Repetitive Behaviours Questionnaire-3 (RBQ-3)
A 20-item measure of restricted and repetitive behaviours in adults, across sameness and sensory-motor domains.
Last reviewed: May 2026
The RBQ-3 (Repetitive Behaviours Questionnaire-3) is a 20-item measure of restricted and repetitive behaviours, developed at Cardiff University’s Wales Autism Research Centre by Sue Leekam, Catherine Jones and colleagues. It consolidates the earlier child (RBQ-2) and adult (RBQ-2A) versions into coordinated self-report and informant-report forms usable across the lifespan, and was validated in adults by Jones and colleagues (2024).
Scoring & Interpretation
Items 1 to 19 are rated on a four-point scale from 1 ("never or rarely") to 4 ("serious or extreme"); item 20 uses a three-point format asking about the range of self-chosen activities. The published scoring is a mean of completed items (range 1.00 to 4.00, with up to 10% missing allowed); the equivalent sum runs 20 to 79. Two subscales are scored from items 1–19: Repetitive Sensory-Motor Behaviours (8 items: 2–6, 8, 10 and 19; sum 8–32) and Insistence on Sameness (9 items: 7, 9 and 12–18; sum 9–36) — items 1, 11 and 20 belong to neither subscale but count toward the total. The RBQ-3 has no published norms or clinical cut-off, so the dashboard deliberately shows position within each range rather than a severity band or percentile.
| Score Range | Severity | Clinical Action |
|---|---|---|
| 20–79 (total) | No published cut-off | Higher scores mean more restricted and repetitive behaviour; read as position within the range alongside the trait picture, never as a percentile |
| 8–32 (Repetitive Sensory-Motor) | Subscale | Repetitive movement and sensory habits — rocking, pacing, hand and finger movements, sensory interests (DSM-5 B1) |
| 9–36 (Insistence on Sameness) | Subscale | Routines, rituals and distress at change — sameness at home, re-doing, same music, clothes or foods (DSM-5 B2) |
Jones and colleagues (2024, Molecular Autism) evaluated the RBQ-3 in a clinical diagnostic sample (N = 110) and an online sample of 151 autistic and 151 non-autistic adults. Internal consistency was high for the total score (α = .90) and the Insistence on Sameness subscale (α = .88), and acceptable-to-marginal for the shorter Repetitive Sensory-Motor subscale (α = .69–.77). Self- and informant-report agreed strongly (rs = .71), and the measure converged with the DISCO interview’s repetitive-behaviour domain while diverging from its social-communication domain. Autistic adults scored much higher than non-autistic adults (mean 2.18 vs 1.35 on the 1–4 scale; d = 2.07). No cut-offs or norms are published, and the authors state such questionnaires should not be used on their own to determine a diagnosis.
When to Use This vs Alternatives
Use RAADS-R when…
You want the broad autistic-trait profile across social, sensory and language domains with a published screening cut-off. The RAADS-R gives the screen read; the RBQ-3 adds focused depth on the repetitive-behaviour domain.
View RAADS-R →Use CAT-Q when…
Camouflaging may be flattening the trait picture. The CAT-Q measures the effort spent masking; the RBQ-3 measures repetitive behaviours themselves, which are often less visible to others and under-reported by informants.
View CAT-Q →Use AQ-10 when…
You need a one-minute first-line screen to decide whether fuller assessment is warranted. The AQ-10 flags; the RBQ-3 characterises one specific DSM-5 domain in depth.
View AQ-10 →See It in Action
- 1Total shown as position within the 20–79 range, deliberately without a severity ramp (no published cut-off)
- 2Repetitive Sensory-Motor and Insistence on Sameness subscale bars (published Jones 2024 key)
- 3Item-level responses on the four-point scale, with items 1, 11 and 20 marked as outside both subscales
- 4One-click PDF export and email delivery for the clinical record
What It Measures
The RBQ-3 (Repetitive Behaviours Questionnaire-3) is a 20-item measure of restricted and repetitive behaviours, developed at Cardiff University’s Wales Autism Research Centre by Sue Leekam, Catherine Jones and colleagues. It consolidates the earlier child (RBQ-2) and adult (RBQ-2A) versions into coordinated self-report and informant-report forms usable across the lifespan, and was validated in adults by Jones and colleagues (2024). It covers the DSM-5 restricted-and-repetitive domain that trait screens summarise only briefly: repetitive movement and sensory habits on one hand, and routines, rituals and insistence on sameness on the other.
When to Use the RBQ-3
Use the RBQ-3 when an adult autism assessment needs specific detail on restricted and repetitive behaviours — DSM-5 criteria B1 (stereotyped or repetitive behaviours) and B2 (insistence on sameness) — beyond what a broad trait measure such as the RAADS-R gives. The subscale pattern shows whether the picture is predominantly sensory-motor, predominantly sameness-and-routine, or both. Because the RBQ-3 has no published cut-off, use it to characterise the repetitive-behaviour domain within a structured assessment, not to screen in or out.
Who It's For
Adults — the 2024 validation covered ages 18 to 67, in clinical and community settings. Self-report, with a matching informant-report version for a relative or someone who knows the person well. Designed as a lifespan measure, but not yet validated in children or people with intellectual disability. Freely available from Cardiff University.
Frequently Asked Questions
What is the RBQ-3?
The RBQ-3 (Repetitive Behaviours Questionnaire-3) is a 20-item measure of restricted and repetitive behaviours, developed at Cardiff University’s Wales Autism Research Centre. It consolidates the earlier RBQ-2 and RBQ-2A into matching self-report and informant-report versions, and was validated in adults by Jones et al. (2024, Molecular Autism).
How is the RBQ-3 scored, and what are its subscales?
Items 1–19 are rated 1 ("never or rarely") to 4 ("serious or extreme"); item 20 uses a three-point format. The published score is the mean of completed items (1.00–4.00). Two subscales are reported: Repetitive Sensory-Motor Behaviours (8 items) and Insistence on Sameness (9 items); items 1, 11 and 20 sit outside both subscales but count toward the total.
Is there an RBQ-3 cut-off score?
No. The RBQ-3 has no published cut-off or normative bands, and its authors state such questionnaires should not be used on their own to determine a diagnosis. For context, the validation study reported mean scores of 2.18 in autistic adults versus 1.35 in non-autistic adults on the 1–4 scale — a large group difference, but a group statistic rather than an individual threshold.
What is the difference between the RBQ-2A and the RBQ-3?
The RBQ-3 consolidates the child RBQ-2 and adult RBQ-2A into one lifespan measure with coordinated self-report and informant-report versions, and unifies items 1–19 onto a single four-point scale (the older versions mixed three- and four-point items). Item content is otherwise retained. The RBQ-3 is the version validated for adult clinical practice (Jones et al., 2024).
Is the RBQ-3 free to use?
Yes — the RBQ-3 is freely available to download from Cardiff University’s Wales Autism Research Centre. In ClientForms the items are reproduced verbatim and scored the moment your patient submits, with both subscales reported.
Use the RBQ-3 in your practice
New accounts get 30 days of full Professional access, no card. After that it stays on the Free plan. Scored the moment patients submit.