematicians (Baron-Cohen et al., 2001)
Versions & translations
- The AQ is available in most languages
- Non-adults versions are also available:
- Adolescents (ages 12–15)
- Children (ages 4–11)
Taking the test
The AQ consists of 50 statements, giving you 4 choices for each statement:
- Definitely agree
- Slightly agree
- Slightly disagree
- Definitely disagree
Note: it makes no difference to your score whether you choose slightly or definitely, so treat the statements as a binary choice agree anddisagree.
If you decide to take the test, please consider the information under the sections titled Outdated and Updated below.
Scoring
- Scoring range: 0–50
- Threshold score: 26↑
- Scores 26or greater indicate you might be autistic
- Lower scores mean you likely are not
- 79.3% of autistic people score 32 or higher[2]The Autism-Spectrum Quotient (AQ)—Adolescent Version (Baron-Cohen et al., 2006)
- Most non-autistic males score 17 onaverage
- Most non-autistic females score 15 on average
You can take the test using two methods of scoring:
- Automated-scoring
- Self-scoring, if you want documentation of your answers
Mean scores
In the table below—based on Baron-Cohen’s 2001 paper—you can see the mean AQ scores of autistic people (designated Asperger’s syndrome and high-functioning autism) and controls.[3]The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism, Males and Females, Scientists and Mathematicians (Baron-Cohen et al., 2001)
Mean AQ scores
| Total AQ score | |
|---|---|
| AS/HFA total | 35.8 |
| AS/HFA males | 35.1 |
| AS/HFA females | 38.1 |
| Controls total | 16.4 |
| Controls males | 17.8 |
| Controls females | 15.4 |
And in the table below, you can see the mean AQ scores and subscale scores of autistic people (n: 58), controls (n: 174), students from the University of Cambridge (n: 840), and UK Mathematics Olympiad winners (n: 16).[4]The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism, Males and Females, Scientists and Mathematicians (Baron-Cohen et al., 2001)
Mean AQ scores & subscale scores
| Total AQ score | Communication | Social | Imagination | Local details | Attention switching | |
|---|---|---|---|---|---|---|
| AS/HFA total | 35.8 | 7.2 | 7.5 | 6.4 | 6.7 | 8.0 |
| AS/HFA males | 35.1 | 7.2 | 7.4 | 6.2 | 6.6 | 7.7 |
| AS/HFA females | 38.1 | 7.3 | 7.9 | 7.0 | 6.9 | 8.9 |
| Controls total | 16.4 | 2.4 | 2.6 | 2.3 | 5.3 | 3.9 |
| Controls males | 17.8 | 2.8 | 2.8 | 2.7 | 5.2 | 4.3 |
| Controls females | 15.4 | 2.1 | 2.3 | 1.9 | 5.4 | 3.6 |
| Students total | 17.6 | 2.9 | 2.3 | 2.5 | 5.3 | 4.5 |
| Students males | 18.6 | 3.2 | 2.6 | 2.9 | 5.3 | 4.7 |
| Students females | 16.4 | 2.7 | 2.0 | 2.0 | 5.4 | 4.3 |
| Olympiad total | 24.5 | 3.0 | 5.1 | 4.9 | 6.6 | 4.9 |
And finally, in the table below, you can see the mean AQ scores and subscale scores of the Cambridge students, from various disciplines (highest scores in bold).[5]The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism, Males and Females, Scientists and Mathematicians (Baron-Cohen et al., 2001)
Mean AQ scores of student scientists
| Total AQ score | Communication | Social | Imagination | Local details | Attention switching | |
|---|---|---|---|---|---|---|
| Biological sciences | 14.9 | 2.7 | 1.5 | 1.7 | 4.7 | 4.2 |
| Computer science | 21.1 | 3.4 | 3.7 | 3.4 | 5.7 | 4.8 |
| Engineering | 17.9 | 2.9 | 2.3 | 3.0 | 5.4 | 4.3 |
| Mathematics | 21.5 | 3.8 | 3.6 | 3.3 | 5.8 | 5.1 |
| Medicine | 15.4 | 2.5 | 1.4 | 2.0 | 5.2 | 4.2 |
| Physical sciences | 19.6 | 3.0 | 3.4 | 3.1 | 5.4 | 4.6 |
| Nonspecific science | 18.5 | 3.0 | 2.6 | 2.6 | 5.6 | 4.7 |
Validity
How reliable, accurate, valid, and up to date is the test?
The AQ correctly scores autistics (both male and female) higher than neurotypicals.[6]Negatively phrased items of the Autism Spectrum Quotient function differently for groups with and without autism (van Rentergem, Lever, & Geurts, 2019)
- Test–retest reliability (consistency of scores when a person retakes the test) was found to be good.
- Inter–rater reliability (consistency of scores when two different clinicians provide the test to the same person) was found to be good.
Research shows that the AQ is a quick tool to identify where a personis situated on the continuum from autism to neurotypicality.[7]The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism, Males and Females, Scientists and Mathematicians (Baron-Cohen et al., 2001)
Outdated
In 2017, the following items were proven unrepresentative measures of autistic traits, thus needing revision.[8]Is the Autism-Spectrum Quotient a Valid Measure of Traits Associated with the Autism Spectrum? A Rasch Validation in Adults with and Without Autism Spectrum Disorders (Lundqvist & Lindner, 2017)
Researchers thought that autistics would agree with the following questions, but we don’t necessarily:
- 9. I am fascinated by dates.
- 21. I don’t particularly enjoy reading fiction.
And researchers thought that autistics would disagree with the following questions, but again, we don’t necessarily:
- 29. I am not very good at remembering phone numbers.
- 30. I don’t usually notice small changes in a situation, or a person’s appearance.
- 49. I am not very good at remembering people’s date of birth.
Updated
If you take the test, please interpret the outdated questions as follows:
- 9. I am interested in the patterns or correlations of events.
- 21. I enjoy reading informative literature, but I sometimes like reading fiction as well (and might use it to learn social skills).
- 29. I am not very good at remembering information that is important to me.
- 30. I usually notice small changes in a situation or person’s appearance.
- 49. I am not very good at remembering information that is important to me.
NB: This is our personal supplementation; you won’t find references to this.
Discussion
- Not knowing there are only two options (‘agree’ or ‘disagree’), dramatically increases the time it takes to complete this test.
- Outdatedquestions may reduce the accuracy of the test.
- Some questions are too general for an ‘agree’ or ‘disagree’ answer. For instance, item 21: I don’t particularly enjoy reading fiction.
I very much enjoy particular types of fiction, but did not read fiction when I was younger. I still enjoy reading factual information. (Kendall)
- Some questions are based on outdated assumptions about autism. For example, item8: When I’m reading a story, I can easily imagine what the characters might look like. Older research papers claim we lack imagination, but this is affective alexithymia, not autism. Furthermore, affective alexithymia is not at all common in autism.
To read more on alexithymia and aspects of this construct that are commonly mistaken for autism, have a look at:
Autism Spectrum Quotient
For each statement below, choose one response that best describes how strongly that statement applies to you:
