A screener, not a diagnosis

Autism test for adults

This is the RAADS-14, a short screening scale for autistic traits in adults. Fourteen statements, about three minutes, no sign-up needed. Your answers stay in your browser.

Please answer the questions according to what is true for you. Choose the alternative that best describes you now and when you were young, and check only one column per statement.

Question 1 of 14

It is difficult for me to understand how other people are feeling when we are talking

Your answers stay in your browser and are never sent anywhere. We do not see them and we do not store them.

What this scale is

The RAADS-14 Screen is an abridged form of the Ritvo Autism and Asperger Diagnostic Scale-Revised. The full version runs to 80 items. The short one keeps the fourteen that separated the groups best, and Eriksson, Andersen and Bejerot published it in Molecular Autism back in 2013.

It covers three subscales: reading other people, difficulty in social interaction, and sensory reactivity. That trio is what distinguished the autistic profile from other psychiatric conditions in the original study.

The wording here is reproduced exactly as published. A reworded item stops being the item the psychometrics were measured on, so we leave it alone even where the phrasing is not how we would write it.

How the score works

Each statement is worth nought to three points, and the option covering both now and childhood carries the most. Item six about small talk is scored in reverse, because it describes an ability rather than a difficulty. The scale runs to 42 points in total.

The original threshold is 14. Different language versions of this scale carry different thresholds, because each one is calibrated on the sample it was validated against, and borrowing another version's number would be wrong.

How accurate the result is

In the original study a cut-off of 14 reached a sensitivity of 97%. Specificity depended entirely on the comparison group: 95% against people with no psychiatric diagnosis, 64% against other psychiatric conditions.

One limitation matters especially to our readers. Against a group with ADHD, specificity fell to 46%. Put plainly, a great many people with ADHD score high on this scale without being autistic, and the scale on its own cannot tell the two apart.

So a high score says the statements describe something you recognise. It does not say you are autistic. Working out what actually sits underneath is a job for a clinician, who has instruments longer than fourteen items.

What to do with the result

Adult autism is assessed by psychiatrists and clinical psychologists, ideally ones who work with adults: criteria written around children map poorly onto a grown person.

Bring more than the total. The button under your result copies every answer as text, and it helps to add examples from your life and, if you can, what relatives remember about your childhood.

Bear in mind that in adults, and in women particularly, the picture is often masked by years of adapting to the people around them. A questionnaire cannot see that. A clinician will ask.

Common questions

Can an online test diagnose autism?
No. A screening scale answers one question: whether seeing a specialist is worth doing. A diagnosis follows a long assessment that looks at development from early childhood, gathers history from people who knew you then, and rules out other explanations.
Why do different versions of this scale use different thresholds?
Because a threshold belongs to the sample it came from. The original Swedish work produced 14, while the Russian validation, run on its own sample and its own translation, produced 22 instead. Each number is valid only for the version it came from.
I have ADHD and scored high. Does that mean I am autistic as well?
Not on its own. This scale separates autism from ADHD poorly, with a specificity of 46% against an ADHD group in the original study. The combination is genuinely common, but only an in-person assessment can confirm or rule it out.
My score is low but I am struggling. What now?
See a clinician anyway. Fourteen items do not cover everything, and in adults with years of masking behind them the scale reads low. A low score rules nothing out.
Are my answers stored?
They never reach our servers. Scoring happens in your browser, the draft sits in your own browser storage, and a button wipes it.
Is the test free, and do I need an account?
It is free, no account is needed, and no email is asked for.
How does the RAADS-14 differ from the RAADS-R and the AQ?
The RAADS-R has 80 items and is used clinically. The RAADS-14 is its quick screening form. We deliberately do not publish the AQ: its rights holder forbids commercial use without a paid licence.
Why is one statement scored backwards?
Item six, about making small talk, describes a retained ability. Agreeing with it scores nought and disagreeing scores three. Reversed items exist so that nobody can answer the whole form down a single column on autopilot.

What this test is not

This page is informational. It is not a medical device and not a diagnostic procedure. We do not diagnose, treat or prevent any condition, and we do not replace medical advice, psychotherapy or medication decisions. If you are in distress right now and having thoughts of harming yourself, please reach urgent professional help.

Sources

Scale copyright

RAADS-14 Screen is an abridged version of the Ritvo Autism and Asperger Diagnostic Scale-Revised (RAADS-R). Eriksson JM, Andersen MJ, Bejerot S. RAADS-14 Screen: validity of a screening tool for Autism Spectrum Disorder in an adult psychiatric population. Molecular Autism 2013;4:49.