What is the ADOS-2? A guide to the Autism Diagnostic Observation Schedule

The ADOS test, formally the "Autism Diagnostic Observation Schedule, Second Edition" (ADOS-2), is a semi-structured, standardized observation assessment that trained clinicians use to evaluate communication, social interaction, play, and restricted and repetitive behaviors in people who may be autistic. It is one component of a comprehensive diagnostic evaluation, and it is neither a standalone diagnosis nor an IQ test. Published in 2012 by WPS (Western Psychological Services) and authored by Catherine Lord, Michael Rutter, and colleagues, the ADOS-2 spans ages 12 months through adulthood using five modules matched to a person's age and language level. This article explains where the ADOS-2 came from, how its modules and activities work, what the scores mean, who can administer it, and what its strengths and limitations are.

The ADOS-2 is the second edition of the Autism Diagnostic Observation Schedule, an instrument that WPS describes as already viewed as the gold standard for the observational assessment of autism spectrum conditions. "Semi-structured" means the examiner works from a planned set of activities but delivers them flexibly, responding naturally to the person being assessed rather than reading from a rigid script. "Standardized" means the materials, the way activities are presented, and the coding rules are the same for everyone, so results can be compared against consistent criteria.

According to the publisher, the second edition kept the administration and coding procedures of the original ADOS while adding updated protocols, revised algorithms, a new comparison score, and a completely new Toddler Module for children as young as 12 months. Each module takes about 40 to 60 minutes to administer. The instrument covers an unusually wide range for a single assessment, from 1-year-olds who are not yet speaking in phrases to verbally fluent adults.

A point worth stating plainly at the outset: the ADOS-2 observes behavior. It does not measure intelligence, and a person's ADOS-2 results say nothing about their IQ. Cognitive ability and autism are separate questions, which is why a thorough evaluation usually examines both.

A person being evaluated completes only one module. WPS explains that the module is selected based on chronological age and expressive language level, which keeps the tasks appropriate and prevents language ability from distorting the picture of social communication.

• Toddler Module: for children between 12 and 30 months of age who do not consistently use phrase speech. Its results are deliberately cautious, flagging levels of concern rather than assigning a classification.

• Module 1: for children 31 months and older who do not consistently use phrase speech.

• Module 2: for children of any age who use phrase speech but are not yet verbally fluent.

• Module 3: for verbally fluent children and young adolescents, mixing play-based tasks with conversation.

• Module 4: for verbally fluent older adolescents and adults, which leans on interview-style questions and conversation more than play.

Each module engages the person in a series of activities built around interactive materials included in the test kit. The publisher's published activity list for Module 3 gives a good sense of the flavor: a construction task, make-believe play, joint interactive play, a demonstration task, describing a picture, telling a story from a book, discussing cartoons, conversation about emotions, social difficulties, friendships and relationships, and creating a story. To a child, much of this feels like structured play and chatting. To the examiner, every activity is a designed opportunity, or "press," to observe specific social and communication behaviors, such as how the person shares enjoyment, gestures, asks for help, or shifts attention between people and objects.

During the session the clinician runs the activities, observes, and takes notes. Immediately afterward, they code the behaviors they observed using detailed criteria in the module's protocol booklet, then transfer selected codes to an algorithm form. "Algorithm scores" are the subset of coded items that research has shown best distinguish autism-related behavior; they are summed within two domains that mirror current diagnostic thinking, social affect and restricted and repetitive behaviors.

In Modules 1 through 4, WPS reports that the algorithm total is compared with empirically derived cutoff scores to yield one of three classifications: autism, autism spectrum, or non-spectrum. The difference between the first two is one of severity, with the autism classification indicating more pronounced observed symptoms. The Toddler Module works differently. Its algorithms produce "ranges of concern" rather than classifications, because formal classification may not be appropriate at such a young age; the ranges quantify risk and guide monitoring and follow-up.

The second edition also introduced a "comparison score" for Modules 1 through 3. According to the publisher, this score lets a clinician compare a child's overall level of autism spectrum related symptoms with those of children diagnosed with autism who are the same age and have similar language skills, and it makes it easier to track symptoms over time. Note what these scores are and are not: they describe observed behavior during one structured session, and they feed into clinical judgment rather than replacing it.

The ADOS-2 is a restricted clinical instrument. WPS lists it at qualification level C, the publisher's most restricted purchasing tier, and sells extensive training for it, including a multi-module video training program and clinical workshops. In practice, administrators are clinicians such as psychologists, developmental pediatricians, psychiatrists, and speech-language pathologists who have completed instrument-specific training, because accurate administration and coding take supervised practice. A 2018 study in European Child and Adolescent Psychiatry led by Inge Kamp-Becker found high variance when 189 clinicians coded the same recorded ADOS sessions, and concluded that coding accuracy depends heavily on the coder's experience with the instrument and the quality of the administration.

Just as important, an autism diagnosis is made by qualified clinicians, never by a single score. The CDC states that there is no medical test, such as a blood test, for autism; specialists evaluate developmental history and behavior. The National Institute of Mental Health describes the diagnostic evaluation as the work of a team that may include child neurologists, developmental behavioral pediatricians, psychologists, psychiatrists, and speech-language pathologists, and lists its typical components: medical and neurological examination, assessment of cognitive abilities, assessment of speech and language, behavioral observation, an in-depth caregiver interview, and assessment of daily living skills.

The ADOS-2 supplies the structured behavioral observation piece of that puzzle. It is commonly paired with a detailed caregiver interview such as the "ADI-R" (Autism Diagnostic Interview, Revised), a rating scale such as the SRS-2, an adaptive behavior measure such as the Vineland-3, and individually administered cognitive testing. Because autism occurs across the full range of intellectual ability, understanding how IQ and autism relate helps teams interpret ADOS-2 results in context and plan supports that fit the whole person.

The instrument's reputation rests on real evidence. Kamp-Becker and colleagues describe the ADOS as part of the gold standard for diagnostic evaluation, and a 2026 meta-analysis in Frontiers in Psychiatry pooling six clinical studies with 3,021 participants reported an overall "sensitivity" (the proportion of autistic people the test correctly identifies) of 0.88 and a "specificity" (the proportion of non-autistic people it correctly rules out) of 0.74 in real-world referral settings.

Those same numbers point to the limitations. In that meta-analysis, specificity fell to 0.67 for Module 3 and 0.63 for Module 4, meaning elevated scores were fairly common among verbally fluent adolescents and adults who turned out not to be autistic, often because other psychiatric conditions can produce similar behavior in a single session. The authors concluded that the ADOS-2 should not be used in isolation for diagnostic determination and belongs inside a multidisciplinary evaluation. Other honest caveats: it samples behavior from one session on one day, its accuracy depends on the examiner's training, and its classifications describe observed symptoms rather than a person's abilities or worth.

No. The ADOS-2 is an observational assessment of social communication and restricted and repetitive behaviors. It produces no cognitive score, and evaluators typically administer a separate, individually administered intelligence test when cognitive ability needs to be measured.

No. Diagnosis is a clinical judgment made by qualified professionals who integrate the ADOS-2 with developmental history, caregiver interviews, cognitive and language testing, and medical examination.

According to WPS, each module takes about 40 to 60 minutes to administer, plus time afterward for coding and scoring.

Clinicians with graduate-level assessment training who have also completed ADOS-2 specific training, such as psychologists, developmental pediatricians, and speech-language pathologists. WPS restricts purchase to its highest qualification level.

From 12 months through adulthood. The Toddler Module serves children aged 12 to 30 months, and Modules 1 through 4 cover everyone older, matched by language level.

No. It is an in-person, interactive assessment built around live activities and materials, administered and coded by a trained clinician. No reputable online test screens for or detects autism.

If your family is going through an autism evaluation, the cognitive testing that accompanies the ADOS-2 will be an individually administered test chosen and interpreted by the evaluating clinician. Online measures serve a different purpose: measuring reasoning ability in adults who want a rigorous, normed score outside a clinical context. The RIOT IQ test (the Reasoning and Intelligence Online Test), developed by RIOT IQ with psychometrician Dr. Russell T. Warne, is one example: built for adults 18 and older, it includes 15 subtests across six cognitive indices (verbal reasoning, fluid reasoning, spatial ability, working memory, processing speed, and reaction time), takes about 52 minutes, and reports scores on the familiar mean-100, standard-deviation-15 IQ scale. It measures cognitive ability only; it does not screen for autism, and it does not replace an individually administered diagnostic evaluation. If a well-normed measure of adult reasoning ability is what you are after, you can take it at riotiq.com.

1. Western Psychological Services. (2012). (ADOS-2) Autism Diagnostic Observation Schedule, Second Edition [Product page]. wpspublish.com

2. Centers for Disease Control and Prevention. (n.d.). Diagnosing autism spectrum disorder. cdc.gov

3. National Institute of Mental Health. (n.d.). Autism spectrum disorder. nimh.nih.gov

4. Kamp-Becker, I., Albertowski, K., Becker, J., Ghahreman, M., Langmann, A., Mingebach, T., Poustka, L., Weber, L., Schmidt, H., Smidt, J., Stehr, T., Roessner, V., Kucharczyk, K., Wolff, N., & Stroth, S. (2018). Diagnostic accuracy of the ADOS and ADOS-2 in clinical practice. European Child & Adolescent Psychiatry, 27(9), 1193–1207. doi.org

5. Kılıncel, S., Bulut, F., Goksel, P., Usta, M. B., & Kilincel, O. (2026). Module-specific diagnostic accuracy of ADOS-2 in real-world clinical referral populations: An updated systematic review and HSROC meta-analysis. Frontiers in Psychiatry. pmc.ncbi.nlm.nih.gov

Alex1ruff, CC BY-SA 4.0, via Wikimedia Commons