The RAADS-r Full Guide (2026)
This guide goes beyond the test itself. If you’ve already taken the RAADS-R, or you’re deciding whether to, this is where we explore what the test actually measures, how to answer honestly, what recent research says about its accuracy, and — uniquely — what over 100,000 real completions from our own community tell us about how this test plays out in practice.
Why People Take the RAADS-R
Most adults arrive at the RAADS-R after a period of quiet questioning rather than a single moment of realisation. Common starting points include:
- A close friend, partner, or family member is diagnosed as autistic, prompting a closer look at your own patterns
- Autism-related content on social media starts feeling unexpectedly familiar
- Coping strategies that worked for years suddenly stop working
- Burnout, exhaustion, or a sense that rest no longer feels restorative
- A therapist, GP, or another clinician raises autism as a possibility
- Years of being told you’re “too sensitive,” “too intense,” or “hard to read”
None of these on their own confirm anything. But together, they’re the reason the vast majority of people land on this specific test rather than a general quiz.
What the RAADS-R Actually Measures
The RAADS-R is an 80-item self-report questionnaire built around the two core diagnostic domains used in the DSM-5-TR for Autism Spectrum Disorder: Criterion A (social communication) and Criterion B (restricted, repetitive patterns of behaviour). It translates these into four measurable domains.
1. Language (A1 & A2) How you use and interpret language, both spoken and unspoken. This includes literal interpretation of phrases, picking up on tone or sarcasm, and reading facial expressions or gestures.
2. Social Relatedness (A3) How social interaction and relationships feel to you. This covers whether social situations feel intuitive or require conscious effort, how conversations flow, and how relationships develop and are sustained over time.
3. Sensory & Motor (B1 & B4) Sensory processing and repetitive movement. This includes sensitivity to sound, light, touch, or smell, feeling overwhelmed in certain environments, and using movement (stimming) to self-regulate.
4. Circumscribed Interests (B2 & B3) Patterns of routine, sameness, and focused interest. This covers a need for predictability and the depth or intensity of specific interests.
How to Answer RAADS-R Questions Honestly
One of the RAADS-R’s distinguishing features is that it asks about development over time — not just how you feel now, but whether a trait was present when you were young. This is deliberate: autism is a lifelong neurotype, and the test is designed to capture patterns that may have existed since childhood, even if you’ve since adapted, masked, or built coping strategies around them.
Each item has four response options:
- True now and when I was young
- True now only
- True only when I was younger than 16
- Never true
A note on interpreting the questions. Some of the RAADS-R’s statements are worded ambiguously by design, and many autistic adults report finding this genuinely difficult — not because the questions are unclear, but because precise, literal interpretation is itself a common autistic trait. Questions like “I only like to talk to people who share my special interests” can prompt genuine uncertainty: does only mean only? How is “special interest” being defined here?
This difficulty interpreting the wording is not itself diagnostic, but it is a pattern reported more often by autistic respondents than non-autistic ones. If you find yourself overthinking a question, that’s a normal part of the experience for many people who go on to score highly — try to answer based on your gut instinct rather than searching for a “correct” interpretation.
Recent research (Sturm et al., 2024) has also raised a fair critique: holding four response options and two timeframes in mind across 80 items can be cognitively demanding, and may reduce consistency for some respondents. Some researchers now suggest a simplified two-option approach (generally true / never true) as a more reliable alternative. NeuroDirect’s version of the test uses the original, validated four-option format, in line with the instrument as it was designed and studied — but it’s worth knowing this critique exists if you find the four options difficult to distinguish.
What a "Good" Cutoff Score Actually Means
The RAADS-R’s original validation research set 65 as the clinical threshold — scores above this were considered consistent with an autism spectrum profile. This is still the most widely cited cutoff, and it’s the one used on our own test results page.
But it’s worth understanding the full picture. The 65 cutoff was optimised for sensitivity — correctly identifying autistic individuals — rather than specificity — correctly ruling out non-autistic individuals with overlapping traits. In practice, this means scores in the 65–105 range can also appear in people with ADHD, anxiety disorders, trauma-related conditions, or other neurodevelopmental profiles, without those individuals being autistic.
More recent research (Jones et al., 2021) has proposed higher, more specific cutoffs — 81, 106, and 119 have each been suggested as points where the test more reliably distinguishes autism from these overlapping conditions. None of these newer cutoffs have replaced 65 as the standard reference point, but they add useful context: a score of 70 and a score of 150 are not the same kind of result, even though both sit “above threshold.”
What our data shows: Across more than 100,000 completions on NeuroDirect, the average score is 114.09 — comfortably above even the more conservative 106 cutoff proposed in recent research.
What the RAADS-R Doesn't Tell You
No screening tool is a complete picture, and the RAADS-R has known limitations worth understanding before you interpret your own result.
It measures autistic-like traits, not autism itself. Elevated scores can also occur in people with ADHD, anxiety disorders, trauma histories, or other neurodevelopmental profiles. The test is highly sensitive (it rarely misses autistic traits when they’re present) but its specificity — its ability to rule out these other explanations — is more limited, particularly using the standard 65 cutoff.
It relies entirely on self-report. Your answers depend on your own interpretation of each question and your own insight into your patterns over time. This isn’t a flaw unique to the RAADS-R — it’s true of every self-report screening tool — but it’s worth remembering that the result reflects your self-perception, not an external clinical observation.
It isn’t designed to stand alone. Within a full diagnostic assessment, the RAADS-R is one input among several — typically combined with a clinical interview, developmental history, input from people who know you well, and other psychometric tools. A score from this test, however high, is a starting point for that conversation, not a substitute for it.
What Happens to This Test in Clinical Practice
Within a formal autism assessment, the RAADS-R typically contributes to:
- Identifying patterns consistent with autistic traits worth exploring further
- Supporting the clinician’s overall formulation of your presentation
- Guiding which areas the clinical interview should focus on
It does not function as a standalone diagnostic tool anywhere in clinical practice. A qualified clinician will always integrate it with a structured interview, your developmental history, and — where possible — information from people who’ve known you across different periods of your life.
What Our Data Shows
NeuroDirect has published a full statistical analysis of the RAADS-R, based on over 100,000 completions collected between August 2025 and July 2026 — one of the largest independent datasets of its kind, roughly 130 times the size of the original 2011 validation study.
A few headline findings:
- The average score across all completions is 114.09 — well above the standard 65 cutoff, and above the more conservative 106 cutoff proposed in recent research
- 81% of test-takers scored above the clinical threshold
- One third scored above 140 — a range consistently associated with a pronounced autistic profile
- The community average sits almost exactly between the clinically diagnosed autistic group and non-autistic control group from the original validation research
These figures reflect a self-selecting population — people who sought out this specific test already suspected something. They can’t be used to estimate how common autism is in the general population. What they do show is that when adults act on that suspicion and take a validated screening test, the results overwhelmingly confirm it.
Recommended Next Steps
Already taken the test? Read our score interpretation guide for a full breakdown of every score band, or unlock your Personalised Report for a detailed analysis across all four domains.
Haven’t taken it yet? Take the free RAADS-R test →
Want to compare it with other tools?
- AQ Test — a widely used 50-item screener, particularly useful for a faster initial check
- CAT-Q Test — specifically measures masking and camouflaging, useful if you suspect your RAADS-R score may be lower than your lived experience due to compensation strategies
- RAADS-14 — a shorter version of this test, useful if 80 questions feels like too much right now
Considering a formal assessment?
Our directory of neurodivergent-affirming specialists can help you find an assessor, and our guide to what happens during an adult autism assessment walks through what to expect.
References
- Ritvo, R.A., Ritvo, E.R., Guthrie, D., et al. (2011). The Ritvo Autism Asperger Diagnostic Scale-Revised (RAADS-R). Journal of Autism and Developmental Disorders, 41(8), 1076–1089:
https://pmc.ncbi.nlm.nih.gov/articles/PMC3134766/ - Jones, S.L., Johnson, M., Ames, D., & Bush, A.I. (2021). The effectiveness of RAADS-R as a screening tool for adult ASD populations. Autism Research, 14(9), 2019–2027:
https://onlinelibrary.wiley.com/doi/full/10.1155/2021/9974791 - Sturm, A., et al. (2024). Psychometric exploration of the RAADS-R with autistic adults:
https://pubmed.ncbi.nlm.nih.gov/38305196/ - RAADS-14 short-form development (2023):
https://www.mdpi.com/2254-9625/13/11/188 - German validation study (2024). European Archives of Psychiatry and Clinical Neuroscience:
https://link.springer.com/article/10.1007/s00406-024-01894-w - NHS – Signs of autism in adults:
https://www.nhs.uk/conditions/autism/signs/adults/
About the Author
Harry Sherman
Harry Sherman is the owner of NeuroDirect, a UK-based platform that provides online neurodiversity screening tools, resources, and access to qualified specialists. In this role, Harry oversees the creation, accuracy, and accessibility of content, ensuring that NeuroDirect’s tools, including the RAADS-R test, are easy to use and responsibly presented.
With professional experience in digital strategy, online platforms, and health-related technology projects, Harry has a strong background in making complex information accessible to the public. He works closely with clinicians, researchers, and advocacy groups to ensure NeuroDirect reflects current knowledge and best practice in the field of neurodiversity.
Although not a clinician himself, Harry is passionate about bridging the gap between research, digital tools, and public understanding. His goal is to create a trusted, stigma-free space where people can explore self-assessments, learn more about autism and related conditions, and take informed next steps toward professional evaluation if needed.