Is the RAADS-R Test Accurate? Scoring, Cut-Offs and What Research Shows

Updated October 2026. Written by Dror Arbel, a parent, not a clinician. This page explains published research; it is not medical advice.

What is the RAADS-R?

The RAADS-R (Ritvo Autism Asperger Diagnostic Scale – Revised) is an 80-item self-report questionnaire for adults, developed by Dr. Riva Ariella Ritvo and colleagues and published in 2011 (Ritvo et al., Journal of Autism and Developmental Disorders, 2011). It revised the original 78-item RAADS from 2008. It was designed to help clinicians identify autistic adults who were missed in childhood, especially adults of average or above-average intelligence.

Each statement asks whether something is true for you now and/or was true when you were younger than 16. That “then and now” format is deliberate: autism is a lifelong developmental condition, so the scale looks for traits that were present in childhood, not only current difficulties. Most people take roughly 15–20 minutes to complete it.

How the RAADS-R is scored

Each item has four answer options. For 63 items, “true now and when I was young” scores 3, “true only now” scores 2, “true only when I was younger than 16” scores 1, and “never true” scores 0. Seventeen “normative” items (things most non-autistic people agree with) are scored in reverse. The total ranges from 0 to 240.

What does my RAADS-R score mean?

• Below 65: below the original research threshold. This does not rule autism out. In the 2011 study about 3% of autistic participants scored below 65, mostly younger adults who did not report traits their families described.
• 65 or above: at or above the original threshold, meaning you report many traits associated with autism. On its own this is not a diagnosis, especially if you also live with anxiety, depression, ADHD, trauma or a personality disorder (see the accuracy section below).
• Well above 65: autistic adults in the original study averaged around 130. Higher scores mean more reported traits, but there is no official “severity” scale for the RAADS-R.

You may see score bands online such as “65–90 = some traits” or “160+ = very strong evidence”. These are informal rules of thumb from websites, not cut-offs from the original research, so treat them with care.

How accurate is the RAADS-R? What the research shows

The original study (2011)

Ritvo and colleagues tested the RAADS-R with 779 adults at nine centres on three continents: 201 autistic adults and 578 comparison participants (some with other psychiatric diagnoses, some without). At a total score of 65 they reported 97% sensitivity (autistic adults correctly flagged) and 100% specificity (no comparison participants flagged), with test-retest reliability of 0.987 (Ritvo et al., 2011).

Those are excellent numbers, but there is an important catch: the autistic participants had already been diagnosed, and the comparison group was clearly defined. Screening tools almost always perform worse when used with people who are genuinely uncertain, which is exactly who takes the test online or is referred for an adult assessment.

Later studies in real clinical settings

Independent studies with adults referred to mental health or autism assessment services found that many non-autistic people also score above 65 (summarised by NovoPsych):

• Sizoo et al. (2015, Netherlands): adults with other psychiatric conditions scored well above 65 on average; the authors suggested a higher cut-off for clinical samples.
• Brugha et al. (2020, UK): in mental health services, more than half of service users scored above 65, so specificity was very low at the original cut-off.
• Jones et al. (2021, UK): in a specialist adult autism assessment service, RAADS-R scores did a poor job of predicting who was eventually diagnosed.

The takeaway: the RAADS-R is good at picking up autistic traits (few autistic adults score low), but in people who already have other mental health difficulties it produces many false positives. Social anxiety, depression, ADHD, trauma and some personality disorders can all raise scores.

Why online RAADS-R results can mislead

• Self-report: it measures how you see yourself. People who mask heavily, or who have limited insight into childhood behaviour, may under- or over-report.
• Who takes it: people who search for an autism test usually already suspect something, which is different from the original research groups.
• No clinician: the scale was designed to be used alongside a clinical assessment, not as a stand-alone online quiz.
• Language and culture: it was developed in English-speaking countries; translations exist, but norms may differ.

RAADS-R vs other adult autism screeners

None of these tools diagnose autism. A diagnosis comes from a clinician, usually combining a developmental history, an interview (such as the ADI-R), and sometimes a structured observation such as the ADOS-2.

What to do with your RAADS-R score

• Write down your total and subscale scores, plus a few real-life examples for each area (social, sensory, interests, language), from childhood and now.
• Ask a parent or someone who knew you as a child what they remember; clinicians value this history.
• Bring the score to your doctor or a psychologist as a starting point for a conversation, not as proof. Our guide to testing for autism in adults explains the process.
• If you are weighing self-identification, read our page on autism self-diagnosis in adults for the pros and limits.

For the short version of the scale, see our guide toRAADS-14 scoring. If a clinician has given you or your child the Social Responsiveness Scale, see our SRS-2 scoring and T-score guide.

Frequently asked questions

Is the RAADS-R test accurate?

In the original 2011 study it showed 97% sensitivity and 100% specificity at a cut-off of 65. In later studies with adults referred to mental health or autism services, many non-autistic people also scored above 65, so it is much less specific in real clinical use. It is a screening tool, not a diagnosis.

What is a good or normal RAADS-R score?

There is no “good” score. In the original research, no non-autistic comparison participant scored 65 or above, and autistic adults averaged around 130. A score below 65 does not rule out autism, and a score above 65 does not confirm it.

Can the RAADS-R diagnose autism?

No. It is a self-report questionnaire designed to support a clinical assessment. Only a qualified clinician can diagnose autism, using developmental history, interviews and observation.

Can anxiety, ADHD or depression raise a RAADS-R score?

Yes. Studies in mental health services found that people with other psychiatric conditions often score above 65, which is why clinicians interpret the score alongside a full assessment.

How long does the RAADS-R take?

It has 80 statements and usually takes about 15 to 20 minutes.

Who created the RAADS-R?

It was developed by Riva Ariella Ritvo and colleagues and published in the Journal of Autism and Developmental Disorders in 2011, revising the original RAADS (2008).

References

• Ritvo RA, Ritvo ER, Guthrie D, et al. (2011). The Ritvo Autism Asperger Diagnostic Scale-Revised (RAADS-R): a scale to assist the diagnosis of autism spectrum disorder in adults: an international validation study. Journal of Autism and Developmental Disorders, 41(8), 1076–1089. doi:10.1007/s10803-010-1133-5
• Eriksson JM, Andersen LM, Bejerot S. (2013). RAADS-14 Screen: validity of a screening tool for autism spectrum disorder in an adult psychiatric population. Molecular Autism, 4, 49.
• Sizoo BB et al. (2015); Brugha TS et al. (2020); Jones et al. (2021): clinical-sample studies of the RAADS-R, as summarised by NovoPsych.
• Scoring details: Embrace Autism, RAADS-R research overview. See also our guide to the Embrace Autism tests (free tests, paid adult assessments, and how accurate they are).