Introduction
Methods
Search Strategy and Selection for Identification of Studies
-
The target population had to be 4–12 years of age. The age was above the range in 27 out of 48 studies. Of those studies, 3.7% exceeded the age limit by 1 year, 7.4% by 2 years, 22.2% by 3 years, 25.9% by 4 years, 29.6% by 5 years, 7.4% by 6 years, and 3.7% by 7 years. Still, we included these studies in our review, as the results from younger children in those studies are important for our review. Whenever possible, only the results from primary school-aged children were extracted, and the results from secondary school-aged children were omitted.
-
Studies had to assess the psychometric properties.
-
Studies had to use the parent and/or teacher SDQ version but not self-report.
-
Reports had to be available in English.
Nr | Study | N | Age (years) | Informant | Source population | Country | General study aim | Psychometric properties assessed | Standard comparison diagnosis | Diagnostic domains assessed |
---|---|---|---|---|---|---|---|---|---|---|
1 | Matsuishi et al. (2008) | 2,899 | 4–12 | P | CO | Japan | Establish norms, factor structure | Internal consistency, construct validity | None | Total, prosocial |
2 | Perren et al. (2007)* | 160 | 5–6 | P, T | CO | Switzerland | Predictive value of prosocial behavior | Predictive validity | Berkeley Puppet Interview | Total, prosocial |
3 | Hill and Hughes (2007) | 374 | 6 | P, T | CL | United States (Texas) | Factor structure | Construct validity | Class play | Total, prosocial |
4 | Van Leeuwen et al. (2006) | 3,169 | 4–8 | P, T | CO | Belgium | Psychometric properties Dutch version | Internal consistency, inter-rater agreement, construct and concurrent validity | CBCL | Prosocial, hyperactivity, internalizing |
5 | Samad et al. (2005) | 112 | 4–16 | P | CL | Pakistan | Validity in Pakistan | Capacity to discriminate | ICD-10, Pediatric vs. Psychiatric | Total, prosocial |
6 | Bourdon et al. (2005) | 9,878 | 4–17 | P | CO | United States | Establish norms United States, evaluate SDQ | Internal consistency | Service use | Total, prosocial |
7 | Dickey and Blumberg (2004) | 9,574 | 4–17 | P | CO | United States | Establish factor structure in Unites States | Construct validity | None | Total, prosocial |
8 | Goodman et al. (2004) | 1,028 | 5–17 | P, T, S | CL | United Kingdom | Assess SDQ for improving detecting community psychiatric disorders | Capacity to discriminate | None | Total, prosocial |
9 | Becker et al. (2004) | 543 | 5–17 | P, T | CL | Germany | Examine reliability and validity of German SDQ | Internal consistency, construct, concurrent and capacity to discriminate | CBCL & clinical diagnoses (ICD-10) | Total, prosocial |
10 | Hawes and Dadds (2004) | 1,359 | 4–9 | P, T | CO | Australia | Psychometric properties Australian version | Internal consistency, concurrent validity | Diagnostic interviews | Total, prosocial |
11 | Malmberg et al. (2003) | 493 | 5–15 | P | CO, CL | Sweden | Validity of Swedish SDQ | Capacity to discriminate | Community versus Psychiatric | Total, prosocial |
12 | Glazebrook et al. (2002) | 10,745 | 5–15 | P | CO, CL | United Kingdom | Screen for utility of SDQ in pediatric clinics | Capacity to discriminate | Community versus Clinic | Total, prosocial |
13 | Mullick and Goodman (2001) | 261 | 4–16 | P, T, S | CO, CL | Bangladesh | Suitability of the SDQ for detecting psychiatric problems in Bangladesh | Capacity to discriminate | Community versus Clinic | Hyperkinesis, conduct disorder, emotional disorders or any psychiatric disorders |
14 | Goodman (2001) | 10,438 | 5–15 | P, T, S | CO | United Kingdom | Describe psychometric properties of SDQ | Internal consistency, inter-rater agreement, test retest reliability, construct and concurrent validity | DSM diagnoses | Total, prosocial |
15 | Klasen et al. (2000) | 273 | 4–13 | P | CO, CL | Germany | Compare SDQ with CBCL in Germany | Concurrent validity | CBCL and ICD diagnosis | Total, prosocial |
16 | Goodman et al. (2000b)
| 7,984 | 5–15 | P, T, S | CO | United Kingdom | SDQ as a means for improving detection of child psychiatric disorders in community | Capacity to discriminate | SDQ versus DAWBA | Total, prosocial |
17 | Goodman and Scott (1999) | 132 | 4–7 | P | CL | United Kingdom | Compare SDQ with CBCL in United Kingdom. | Concurrent and capacity to discriminate | CBCL | Total, prosocial |
18 | Goodman (1999) | 699 | 5–15 | P, T | CO, CL | United Kingdom | Validation study into extended SDQ version: impact scale | Capacity to discriminate | Community versus Clinic | Total, prosocial |
19 | Goodman (1997) | 403 | 4–16 | P, T | CO, CL | United Kingdom | Compare SDQ with Rutter scales | Concurrent and capacity to discriminate | Rutter questionnaires | Total, prosocial |
20 | Du et al. (2008) | 1,965 | 3–17 | P, T, S | CO, CL | China (Shanghai) | Assess norms, reliability, validity, factor structure | Internal consistency, inter-rater agreement, test retest reliability, construct, discriminant and capacity to discriminate | None | Total, prosocial |
21 | Van Roy et al. (2008) | 32,914 | 10–19 | P/proxy, S | CO | Norway | Construct validity of five-factor structure pre-, early, late adolescence | Construct validity | None | Total, prosocial |
22 | Parkes et al. (2008) | 818 | 8–12 | P | CL | Europe (8 regions) | Describe psychological symptoms in children with cerebral palsy & assess psychometric properties | Internal consistency | None | Total, prosocial |
23 | Zwirs et al. (2008)* | 2,439 | 6–10 | P, T | CO | The Netherlands (ethnically diverse) | Develop & validate screening instrument for externalizing disorders | Internal consistency, capacity to discriminate | None | Externalizing disorder |
24 | Mellor and Stokes (2007) | 914 | 7–17 | P, T, S | CO | Australia | Factor structure | Construct validity | None | Total, prosocial |
25 | Palmieri and Smith (2007) | 733 | 4–17 | P | CL | United States | Structural validity of P version for custodial grandmothers | Internal consistency, construct validity | None | Total, prosocial |
26 | Hysing et al. (2007) | 1,040 | 7–9 | P | CO | Norway | Evaluate sensitivity & specificity of SDQ in children with chronic illness | Concurrent validity | DAWBA | Total, prosocial |
27 | Becker et al. (2006) | 1,459 | 6–18 | P, T | CO | Europe (10 countries) | Examine SDQ P version in several European countries | Internal consistency, construct and concurrent validity | None | Total, prosocial |
28 | Sharp et al. (2005)* | 659 | 7–11 | P, T | CO | United Kingdom | Establish predictive validity for P and T ratings | Predictive validity | Parental help-seeking behavior and worry | Total, prosocial |
29 | Kashala et al. (2005) | 1,187 | 7–9 | T | CO | Congo (Kinshasa) | Pilot SDQ, investigate mental health and association with school problems, etc. | Internal consistency, construct validity | None | Total, prosocial |
30 | Edmunds et al. (2005) | 278 | 5 | P | CO | United Kingdom (London) | Evaluate reliability & validity for SDQ in a health study among young children | Internal consistency, concurrent validity | Child Health Questionnaire | Total, prosocial |
31 | Mellor (2004) | 917 | 7–17 | P, T, S | CO | Australia | Reliability with younger respondents | Internal consistency, inter-rater agreement test retest reliability | None | Total, prosocial |
32 | Widenfelt et al. (2003) | 1,686 | 8–16 | P, T, S | CO | The Netherlands | Psychometric properties of Dutch SDQ | Internal consistency, inter-rater agreement, concurrent validity | CBCL, YSR | Total, prosocial |
33 | Muris et al. (2003) | 562 | 9–15 | P, S | CO | The Netherlands | Examine psychometric properties | Internal consistency, test retest reliability, concurrent validity | CBCL, CDI, RCMAS, ADHDQ | Total, prosocial |
34 | Koskelainen et al. (2001) | 735 | 7–16 | P, T, S | CO | Finland | Evaluate psychometric properties | Internal consistency, inter-rater agreement, concurrent validity | CBCL, help-seeking questions | Total, prosocial |
35 | Goodman et al. (2007) | 400 | 5–7 | P | CO | United Kingdom | Check whether minor changes to questionnaire alter psychometrics | Capacity to discriminate | Rutter questionnaires | Total, prosocial |
36 | Smedje et al. (1999) | 900 | 6–10 | P | CO | Sweden | Validation study of Swedish SDQ | Internal consistency, construct validity | None | Total, prosocial |
37 | Kaptein et al. (2008) | 967 | 6–12 | P | CO, CL | The Netherlands | Assess differences in mental health of ID and non-ID children | Internal consistency, capacity to discriminate | None | Total, prosocial |
38 | Shojaei et al. (2008) | 1,348 | 6–11 | P | CO | France | Examine psychometric properties | Internal consistency | Socio-demographic data | Total, prosocial |
39 | Rothenberger et al. (2008) | 2,406 | 7–16 | P | CO | Germany | Examine psychometric properties of parent SDQ | Internal consistency, construct validity | None | Total, prosocial |
40 | Goodman et al. (2000b)
| 190 | 4–16 | P, T, S | CL | United Kingdom and Bangladesh | Predict type of disorder from the SDQ | Concurrent validity | Clinical diagnosis | Total, prosocial |
41 | Mathai et al. (2004) | 130 | 4–14 | P, T, S | CL | Australia | Examine agreement between SDQ and CAMHS | Concurrent and capacity to discriminate | Community child and adolescent mental health service (CAMHS) | Hyperactivity/inattention, conduct and emotional problems |
42 | Alyahri and Goodman (2006) | 187 | 5–12 | P, T | CO, CL | Yemen | Validate Arabic version of SDQ | Capacity to discriminate | DAWBA | Total, prosocial |
43 | Mathai et al. (2002) | 130 | 4–14 | P, T, S | CL | Australia | Examine usefulness of the SDQ as a screening measure | Inter-rater agreement and concurrent validity | CAMHS | Total, prosocial |
44 | Sanne et al. (2009) | 8,999 | 7–9 | P, T | CO | Norway | Test five-factor model | Internal consistency, construct validity | None | Total, prosocial |
45 | Lai et al. (2009) | 3,722 | 6–12 | P,T | CO, CL | China | Validate Chinese version of the SDQ | Internal consistency, test–retest reliability, capacity to discriminate | None | Total, prosocial |
46 | Janssens and Deboutte (2009) | 292 | 3–18 | P, caregiver, S | CL | Belgium | Usefulness of SDQ for screening children entering welfare | Internal consistency, concurrent validity | CBCL | Total, prosocial |
47 | Syed et al. (2009) | 556 | 5–11 | P | CO | Pakistan | Compare SDQ with CBCL | Concurrent validity | CBCL | Total, prosocial |
48 | Vogels et al. (2009) | 2,066 | 7–12 | P, S | CO | The Netherlands | Compare three questionnaires | Capacity to discriminate | CBCL | Total, prosocial |
Strategy for Analysis
Results
Internal Consistency
Informant | ||||
---|---|---|---|---|
Parent | Range | Teacher | Range | |
Prosocial behavior | 0.67 | 0.54–0.84 | 0.82 | 0.79–0.86 |
Hyperactivity/inattention | 0.76 | 0.58–0.85 | 0.83 | 0.66–0.89 |
Emotional symptoms | 0.66 | 0.60–0.76 | 0.73 | 0.63–0.80 |
Conduct problems | 0.58 | 0.46–0.76 | 0.70 | 0.63–0.84 |
Peer problems | 0.53 | 0.30–0.76 | 0.63 | 0.35–0.77 |
Total difficulties | 0.80 | 0.53–0.84 | 0.82 | 0.62–0.85 |
Impact scores | 0.81 | 0.69–0.87 | 0.85 | – |
N
| 53,691 | – | 21,866 | – |
Test–Retest Reliability
Informant | ||||
---|---|---|---|---|
Parent | Range | Teacher | Range | |
Prosocial behavior | 0.65 | 0.43–0.78 | 0.79 | 0.50–0.84 |
Hyperactivity/inattention | 0.71 | 0.48–0.85 | 0.85 | 0.64–0.89 |
Emotional symptoms | 0.66 | 0.47–0.82 | 0.72 | 0.40–0.80 |
Conduct problems | 0.66 | 0.52–0.89 | 0.77 | 0.58–0.86 |
Peer problems | 0.66 | 0.61–0.91 | 0.77 | 0.58–0.82 |
Total difficulties | 0.76 | 0.72–0.86 | 0.84 | 0.55–0.90 |
Impact scores | 0.57 | – | 0.68 | – |
N
| 2,852 | – | 1,693 | – |
Inter-Rater Agreement
Weighted total | Range | |
---|---|---|
Prosocial behavior | 0.26 | 0.22–0.30 |
Hyperactivity/inattention | 0.47 | 0.44–0.61 |
Emotional symptoms | 0.28 | 0.23–0.41 |
Conduct problems | 0.34 | 0.27–0.65 |
Peer problems | 0.35 | 0.27–0.59 |
Total difficulties | 0.44 | 0.37–0.62 |
Impact scores | – | – |
N
| 14,811 |
Construct Validity
Frequencies | ||||||||
---|---|---|---|---|---|---|---|---|
Parent N = 43,274 (13 studies) | Teacher N = 19,105 (6 studies) | |||||||
<.40 | ≥.40–<.70 | ≥.70 |
M
| <.40 | ≥.40–<.70 | ≥.70 |
M
| |
Prosocial behavior | ||||||||
1 Considerate | 0 | 12 | 4 | 0.65 | 0 | 5 | 2 | 0.70 |
4 Shares | 1 | 15 | 0 | 0.56 | 0 | 2 | 5 | 0.71 |
9 Caring | 0 | 9 | 7 | 0.68 | 1 | 0 | 6 | 0.80 |
17 Kind to kids | 1 | 14 | 1 | 0.60 | 0 | 2 | 5 | 0.74 |
20 Helps out | 1 | 9 | 4 | 0.63 | 0 | 1 | 6 | 0.76 |
Hyperactivity/inattention | ||||||||
2 Restless | 1 | 8 | 7 | 0.63 | 0 | 1 | 5 | 0.80 |
10 Fidgety | 0 | 8 | 8 | 0.60 | 0 | 2 | 4 | 0.81 |
15 Distractible | 1 | 4 | 11 | 0.74 | 0 | 2 | 5 | 0.77 |
21 Reflective* | 1 | 14 | 1 | 0.57 | 0 | 6 | 1 | 0.56 |
25 Persistent* | 1 | 6 | 9 | 0.70 | 0 | 2 | 5 | 0.73 |
Emotional symptoms | ||||||||
3 Somatic complaints | 4 | 10 | 1 | 0.47 | 1 | 6 | 0 | 0.48 |
8 Worries | 0 | 7 | 9 | 0.70 | 0 | 3 | 4 | 0.73 |
13 Unhappy | 0 | 14 | 2 | 0.63 | 0 | 6 | 1 | 0.65 |
16 Clingy | 0 | 14 | 2 | 0.65 | 0 | 4 | 3 | 0.73 |
24 Fears | 0 | 11 | 5 | 0.66 | 0 | 3 | 4 | 0.80 |
Conduct problems | ||||||||
5 Tempers | 3 | 12 | 1 | 0.52 | 0 | 5 | 1 | 0.67 |
7 Obedient* | 3 | 11 | 1 | 0.46 | 2 | 0 | 2 | 0.43 |
12 Fights | 1 | 11 | 4 | 0.61 | 0 | 3 | 4 | 0.72 |
18 Lies, cheats | 3 | 10 | 2 | 0.62 | 1 | 4 | 2 | 0.63 |
22 Steals | 2 | 12 | 1 | 0.56 | 0 | 6 | 0 | 0.59 |
Peer problems | ||||||||
6 Solitary | 3 | 12 | 1 | 0.61 | 2 | 2 | 2 | 0.67 |
11 Good friend* | 5 | 9 | 1 | 0.52 | 1 | 3 | 2 | 0.64 |
14 Popular* | 0 | 11 | 3 | 0.67 | 1 | 2 | 2 | 0.58 |
9 Picked on, bullied | 1 | 12 | 2 | 0.58 | 1 | 4 | 1 | 0.51 |
23 Best with adults | 2 | 9 | 3 | 0.63 | 2 | 2 | 2 | 0.68 |
Concurrent Validity
Informant | ||||
---|---|---|---|---|
Parent | Range | Teacher | Range | |
Conduct problems/externalizing | 0.71 | 0.60–0.84 | 0.79 | 0.74–0.86 |
Hyperactivity/attention problems | 0.69 | 0.64–0.78 | 0.77 | 0.76–0.80 |
Emotional symptoms/internalizing | 0.64 | 0.44–0.77 | 0.58 | 0.40–0.80 |
Peer problems/social problems | 0.52 | 0.41–0.75 | 0.57 | 0.48–0.71 |
Total/total | 0.76 | 0.70–0.87 | 0.76 | 0.68–0.87 |
Impact/total | 0.46 | 0.44–.051 | 0.53 | – |
N
| 4,590 | – | 784 | – |
SDQ Correlations with Measures of General Psychopathology
SDQ Correlations with Measures of Specific Psychopathology
Associations of the SDQ with the DAWBA, DMS-IV Diagnoses, and Risk Factors in Community Samples
Capacity to Discriminate
Informant | ||||
---|---|---|---|---|
Parent | Range | Teacher | Range | |
Prosocial behavior | 0.71 | 0.39–0.82 | 0.65 | 0.64–0.67 |
Hyperactivity/inattention | 0.90 | 0.76–0.97 | 0.95 | 0.90–0.95 |
Emotional symptoms | 0.79 | 0.69–0.85 | 0.84 | 0.65–0.88 |
Conduct problems | 0.92 | 0.68–0.97 | 0.86 | 0.82–0.87 |
Peer problems | 0.71 | 0.49–0.78 | 0.57 | 0.45–0.69 |
Total difficulties | 0.87 | 0.64–0.91 | 0.83 | 0.65–0.91 |
Impact scores | 0.86 | 0.83–0.87 | 0.88 | 0.85–0.89 |