Is the ASRS ADHD Test Accurate? What the Science Actually Says

If you have taken an ADHD screening quiz anywhere serious — including the one on this site — it was probably the ASRS v1.1: the Adult ADHD Self-Report Scale. Before you trust any test with a question about your own brain, it is fair to ask how good it actually is. Here is the honest answer.

Where the ASRS comes from

The ASRS was developed by the World Health Organization with a workgroup of ADHD researchers from institutions including Harvard and NYU, as part of the WHO World Mental Health Survey Initiative. The full checklist has 18 questions mirroring the 18 diagnostic symptoms of ADHD. Its famous short form — Part A — is just six questions, statistically selected because they predicted an ADHD diagnosis better than any other combination.

That is worth pausing on: the six questions were not chosen because they sound like ADHD. Several barely mention attention at all (finishing the fiddly ends of projects, delaying big-thinking tasks, feeling driven by a motor). They were chosen because, mathematically, they discriminate — they separate adults with ADHD from adults without it better than the obvious questions do.

The accuracy numbers, honestly

In the original validation and later general-population studies, the Part A screener at the standard threshold (4+ positive answers) showed sensitivity around 65–70% and specificity around 94–99% — meaning it misses some true cases, but a positive screen is meaningful because false alarms are relatively rare in the general population. Screening studies in clinical settings show higher sensitivity with lower specificity: among people already seeking help, more non-ADHD conditions (anxiety, depression, sleep deprivation) can push scores up.

Translated out of statistics:

• A positive screen is a strong signal worth acting on — it says "this pattern looks enough like ADHD that a professional should look."
• A negative screen makes ADHD less likely but does not rule it out — particularly for people (often women) whose symptoms are predominantly inattentive and internal, a presentation every screener catches less reliably.
• Neither result is a diagnosis. Other conditions genuinely mimic ADHD — see ADHD vs anxiety — and only a clinical interview covering childhood history and current impairment can separate them.

How to take it so the result means something

Answer for the last six months, not your worst week ever; answer as you actually are, not as you are when deploying maximum coping effort; and do it in one sitting without overthinking — first instincts on frequency questions are usually the most accurate. Then keep the result: our version runs entirely in your browser, sends nothing to any server, and can be printed to bring to an appointment, which is exactly what screeners are for. If you are in Sri Lanka, the practical next steps are in our diagnosis guide.

A six-question test cannot know you. It can tell you — with respectable, published accuracy — whether the question deserves a professional answer. That is not a small thing.

Frequently asked questions