ADOS-2 Report Writing: What You Can (and Can't) Include

The ADOS-2 Section Is Where Autism Reports Go Wrong

Most autism evaluation reports live or die on one section: the write-up of the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2). It is the section schools scrutinize, payers audit, and future evaluators re-read. It is also the section where clinicians most often make one of two mistakes — copying protected scoring content into a document that will be photocopied and emailed for years, or writing a bare classification statement so thin that no reader can tell what was actually observed.

This guide covers how to write the ADOS-2 section of a report: what may and may not appear in a document that leaves your office, how to describe observed behavior narratively, how to phrase classification results without reproducing scoring content, and how to integrate the ADOS-2 with the rest of a multi-method evaluation. For the structure of the full evaluation report — history, cognitive and adaptive testing, DSM-5-TR criteria mapping, and a complete sample report — see the companion guide to the autism diagnostic evaluation report.

What the ADOS-2 Is (and Isn't)

The ADOS-2 (Lord, Rutter, DiLavore, Risi, Gotham, & Bishop, 2012; published by Western Psychological Services) is a semistructured, standardized observational assessment of communication, social interaction, play and imagination, and restricted and repetitive behaviors. It consists of five modules, selected on the basis of the examinee's chronological age and expressive language level — including a Toddler Module for minimally verbal children roughly 12 to 30 months old. Relative to the original ADOS, the second edition introduced revised algorithms, a comparison score for several modules, updated administration and coding guidance, and the Toddler Module. It is widely described as a gold-standard observational measure in autism assessment.

Two boundaries define how it should appear in your reports:

1. The ADOS-2 yields a classification, not a diagnosis. Results place an individual's presentation within a classification range and, for some modules, a comparison score describing the level of autism-related features relative to peers of similar age and language level. Autism spectrum disorder itself is a DSM-5-TR diagnosis made by a clinician integrating history, observation, and multi-informant data. Your report language must preserve this distinction throughout.

2. The instrument's content is protected. Item content, activity descriptions, coding definitions, algorithms, and cutoff values are copyrighted test materials. They belong in the protocol booklet and your test file — not in a report that will circulate among non-clinicians.

Who May Administer It — and What to Say About Qualifications

Because reports are read by IEP teams, attorneys, and payers who may question the evaluation's validity, a brief statement of the administrator's qualifications strengthens the section. The relevant background, per the publisher and major training centers:

• Professional background. Typical users are clinical and school psychologists, psychiatrists, developmental pediatricians, speech-language pathologists, and occupational therapists with graduate training in individually administered assessment and experience in ASD evaluation. Training programs commonly expect at least a master's degree in psychology, counseling, social work, speech-language pathology, occupational therapy, education, or a related field.
• Clinical training tier. An introductory clinical workshop (typically two days) — or the publisher's video-guided self-study package — prepares a qualified clinician for clinical administration. As the Weill Cornell CADB FAQ notes, attendance alone is not sufficient to ensure competent use; supervised practice administrations are expected before independent clinical use.
• Research reliability tier. Research use requires advanced training and establishing inter-rater reliability (conventionally about 80 percent agreement with a certified trainer across multiple administrations). Clinical users are encouraged, but not required, to reach this standard.

In the report itself, a single sentence suffices: "The ADOS-2 was administered by [clinician], a licensed psychologist trained in ADOS-2 administration with experience in autism spectrum evaluations." If the administration was for research purposes, note research reliability status instead.

What May — and May Not — Appear in the Written Report

This is the compliance core of the section. A written report leaves your control the moment it is released; assume every report will eventually be read by a parent, a school team, opposing counsel, and another evaluator.

Do not include:

• Item scores, coding values, or coding definitions
• Algorithm composition, algorithm totals, or cutoff values
• Comparison-score conversion details or numeric comparison scores presented against thresholds
• Activity or task names, descriptions of administration procedures, or the specific social presses used
• Reproductions of any portion of the protocol booklet or manual

Two obligations drive this. First, copyright: WPS's terms of use prohibit reproducing test items, scales, scoring mechanisms, or manual content, and reproduction permission runs through the publisher's rights department. Second, test security: APA Ethics Code Standard 9.11 requires psychologists to make reasonable efforts to maintain the integrity and security of test materials. A parent who has read a description of the tasks and scoring thresholds cannot provide a naïve second-opinion administration for their child, and coached familiarity degrades the instrument for everyone.

Do include:

• The instrument's full name and edition, the module administered, and the rationale for module selection (expressive language level and age)
• The setting, the examinee's cooperation and engagement, and any factors bearing on validity
• Narrative behavioral observations, in your own words, organized by the domains the instrument samples
• The classification the results fell within, stated by name (for example, "results fell within the autism spectrum classification range")
• A qualitative characterization of the comparison score where applicable (for example, "a moderate level of autism spectrum–related features relative to same-age peers with similar language")
• Any deviations from standardized administration and their implications

The phrasing pattern to internalize: name the classification; never publish the arithmetic that produced it. "Ms. R.'s results met the ADOS-2 classification of autism" communicates everything a reader needs. "She scored X against a cutoff of Y" communicates nothing additional to a lay reader and breaches both obligations above.

Writing the ADOS-2 Section Step by Step

Step 1: State the module and why. One or two sentences: which module was administered and the expressive-language and age rationale. Module selection errors are a recognized threat to validity, so showing your reasoning preempts challenge. ("Module 3 was selected because J. speaks in fluent, complex sentences.")

Step 2: Establish the observation conditions and validity. Where the assessment occurred, who was present, how the examinee engaged, and anything that could bear on interpretation — illness, refusal, interpreter use, or interruptions. If everything was standard, say so: "Administration followed standardized procedures, and results are considered a valid sample of current functioning."

Step 3: Describe social communication and social interaction narratively. This is the heart of the section and should be its longest part. Write what you saw, in plain behavioral language: use of eye contact integrated (or not) with gestures and speech; range of facial expressions directed to the examiner; initiation and response in back-and-forth conversation; shared enjoyment; gestures; requesting and showing behaviors. Anchor each observation in a concrete moment without naming the task that elicited it — "when offered an opportunity for a shared activity," "during an unstructured conversational exchange." Specific, individualized description is what distinguishes a defensible report from boilerplate.

Step 4: Describe restricted and repetitive behaviors observed. Motor mannerisms, repetitive use of objects or speech, unusual or intense preoccupations, and sensory-seeking or sensory-avoidant behavior observed during the session. If little was observed, say so and note that the observational window is brief — restricted and repetitive behaviors often appear more clearly in history and collateral report than in a single structured session.

Step 5: State the classification — carefully. One short paragraph: the classification range the results fell within, a qualitative comparison-score characterization if applicable, and an explicit reminder of what the result is: "The ADOS-2 yields a classification based on directly observed behavior; it is not, by itself, a diagnosis."

Step 6: Note limitations. A single observational session samples behavior at one point in time. Strong verbal or cognitive ability, camouflaging and masking (particularly in females and adults), acute anxiety, and unfamiliar settings can suppress observable features; fatigue or noncompliance can inflate them. If administration deviated from standard conditions — including any remote format, which is not validated — state it plainly and qualify the results accordingly.

Integrating the ADOS-2 With the Rest of the Evaluation

The ADOS-2 section should never stand alone as the basis for the diagnostic conclusion. In the integration or formulation section of the report, connect the observational findings to the other data streams:

• Developmental history and caregiver interview (such as the ADI-R): do caregiver-reported early social communication patterns converge with what you observed directly?
• Rating scales and collateral report: do teachers and other informants describe the same behaviors across settings?
• Cognitive and adaptive data: do intellectual and adaptive profiles contextualize the observations (see the guides to the cognitive assessment report and psychological evaluation)?

When all sources converge, integration is a short paragraph. When they diverge, the report earns its fee: document which findings you weighted and why. An examinee may fall below the classification range while history, collateral data, and adaptive findings still support a DSM-5-TR diagnosis — commonly with high verbal ability or long-practiced masking. Conversely, an above-range classification can reflect anxiety, trauma-related presentation, language disorder, or ADHD rather than autism. Either way, the reasoning belongs in writing: "Although the ADOS-2 classification was non-spectrum, the diagnosis of autism spectrum disorder is supported by [specific converging evidence], and the observational result likely reflects [specific, individualized explanation]." The full autism evaluation guide covers the DSM-5-TR criteria mapping this reasoning feeds into.

Fictional Example: An ADOS-2 Results Paragraph

The excerpt below is entirely fictional and shows the target level of detail — individualized behavior description, named classification, no scores, no cutoffs, no task content.

Common Mistakes

1. Reproducing scores, algorithm totals, or cutoffs in the report. This is the most consequential error: it violates the publisher's copyright terms and the test-security obligation of APA Standard 9.11, and it adds nothing a narrative classification statement doesn't convey better. If a records request or court order seeks raw test data, handle it through the applicable test-data release rules — not by publishing scoring content in the report itself.

2. Treating the classification as the diagnosis. "ADOS-2 positive, therefore ASD" reasoning appears in reports more often than it should. The classification is one data source; the diagnosis is a clinical judgment on the whole record, and the report must show that judgment.

3. Boilerplate behavioral observations. If the observations paragraph could describe any child, it fails. Templated phrases ("poor eye contact, limited reciprocity") without individualized, concrete moments read as unexamined and are vulnerable in due-process hearings and independent-review contexts.

4. Naming tasks and administration procedures. Describing the specific activities or social presses puts protected material into circulation and coaches future administrations. Describe the behavior, not the elicitation method.

5. Wrong module, unexplained. Module choice follows expressive language level, not just age. If the report doesn't state the rationale, a reviewer cannot verify validity — and if the module genuinely was mismatched, the results section inherits the problem.

6. No rationale when judgment diverges from the classification. Falling below or above the classification range without discussion invites the reader to conclude the diagnosis is unsupported. Divergence is clinically common and defensible — but only if the reasoning is documented.

Attribution and Trademark Etiquette

Refer to the instrument accurately and only as needed to identify it: full name and edition on first use, "ADOS-2" thereafter, with authorship and publisher (Lord et al., 2012; Western Psychological Services) in the measures list. Do not present any report shell or practice materials as publisher-issued forms, and do not imply endorsement by the publisher.

ADOS-2 is a trademark of Western Psychological Services (WPS). This page is not affiliated with or endorsed by WPS.