Selective Mutism Questionnaire: What the SMQ Measures
Updated: Oct 8
Author: Ukrainian Psychological Hub · Published: October 1, 2026 · Editorial Policy
The Selective Mutism Questionnaire (SMQ) is a parent- or caregiver-report measure designed to quantify how often a child speaks in situations where speech is expected. Its core purpose is to turn a highly context-dependent behavior—speaking freely in some settings and speaking little or not at all in others—into a structured profile that can support assessment, severity measurement, research, and monitoring over time.
The central scoring idea is simple: the caregiver rates speaking frequency across specific school, home or family, and public or social situations. In the widely used 17-item symptom scoring convention, responses run from 0 for never speaking to 3 for always speaking. Lower scores represent less speaking in the rated situations; higher scores represent more speaking. A raw total can range from 0 to 51 when the 17 items are summed, although some studies calculate domain means, transform scores, or reverse the direction for particular analyses.
An SMQ score is not a diagnosis of selective mutism. The questionnaire does not establish whether diagnostic criteria are met, determine why a child is not speaking, rule out language or communication problems, assess hearing, identify autism or another neurodevelopmental condition, distinguish every anxiety presentation, or replace a clinical history across settings. The American Speech-Language-Hearing Association's selective mutism guidance describes screening and comprehensive assessment as multi-method processes that can include caregiver and teacher measures, interviews, observation, speech-language assessment, and hearing screening.
This article explains exactly what the SMQ measures, how its scored items and domains work, what low and high scores can mean, what psychometric research supports, why published scoring conventions sometimes look different, how the scale is used to follow change, and where interpretation must stop. For the broader disorder itself, see Selective Mutism: Symptoms, Causes, Diagnosis, and Treatment.
What Is the Selective Mutism Questionnaire?
The SMQ was developed because selective mutism is unusually difficult to capture with a single office-based observation. A child may speak normally with family and become unable to speak at school, with unfamiliar adults, during medical appointments, or in other socially demanding situations. A measure therefore has to ask about behavior across environments rather than assume that one encounter represents the child's full communication pattern.
In the original 2008 development study, Bergman and colleagues introduced the SMQ as a 17-item parent-report measure of failure to speak associated with selective mutism. Their first study included 589 parents of children ages 3 to 11, and a second study included 66 children; the results supported a clinically meaningful factor structure, good internal consistency, and the ability of scores to distinguish a clinic-referred selective mutism group from children with other anxiety disorders.
An independent 2008 psychometric study examined the same 17-item measure in 102 children with selective mutism and 43 children without selective mutism, also ages 3 to 11. Its exploratory analysis produced a 13-item solution with three factors covering social situations outside school, school situations, and home or family situations. That finding is important because it shows that the SMQ has a stable contextual logic while the exact item structure reported in research is not perfectly identical across every study.
A later critical review of selective mutism assessment measures found the SMQ to be among the most frequently used standardized instruments in the literature. The review also emphasized that psychometric evidence across selective mutism measures remains developing and that diagnosis should draw on multiple sources of information rather than any one questionnaire.
What Does the SMQ Actually Measure?
The SMQ primarily measures the frequency of speaking behavior in specific contexts. It is best understood as a behavioral severity measure tied to the defining situation-specific communication pattern of selective mutism. The score answers a question such as, “How often is this child speaking in the situations represented by these items?” It does not directly answer, “How anxious is this child?” or “Does this child have selective mutism?”
That distinction matters because speaking frequency and subjective anxiety are related but not interchangeable. A child may experience intense anxiety and still produce some speech. Another child may appear calm while speaking remains inaccessible in a particular setting. The SMQ records the observable or caregiver-known speaking pattern; it does not provide a direct physiological or subjective anxiety-intensity index.
The scale is also context-sensitive. A single overall score compresses information, but the domain pattern can reveal whether difficulty is concentrated at school, extends into community or public situations, or is also visible in home and family contexts. For selective mutism, where setting is part of the clinical phenomenon, that contextual map can be more informative than a total score alone.
Speaking Frequency, Not Intent or Willingness
The SMQ does not measure whether a child is “choosing” not to speak, being oppositional, refusing to cooperate, or intentionally withholding speech. Those interpretations require evidence that a frequency rating cannot provide. Modern clinical descriptions of selective mutism focus on the persistent situation-specific failure to speak despite speech in other situations, not on an assumption of voluntary refusal.
This is one reason the language of the measure should remain behavioral. “Never,” “seldom,” “often,” and “always” describe frequency. They do not assign motive.
A Profile Across Environments
The three-context structure is one of the most useful features of the SMQ. The measure repeatedly returns to the same clinical fact from different directions: speech can be accessible in one environment and sharply restricted in another. This lets clinicians and researchers examine whether a child's communication profile changes by setting, communication partner, familiarity, or social demand.
A parent or caregiver, however, does not directly observe every school interaction. For that reason, a parent-rated SMQ should be interpreted alongside school information when school functioning is central. A teacher report, observation, or another school-specific measure can add information that the caregiver cannot reliably witness.
Who Completes the SMQ?
The SMQ is a parent- or caregiver-report questionnaire. The respondent rates the child's speaking behavior based on what they know about the child's communication across the situations named in the measure. That design is distinct from a child self-report of anxiety and distinct from a teacher-only rating scale.
The distinction between respondent types is clinically important. The caregiver may have detailed knowledge of home, relatives, community activities, appointments, and the child's reports about school, while a teacher has direct access to classroom and school speech. A discrepancy between informants can therefore be informative rather than automatically indicating that one report is wrong.
ASHA's assessment guidance recommends combining caregiver and teacher information with interviews, observations, and relevant speech-language and hearing assessment when selective mutism is suspected. The SMQ can contribute one structured layer to that process, but it does not replace the other layers.
What Age Is the SMQ For?
The strongest foundational validation evidence is pediatric. The original development study and the independent U.S. psychometric study both focused on children ages 3 to 11. A Spanish validation study included children ages 3 to 10, while a Dutch validation extended the age range to 3 to 17.
The Spanish validation study included 110 pairs of parents of Spanish-speaking children diagnosed with selective mutism and supported the translated measure's factor structure, reliability, and concurrent validity. The Dutch validation study included 303 children ages 3 to 17, including 106 in a selective mutism group and 197 controls, and reported strong reliability and the expected factor structure for the Dutch version.
These studies support use in children and, for the Dutch adaptation, adolescents. They do not establish the original SMQ as a validated adult diagnostic instrument. An adult may recognize their own experiences in SMQ-style situations, but using a pediatric parent-report measure informally in adulthood is not the same as having an adult-validated assessment tool. For the evidence on persistence beyond childhood, see Selective Mutism in Adults: What Is Known and What Remains Uncertain.
The 17 Scored Speaking Items and the Three Contexts
The SMQ is commonly described as a 17-item symptom measure. In a widely used scoring structure, those 17 speaking-frequency items are distributed across three domains: six school items, six home or family items, and five public or community items. Each item asks about a situation in which speaking may be expected, and the caregiver rates how often the child speaks.
The 2020 Oerbeck and colleagues study used six scored school items, six home items, and five public items. The authors described the broader questionnaire material as containing 32 items overall, while 17 items were used to calculate the three scored speaking subscales. This is one reason readers may encounter different numbers when looking at research papers, forms, or supplementary materials.
The 2008 independent factor-analysis paper also retained 13 of the 17 items in its statistical solution. That does not mean that every clinical or research user should simply delete four items. It means that the psychometric literature has examined alternative item structures. The correct scoring procedure is the one specified for the exact validated version or adaptation being used.
For practical interpretation, the three-domain concept is more stable than any assumption that every publication must present the instrument in an identical layout. The core idea remains measurement of context-specific speech across school, home or family, and social or public settings.
How SMQ Scoring Works
In the common 0-to-3 scoring convention, each scored speaking item is rated 0 for never, 1 for seldom, 2 for often, and 3 for always. When all 17 scored symptom items are simply summed, the theoretical total runs from 0 to 51. Under this direction, a higher score means more frequent speech across the rated situations, while a lower score means less frequent speech and therefore greater speech restriction.
Researchers do not always report the score in exactly the same way. Some calculate a mean for each domain. Oerbeck and colleagues divided scores by the relevant item counts so results could remain on the familiar 0-to-3 scale. Other analyses may sum the items. The Dutch validation reversed the score direction for its receiver-operating-characteristic analysis so that higher transformed scores represented more problems.
That creates a simple rule for reading research: never infer the meaning of “higher” or “lower” from the number alone. Check the scoring direction used in that paper or version. A raw 0-to-51 symptom total scored in the conventional direction is not directly interchangeable with a reverse-coded research score.
Missing responses add another complication. A total produced from all 17 completed items is straightforward; a mean or prorated score may use a different rule when an item is missing. Clinical services and researchers should follow the instructions attached to the version they administer rather than inventing a local scoring rule after the fact.
What Does a Low SMQ Score Mean?
With the common 0-to-3 speech-frequency scoring direction, a low score means the child is reported to speak less often in the situations covered by the questionnaire. Within a child already being assessed or treated for selective mutism, a lower score can indicate greater severity of the core speaking restriction.
It cannot, by itself, explain why the child is speaking less. Low speaking frequency can occur in a clinical picture that requires consideration of language proficiency, speech or language disorders, hearing, autism or other neurodevelopmental differences, other psychiatric conditions, unfamiliarity with the setting, cultural or linguistic transition, and contextual factors. The questionnaire records the behavior pattern; differential diagnosis interprets that pattern.
A low score also should not be treated as a universal statement about the child's communication ability. A child can have rich, age-appropriate speech in comfortable settings and a very low score in settings where speech becomes inaccessible. That contrast is central to selective mutism.
What Does a High SMQ Score Mean?
Under the conventional direction, a high score means more frequent speech in the rated contexts. In treatment or follow-up research, an increase over time can therefore represent broader access to speech across situations.
A high score does not prove that selective mutism is absent in every clinically important context. The total may hide a sharply restricted school domain, a single high-impact situation, or a setting not represented well by the items. Nor does a higher score automatically mean that distress, anticipatory anxiety, social avoidance, or functional impairment has disappeared.
Score interpretation is strongest when the total is read together with domain scores, concrete examples of functioning, and reports from people who observe the child in different settings.
Is There a Diagnostic SMQ Cutoff?
There is no single universal SMQ cutoff that can be applied across every language, population, age range, scoring convention, and clinical setting as if it were a diagnostic laboratory value.
The original U.S. development work showed that SMQ scores could discriminate groups, but that is different from establishing a universal diagnostic threshold. Oerbeck and colleagues reported reference values in a small sample of 32 typically developing children and 32 children with selective mutism; those group averages are useful research data, not a stand-alone rule for deciding whether an individual child has the disorder.
The Dutch validation performed receiver-operating-characteristic analyses and proposed a screening threshold for its Dutch sample after reversing the scoring direction. That is valuable population-specific validation. It should not be silently exported to a different language version, a different score direction, an adult population, or a different clinical setting.
A cutoff can optimize sensitivity and specificity for a particular research sample while still producing false positives and false negatives in real practice. The clinically meaningful question is therefore not “Did the score cross one magic number?” but “What does the score add to the full cross-setting assessment?”
What the Psychometric Evidence Shows
The evidence base supports the SMQ as a useful standardized measure, especially for describing selective-mutism-related speaking behavior and severity in children. It also supports caution about overclaiming what one score can establish.
The original development study found good internal consistency and meaningful validity evidence. Scores distinguished clinic-referred children with selective mutism from children with other anxiety disorders, and they correlated with clinically relevant dimensions. That supports the instrument's ability to capture something specific and useful about the selective mutism phenotype.
The independent measurement study found moderate-to-high internal consistency across its factors and total scale, along with convergent and incremental validity. It also found a 13-item solution rather than simply reproducing every item in the original structure. This is evidence for the underlying construct, while also showing that measurement structure is an empirical question rather than a fixed assumption.
The 2023 critical review placed those findings in a wider assessment landscape. It identified the SMQ, the Anxiety Disorders Interview Schedule, and the School Speech Questionnaire as the measures used most often in the reviewed literature, while concluding that psychometric evidence for selective mutism assessment tools was still emerging. The review therefore supports use of standardized measures within a broader clinical assessment, rather than elevating one questionnaire into a diagnostic authority.
Cross-language studies extend the evidence but do not erase population differences. The Spanish study supported the translated version in children ages 3 to 10. The Dutch study supported its adaptation across ages 3 to 17 and reported high internal consistency in its sample. Translation is therefore an empirical measurement issue: a translated wording becomes more trustworthy when its reliability, structure, and validity have been studied in the population where it will be used.
Can the SMQ Be Used to Track Treatment Progress?
Yes. Repeated SMQ measurement is one of the scale's important uses. Because the items describe specific speaking situations, the questionnaire can provide a baseline and then show whether speech becomes more frequent across the same types of contexts over time.
The Oerbeck study of typically developing and treated children found stable high speaking scores in the typically developing comparison group and significant increases in speaking among children with selective mutism after treatment. The authors concluded that the SMQ and the teacher-rated School Speech Questionnaire could be useful for assessing baseline severity and potential treatment effects.
A rising score should still be interpreted as a change in rated speaking behavior, not as proof that every treatment goal has been met. A child may speak more and still experience substantial distress. Speech may improve at home or in public before it generalizes to school. A total score can also improve while one high-impact context remains difficult.
For that reason, treatment monitoring is strongest when score change is combined with functional goals, school information, direct observation where feasible, and the child's own experience when developmentally appropriate.
SMQ Screening, Clinical Assessment, and Diagnosis Are Different Things
Screening asks whether a pattern is concerning enough to justify closer evaluation. Severity measurement asks how much of a target behavior or symptom is present. Clinical assessment integrates history, impairment, development, context, observation, informants, differential diagnosis, and sometimes standardized instruments. Formal diagnosis applies a recognized classification framework to the whole pattern.
The SMQ can contribute to screening, severity measurement, and monitoring. It can also be one component of a clinical assessment. It does not perform the final diagnostic step by itself.
For the wider assessment process, see How Is Anxiety Diagnosed? Assessment, History, and Differential Diagnosis. For the selective-mutism-specific diagnostic framework, the Hub's DSM-5-TR criteria explainer and ICD-11 diagnostic requirements explainer keep the classification criteria separate from questionnaire scoring.
The WHO ICD-11 clinical diagnostic manual is another reminder of the distinction. Diagnostic classification considers the persistent situation-specific speaking pattern, functional consequences, language competence, duration, and relevant exclusions. A parent rating total cannot substitute for that reasoning.
SMQ vs the School Speech Questionnaire
The SMQ and the School Speech Questionnaire (SSQ) are related but different tools. The SMQ is parent- or caregiver-rated and spans multiple environments. The SSQ is teacher-rated and focuses on school speaking behavior.
This difference is useful because selective mutism often creates an information asymmetry. The parent may see fluent speech at home and hear that the child does not speak at school; the teacher may see the school restriction directly but have little knowledge of home speech. Parent and teacher measures can therefore describe different slices of the same cross-setting pattern.
A teacher should not automatically be treated as the respondent for the parent-rated SMQ just because school is the most impaired setting. If a service wants a teacher measure, it should use an instrument and scoring procedure designed or validated for teacher report, or clearly document any nonstandard adaptation.
Selective Mutism, Multilingual Children, and Language Context
Language context can change the meaning of reduced speech. A child who is still acquiring the language expected at school may speak less because of limited proficiency, unfamiliarity, or normal adaptation. That is not equivalent to selective mutism.
ASHA specifically recommends gathering information about languages used in the home and other environments, the child's exposure and proficiency, and communication across settings. The diagnostic question is whether the situation-specific failure to speak is better explained by lack of knowledge of or comfort with the required language.
The SMQ can help map where speaking occurs, but it cannot independently determine why the pattern exists. In multilingual assessment, the clinician needs to understand the child's speech in each relevant language and setting. A low school score in a newly learned language carries different interpretive possibilities from a persistent pattern that appears across settings where the child is demonstrably proficient.
Validated translations also matter. Evidence from Spanish and Dutch adaptations shows that researchers can test whether translated versions retain acceptable measurement properties. A translation that has never been studied should not automatically inherit every reliability estimate or cutoff from the English-language research.
What the SMQ Does Not Measure
The SMQ does not directly measure a child's internal anxiety intensity. It does not establish whether silence is caused by fear of social evaluation, separation anxiety, language difficulty, sensory overload, autistic social-communication differences, trauma-related processes, depression, psychosis, neurological disease, medication effects, or another explanation. It does not assess hearing. It does not provide a complete speech-language evaluation.
It also does not convert selective mutism into a single continuum where every low-speaking child is assumed to have the disorder. The same observable behavior can arise through different mechanisms, and diagnosis depends on the pattern, duration, impairment, developmental history, language context, and differential assessment.
The questionnaire is therefore most powerful when it stays within its measurement territory: structured information about speaking behavior across situations.
Common SMQ Interpretation Mistakes
Mistake 1: Treating the Total Score as a Diagnosis
A total score summarizes rated speaking frequency. It cannot confirm selective mutism by itself. A child can score low for reasons that require a different clinical explanation, and a child with clinically significant selective mutism can have a pattern whose total score looks less dramatic because speech is available in many other settings.
Mistake 2: Assuming Every Published Score Runs in the Same Direction
The common raw scoring direction gives more points for more speech. Some research reverses the score for analysis so higher numbers represent greater difficulty. Always verify the scoring convention before comparing numbers.
Mistake 3: Using a Group Average as an Individual Cutoff
Research means from selective mutism and comparison groups describe samples. They are not automatically diagnostic thresholds for an individual child.
Mistake 4: Ignoring Domain Scores
A total can hide the defining contextual pattern. School, home or family, and public or social scores can show where speech is accessible and where it is restricted.
Mistake 5: Comparing Different Language Versions as If They Were Identical
Translation can alter item meaning, response behavior, and thresholds. Population-specific validation should guide interpretation when available.
Mistake 6: Assuming a Higher Score Means the Child Is No Longer Struggling
More speech is meaningful, but functioning also includes distress, avoidance, participation, relationships, learning, and the effort required to communicate. Improvement in speech frequency is one outcome, not the whole clinical picture.
How the SMQ Fits Into a Comprehensive Selective Mutism Evaluation
A careful evaluation starts with the pattern itself: where the child speaks, where speech becomes limited or absent, with whom, under what demands, for how long, and with what effect on school, social communication, healthcare, and everyday functioning.
A structured parent measure such as the SMQ can organize that information. Teacher data can describe school functioning. Interviews can clarify onset, developmental history, anxiety, family context, language exposure, and accommodations. Observation can show how communication changes with familiarity, demand, and support. Speech-language assessment can examine communication skills in conditions where the child is able to demonstrate them. Hearing screening helps exclude a hearing problem from the communication picture.
The evaluation also asks whether another condition better explains the speaking pattern or co-occurs with selective mutism. This is why the SMQ should be integrated into clinical reasoning rather than interpreted as an isolated test.
The broader diagnostic owner for the cluster is Anxiety Disorders: Symptoms, Types, Diagnosis, and Treatment. Within the Selective Mutism branch, age-specific pages cover children, teens and adults without turning one pediatric questionnaire into a lifespan diagnostic instrument.
Frequently Asked Questions
What does SMQ stand for?
SMQ stands for Selective Mutism Questionnaire. It is a standardized caregiver-report measure used to rate how often a child speaks in specific situations associated with selective mutism.
How many scored items are in the SMQ?
The core symptom scale is commonly described as 17 scored speaking-frequency items. Some study materials contain additional contextual questions, and psychometric research has also examined reduced item solutions. Use the instructions for the exact version being administered.
How is the SMQ scored?
In the widely used raw scoring convention, each scored item is rated from 0 to 3: never, seldom, often, and always. When 17 items are summed, the theoretical raw total is 0 to 51. Higher raw scores mean more speaking. Some studies use means, transformed scores, or reverse coding, so the study's scoring method must be checked.
What is a normal SMQ score?
There is no single universal “normal” number for every age, language, setting, and version. Research provides group reference data, but an individual score should be interpreted in the context of the validated version, domain pattern, functioning, and the purpose of assessment.
What SMQ score indicates selective mutism?
No universal raw SMQ cutoff can diagnose selective mutism. Some language-specific validation studies have evaluated screening thresholds, but those thresholds depend on the scoring convention and study population. Diagnosis requires clinical assessment.
Can the SMQ diagnose selective mutism?
No. It can provide standardized evidence about speaking frequency and severity, support screening, and contribute to assessment. Diagnosis requires the full clinical pattern, impairment, duration, language context, and differential diagnosis.
What age range is the SMQ validated for?
The foundational U.S. validation studies focused on ages 3 to 11. A Spanish validation included ages 3 to 10, and a Dutch validation included ages 3 to 17. Evidence for the original parent-report SMQ should not be assumed to extend unchanged to adults.
Is the SMQ completed by parents or teachers?
The SMQ is a parent- or caregiver-report measure. The School Speech Questionnaire is a separate teacher-rated instrument. Multi-informant assessment can be especially useful because speech may differ sharply between home and school.
Can the SMQ be used during treatment?
Yes. Repeated scores can help track changes in speaking frequency across settings, and research supports its use as a treatment-outcome measure. Score change should be interpreted alongside functioning and other clinical information.
Can the SMQ be used for adults?
The original measure is pediatric and parent-rated. Its foundational validation does not establish it as a validated adult diagnostic test. Adult selective mutism requires adult-appropriate clinical assessment, and the adult-specific evidence base remains limited.
Does the SMQ measure anxiety?
It measures speaking behavior associated with selective mutism rather than providing a direct measure of subjective anxiety intensity. Anxiety can be clinically relevant to the speaking pattern, but the SMQ's scored core is about how often the child speaks in specified situations.
The Bottom Line
The Selective Mutism Questionnaire measures context-specific speaking frequency. Its core scored items organize caregiver observations across school, home or family, and public or social situations, creating a structured estimate of how available speech is across environments.
The evidence supports the SMQ as a useful pediatric assessment and severity tool with established psychometric foundations and several cross-language validations. It is also widely used to monitor change. At the same time, the research does not justify turning one raw score into a universal diagnostic cutoff or treating the measure as a substitute for multi-informant clinical assessment.
The most accurate interpretation is therefore straightforward: the SMQ tells us how often a child is reported to speak in the situations it samples. Clinical assessment determines what that pattern means.
Related Articles
References
American Speech-Language-Hearing Association. (n.d.). Selective Mutism [Practice Portal]. https://www.asha.org/Practice-Portal/Clinical-Topics/Selective-Mutism/.
Bergman, R. L., Keller, M. L., Piacentini, J., & Bergman, A. J. (2008). The development and psychometric properties of the Selective Mutism Questionnaire. Journal of Clinical Child & Adolescent Psychology, 37(2), 456–464. https://doi.org/10.1080/15374410801955805.
Letamendi, A. M., Chavira, D. A., Hitchcock, C. A., Roesch, S. C., Shipon-Blum, E., & Stein, M. B. (2008). Selective Mutism Questionnaire: Measurement structure and validity. Journal of the American Academy of Child & Adolescent Psychiatry, 47(10), 1197–1204. https://doi.org/10.1097/CHI.0b013e3181825a7b.
Oerbeck, B., Overgaard, K. R., Bergman, R. L., Pripp, A. H., & Kristensen, H. (2020). The Selective Mutism Questionnaire: Data from typically developing children and children with selective mutism. Clinical Child Psychology and Psychiatry, 25(4), 754–765. https://doi.org/10.1177/1359104520914695.
Olivares-Olivares, P. J., Rosa-Alcázar, Á., Nuñez, R., & Olivares, J. (2021). Psychometric properties of the Selective Mutism Questionnaire in Spanish children. International Journal of Clinical and Health Psychology, 21(3), 100249. https://doi.org/10.1016/j.ijchp.2021.100249.
Rodrigues Pereira, C., Ensink, J. B. M., Güldner, M. G., de Jonge, M. V., Lindauer, R. J. L., & Utens, E. M. W. J. (2023). Diagnosing selective mutism: A critical review of measures for clinical practice and research. European Child & Adolescent Psychiatry, 32(10), 1821–1839. https://doi.org/10.1007/s00787-021-01907-2.
Rodrigues Pereira, C., Ensink, J. B. M., Güldner, M. G., Kan, K. J., de Jonge, M. V., Lindauer, R. J. L., & Utens, E. M. W. J. (2024). The validation of the Selective Mutism Questionnaire for use in the Dutch population. Child Psychiatry & Human Development, 55, 82–93. https://doi.org/10.1007/s10578-022-01387-8.
World Health Organization. (2024). Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders. https://www.who.int/publications/i/item/9789240077263.