Child development

The Q-CHAT Autism Checklist, and How It Differs From the M-CHAT

Save

The Q-CHAT was built as a quantitative revision of an older checklist, designed to measure autistic traits as a continuum rather than to divide toddlers into two groups. That design choice is also its main limitation for a parent: a continuum has no line on it. Here is how it is scored, what its first study did and did not establish, and how it compares with the screen a pediatrician actually uses.

Last updated: July 2026

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

What the Q-CHAT is

The Q-CHAT, the Quantitative Checklist for Autism in Toddlers, is a twenty-five-item parent questionnaire developed as a major quantitative revision of the earlier CHAT. Each item is answered on a five-point frequency scale from zero to four, half the items are reverse-scored, and the answers are summed into a single total that runs from zero to one hundred 1.

The direction is stated in the paper without hedging: the scores from all items are summed to obtain a total Q-CHAT score, with higher scores indicative of more autistic traits 1. The target age band is eighteen to twenty-four months 1.

A frequency scale asks how often something happens rather than whether it happens at all. That is the mechanical difference between the Q-CHAT and a yes-or-no screen, and it is where the extra information in a Q-CHAT total comes from 1.

That design decision runs through everything else about the instrument, including the reason it cannot tell a parent what they came looking for.

Why quantitative is the whole point

Most autism screens are built to sort. A child lands above the threshold or below it, and the questionnaire's job is finished. The Q-CHAT was built to measure instead. In an unselected sample of 754 toddlers its totals were close to normally distributed, with a mean of 26.7 and a standard deviation of 7.8, which is the signature of a trait spread continuously across a population 1.

The first report gives the rest of its psychometrics candidly. Test-retest reliability was an intraclass correlation of 0.82 across 330 children over a mean interval of thirty-eight days. Internal consistency was 0.67 in the unselected group and 0.83 in the group with an autism spectrum condition 1.

The sex difference in average score was small: boys 27.5 against girls 25.8, a Cohen's d of 0.2 1.

A dimensional instrument is genuinely useful for research. It lets a study ask how a trait varies, how it tracks with something else, how it moves across a population. What it does not do is answer the question a worried parent is holding, which is a question about one child on one afternoon.

How the Q-CHAT differs from the M-CHAT

The M-CHAT-R/F is the screen a pediatric office in the United States is most likely to hand over, and it is a different machine entirely. It is a twenty-item parent-report screen for toddlers between sixteen and thirty months, available free from its copyright holders at mchatscreen.com, and it carries a structured follow-up interview that runs as a second stage when a screen comes back positive. It is a screen, not a diagnostic test 2.

Its evidence base is on a different scale as well. The M-CHAT-R/F validation ran in 16,071 toddlers and showed improved detection compared with the original M-CHAT alongside a reduced follow-up burden, with screened children diagnosed roughly two years earlier than the national median age of diagnosis 3.

Q-CHATM-CHAT-R/F
Items25, each on a 0-4 frequency scale 120, parent-report 2
Target age18-24 months 116-30 months 2
Result shapeone continuous total, 0-100, higher means more traits 1a positive or negative screen, with a structured follow-up interview for positives 2
Founding sample754 unselected toddlers, described as preliminary 116,071 toddlers 3

The two instruments were built to do different jobs, and the table is the reason a Q-CHAT total and an M-CHAT result are not two versions of the same answer.

The Q-CHAT publishes no cut-off score

The originating paper reports no cut-off, no sensitivity, and no specificity, and it is explicit about why. It describes itself as preliminary, saying it is not possible to report the full range of test properties at that stage, and it defers thresholds by name: future studies would test whether children scoring above a cut-off are at risk for a diagnosis, a test the authors state was beyond the scope of the initial study 1.

There was a comparison group. A group of 160 children with an autism spectrum condition scored higher on average, with a mean of 51.8 and a standard deviation of 14.3. The authors caution that this group was older than the eighteen-to-twenty-four-month band the instrument targets, and was diagnostically unverified, so the separation between the groups is preliminary rather than established 1.

There is no Q-CHAT score at which a child counts as screen-positive. The paper that created the instrument says testing for one was beyond its scope 1.

One more reading the paper does not support. Between two administrations, scores differed by an average of −0.33 with a standard deviation of 4.66 — reliability evidence, not a minimal detectable change 1. No minimal important difference is defined for the Q-CHAT either, so a shift in a child's total between two occasions has no published interpretation at all.

What that means for a parent holding a number

It means the number is not a result. With a group mean of 26.7 in unselected toddlers and 51.8 in an already-diagnosed group, and standard deviations wide enough that the two ranges overlap substantially, a single total in the thirties or forties sits in territory occupied by children from both groups 1. There is no band it falls into, because no bands were ever published.

That is why this page prints no interpretation of any Q-CHAT total, and it is worth knowing when an online version returns one. A site that converts a Q-CHAT sum into a risk level is doing something the instrument's own authors declined to do, using a threshold they said they had not tested.

Even for the screen that does publish its scoring, m-chat score bands are best read alongside the clinician who administered the questionnaire, because the follow-up stage exists precisely to change what a first-stage result means.

A questionnaire total changes nothing about a child that was not already true this morning. What changes things is a conversation with someone who can watch them.

The useful thing a parent can carry into that conversation is not a score but the observations underneath it: what happens when a name is called, what happens when something interesting appears, what a shared game looks like.

Where autism screening actually sits in a pediatric visit

Current US pediatric guidance recommends standardized autism-specific screening at eighteen and twenty-four months, layered on top of developmental surveillance that continues at every well-child visit, and holds that autism can be diagnosed as early as eighteen months 4. That is why the eighteen-month appointment carries the weight it does.

The instruments used for that job are named and validated ones, and the rule attached to every one of them is identical: a positive screen is not a diagnosis but an indication for further evaluation 5. Among the instruments named for that window is the m-chat autism screening. Broadband instruments such as the csbs infant-toddler checklist sit alongside them and answer a wider developmental question rather than an autism-specific one 5.

A parent reading about autism signs at 18 months is reading about the same window the schedule was designed around. The Q-CHAT, for all its research value, is not part of that schedule in routine US primary care.

Is universal screening settled? What the USPSTF actually said

In 2016 the US Preventive Services Task Force issued an I statement on autism screening, meaning it judged the evidence insufficient to weigh the balance of benefits and harms. The scope of that statement is narrow and is constantly misquoted: it covers universal screening of children aged eighteen to thirty months who have no signs of autism and about whom no concern has been raised 6.

Two things it is not. It is explicitly not a recommendation against screening. And it explicitly does not apply to a child who has signs, or about whom a parent or a clinician has raised a concern 6. Those two exclusions cover most of the children whose parents end up reading a page like this one.

Pediatric guidance in the United States continues to recommend the eighteen- and twenty-four-month screens 4, so the two documents sit beside each other rather than cancelling out. The disagreement is about universal screening as a population policy. It was never about whether a specific worry is worth acting on.

Common questions

On its own, very little. The total runs from zero to one hundred and higher means more autistic traits, but the paper that created the instrument published no cut-off, no sensitivity and no specificity. Without a validated threshold there is no score at which a child counts as screen-positive, so a total is a research measurement rather than a result about one child.

No. They share ancestry and an age range that partly overlaps, and nothing else. The M-CHAT-R/F is a twenty-item screen for sixteen to thirty months that returns a positive or negative result and carries a structured follow-up interview. The Q-CHAT is twenty-five items scored on a frequency scale and returns a continuous number with no published threshold.

Eighteen to twenty-four months. Worth noting that the comparison group in the original study, the children already diagnosed with an autism spectrum condition, was older than that band and diagnostically unverified. The authors say so themselves, which is part of why they describe the instrument's discriminant validity as preliminary rather than established.

Versions circulate that return one. That risk level is not coming from the instrument's own evidence, because no threshold was published or tested in the founding study. A site that converts a Q-CHAT sum into a category has supplied a line the authors specifically said was beyond the scope of their work.

Because it was designed to measure rather than to sort. A dimensional instrument is built to show how a trait varies across a population, and the first paper reported the distribution, the reliability and the internal consistency while explicitly deferring threshold-setting to later studies. That is a normal sequence in test development, not an oversight.

No, and the misquote is common. Its 2016 statement found the evidence insufficient to weigh benefits against harms for universal screening of eighteen- to thirty-month-olds with no signs and no raised concerns. It is not a recommendation against screening, and it does not apply at all to a child with signs or a concern already on the table.

Related

Say it back

How would you explain this to someone you love?

Two or three sentences, just as you’d say it. Gale reflects back what you focused on — a mirror, not a quiz.

Talk to a clinician

Gale can help you find a clinician in your state and request a visit.

Find care →

When a toddler concern needs an appointment rather than a questionnaire

  • —Loss of words, gestures, or social responsiveness a toddler previously had, at any age
  • —A toddler not yet using words whose hearing has never been formally tested
  • —A concern raised at the eighteen-month visit with no standardized screen completed
  • —A positive screen that has gone weeks without a referral being made

This page explains what a research questionnaire measures and what its founding study did and did not establish. It is not a screen, it produces no score, and it says nothing about any particular child. That is the work of a clinician who has evaluated them.

References

  1. 1.Allison C, Baron-Cohen S, Wheelwright S, et al. (2008). The Q-CHAT (Quantitative CHecklist for Autism in Toddlers): A Normally Distributed Quantitative Measure of Autistic Traits at 18–24 Months of Age: Preliminary Report. Journal of Autism and Developmental Disorders. 2008 Sep;38(8):1414–1425.. doi:10.1007/s10803-007-0509-7 ✓The Q-CHAT's structure (25 items, 5-point frequency scale from 0 to 4, half reverse-scored, summed to a 0-100 total, higher scores indicating more autistic traits) and its status as a quantitative revision of the CHAT for 18-24 months; its first-report psychometrics (near-normal distribution in 754 unselected toddlers, mean 26.7 SD 7.8; ASC group n=160, mean 51.8 SD 14.3; test-retest ICC 0.82 in 330 children over a mean 38-day interval; internal consistency 0.67 and 0.83; sex difference of 27.5 versus 25.8, Cohen's d 0.2); and the paper's explicit limits — no cut-off, no sensitivity or specificity, preliminary status, an older and diagnostically unverified ASC group, the test-retest pair difference of −0.33 (SD 4.66) as reliability evidence rather than an MDC, and no MCID.
  2. 2.Robins DL, Fein D, Barton M (2009). M-CHAT-R/F (Modified Checklist for Autism in Toddlers, Revised, with Follow-Up) — official screening instrument. mchatscreen.com (Robins, Fein & Barton, copyright holders). link ✓What the M-CHAT-R/F is: a 20-item parent-report screen for toddlers 16-30 months, freely available from its copyright holders, with a two-stage structured follow-up interview for positive screens, and that it is a screen rather than a diagnostic test.
  3. 3.Robins DL, Casagrande K, Barton M, Chen CA, Dumont-Mathieu T, Fein D (2014). Validation of the Modified Checklist for Autism in Toddlers, Revised With Follow-up (M-CHAT-R/F). Pediatrics. doi:10.1542/peds.2013-1813 ✓The M-CHAT-R/F validation in 16,071 toddlers, showing improved detection versus the original M-CHAT with reduced follow-up burden, and screened children diagnosed roughly two years earlier than the national median.
  4. 4.Hyman SL, Levy SE, Myers SM; AAP Council on Children With Disabilities, Section on Developmental and Behavioral Pediatrics (2020). Identification, Evaluation, and Management of Children With Autism Spectrum Disorder. Pediatrics (AAP clinical report). doi:10.1542/peds.2019-3447 ✓The recommendation of standardized autism-specific screening at 18 and 24 months alongside ongoing developmental surveillance, and that autism can be diagnosed as early as 18 months.
  5. 5.Centers for Disease Control and Prevention (2024). Clinical Screening for Autism Spectrum Disorder. CDC — Autism Spectrum Disorder (ASD), Healthcare Providers. linkThat primary-care autism screening uses named validated instruments including the M-CHAT-R/F and the CSBS DP Infant-Toddler Checklist, and that a positive screen is not a diagnosis but an indication for further evaluation.
  6. 6.U.S. Preventive Services Task Force (2016). Autism Spectrum Disorder in Young Children: Screening — Final Recommendation Statement. United States Preventive Services Task Force. link ✓The 2016 'I' statement that evidence is insufficient to assess the balance of benefits and harms of universal screening in children aged 18-30 months with no signs and no raised concerns, that it is not a recommendation against screening, and that it does not apply to children with signs or concerns.

6 sources, numbered by first appearance. General health information, not medical advice. AI-assisted editorial content — citations link their sources. Editorial policy

← Back to the health library