This is Everything You Need to Know about the RAADS-R Autism Test

By: Julie Landry

If you've been researching adult autism assessment or are in the process of self-diagnosis, you’ve probably seen the free online tests, which include the RAADS-R test. Maybe you have questions about the test, how it works, and whether it’s reliable. We had similar questions, so we did a deep dive so you don’t have to. 
 

What is the RAADS-R test?

The Ritvo Autism Asperger Diagnostic Scale-Revised (RAADS-R) is a self-report questionnaire designed to assess autistic traits in adults. It’s based on the ICD-10 and DSM-5 diagnostic criteria for Autism Spectrum Disorder (*which we call autism since it’s not actually a disorder). It has 80 statements and four subscales, including language, social relatedness, sensory-motor, and circumscribed interests. 

The RAADS-R is intended for use in an outpatient setting. You can find the test and automatic scoring online, but the test is designed to be administered by a clinician in a clinical setting. The evaluator reads each question to the client and records the score. The idea is that the evaluator can ensure the client understands the questions, and according to the test authors, it also “gives the diagnostician a chance to identify and discuss specific symptoms during administration.” 
 

Is the RAADS-R an “official” test?

The RAADS-R isn’t diagnostic in and of itself, but some providers use it as part of their assessment battery. Many people also use it for self-assessment purposes or exploration, along with the other free screeners available online.

Interesting fact: A recent study found people with an official autism diagnosis and those who are self-diagnosed responded to the RAADS-R questions in a very similar way, but people questioning whether they were autistic were more different in their responses. 
 

Who is the test for?

The RAADS-R was designed for adults 18 and up with an IQ in the average range or above (>80) and was created in response to the lack of diagnostic services available for adults. The authors describe it as “... a useful clinical tool to assist clinicians with the diagnosis of this growing population of higher functioning individuals in adulthood.”
 

How long does the RAADS R test take?

The test typically takes 30 minutes to complete, but the time will vary depending on your reading and comprehension speed and how long you spend thinking about your answers. 
 

Scoring of the RAADS-R test

Scoring ranges from 0 to 240. Sixty-four of the 80 questions are intended to identify “symptomatic behaviors.” They’re scored on a 4-point Likert scale that considers developmental factors (True now and when I was young = 3, True only now = 2, True only when I was younger than 16 years old = 1, and Never true = 0). The RAADS also contains sixteen “non-symptomatic” questions, scored in reverse order (True now and when I was young = 0, True only now = 1, True only when I was younger than 16 years old = 2, and Never true = 3). 

A total RAADS-R score of 64 or higher is considered clinically significant and indicative of autism. 
 

Is the RAADS-R test accurate?

Based on several studies, the RAADS-R is considered valid and reliable. The test authors describe it as “a highly specific (100%) and sensitive (97%) instrument that is useful as an adjunct clinical diagnostic tool.” (As a refresher, sensitivity refers to the test’s ability to correctly identify autistic adults, and specificity refers to its ability to correctly identify neurotypical individuals.)

However, other research hasn’t been as positive. Jones et al. found when used as a self-report screening tool, the RAADS-R was unable to discriminate between autistic and allistic individuals. The RAADS-R only had a 3.03% chance of detecting the absence of autism in their study sample. Similarly, Sizoo and Horwitz concluded the RAADS-R had poor predictive validity. The neuronormative group in their research scored higher on average than the autistic group. In another study, Conner et al. found participants with a formal diagnosis of autism scored higher on the ADOS than those who did not receive a diagnosis, but there were no differences between outcome groups on the RAADS-R assessment. 
 

Can you score high on the RAADS test and not be autistic?

A high score doesn’t necessarily mean you’re autistic.
 

We asked our team to take the RAADS and share their thoughts. 

An anonymous NeuroSpark team-member felt some of the questions were confusing. Overall, they thought the test was too focused on empathy and social relationships and didn’t take masking into account. They didn't like how the Likert scale was designed and often wanted to answer "Sometimes." They would have preferred something like a "Often-Sometimes-A Bit-Not at all" option. 

Cat also found the statements too broad and confusing because they needed more context. She thought descriptions of the situation would have been helpful and suggested a better use of the test would be to have someone take it and then describe what they found confusing and disliked about it. 

Lex got caught up in how literally to interpret these questions (“like when would I have been going out to a restaurant by myself before the age of 16...?”). They were surprised that their sensory/motor subtotal wasn’t higher. Lex also pointed out that the international validation study didn't collect any information on participants' race, ethnicity, or language(s) spoken at home and only used binary gender options.

When we heard from Jennifer, she was only on question 10, but she already hated it. She found it too rigid. 

Dani thought the questions were too binary and didn’t capture the nuances of experience like an assessment can. She doesn’t think the RAADS-R is a good option for intelligent or emotionally intelligent autistic people. She also found many of the questions to be inaccurate and downright offensive, especially a lot of the relationship questions which seem to be designed around the harmful assumption that autistic people cannot (or don’t want to) engage in relationships, friendships or intimacy. She says, “So many autistic people of all support needs want relationships AND sex!”

For most of the statements, I had a hard time thinking back to before I was 16 and primarily chose true only now and never true responses as a result. I also felt several of the statements were based on outdated stereotypes. For example, the statement ‘I am a sympathetic person’ is intended to identify allistic adults and is referred to as “non-symptomatic (normative) behaviors” statements.
 

So, do we recommend it?

Nah. Despite its success rate in several studies, we’re not fans of the RAADS-R autism test for all the reasons above. PLUS, based on my lit review and research, the test creators are not ND-affirming and subscribe to the medical model, which pathologizes autism (i.e., “developmental pathology,” “high-functioning,” “symptoms of ASD,” etc.) and autistic people.