|Year : 2019 | Volume
| Issue : 1 | Page : 317-322
Screening intelligence of primary school children using Draw-a-Person test
Ali M El-Shafie1, Dalia M ElLahony1, Mohamed A Samir1, Khaled M Biram2
1 Paediatric Department, Faculty of Medicine, Menoufia University, Menoufia, Egypt
2 Paediatric Department, El-Amreya General Hospital, Alexandria, Egypt
|Date of Submission||19-Aug-2017|
|Date of Acceptance||29-Nov-2017|
|Date of Web Publication||17-Apr-2019|
Khaled M Biram
Pediatric Department, El-Amreya General Hospital, 23 Ali Heba Street, Sidi Bishr, Alexandria
Source of Support: None, Conflict of Interest: None
The aim of this study was to screen intelligence among primary school children using Draw-a-Person test (DAP) in Alexandria, Egypt.
DAP test was created by Dr. Florence Goodenough in 1926. It measures multiple developmental streams of progress like learning, cognitive, visual, and motor by having the child to draw any person, with scoring based according to the presence and quality of features. By comparing the raw score to the typical rates of drawing, the child's intelligence quotient (IQ) is obtained.
Patients and methods
This study was carried out on 1000 apparently healthy primary school children aged from 6 to 12 years in Alexandria, Egypt. Students were subjected to medical evaluation. Parents were given a personal/family data questionnaire by the students to assess their socioeconomic standard (SES), which also contained a part for the student's medical history to exclude any serious medical problems or chronic diseases that might affect test score. School achievement data were obtained from student's school file. Thereafter, the students performed the test.
The study showed that the number of children with gifted intelligence was 24 (2.4%), with superior intelligence was 85 (8.5%), with average intelligence was 816 (81.6%), with borderline intellectual function was 41 (4.1%), and lastly, with mild and moderate mental retardation was 34 (3.4%). Positive correlations were found between IQ and SESs, school achievement, and residence. No correlation was found between IQ and sex.
The DAP test provides pediatricians with an easy and low-cost developmental screening device, alerting them to the possibility of developmental disorders.
Keywords: achievement, cognition, intellectual disability, intelligence tests, pediatricians
|How to cite this article:|
El-Shafie AM, ElLahony DM, Samir MA, Biram KM. Screening intelligence of primary school children using Draw-a-Person test. Menoufia Med J 2019;32:317-22
|How to cite this URL:|
El-Shafie AM, ElLahony DM, Samir MA, Biram KM. Screening intelligence of primary school children using Draw-a-Person test. Menoufia Med J [serial online] 2019 [cited 2020 Jun 2];32:317-22. Available from: http://www.mmj.eg.net/text.asp?2019/32/1/317/256127
| Introduction|| |
Draw-a-Person test (DAP) is a projective test that allows an examinee to respond to questions through drawings. Projective tests can be applied in various settings such as schools, corporations, and private practices to assess many psychological aspects including personality, family background, intelligence, physical and emotional abuse, and depression. .
Dr. Goodenough first became interested in figure drawing when she wanted to find a way to supplement the Stanford–Binet intelligence test with a nonverbal measure. The test was developed to assess the maturity in young people. Dr. Goodenough concluded that the amount of detail involved in a child's drawing could be used as an effective tool. This led to the development of the first official assessment using figure drawing, which was the DAP test. Over the years, the test has been revised many times with added measures for assessing intelligence .
Harris later revised the test, with features including drawings of a woman and of themselves. Now considered the Goodenough-Harris test, it has guidelines for assessing children from age 6 to 17 years. Harris assumes that changes in the child's drawings of a man or a woman represent the development of cognitive complexity or intellectual maturity expressed by increasingly complex representations of the human figure. He regards the child's concept of a human figure as an index or sample of their concept generally. As Cox  in a comprehensive review notes that in Western cultures 'children progress from a period of scribbling to the production of tadpole forms, and then to conventional forms, at first composed of segmented body parts, and then later of more contoured and integrated sections'.
The aim of our study is to screen intelligence among primary school children using DAP test in Alexandria, Egypt.
| Patients and Method|| |
After approval of the Local Institutional Ethical Committee of educational affairs of Alexandria and obtaining written consents from all student's parents to participate in our study, the study was carried out on 1000 apparently healthy primary school children aged from 6 to 12 years from July 2015 to October 2016. This study involved a mixture of both rural and urban cultures.
Parents were given a questionnaire by the students inquiring about their telephone number, full name, date of birth, and any serious medical problems in the past, and it also included degree of parents' education, occupation of husband, family size, and monthly family income to assess socioeconomic standard (SES).
All study children were subjected to medical evaluation to exclude any chronic clinical problems.
School achievement was obtained by the marks of the previous year using subject's school records. Marks of the children from junior one class were obtained from the midterm, and then school achievement was classified according the most common grading scale: grade A (90–100, excellent), B (80–89, above average), C (70–79, average), D (60–69, usually the minimum passing grade), and F (0–59).
DAP test was performed with test instructions, and test setting included a child seated comfortably at an individual table with enough space to draw freely. Before asking the student to complete the DAP test, we provided the following: full understanding of the instructions to the children involved in study, sufficient lighting, avoidance of distractions and noises, and materials such as a pencil with soft lead and a sheet of paper placed on the table.
Administrator requested the children to DAP with no time limit and did not make any comment on the drawing or ask the child to correct any details, as this is not an art lesson but an attempt to ascertain the child's concept of the human figure.
Children who refused to draw were encouraged, and if they did not respond, they were given another time for testing.
After children had completed the test, drawings were rated according to the DAP scoring system to be converted to intelligence quotient (IQ) by modified Harris scoring guide .
The following features were evaluated in the DAP test:
Head present, legs present, arms present, trunk present, length of trunk greater than breadth, and shoulders are indicated (abrupt broadening of the trunk below the neck).
Both arms and legs attached to the trunk, arms and legs attached to the trunk at the correct points, neck present, and outline of neck, continuous with that of head, trunk, or both.
Eyes present (one or two), nose present, mouth present, nose and mouth in two dimensions, two lips shown, nostril shown, hair shown, and hair on more than circumference of head and nontransparent – better than scribble.
Clothing presents (any clear representation of clothing), two articles of clothing nontransparent (e.g., hat and trousers), the entire drawing free from transparencies – sleeves and trousers must be shown, four articles of clothing definitely indicated such as hat, shoes, coat, shirt, necktie, belt and trousers, and lastly costume complete with incongruities such as business suit and soldier's costume, and hat, sleeves, trousers, and shoes must be shown.
Fingers present (any indication), correct number of fingers shown, fingers in two dimensions – length greater than breadth, the angle subtended not greater than 180 degrees, opposition of thumb clearly defined, and hand shown distinct from fingers and arm.
Arm joint shown – elbow, shoulder, or both, and leg joint shown – knee, hip, or both.
Head not more than half or less than one-tenth of trunk, arms equal to trunk, but not reaching knee, legs not less than trunk not more than twice trunk size, feet in two dimensions – not more than one-third or less than one-tenth of leg, and both arms and legs in two dimensions.
Lines firm without marked tendency to cross, gap, or overlap; all lines firm with correct joining, outline of head without obvious irregularities. Develop beyond the first crude circle, conscious control apparent, trunk outline. Score same as 3, arms and legs without irregularities. Two dimensions and no tendency to narrow at the point of junction with trunk, and features symmetrical (more likely to credit in profile drawings).
Fine head details
Ears present (two in full face, one in profile). Ears present in correct position and proportion. Eye details – brow or lashes and pupil shown; proportion: length greater than width; eye detail – glance – only plus in profile; and chin and forehead shown.
Projection of chin shown in profile, and heel clearly shown. Body profile – head, trunk, and feet without error, and the figure shown in true profile without error or transparency.
For example, if a boy is aged eight years, and his raw drawing score was 40, then the IQ was 112 by data in reference table.
Then, the student's IQ was classified according to Sattler. The Stanford–Binet Intelligence Scale fourth edition is as follow: very gifted, 145–160; gifted, 130–144; superior intelligence, 120–129; high average intelligence, 110–119; average intelligence, 90–109; low average, 80–89; borderline impaired, 70–79; mildly impaired, 55–69; and moderately impaired, 40–54.
In our study, we further classified the obtained IQ scores into five classes: gifted group included very gifted and gifted scores, 130–160; superior intelligence, 120–129; average included high and flow average, 80–119; borderline impaired, 70–79; and mild and moderate impaired, 40–69; the results were statistically analyzed by SPSS, version 20 (SPSS Inc., Chicago, Illinois, USA).
Two types of statistics were done: descriptive was done using percentage (%), mean, and SD, and analytical one using Student's t-test, which is a single test used to collectively indicate the presence of any significant difference between two groups for a normally distributed quantitative variable.
One-way analysis of variance (ANOVA) (F test): ANOVA is a single test used to collectively indicate the presence of any significant difference between several groups for a normally distributed quantitative variable.
Post-hoc test: it is a test used after one-way ANOVA (F test) or Kruskal–Wallis test to show any significant difference between the individual groups.
P value is considered significant if P is less than or equal to 0.05.
| Results|| |
A total of 1000 patients were involved: 423 (42.3%) females and 577 (57.7%) males. The age range of the patients was 6–12 years, with the mean age of 8.17 ± 1.6 years. A total of 256 (25.6%) of them were rural residents and 744 (74.4%) were urban residents. Children were classified into low SES children 327 (32.7%), average SES 418 (41.8%), and children with high SES 255 (25.5%). As regarding school achievement, children were divided into four groups: grades A, B, C, or D, accounting for 187 (18.7%), 378 (37.8%), 263 (26.3%), and 172 (17.2%), respectively [Table 1].
|Table 1: Distribution of the studied children regarding their characteristics|
Click here to view
The study revealed some points regarding IQ as follows:
First, the number of children with gifted intelligence was 24 (2.4%), superior intelligence was 85 (8.5%), those having average intelligence was 816 (81.6%), borderline intellectual function group was 41 (4.1%), and lastly, 34 (3.4%) children had mild and moderate mental retardation [Table 2].
|Table 2: Distribution of the studied children regarding intelligence quotient level|
Click here to view
Second, strong positive correlation was found between SES and IQ levels; those with higher SES have higher IQ scores (111.25) in comparison with those with average (104.18) and low SES (92.67) [Table 3].
|Table 3: Distribution of sex, residence, socioeconomic standard and school achievement regarding intelligence quotient level (n=1000)|
Click here to view
Third, children living in urban areas had higher IQ levels (104.32) compared with those living in rural areas (97.52) [Table 3].
The fourth finding was the positive correlation between IQ levels and school achievement; children with high school achievement (grade A) achieved the highest DAP test score (112.16) and IQ level among the whole studied sample, and vice versa [Table 3].
The fifth point is there was no significant difference in IQ levels between males and females [Table 3].
| Discussion|| |
The assessment of intelligence has a long controversial history. However in recent years, one area that has received more intense focus on the study of it has been the assessment of neural and psychological functions in children and adolescents .
Many studies have been conducted all over the world to find the effect of many risk factors on child cognition. The education, occupation, and income of parents – indices of the family's SES – have been found to moderate the heritability of their children intelligence, as stated by Turkheimer et al. .
Because there are many risk factors that contribute to cognitive achievement, examining these factors in a cumulative risk model is valuable because the cumulative risk may be more influential than any specific risk factor alone in predicting negative child development outcomes .
Cox and Howarth  reported significant differences in drawing by normal children and by children with developmental delays lagging 4 years behind the normal.
This study, regarding IQ results, revealed some findings.
First, from [Table 2], the mean IQ (99.73 ± 15.47) is considered higher than IQ measured by many previous studies that performed to assess international IQ in many countries worldwide. For example, according to one study (Flynn effect in Turkey), the population was estimated to have a mean IQ of 87.41 points by DAP test in turkey in 1977 and a mean IQ of 90.93 points . Moreover, in another study (the DAP test for Ghana), the estimated mean IQ was 88 (male) and 84 (female) by DAP test in Ghana .
The second point from [Table 3] was no significant difference in IQ levels between male and female. This is similar to a study conducted by Flynn et al. , who found that male and female obtained roughly equal IQ scores on Raven's progressive matrices after reviewing recent standardization samples in five modernized nations.
On the contrary, Lynn  stated that although it is correct that there is no sex difference in average intelligence between the ages of 5 and 15 years, from the age of 16 years, males begin to have greater average intelligence than females.
This disagrees with a 2004 meta-analysis by Irwing et al.  who found that the mean IQ of male exceeded that of females by up to five points on the Raven's progressive matrices test.
The third finding [Table 3] states that children who lived in an urban area had higher IQ levels in comparison with those lived in rural areas, which may indicate strong positive correlation between residency and cognitive functions. This supports the study conducted by Emmett, which showed that the rural school children obtained lower scores than urban children, and that the spread in intelligence among them was also smaller .
Tabriz et al.  found that the children's IQ, as determined by the Wechsler Intelligence Scale for children, administered as part of the study, was positively related to living in metropolitan and urban areas and the level of the children's fathers' education.
On the contrary, Naomi et al.  in a study 'Stability and Change in Children's Intelligence Quotient Scores' found that IQ was lower among urban children than those of rural residence. Further research directed to investigate the relation between children IQ and residence may be helpful in recognizing the residence-related causes that may affect IQ levels, such as quality of education, nutritional status, genetic admixture, and pollution.
The fourth finding from [Table 3] also shows that a strong positive correlation was found between SESs and IQ levels; those children living in high SES have higher DAP test scores and IQ levels in comparison with those with average and low SES.
This finding agrees with Fernald et al. , who found a positive correlation between SESs and IQ levels; children who live in high SES have a higher IQ. Children from disadvantaged family backgrounds score on average lower on intelligence tests than their higher SES peers , and their performance has been suggested to worsen over time, even if they did relatively well in early assessments. Conversely, high SES children are thought to gain in intelligence over time, even if they initially had a lower test score .
We think that this may be because of specific risk factors related to the SES, such as lack of family resources and parental support, which may lead to low IQ levels.
Sirin  conducted a meta-analysis based on data of roughly 100 000 students, and he found a weak to moderate relation between cognitive ability and parental SES. This finding is similar to what was reported by 1996 American Psychological Association task force's report on intelligence. Given this correlation, the question arises as to whether it is IQ or SES which causes a variation in the other variable. The answer is almost certainly both .
The fifth finding from [Table 3] was the positive correlation between IQ levels and school achievement; those children had higher school achievement 'grade A' had the highest DAP test scores and IQ levels among the whole studied sample, and vice versa.
Our study came in line with a study in UK schools was to relate scores on statewide standardized achievement tests to measures of cognitive skills in a large and representative sample of students in a city that includes traditional district, exam, and charter public schools. They found substantial positive correlations between cognitive skills and achievement test scores, especially in math. These correlations are consistent with prior studies relating working memory to academic performance (grades) in UK schools .
According to Laidra et al. , students 'achievement relies most strongly on their cognitive abilities through all grade levels.
| Conclusion|| |
From the results of the study, IQ levels obtained by DAP test were positively correlated with SESs, residences, and school achievement. No correlation was found between IQ levels and sex.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Fan RJ. A study on the kinetic family drawings by children with different family structures. Int J Art Des Educ2012; 10
Weiner I, Greene R. Handbook of personality assessment
. Hoboken, NJ: John Wiley and Sons; 2008. 483.
Cox MV. Drawings of people by australian aboriginal children: the intermixing of cultural styles. Int J Art Des Educ 1998; 17
Naglieri JA. Draw-a-person a quantitative scoring system manual
. San Antonio, TX: The Psychological Corporation; 1988.
Sparrow SS, Davis SM. Recent advances in the assessment of intelligence and cognition. J Child Psychol 2000; 41
Turkheimer E, Thomas CF, Oltmanns T. factorial structure of pathological personality as evaluated by peers. JAbnorm Psychol 2003; 112
Stevens GD. Gradients in the health status and developmental risks of young children: the combined influences of multiple social risk factors. Matern Child Health J 2006; 10
Cox MV, Howarth C. Human figure drawing of normal children and those with severe learning difficulties. Brit J Dev Psychol 1989; 7
Rindermann HI, Schott T, Baumeister AE. Flynn effect in Turkey: a comment on intelligence. Int Kagitcibasi Biricik 2011; 41
Flynn J, Rossi-Casé L. Modern women match men on Raven's Progressive Matrices. Pers Indiv Differ 2011; 50
Lynn R. Sex differences in brain size and intelligence. A paradox resolved. Pers Indiv Differ 1994; 17
Irwing P, Lynn R. Sex differences in means and variability on the progressive matrices in university students: a meta-analysis. Br J Psychol 2005; 96
Tabriz A, Sohrabi MR, Roodaki A, Parasay S, Abadi A, Kiapour N, et al
. Relation of intelligence quotient and body mass index in preschool children: a community-based cross-sectional study. Nutr Diabetes 2015; 5
Naomi B, Howard D, Ezra S, Thomas M, Kung-Yee L, Edward L. Stability and change in Children's Intelligence Quotient Scores: a comparison of two socioeconomically disparate communities. Am J Epidemiol 2001; 154
Fernald LC, Weber A, Galasso E, Ratsifandrihamanana L. Socioeconomic gradients and child development in a very low income population: evidence from Madagascar. Dev Sci 2011; 14
Schoon I, Jones E, Cheng H, Maughan B. Family hardship, family instability, and cognitive development. J Epidemiol Community Health 2012; 66
Feinstein L. Very early cognitive evidence. London School of Economics, Centre for Economic Performance. Center Piece; 2003, pp. 24–30.
Selcuk RS. Socioeconomic status and academic achievement: a meta-analytic review of research. Rev Educ Res 2005; 75
Neisser U, Boodoo G, Bouchard JR, Boykin AW, Brody N, Ceci SJ, et al
. Intelligence: knowns and unknowns. Am Psychol 1996; 51
Alloway TP, Passolunghi MC. The relations between working memory and arithmetical abilities: a comparison between Italian and British children. Learn Individ Differ 2011; 21
Laidra K, Pullmann H, Allik J. Personality and intelligence as predictors of academic achievement: a cross-sectional study from elementary to secondary school. Pers Individ Dif 2007; 42
[Table 1], [Table 2], [Table 3]