Goodenough-harris Drawing Test

There is a need for simple, cost-effective tools to detect developmental delay in preschool children in Low- and Middle-income countries (LMIC) in both research and clinical contexts (Sabanathan, Wills, & Gladstone, 2015). In such resource-constrained settings, imported equipment and copyrighted assessment booklets are often unaffordable, parents generally have limited funds for transport or time off work to attend more comprehensive assessments, and clinics have long waiting times. Human figure drawing (HFD) has a long history both as an informal developmental screening tool and in standardised test batteries such as the Griffiths Mental Developmental Scales (GMDS) (Luiz, Foxcroft, & Tukulu, 2004). It was itemised in the Eye and Hand coordination subscale of the Extended Revised GMDS (GMDS-ER) and is retained in the most recent version version (Griffiths III) (Stroud, Foxcroft, & Green Bloomfield, 2016). Further applications of the HFD have included assessment of mental maturity and as a projective instrument to assess emotional and personality traits (Koppitz, 1968). Koppitz also validated a simplified scoring system as a developmental screening tool for children 6 years and older (Koppitz, 1968).

There is a resurgence of interest in using the HFD for research. A recent study of over 4000 pairs of twins reported that HFD at 4 years old correlated significantly with heritable factors and intelligence at both 4 and 14 years of age (Arden, Trzaskowski, Garfield, & Plomin, 2015). Tükel, Eliasson, Bőhm and Smedler (2018) found that the HFD was influenced by visual perception and visuomotor control and could be used to screen for developmental delay in Swedish preschool children.

Frances Goodenough originally developed the 'Draw-a-Man' (Draw-A-Person/DAP) scoring system in 1926 and found it correlated well with concurrent standard intelligence quotient (IQ) tests (Goodenough, 1926). Subsequent iterations of the test were widely used in South African research (Hunkin, 1950; Oates, 1938; Reynolds & Hickman, 2004; Sherr et al., 2017): DAP scores obtained from the McCarthy scales (McCarthy, 1972) for under 8-year-olds were comparable with historical samples and demonstrated validity both as 'a measure of intellectual functioning' and school performance in African children (Richter, Griesel, & Wortley, 1989). Venter and Bham (2003) found that the Goodenough-Harris DAP (Harris, 1963) predicted academic achievement in first-grade African pupils. Burger applied the international norms of the DAP:IQ version (Reynolds & Hickman, 2004) and found them appropriate for 5–7-year-olds, suggesting that it might be a suitable screening tool to detect mild to moderate global developmental delay in South African children (Burger, 2008).

The DAP scoring system was refined by Ireton, Quast, and Gantsher (1971) to produce an age-equivalent standardised score called the 'Index of Psychological Function' (IPF). Children aged 4–10 years with an IPF below 85 had significant risk of a developmental disorder or delay. In this way, the Goodenough DAP was used to screen for developmental delay in a cohort of HIV-infected (HIV+) South African children (Zeegers et al., 2009).

The Goodenough DAP is easy to administer and transcends language barriers. The instructions are simple, and can be conveyed to the child by a translator, an advantage where language difficulties are present. The tool is self-explanatory and without formal training or certification requirements. In a review of 48 standardised instruments, the DAP fulfilled six 'usability' criteria, that is, (1) rapid test administration, (2) scoring time, (3) low level of assessor qualification, (4) minimal training required and being the only test available (5) free of charge, and (6) not requiring expensive equipment or manuals. All of these are important considerations for LMIC. However, it fell short on psychometric properties including recent standardisation, validity, and reliability (Miles, Fulbrook, & Mainwaring-Magi, 2016).

Questions remain regarding the construct validity of the Goodenough DAP test (Kamphaus & Pleiss, 1991). Richter, Mabaso, and Hsiao (2015) also found that the DAP administered at age 7 years had insufficient predictive power to identify learners requiring intervention to prevent grade repetition, and it underestimated abilities in children above 8 years of age (Richter et al., 1989).

Controversy regarding use of DAP (and other versions of HFD) largely relates to its inaccuracy in predicting intelligence, as it does not detect children with above average or borderline IQ (Imuta, Scarf, Pharo, & Hayne, 2013). Imuta et al. (2013) assessed 100, 5-year-old children, and reported that none of the five children with a DAP:IQ standard score below 80 showed borderline functioning on the Wechsler Preschool Scale of Intelligence (WPSI), and neither of the two children with borderline functioning on the WPSI were delayed on the DAP:IQ test. However, no participants in their cohort had intellectual disability (IQ < 70).

The DAP has not been validated for South African preschoolers. Diagnostic accuracy (i.e. sensitivity, specificity, and positive and negative predictive values measured against a gold standard) must be established before its use can be recommended. Validation of its potential value in research could affirm or guide its use as a clinical tool. Because of this uncertainty, we aimed to explore the potential value of HFD as a research tool in South Africa. Moreover, the study was justified in order to determine the construct and predictive validity of the DAP for a specific age group. Our hypothesis was that the Goodenough DAP would correlate best with the eye-hand quotient (EHQ) of the GMDS-ER, which was selected as the gold standard.

The objectives were to determine (1) whether the DAP correlated best with the EHQ versus other GMDS-ER quotients and (2) the diagnostic accuracy of Goodenough DAP in detecting developmental delay in South African preschool children.

Method

This was a cross-sectional neurodevelopmental sub-study linked to the Children with HIV Early Antiretroviral Therapy (CHER) trial (Cotton et al., 2013). The primary aim of the CHER study was to evaluate different antiretroviral regimens over time in children from Cape Town and Soweto.

Participants

The participants were enrolled in two prospective interlinking studies during 2005–2006 at the Children's Infectious Diseases Clinical Research Unit (KID-CRU) (Laughton et al., 2012; Madhi et al., 2010). HIV+ children enrolled in the CHER study at the Cape Town site and a control group of HIV-exposed uninfected (HEU) and HIV-unexposed (HU) children from a concurrent vaccine trial (Madhi et al., 2010) were recruited for a longitudinal neurodevelopmental sub-study. The neurodevelopmental study inclusion criteria were birth weight greater than 2000 g, and no dysmorphism or central nervous system insults such as foetal alcohol syndrome, perinatal asphyxia, or metabolic abnormalities. HIV+ children were enrolled in the CHER study before 12 weeks of age with a normal neurological examination at that time. A total of 128 participants from the neurodevelopmental sub-study with Griffiths assessments that included HFDs at 5 years of age were eligible for the study. Three drawings were excluded as they consisted of illegible scribbles. Mean age was 60.8 months ranging from 59 to 66 months, and 48.8% were boys (Table 1). The majority (76%) were attending a daycare or preschool facility.

Table

Table 1. Characteristics of study participants (n = 125).

Instruments

The GMDS-ER is a comprehensive paediatric developmental assessment tool from the United Kingdom (UK), which has been adapted and used extensively in South Africa (Luiz et al., 2006). Training and certification are required to administer the tests. It is a criterion and norm-referenced tool that assesses various developmental domains in children from 2 to 8 years of age based on six subscales detailed in Table 2. The raw scores are converted to z-scores, percentiles, and age-equivalents derived from UK normative data. A subscale quotient is estimated by dividing the age-equivalent by the chronological age expressed as a percentage. The general quotient (GQ) is calculated by averaging all six subscale quotients.

Table

Table 2. Correlation between Draw-a-person test and Griffiths mental developmental-extended revised subscale quotients.

Developmental delay is defined as a score more than two standard deviations (−2 SD) below the expected mean (Davies et al., 2011). Thus, a general or subquotient score below 70 indicates definite developmental impairment by British norms. However, there is no universally accepted definition of delay, with uncertainty as to which cut-off score is most relevant in the South African context. The referral criteria vary, and are set by therapists based on local experience. For this analysis, the conservative EHQ threshold of 75 (as opposed to 70) was set as the score requiring referral for further diagnostic evaluation, to allow for the imprecision of subquotient determination.

The human figure in the Goodenough DAP instrument is scored against 51 specified characteristics and applies a quantitative scoring system to the drawing (Goodenough, 1926). This is expressed as a mental age and converted to a standardised score or quotient. It therefore provides a non-verbal measure of mental development for children 3–10 years old. A standard score below 85 was taken to represent significant delay requiring referral (Ireton et al., 1971).

Procedure

All children were tested with the GMDS-ER at 5 years of age by one of two developmental paediatricians (B.L. and H.S.) assisted by a Griffiths-trained translator (L.R.K.). The GMDS-ER Eye and Hand Coordination Subscale (EHQ) includes an HFD which is scored differently to the DAP. Participants were tested individually in the presence of their caregiver, who sat in the background but was requested not to help or make comments. The children were instructed in their home language to draw a person with a pencil in the Griffiths (1970) record book using the standardised administration instructions outlined in the manual. They sat in a quiet room and the assessor encouraged them 'to draw the best possible person' but did not offer assistance and applied no time limit. Vision was assessed using a Snellen's picture chart and tiny cake decorations. Participants were excluded if any item on the EHQ was missing.

Copies of each drawing were placed in three separate files with no identifying information apart from a participant code. Drawings were then scored independently according to the Goodenough DAP scoring system (Goodenough, 1926) by a third developmental paediatrician (P.S.) and two medical officers (E.K. and H.E.). The assessors did not collaborate and were blinded to GMDS-ER findings and to the HI and HIV-exposure status of the children. The medical officers had no experience with the Goodenough DAP scoring system prior to the study and used the DAP instruction sheet for scoring. The DAP scores together with the de-identified GMDS-ER data were converted into quotients and entered into an Excel database.

Ethical considerations

The study was approved by the Stellenbosch University Health Research Committee (N05/05/092). Informed consent was obtained in person from the child's legal guardian in their preferred language according to Good Clinical Practice guidelines. Parents remained with their children for reassurance during the assessments and only children who remained cooperative were tested. Children with developmental problems were referred with parental consent to relevant services. Clinical files were stored at the study site. No identifying data were accessed during the analysis and data were stored on a password-protected database.

Data analysis

Data were analysed using Stata version 14 (StataCorp LP, College StationTX). Continuous data were described using mean and standard deviation (SD) if symmetrically distributed; otherwise the median and interquartile range (IQR). Pearson's correlation analysis of all the GMDS-ER subscales was used to assess the relationship between continuous variables, and the associated confidence interval (CI) was derived using Fisher's tranformation. CIs which show the likely degree of correlation in the population were calculated rather than p is values, which only test whether the correlation is significantly different from zero. If the 95% CI provided does not include a zero, the correlation is significantly different from zero at the 5% level of significance.

Intraclass correlations (ICCs) between the DAP and the EHQ and GQ were also calculated. The correlation coefficients were used to determine strength of the relationship between variables. Cohen (1992) suggested that effect sizes could be categorised as small (r = 0.10), medium (r = 0.30), and large (r = 0.50). We used two-way random effects ICC to measure the inter-rater agreement between DAP values by the three scoring clinicians. Sensitivity and specificity and 95% CI and positive and negative predictive values were calculated for DAP quotient less than 85 versus EHQ using an EHQ threshold of 75.

Optimal cut-off points for DAP versus the EHQ as gold standard were explored using receiver operating characteristic (ROC) curves. The area under the ROC curve and its significance were examined using a non-parametric assumption. Various thresholds were tested to identify the optimal values.

Results

Participant characteristics

A total of 128 drawings were available. Three consisted of illegible scribbles and were excluded and these children all had EHQs below 75 (i.e., 62, 72, and 58.3), confirming definite developmental delay. The DAP scores was calculated on the remaining 125 drawings. GMDS-ER EHQ scores were available for all 125, and GQ scores for 123 participants.

Correlation between DAP and GMDS-ER quotients

The median DAP score was 94 (IQR:85–103) and 28 children (22%) had a DAP score below 85, that is, meeting our definition for developmental delay. There were no differences in median DAP score between the HIV+, HEU, and HU groups (p = .25). The median EHQ of the cohort was 83.3 (IQR 76.9–88.8), while median GQ was 82.8 (IQR 77.5–88.1). Most participants (93%) had numerically higher DAP than EHQ and GQ quotients. The correlation between DAP and EHQ was strong (Pearson's r = 0.69 95% CI = [0.58–0.77]), less so for GQ, r = 0.57 95% CI = [0.44–0.68], and moderate for other GMDS-ER subscale quotients (Table 2). HIV status had no significant effect on these correlations (p = .22).

Inter-rater agreement

Agreement on overall DAP scoring between the developmental paediatrician and medical officers was fair (ICC = 0.73, 95% CI = [0.43–0.86]. The medical officers tended to give higher scores than the developmental paediatrician. For participants with DAP less than 85 the ICC was 0.66, 95% CI = [0.56–0.81], whereas amongst those with DAP ⩾ 85, ICC was 0.58, 95% CI = [0.18–0.78].

Diagnostic accuracy

Using a DAP threshold of 85 as specified by Ireton to indicate developmental delay we generated optimal sensitivity and specificity for an EHQ below 75. The threshold of 85 was tested using the experienced scorer (P.S.). Sensitivity of the DAP:EHQ was 80% and specificity 89%. In all, 19 (68%) of the 28 participants with developmental delay on the DAP were also delayed on EHQ, while 91 of 97 (93.8%) participants scoring above 85 on the DAP were not delayed on the EHQ (Table 3). Thus, when comparing the DAP with the GMDS-ER there was concordance between the two tests in classifying participants as delayed versus non-delayed in 110 (92.8%) participants, that is, giving positive and negative predictive values of 64.5% and 94.7% (Tables 3).

Table

Table 3. Diagnostic properties of the Goodenough Draw-a-Person score <85 when predicting scores <75 on Griffiths eye-hand coordination subscale.

There were six false negative results using the DAP threshold of 85 i.e., DAP greater than/equal to 85 but EHQ less than 75 (range 69.9–74.8). Two of these children had a borderline DAP score of 85. Only one child with DAP ⩾ 85 had an EHQ less than 70 (definite developmental delay). This child also had severe language delay (language subquotient 56.9), an EHQ of 69.9, and GQ of 70.7 but a DAP score of 102.

We identified optimal cut-off points for DAP versus the EHQ gold standard using ROC curves. Various thresholds were tested to identify the optimal values. Using the area under the ROC curve the variable DAP threshold of 85.6 was the best predictor for EHQ < 75 (ROC = .87; p =< .001) versus the DAP threshold of 85 (ROC = .84, p < .001) (Table 3).

Diagnostic accuracy was initially also estimated for the DAP versus GQ, however, the low prevalence of global developmental delay (3%) in the cohort precluded the diagnostic accuracy analysis of DAP/GQ, as it undermined positive predictive value determinations.

Discussion

We assessed the diagnostic accuracy of the Goodenough DAP instrument as a research tool, to detect developmental delay, against the gold standard GMDS-ER assessment in 125, 5-year-old South African children from low-income families. The correlation between the DAP and the EHQ (eye-hand coordination quotient) was the strongest according to Cohen's criteria (Cohen, 1992), and less so for the GMDS-ER GQ. Burger, who tested 30 children with a wider age range using the DAP:IQ version, found a strong correlation with GMDS-ERGQ (r = .76) (Burger, 2008). However although the DAP:IQ has better psychometric properties than the Goodenough DAP (Miles et al., 2016), it requires longer administration time and the need to import and pay for manuals, often unfeasible in LMIC.

The DAP threshold score of 85 yielded optimal sensitivity and specificity and a high negative predictive value for the EHQ, demonstrating its value to screen for delayed eye-hand coordination. By identifying children with a DAP score <85 we would have detected 76% of those with significant developmental delay as indicated by an EHQ < 75. Of the remaining 34%, only one had severe delay, that is, EHQ < 70.

Inter-rater agreement was good when identifying DAP scores < 85 but less so on higher scores. The three assessors interpreted certain descriptors differently, for example, 'Firm lines without overlapping at junctions'. These disparities in scoring require further analysis.

The mean Griffiths GQ and EHQ of our cohort were significantly below average, that is, more than one SD below the standardised UK norms (Griffiths, 1970). Similar findings were reported by Lowick, Sawry, and Meyers (2012), who compared neurodevelopment of 31 HIV+ South African children (mean EHQ 77.2) with 30 'apparently healthy' children (mean EHQ 82.8). Allan (1992) reported that lower socio-economic status adversely affected performance, with South African children from a higher socio-economic bracket performing better. Our cohort were all of lower socio-economic status, some with additional risks for cognitive delay, that is, HIV infection (50%), and exposure (20%), and would thus require developmental surveillance (Potterton, Hilburn, & Strehlau, 2015).

The South African Road to Health Book (RTHB) is a caregiver-held document in which the child's routine immunisations, growth parameters, and unscheduled clinic visits are recorded. It includes a developmental milestone checklist to be completed by primary healthcare nurse practitioners (Western Cape Government Department of Health, 2011). In the 2011 version, it listed 'Ability to draw a stick person' as the only developmental criterion for fine motor development at 5–6 years, however, there were no accompanying administration or scoring guidelines. It would, however, be impractical to use the Goodenough DAP in this setting, as the scoring system is time-consuming and requires standardised instruction sheets and training.

The DAP is currently used in developmental assessment clinics in South Africa as an informal component of a more comprehensive assessment. However, the DAP may have additional clinical applications, for example, for doctors working in busy community-based paediatric clinics. Children with a DAP below 85 could be referred to an occupational therapist for further assessment and therapy. Validation of the DAP as a research tool could thus inform its clinical use. Children with mild versus severe delay comprise 85% of those with developmental delay and intellectual disability and have the potential to make significant gains with early intervention (Sadock, Sadock, Ruiz, & Kaplan, 2009).

A recent study involving 345 American preschoolers found that while the DAP:IQ scores were not a valid measure of cognitive ability, they were useful to screen for fine motor delay in at-risk children (Rehrig & Stromswold, 2018). Likewise, Tükel et al. (2018) also reported that the HFD was influenced largely by visual perception and visuomotor control at 5.5 years in a group of Swedish children. These latter two studies concur with our findings.

The strengths of our study are as follows: first the DAP scores were calculated by independent assessors blinded to the child's GMDS-ER results as well as their HIV infection or exposure status. Second, although the age-band of the cohort was narrow, it represents an age at which children require school readiness testing and would benefit from intervention prior to school entry. In addition, this cohort was representative of children needing developmental surveillance, that is, HIV+, HEU, and HU children from LMICs. Children from economically disadvantaged environments with developmental impairment who receive therapeutic input prior to school entry show improved scholastic achievement (Heckman, 2006). Finally, children scoring below the cut-off point (DAP less than 85) had mild to moderate (vs severe) developmental delay on the GMDS-ER, a group often undetected at an earlier age.

The study was, however, limited by the low prevalence of global developmental delay in the sample (3%). Paediatric HIV clinics have previously reported higher prevalence of global delay (55%) which probably relates to much later antiretroviral therapy initiation than in the CHER cohort (Potterton et al., 2015). Also, our cohort included HEU and HU children not expected to have severe developmental problems. Second, the findings are not generalisable to all South African preschool children as this cohort had a narrow age range. Finally, the inclusion of the HFD task in the GMDS-ER, even though the scoring method differs from the DAP, could have confounded the diagnostic assessments.

Conclusion

Our findings support using the DAP as a research tool to detect fine motor and visuoperceptual delay in 5-year-old children. It could complement a more comprehensive assessment including verbal and non-verbal tasks. In addition there are clinical contexts where the DAP may be useful, that is, for medical practitioners in resource-constrained outpatient settings. Children with DAP scores ⩽ 85 or those unable to draw a person could be referred to an occupational therapist for assessment and intervention where indicated. However, we recommend that future studies should initially focus on a full standardisation of the DAP to include a broader cross-section of South African society with a wider age range. One could then evaluate the ability of the DAP to predict later scholastic performance in reading, writing, and mathematics.

Acknowledgements

The authors acknowledge Dr Henriette Saunders and Ms Lungiswa Khethelo for their assistance with the Griffiths assessments, Dr Hesti van Huyssteen for her assistance with the scoring of the human figure drawings and Ms Tonya Esterhuizen for her statistical support.

Funding
The authors declared receipt of the following financial support for the research, authorship, and/or publication of this article: Support for the CHER study, which provided the infrastructure for the neurodevelopmental sub-study, was provided by the US National Institute of Allergy and Infectious Diseases through the CIPRA network, Grant U19 AI53217; the Departments of Health of the Western Cape and Gauteng, South Africa; and GlaxoSmithKline. Additional support was provided with Federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health, United States Department of Health and Human Services, under Contract No. HHSN272200800014C. Permission to conduct the neurodevelopmental sub-study on this cohort was granted by Avy Violari, Shabir Madhi, Mark Cotton and the CHER steering committee. Neurodevelopmental assessments from were funded through grants from the Harry Crossley Foundation and the South African Medical Research Council (MRC), the National Research Foundation of South Africa and CIPRA-SA.

ORCID iD
Priscilla Estelle Springer https://orcid.org/0000-0001-8882-5688

References

	Allan, M. M. (1992). The performance of South African normal pre-school children on the Griffiths Scales of mental development: A comparative study (Unpublished MA dissertation). Nelson Mandela Metropolitan University, Port Elizabeth, South Africa. Google Scholar
	Arden, R., Trzaskowski, M., Garfield, V., Plomin, R. (2014). Genes influence young children's human figure drawings and their association with intelligence a decade later. Psychological Science, 25, 1843–1850. Google Scholar \| SAGE Journals \| ISI
	Burger, R. (2008). Human figure drawings and the general mental development of South African children (Unpublished MA dissertation), Nelson Mandela Metropolitan University, Port Elizabeth, South Africa. Google Scholar
	Cohen, J. (1992). Quantitative methods in psychology. Psychological Bulletins, 112, 155–159. Google Scholar \| Crossref \| Medline \| ISI
	Cotton, M. F., Violari, A., Otwombe, K., Panchia, R., Dobbels, E., Rabie, H., Lazarus, E. (2013). Early time-limited antiretroviral therapy versus deferred therapy in South African infants infected with HIV: Results from the children with HIV early antiretroviral (CHER) randomised trial. The Lancet, 382, 1555–1563. doi:10.1016/S0140-6736(13)61409-9 Google Scholar \| Crossref \| Medline \| ISI
	Davies, L., Dunn, M., Chersich, M., Urban, M., Chetty, C., Olivier, L., Viljoen, D. (2011). Developmental delay of infants and young children with and without fetal alcohol spectrum disorder in the Northern Cape Province, South Africa. African Journal of Psychiatry, 14, 298–305. Google Scholar \| Crossref
	Goodenough, F. (1926). Measure of intelligence by drawing. New York, NY: World Book. Google Scholar
	Griffiths, R. (1970). The abilities of young children (Child Development Research Centre). London, England: University of London Press. Google Scholar
	Harris, D. (1963). Children's drawings as measures of intellectual maturity (A revision and extension of the goodenough draw-a-man Test (DAP)). New York, NY: Harcourt Brace Jovanovich. Google Scholar
	Heckman, J. J. (2006). Skill formation and the economics of investing in disadvantaged children. Science, 312, 1900–1902. Google Scholar \| Crossref \| Medline \| ISI
	Hunkin, V. (1950). Validation of the goodenough draw-a-man test for African children. Journal for Social Research, 1, 52–63. Google Scholar
	Imuta, K., Scarf, D., Pharo, H., Hayne, H. (2013). Drawing a close to the use of human figure drawings as a projective measure of intelligence. PLoS ONE, 8(3), e58991. doi:10.1371/journal.pone.0058991 Google Scholar \| Crossref \| Medline \| ISI
	Ireton, H., Quast, W., Gantsher, P. (1971). The draw-a-man test as an index of developmental disorders in a pediatric outpatient population. Child Psychiatry and Human Development, 2, 42–49. Google Scholar \| Crossref \| Medline
	Kamphaus, R. W., Pleiss, K. L. (1991). Critiques of school psychological materials draw-a-person techniques tests in search of a construct. Journal of School Psychology, 29, 395–401. Google Scholar \| Crossref \| ISI
	Koppitz, E. M. (1968). Psychological evaluation of children's human figure drawings. New York, NY: Grune & Stratton. Google Scholar
	Laughton, B., Cornell, M., Grove, D., Kidd, M., Springer, P. E., Dobbels, E., Cotton, M. F. (2012). Early antiretroviral therapy improves neurodevelopmental outcomes in infants. AIDS, 26, 1685–1690. Google Scholar \| Crossref \| Medline
	Lowick, S., Sawry, S., Meyers, T. (2012). Neurodevelopmental delay among HIV-infected preschool children receiving antiretroviral therapy and healthy preschool children in Soweto, South Africa. Psychology, Health & Medicine, 17, 599–610. doi:10.1080/13548506.2011.648201 Google Scholar \| Crossref \| Medline
	Luiz, D. M., Faragher, B., Barnard, A., Knoesen, N., Kotras, N., Burns, L. E., Challis, D. (2006). Griffiths mental development scales-extended revised, (GMDS-ER) two to eight years (Analysis manual). Oxford, UK: Hogrefe. Google Scholar
	Luiz, D. M., Foxcroft, C. D., Tukulu, A. N. (2004). The Denver II scales and the Griffiths Scales of mental development: A correlational study. Journal of Child and Adolescent Mental Health, 16, 77–81. doi:10.2989/17280580409486573 Google Scholar \| Crossref \| Medline
	Madhi, S. A., Adrian, P., Cotton, M. F., Mcintyre, J. A., Jean, P., Meadows, S., Klugman, K. P. (2010). Effect of HIV infection status and ant-retroviral treatment on quantitative and qualitative antibody responses to pneumococcal conjugate vaccine in infants. The Journal of Infectious Disease, 202, 355–361. doi:10.1086/653704 Google Scholar \| Crossref \| Medline
	McCarthy, D. (1972). Manual for the McCarthy scales of children's abilities. New York, NY: Psychological Corporation. Google Scholar
	Miles, S., Fulbrook, P., Mainwaring-Magi, D. (2016). Evaluation of standardized instruments for use in universal screening of very early school-age children: Suitability, technical adequacy, and usability. Journal of Psychoeducational Assessment, 36, 99–119. doi:10.1177/0734282916669246 Google Scholar \| SAGE Journals
	Oates, L. (1938). Goodenough drawing test and its application to Zulu children (MA dissertation). University of Stellenbosch, Stellenbosch, South Africa. Google Scholar
	Potterton, J., Hilburn, N., Strehlau, R. (2015). Developmental status of preschool children receiving cART: A descriptive cohort study. Child: Care, Health and Development, 42, 410–414. doi:10.1111/cch.12321 Google Scholar \| Crossref
	Rehrig, G., Stromswold, K. (2018). What does the DAP: IQ measure? Drawing comparisons between drawing performance and developmental assessments. Journal of Genetic Psychology, 179, 9–18. Google Scholar \| Crossref \| Medline
	Reynolds, C. R., Hickman, J. (2004). Draw-a-person intellectual ability test for children, adolescents, and adults, examiners manual (DAP:IQ). Austin, TX: Pro-ed. Google Scholar
	Richter, L., Mabaso, M., Hsiao, C. (2015). Predictive power of psychometric assessments to identify young learners in need of early intervention data from the birth to twenty plus cohort, South Africa. South African Journal of Psychology, 46, 1–16. doi:10.1177/0081246315599476 Google Scholar \| SAGE Journals
	Richter, L. M., Griesel, R. D., Wortley, M. (1989). The draw-a-man test: A 50-year perspective on drawings done by black South African children. South African Journal of Psychology, 19, 1–5. Google Scholar \| SAGE Journals
	Sabanathan, S., Wills, B., Gladstone, M. (2015). Child development assessment tools in low-income and middle-income countries how can we use them more appropriately. Archives of Disease in Childhood, 100, 482–488. doi:10.1136/archdischild-2014-308114 Google Scholar \| Crossref \| Medline
	Sadock, B. J., Sadock, V. A., Ruiz, P., Kaplan, H. I. (2009). Kaplan & Sadock's comprehensive textbook of psychiatry (9th ed., pp. 3444–3474). Philadelphia, PA: Lippincott Williams & Wilkins. Google Scholar
	Sherr, L., Macedo, A., Cluver, L. D., Meinck, F., Skeen, S., Hensels, I. S., Tomlinson, M. (2017). Parenting, the other oldest profession in the world – A cross-sectional study of parenting and child outcomes in South Africa and Malawi. Health Psychology and Behavioral Medicine, 5, 145–165. doi:10.1080/21642850.2016.1276459 Google Scholar \| Crossref \| Medline \| ISI
	Stroud, L., Foxcroft, C., Green Bloomfield, E. C. J. (2016). Griffiths III. Oxford, UK: Hogrefe. Google Scholar
	Tükel, Ş., Eliasson, A. C., Bőhm, B., Smedler, A. C. (2018) Simple categorization of human figure drawings at 5 years of age as an indicator of developmental delay. Developmental Neurorehabilitation. Retrieved from https://www.tandfonline.com/doi/abs/10.1080/17518423.2018.1532969?journalCode=ipdr20 Google Scholar
	Venter, A., Bham, A. (2003). The usefulness of commercially available culture fair tests in the assessment of educational success in Grade 1 Black pupils in South Africa — An explorative study. Journal of Child and Adolescent Mental Health, 15, 33–37. doi:10.2989/17280580309486538 Google Scholar \| Crossref \| Medline
	Western Cape Government Department of Health . (2011). Minister Botha launches road to health booklet. Jonga. Retrieved from www.exploregov.za Google Scholar
	Zeegers, I., Rabie, H., Swanevelder, S., Edson, C., Cotton, M., van Toorn, R. (2009). Attention deficit hyperactivity and oppositional defiance disorder in HIV-infected South African children. Journal of Tropical Pediatrics, 56, 97–102. doi:10.1093/tropej/fmp072 Google Scholar \| Crossref \| Medline