The Strengths and Difficulties Questionnaire (SDQ) is a 25-item behavioural and emotional screening measure for children and young people aged 2 to 17 years (Goodman, 1997). It assesses four problem domains (emotional symptoms, conduct problems, hyperactivity/inattention, and peer relationship problems) together with a fifth prosocial behaviour scale.
An impact supplement (Goodman, 1999) extends the questionnaire with items on chronicity, distress, interference with everyday life, and burden on others, administered only when the respondent reports the child has difficulties on the supplement’s screening item. The SDQ is available in parent-report, teacher-report, and youth self-report formats across age bands spanning early childhood through adolescence, with follow-up versions for tracking change over time. It is one of the most widely used brief measures of child and adolescent mental health worldwide, applied in clinical, educational, and research settings.
Behavioural and emotional difficulties in childhood and adolescence are common, affecting approximately 10% of the general population at any given time (Goodman et al., 2000; Polanczyk et al., 2015). Early identification of these difficulties is associated with better outcomes, yet many children with significant problems are not identified until difficulties become entrenched. Brief, validated screening instruments play a key role in supporting early identification across community, educational, and clinical settings. The SDQ is one of the most widely used instruments for this purpose, with translations in over 80 languages and extensive use in epidemiological surveys, clinical services, and outcome monitoring programmes internationally. Extensive support materials are available on the SDQ website: https://www.sdqinfo.org/
The SDQ comprises five subscales (25 items) included in all versions, with wording slightly modified depending on the respondent.
Standard versions of the SDQ on NovoPsych include an Impact Supplement, which first assesses whether the respondent thinks the child has any difficulties, and if so, asks how long they have been present as well as whether they are causing distress, interfering with everyday functioning (home life, friendships, classroom learning, leisure activities), or placing a burden on others.
The Impact Supplement adds important clinical context by distinguishing between children with elevated symptom scores who are functioning well and those whose difficulties cause meaningful impairment.
Unlike the standard version, which asks about behaviour over the past 6 months, the follow-up versions ask about ‘the last month’ to aid in detecting change over time. Follow-up versions, which also include the Impact Supplement, omit the chronicity of problems item and include two additional items asking the respondent to rate the change in the child’s problems since coming to the service and how helpful coming to the service has been.
Fourteen versions of the SDQ are available on NovoPsych and reflect the ‘English (Australian)’ translation and normative thresholds. All versions in NovoPsych include the Impact Supplement and are available in both the standard and follow-up forms. They cover three informant types and multiple age bands.
The SDQ is designed as a first-stage screening instrument and is not intended to be the sole basis for clinical diagnosis (Goodman et al., 2000). Elevated scores indicate areas of concern that warrant further assessment.
Common clinical uses include:
All versions of the SDQ in NovoPsych follow the same scoring structure and produce scores at three levels: Total Difficulties score, five subscale scores, and an Impact score. A higher score indicates greater difficulty in all areas except Prosocial Behaviour, where a higher score indicates greater prosocial strength.
Each informant version is available as a Standard form (typically for initial/baseline administration; recall window: past six months or current school year) and a Follow-Up form (typically for re-administration to track change; recall window: past month).
The Total Difficulties score (20 items; range 0-40) reflects the overall level of behavioural and emotional difficulties and is the sum of the four difficulty subscales (5 items each; items marked (R) are reverse-scored):
A fifth subscale, Prosocial Behaviour (items 1, 4, 9, 17, 20; range 0-10), assesses positive social attributes. This is a strength scale, not included as part of the Total Difficulties score, with reversed interpretation: low scores indicate fewer prosocial behaviours and high scores indicate prosocial strengths.
The Impact score is calculated from the Impact Supplement, which is administered on both Standard and Follow-up versions. The Supplement begins with a gateway question asking whether the respondent perceives the child to have any difficulties in emotions, concentration, behaviour, or social functioning.
When the respondent answers “No”: The Supplement is not administered, and an Impact score is not calculated. The results table shows “N/A – No difficulties reported” for the Impact row, and the interpretive narrative notes that the respondent reported no overall difficulties.
When the respondent answers “Yes”: The Impact score is the sum of one distress item and multiple interference items. Each item uses a four-point response scale: Not at all (0), Only a little (0), A medium amount (1), and A great deal (2). As such, only responses indicating at least “A medium amount” of distress or interference contribute to the score.
The number of interference items and the resulting score range vary by informant version:
Impact Supplement Items not included in the Impact score. Two items appear in the Impact Supplement but do not contribute to the Impact score.
Wording: For the 2-4 age band, three items have different wording from the 4-17 versions: item 18 (“Often argumentative with adults” replaces “Often lies or cheats”), item 22 (“Can be spiteful to others” replaces “Steals from home, school or elsewhere”), and item 21 is softened (“Can stop and think things out before acting” replaces “Thinks things out before acting”). Scoring is otherwise identical. Separate normative banding thresholds apply to the 2-4 age band versus 4-17.
Follow-Up Versions: Follow-Up versions ask respondents to rate items over the past month rather than the past six months/school year used on the Standard versions, omit the duration (chronicity) item, and instead include two additional questions asking the respondent to rate how the child’s problems have changed since first coming to the service and how helpful it has been to come to the service. The respondent’s verbatim ratings on these two items are quoted in the interpretive narrative.
All scores are classified using the author-endorsed four-band system (see https://www.sdqinfo.org/):
The Prosocial Behaviour subscale requires separate interpretive attention because it operates in the opposite direction to the difficulty subscales. A high Prosocial score is positive, indicating that the child is considerate, helpful, and kind. A low Prosocial score may indicate difficulties with social engagement, cooperation, or helping behaviour, and warrants further exploration. The Prosocial subscale is not included in the Total Difficulties score, so a child can have a low Total Difficulties score (few problems) but still have a Prosocial score that warrants attention.
The Impact score contextualises the symptom scores by indicating whether identified difficulties are causing functional impairment. For example, a child may score in the High or Very High range on one or more difficulty subscales but have a Close to Average Impact score, suggesting that the difficulties are present but not yet causing significant distress or interference. Conversely, a Slightly Raised symptom profile with a High Impact score may indicate that even moderate difficulties are having a disproportionate effect on the child’s functioning. The Impact score should be interpreted alongside the symptom subscales rather than in isolation.
All 14 versions of the SDQ include an optional free-text item (Q26) that asks “Do you have any other comments or concerns?” This item provides the respondent with a place to add qualifying information, describe difficulties that the rating-based items did not capture, or raise concerns that the clinician may wish to explore in follow-up discussion. It is non-required and is not part of any scored scale. It does not contribute to Total Difficulties, the Impact score, any of the five subscales, or to the four-band categorisation. When the respondent enters text, the response is presented within the interpretive text and is reproduced verbatim in an “Open-Ended Text Response” section at the end of the Client Responses section of the report. The free-text response is presented for clinical context only.
First administration. On a first administration for a given informant, the report displays a Total Difficulties bar chart alongside an Impact bar chart (side by side), each with background shading indicating the four-band boundaries. When the gateway question is answered “No”, the Impact Supplement is not administered, and only the Total Difficulties chart is shown. A “Subscale Scores” bar chart then displays all five subscale scores on a single chart, with the four difficulty subscales using the standard four-band shading and the Prosocial subscale using reversed band shading on the same chart, reflecting its distinct interpretive role as a strength rather than a difficulty. This layout enables clinicians to compare the relative contribution of each domain at a glance. On Standard versions, a caption beneath the Impact chart notes how long the difficulties have reportedly been present.
Repeat administration. When more than one administration exists for the informant whose report is being generated (regardless of whether it is a Standard or Follow-Up version), multi-administration line plots replace the first-administration bar charts: a Total Difficulties line plot alongside an Impact line plot (side by side), a Prosocial Behaviour line plot, and a 2×2 grid of the four difficulty subscale line plots on a subsequent page. All plots display raw scores across time points with the four-band severity threshold shading in the background, enabling clinicians to track change trajectories.
Standard and Follow-up administrations from the same informant are displayed as a single time series over time. This combines the initial administration (recall: past six months/school year) with subsequent follow-up administrations (recall: past month) on a single plot. Raw score comparisons remain valid per Goodman scoring conventions, but the recall window difference between Standard and Follow-up administrations should be considered when interpreting change.
Band changes (for example, a subscale moving from High to Slightly Raised) and band stability across administrations (for example, “remains in the Close to Average range”) are noted in the interpretive text. For the Impact plot, administrations at which the respondent answered “No” to the gateway question are marked with an x to distinguish them from administrations at which a measured Impact score of zero was recorded.
Change descriptions in the interpretive narrative are referenced to two anchors at the Total Difficulties and Impact level: (i) the immediately prior administration, to support short-term clinical decision-making, and (ii) the first administration (baseline), to convey cumulative change. Both comparisons are presented when more than two administrations exist. Subscale-level change descriptions (Emotional Symptoms, Conduct Problems, Hyperactivity, Peer Problems, Prosocial Behaviour) reference the immediately prior administration only.
Age band transitions. When a client has moved between the 2-4 and 4-17 age bands (e.g., the child is age 4 at an initial administration but then turns 5 years old upon a subsequent administration), a dashed vertical transition line marks the administration at which the age band changed, and a caption above the plots explains the convention. Raw score comparisons across administrations remain valid, but band labels are interpreted against each administration’s own age-specific thresholds. When there are more than 20 administrations, each plot is split into two halves (newer half first) so that individual data points remain legible.
Cross-informant reporting. Combining multiple informants’ perspectives improves screening accuracy and surfaces context-specific differences in the child’s presentation (Goodman et al., 2004), and is a core clinical use case for the SDQ. When the same client has been assessed by more than one informant (for example, by both a parent and a teacher), the report displays the current informant’s results first (results table, charts, interpretive narrative, and client responses), then appends one supplementary section for each additional informant at the end of the report. A single administration by a supplementary informant is rendered as a stand-alone results table on its own page; multiple administrations are rendered as plots of results over time (Total, Impact, Prosocial, and the four difficulty subscales). Each supplementary section is evaluated against that informant’s own normative thresholds.
Follow-up Supplement. On Follow-up versions, the respondent’s ratings of change in the child’s problems since first coming to the service (“Much better” to “Much worse”) and how helpful coming to the service has been (“A great deal” to “Not at all”) are quoted in the interpretive narrative and listed under the Follow-Up Supplement in the client responses section at the end of the report. These ratings are categorical single-item responses rather than scored scales, so they are not shown in the results table or the charts.
The SDQ was developed by Goodman (1997) as a brief alternative to the Rutter questionnaires for assessing behavioural and emotional difficulties in children. The initial 25 items were constructed to cover five domains (Emotional Symptoms, Conduct Problems, Hyperactivity/Inattention, Peer Problems, and Prosocial Behaviour), with five items per domain selected on the basis of factor analyses and clinical relevance. In the original validation study, 403 children aged 4-16 were assessed in dental and psychiatric clinic settings. The SDQ demonstrated comparable screening accuracy to the longer Rutter questionnaires, with ROC AUC values of .87 for parent-report and .85 for teacher-report (Goodman, 1997).
The “extended” version, incorporating the Impact Supplement, was introduced by Goodman (1999) to improve the identification of clinically significant difficulties. The Impact Supplement assesses overall distress, chronicity, interference across functional domains, and burden to others. In a community sample of 467 children and a clinical sample of 232 children (ages 5-15), impact caseness was a stronger predictor of clinical status than symptom caseness alone (Goodman, 1999).
Internal consistency for the SDQ subscales has been assessed across multiple large samples. In the combined British Child and Adolescent Mental Health Surveys (B-CAMHS 1999 and 2004; N = 18,222 parents, 14,263 teachers, 7,678 youth, ages 5-16), Cronbach’s alpha coefficients were almost all between .65 and .85 across subscales and informants (Goodman et al., 2010). The two weakest subscales were Peer Problems as reported by parents (alpha = .58) and youth (alpha = .44). In an Australian community sample (N = 1,359, ages 4-9, parent-report), alphas ranged from .59 (Peer Problems) to .80 (Hyperactivity), with the Total Difficulties score demonstrating strong internal consistency (alpha = .82) and the Impact score adequate reliability (alpha = .79) (Hawes & Dadds, 2004). A systematic review of 48 studies (N = 131,223; ages 4-12) confirmed satisfactory internal consistency for the parent and teacher versions, with the teacher version generally demonstrating stronger reliability (Stone et al., 2010).
Test-retest reliability has been assessed in a community sample of 34 parents over a 3-4 week interval, yielding intraclass correlation coefficients of .85 for Total Difficulties and .63 for the Impact rating (Goodman, 1999). In the Australian sample, 12-month stability correlations ranged from .61 (Peer Problems) to .77 (Hyperactivity), with Total Difficulties at .77 and Impact at .63 (Hawes & Dadds, 2004). Given the 12-month interval, these values reflect both measurement stability and genuine developmental change.
The hypothesised five-factor structure has been evaluated across multiple samples using confirmatory factor analysis. In the B-CAMHS data, the five-factor first-order model achieved acceptable fit for parent (CFI = .90, TLI = .95, RMSEA = .05), teacher (CFI = .92, TLI = .97, RMSEA = .08), and youth data (CFI = .86, TLI = .90, RMSEA = .06) after minor modifications (Goodman et al., 2010).
The five-factor structure has been replicated internationally. A systematic review found that 15 of 18 studies confirmed the hypothesised structure using exploratory or confirmatory methods (Stone et al., 2010). In the Australian parent-report sample (N = 1,359, ages 4-9), the five-factor solution was supported for both boys and girls, with most items loading moderately to strongly on their predicted factors (Hawes & Dadds, 2004). However, cross-loadings for the item “Generally obedient” (loading onto Prosocial rather than Conduct) have been consistently observed across studies.
The SDQ has demonstrated good convergent validity against diagnostic interview measures. Using the Development and Well-being Assessment (DAWBA) as a criterion, all five subscales monotonically predicted their corresponding DAWBA-derived diagnoses, but multitrait-multimethod analyses showed poor discriminant validity between the behavioural, hyperactivity, and prosocial subscales in low-risk samples (Goodman et al., 2010), particularly at lower scores. In the Australian sample, SDQ subscale scores correlated significantly with corresponding DISCAP diagnostic interview severity ratings, ranging from .33 (Emotional Symptoms with internalising disorders) to .51 (Hyperactivity with hyperactivity diagnoses) (Hawes & Dadds, 2004).
The SDQ also shows good convergence with the Child Behavior Checklist (CBCL). In a head-to-head comparison of 132 children aged 4 to 7 from London psychiatric and dental clinics, equivalent SDQ and CBCL scales correlated strongly (Total Difficulties with Total Problems r = .87; Conduct with Externalising r = .84; Emotional with Internalising r = .74; Hyperactivity with Attention Problems r = .71; Peer with Social Problems r = .59) and ROC areas under the curve were approximately .95 for Total and Externalising scores on both questionnaires, with no significant difference between the two. The SDQ outperformed the CBCL for detecting inattention and hyperactivity against the Parental Account of Childhood Symptoms interview (SDQ r = .43 vs CBCL r = .15, p < .02) and was at least as good for internalising and externalising problems (Goodman & Scott, 1999).
The SDQ’s screening accuracy has been evaluated in several large studies. In a UK community sample (N = 7,984, ages 5-15), a multi-informant SDQ approach achieved sensitivity of 63.3% (95% CI: 59.7-66.9%) and specificity of 94.6% (95% CI: 94.1-95.1%) for any psychiatric disorder, with a positive predictive value of 52.7% and negative predictive value of 96.4% (Goodman et al., 2000). Sensitivity was higher for conduct disorders (76.2%), hyperactivity disorders (86.1%), and depression (74.6%), but lower for specific phobias (30.9%) and separation anxiety (45.5%).
In a study of children in the child welfare system (N = 539, ages 5-17), the SDQ achieved sensitivity of 85% and specificity of 80% (Goodman et al., 2004). In the Australian sample, children scoring in the top 10% on Total Difficulties had an odds ratio of 11.7 for any DSM-IV diagnosis compared to the remaining 90% (Hawes & Dadds, 2004).
The SDQ uses a four-band categorisation system to classify scores:
This system was developed by the SDQ authors (Goodman, 2015a) based on the empirical distribution of scores in the B-CAMHS surveys (Goodman et al., 2010), with thresholds calibrated so that approximately 80% of children fall in the Close to Average band, 10% in Slightly Raised, 5% in High, and 5% in Very High. It replaced an earlier three-band system (Normal, Borderline, Abnormal) that used an 80%/10%/10% split (Goodman, 1997).
Australian vs. UK Threshold Bands. The thresholds applied in NovoPsych’s SDQ scoring is the English (Australian) author-published thresholds (Goodman, 2015a, 2015b), which are numerically identical to the English (UK) thresholds with one exception: the UK revision corrects the teacher Impact upper bound in the ‘Very High’ band from 3-10 to the mathematically correct 3-6. Because the teacher Impact score has a ceiling of 6, this correction is documentary rather than substantive.
Australian vs. US Threshold Bands. Separate US thresholds are also published by the SDQ authors. The US Total Difficulties cut-points are approximately 1 to 2 points below the UK values at each band boundary, and the US Prosocial cut-points are approximately 1 point below the UK values at the lower two bands; the remaining subscale cut-points are broadly similar. The implication is that, all else equal, the US thresholds classify slightly more children into the elevated bands than the UK and Australian thresholds do, reflecting the lower mean scores observed in the US normative sample rather than any difference in the underlying construct.
Provisional Thresholds for Ages 2-4. Separate thresholds are provided for each informant type (parent, teacher, self-report), reflecting the different distributional properties of scores across informants. The 2-4 age band uses separate provisional thresholds, calibrated by the SDQ authors from a sample of 11,592 parent-completed SDQs for children aged 2-3 in Scotland (1,353 in Dumfries and 10,239 aged 30 months in Glasgow) and 10,004 teacher-completed SDQs for children aged 4 (Local Authority and Partnership nursery staff in Glasgow, 2012-2014). These data are author-published rather than peer-reviewed; the provisional thresholds are documented in Goodman (2015b), and the underlying sample is described at https://www.sdqinfo.org/norms/UK3yearNorm.html. The provisional banding for 2-4 year-olds targets a slightly different distributional split (approximately 80% Close to Average, 12% Slightly Raised, 4% High, 4% Very High), reflecting estimated prevalence rates in this younger age group. The authors describe these thresholds as provisional, pending the collection of larger, nationally representative data for this age group.
Threshold Validation Against Australian Data. The four-band system does not stratify by sex. This is consistent with effect size analyses computed from Hawes and Dadds (2004) means and standard deviations (N = 1,359, ages 4-9), where all sex differences fell below the team’s threshold for meaningful stratification (|d| >= 0.5). The largest sex effect was on the Prosocial subscale (Cohen’s d = 0.38, girls higher than boys, weighted across age bands), with Hyperactivity a close second (Cohen’s d = -0.36, boys higher than girls). All other subscales showed |d| <= 0.30.
Cross-verification of the 4-17 parent thresholds against Australian community data (Hawes & Dadds, 2004) confirmed that the author’s thresholds fall at plausible points in the distribution. Under a normal approximation using the weighted means and SDs from Hawes & Dadds (2004) Table 1, the Total Difficulties Close to Average boundary (raw score 13) corresponds to approximately the 80th percentile of the Australian sample, matching the author’s intended target. Deviations for subscales with highly skewed distributions (Conduct, Peer Problems, Impact) are expected, given that the original thresholds were derived from empirical distributions rather than parametric estimates. Effect sizes and percentile mappings reported in this section were computed by NovoPsych using the means and standard deviations from Hawes and Dadds (2004), Table 1.
A four year old sits at the point where two versions overlap, so the practical question is which setting the child is in rather than their age alone. The usual guide is school attendance: use the 4 to 10 version for a four year old who has started school, and the 2 to 4 version for one who has not yet started. This matters because the two versions are not identical; a few items are worded differently and each uses its own normative cut-offs, so matching the version to the child’s stage gives the most accurate result. If a child is right on the cusp, choose the version that best reflects their day-to-day environment and use that same version for later administrations so scores remain comparable.
Each informant sees the child in a different context, so the versions are designed to complement one another rather than to be used in isolation. Combining a parent and a teacher rating, together with a young person’s self-report where appropriate, improves screening accuracy beyond any single perspective and often reveals where difficulties are situation-specific. Self-report is available only from age eleven, reflecting the reading and self-reflection the items require, while parent and teacher versions span the full age range. When more than one informant has completed the SDQ for the same child, NovoPsych brings their results together in a single report so the perspectives can be compared side by side.
Use the standard version for the first or baseline administration; it asks about behaviour over the past six months (or the school year for teachers) and gives a stable picture of the child’s difficulties. The Follow-up version is for re-administration once the child is engaged with a service: it asks instead about the past month, so it is more sensitive to recent change, and it adds two short questions on whether things have improved and how helpful attending the service has been.
The Impact Supplement opens with a gateway question asking whether the respondent thinks the child has any difficulties at all. If they answer “no”, the Impact score is not calculated and the report records this, even when one or more subscale scores fall in an elevated band. This pattern is clinically informative rather than contradictory: it can mean the behaviours are present but not yet causing the respondent concern, that the respondent has a high tolerance for the behaviours, or that they are minimising. It is worth exploring gently in conversation rather than assuming the rating is a mistake.
Disagreement between informants is common and usually meaningful rather than a sign that one of them is wrong. Children often behave differently at home and at school, so a difference can point to a situation-specific difficulty, for example conduct concerns that show up in the classroom but not at home, or anxiety that a parent sees but a teacher does not. Read each informant’s profile against that informant’s own norms and treat the discrepancy itself as a clinical clue about where and when the child struggles. Surfacing these context differences is one of the SDQ’s real strengths.
The SDQ is a broad first-stage screener: it scans emotional symptoms, conduct, hyperactivity and inattention, peer relationships, and prosocial strengths in a single brief questionnaire, which makes it well suited to intake and routine outcome monitoring. When the SDQ flags a specific area, a narrower measure can give you the depth a broadband screen is not designed to provide. For anxiety and depression in particular, the Spence Children’s Anxiety Scale (SCAS) or the Revised Child Anxiety and Depression Scale (RCADS) offer disorder-level detail, and the Depression Anxiety Stress Scales Youth version (DASS-Y) tracks internalising distress in young people. Used together, the SDQ identifies where to look and the focused measures help clarify what is going on.
Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38(5), 581–586. https://doi.org/10.1111/j.1469-7610.1997.tb01545.x
Goodman, A. (2015a, May 11). Scoring the Strengths & Difficulties Questionnaire for age 4-17 [English (Australian) version]. SDQ Information. https://www.sdqinfo.org/
Goodman, A. (2015b, May 11). Scoring the Strengths & Difficulties Questionnaire for 2-4 year olds [English (Australian) version]. SDQ Information. https://www.sdqinfo.org/
Goodman, A., & Goodman, R. (2009). Strengths and Difficulties Questionnaire as a dimensional measure of child mental health. Journal of the American Academy of Child and Adolescent Psychiatry, 48(4), 400–403. https://doi.org/10.1097/CHI.0b013e3181985068
Goodman, A., Lamping, D. L., & Ploubidis, G. B. (2010). When to use broader internalising and externalising subscales instead of the hypothesised five subscales on the Strengths and Difficulties Questionnaire: Data from British parents, teachers and children. Journal of Abnormal Child Psychology, 38, 1179–1191. https://doi.org/10.1007/s10802-010-9434-x
Goodman, R., & Scott, S. (1999). Comparing the Strengths and Difficulties Questionnaire and the Child Behavior Checklist: Is small beautiful? Journal of Abnormal Child Psychology, 27(1), 17–24. https://doi.org/10.1023/A:1022658222914
Goodman, R. (1999). The extended version of the Strengths and Difficulties Questionnaire as a guide to child psychiatric caseness and consequent burden. Journal of Child Psychology and Psychiatry, 40(5), 791–799. https://doi.org/10.1111/1469-7610.00494
Goodman, R., Ford, T., Corbin, T., & Meltzer, H. (2004). Using the SDQ multi-informant algorithm to screen looked-after children for psychiatric disorders. European Child & Adolescent Psychiatry, 13(Suppl. 2), II/25–II/31. https://doi.org/10.1007/s00787-004-2005-3
Goodman, R., Ford, T., Simmons, H., Gatward, R., & Meltzer, H. (2000). Using the Strengths and Difficulties Questionnaire to screen for child psychiatric disorders in a community sample. British Journal of Psychiatry, 177, 534–539. https://doi.org/10.1192/bjp.177.6.534
Hall, C. L., Guo, B., Valentine, A. Z., Groom, M. J., Daley, D., Sayal, K., & Hollis, C. (2019). The validity of the Strengths and Difficulties Questionnaire (SDQ) for children with ADHD symptoms. PLOS ONE, 14(6), e0218518. https://doi.org/10.1371/journal.pone.0218518
Hawes, D. J., & Dadds, M. R. (2004). Australian data and psychometric properties of the Strengths and Difficulties Questionnaire. Australian and New Zealand Journal of Psychiatry, 38, 644–651. https://doi.org/10.1111/j.1440-1614.2004.01427.x
Kankaanpää, R., Töttö, P., Punamäki, R.-L., & Peltonen, K. (2023). Is it time to revise the SDQ? A psychometric evaluation of the Strengths and Difficulties Questionnaire. Psychological Assessment, 35(12), 1069–1084. https://doi.org/10.1037/pas0001265
Khawaja, N. G., & Dhushyanthakumar, L. (2020). Strengths and Difficulties Questionnaire – teacher: Investigating its factor structure and utility with culturally and linguistically diverse students. Journal of Psychologists and Counsellors in Schools, 30, 43–57. https://doi.org/10.1017/jgc.2019.23
Mathai, J., Anderson, P., & Bourne, A. (2003). Use of the Strengths and Difficulties Questionnaire as an outcome measure in a child and adolescent mental health service. Australasian Psychiatry, 11(3), 334–337. https://doi.org/10.1046/j.1440-1665.2003.00544.x
Polancyzyk, G., Salum, G. A., Sugaya, L. S., Caye, A., & Rohde, L. A. (2015). Annual Research Review: A meta-analysis of the worldwide prevalence of mental disorders in children and adolescents. Journal of Child Psychology and Psychiatry, 56(3), 345-365. https://doi.org/10.1111/jcpp.12381
Stone, L. L., Otten, R., Engels, R. C. M. E., Vermulst, A. A., & Janssens, J. M. A. M. (2010). Psychometric properties of the parent and teacher versions of the Strengths and Difficulties Questionnaire for 4- to 12-year-olds: A review. Clinical Child and Family Psychology Review, 13, 254–274. https://doi.org/10.1007/s10567-010-0071-2
