Eur Child Adolesc Psychiatry (2007)

16:389–397 DOI 10.1007/s00787-007-0611-6 ORIGINAL CONTRIBUTION

Francesca Piperno Evaluation of family drawings of physically

Stefania Di Biasi
Gabriel Levi and sexually abused children

j Abstract Objective The aim of drawings generally show clear

Accepted: 18 January 2006
Published online: 30 March 2007 this study is to analyse the family signals of trauma and the majority
drawings of two groups of physi- of the abused children are likely to
cally and/or sexually abused chil- exclude their primary caregiver
dren as compared to the drawings from the drawings. Conclusions
of non-abused children of a The ‘‘drawings of the family’’ of
matched control group. Methods physically and/or sexually abused
The drawings by 12 physically children significantly evidence a
abused, 12 sexually abused and 12 greater emotional distress then the
F. Piperno (&) non-abused children, all aged drawings of the non-abused chil-
Developmental Psychiatry Section, between 5 years-old and 10 years- dren of the matched control
Department of Child and Adolescent old, were assessed and compared. group.
Psychiatry
Trauma and Child Abuse Research Group Family drawings were analysed
University of Rome ‘‘La Sapienza’’ using a specific Screening Inven- j Key words abused children –
via Dei Sabelli, 108 tory (FDI-Family Drawing Inven- sexually – physically –
Rome 00185, Italy tory). This Inventory takes into drawings
Tel.: +39-06/44712211
E-Mail: [email protected]
consideration such qualitative and
quantitative variables as the qual-
S. Di Biasi ity of drawing, the children’s per-
Department of Child and Adolescent
Psychiatry
ception of their family members
Trauma and Child Abuse Research Group and their own perception of
University of Rome ‘‘La Sapienza’’ themselves within the family sys-
via dei Sabelli, 108 tem. Results The results have
Rome 00185, Italy shown significant differences be-
G. Levi tween the abused minors and the
Department of Child and Adolescent control group. Abused children
Neurology and Psychiatry
University of Rome ‘‘La Sapienza’’
are more likely to draw distorted
via dei Sabelli, 108 bodies, the human figure is usually
00185 Rome, Italy represented devoid of details, their

these models are developmental and considers the

Introduction ways in which children interpret and respond to
A number of theorists have recently presented models traumatic experiences at different ages.
for understanding the effects of childhood trauma In order to conduct a comprehensive assessment of
[17]. These models are integrative and consider the the impact of traumatic stress, Koverola [19] suggests
child as a whole person, in addiction they are con- that we look at the child’s functioning in many do-
ECAP 611

textual, considering the child in the context of the mains, including the cognitive, affective, moral,
family, peer group and the larger society. Thirdly, interpersonal, sexual and physical. It is important to
390 European Child & Adolescent Psychiatry (2007) Vol. 16, No. 6
 Steinkopff Verlag 2007

consider how each of these domains might have been Drawing the family is regarded as an elective tool
affected by the trauma. In addiction he points to the of investigation of the mental representation of
need to assess sources of resiliency and competence, attachments ‘‘in tune with the concepts of develop-
such as the child’s coping strategies and self-esteem, mental psychology’’ [31].
which may moderate the risk of psychopathology. The FD projects the image of the family as per-
As a part of a larger assessment, drawing analysis ceived by the child in its growth; it captures the fan-
is one of the elective techniques employed in child tasies that combine the child’s subjective life
psychiatry [3, 9, 11, 16–18, 20, 22–24]. We think that experiences and their meetings with the objective
those techniques could help professionals to under- outside world.
stand the emotional distress of abused children. The use of FDs as a projective assessment instru-
Information gained from drawings may be useful in ment in prospective studies of child abuse is regarded
the overall assessment process of a distressed child by the specific literature to be a particularly signifi-
(physically, sexually abused children). cant device of clinical investigation and research [35,
In the past several studies have questioned the 36]:
validity of drawings analysis as a projective test [14, by drawing itself as an integral part of a family, the
32, 33, 36, 37, 39]. Nevertheless both psychoanalysis child will be able to express otherwise inhibited
and developmental psychology regard drawing as the thoughts and feelings;
channel that allows the child to express his/her dis-
• the clinician may be enabled to approach pro-
comfort, anguish and the self-defensive mechanisms
spective family conflicts;
against pain.
• Abused children often develop speech impediments
‘‘Drawings are often called upon by professionals
and/or difficulties in verbal communication. Draw-
as a method of allowing a child to communicate more
ings can help physically and/or sexually abused
freely, no language being necessarily involved, as well
children to direct their emotions and graphically
as a way of ‘‘breaking the ice’’ between the child and
express their life experiences.
professional’’[39].
Children’s drawings are used occasionally as evi- Veltman and Browne [37, 38], have conducted an
dence of maltreatment in case conferences and court exhaustive review of the literature concerning the
cases [10]. Indeed, The American Bar Association quality of drawings in physically and sexually abused
supports the use of drawings as part of a child’s tes- children. The review produced a total of 317 refer-
timony [7], especially in cases where the maltreatment ences, but the analysed 23 relevant studies. There
is alleged. For these purpose, projective drawing were 15 different drawing techniques used in the
techniques are largely concerned with evaluation of studies.
emotional states of the child. Drawings are useful for The FDs were used in 5 studies and the most
the identification of emotional problems in children. considered variables in the specialized literature [13,
Its use is all the more valid when one considers that 14, 33, 34] were:
children are able to convey in their drawings thoughts
and feelings they cannot express in speech or writing. 1. Size and enhancement of the drawn subjects: the
This may be particularly true of abused children size of the human figure reflects the children’s
(physically and sexually abused children) who feel own self esteem; it is important not to consider
fear to talk about the abuse. the size of the whole human figure only, but also
Only two drawing techniques specifically convinced the size and the emphasis of specific parts of the
for the assessment of physically and sexually abused body.
children were identified: The Favourite Kind of Day 2. The placemant of the human figures in the drawing
Drawing [38] and the Kinetic Family Drawings (FDs) [40]. (affective distance): the distance between members
Family Drawing is widely employed as a test in of the family could be a graphic representation of
psychodiagnostic assessment as it enables the clini- the actual emotional distance between the charac-
cian to make contacts with: terized individuals.
3. Omission of subjects or of parts of the body: the
• the perception child has of itself, omission of themselves reveals a seriously low level
• the perception child has of its parents, of self-esteem.
• some indicators of the development of its mental
organization. Carpenter et al. [4] found that physically abused chil-
dren were more likely to distort the bodies they draw,
The FD Test [8] aims to investigate the relationship use lack of details, poor body image and sexual iden-
that children have established with their parent fig- tification; they were less likely to include their primary
ures and with other family members. caregiver and themselves. It was further found that
 F. Piperno et al. 391
Drawings of the family

sexually abused children used faint lines, physically Table 1 FDI-Family Drawings Inventory
abused children heavy lines, and children from violent
Graphically maturity Adequate, no adequate
homes were found to use heavy outlines [4]. Omissions Parents, Self, no omissions, the total family
The evidence from research found in this review Body Distortion Distortions, no distortions
does indicate that caution should be employed in the Identification Parents, self, others, no identifications
use and overinterpretation of drawings. Emotional Proximity Parents, siblings, nobody
The aim of this study is to analyse the FDs of two
groups of physically and/or sexually abused children
and compared to the drawings of non-abused chil- drawing Inventory) used for analysed the FDs of the
dren of a matched control group. sample.
We intend to investigate whether the FD repre- All subjects completed a semi-structured interview,
sentations will differentiate: a brief version of a psychopathology assessment, the
Kiddie-SADS-patien version [15], The Wechsler Scale
• Drawings of abused children (physically and sexu- od Intelligence Revised (WIPPSI-R; WISC-R). The
ally) from normal ones; Bender Test [1] was used to evaluate the drawing
• Physically abused from sexually abused children. capacity.
Exclusion criteria included children who had
experienced multiple types of abuses (both physical
j Methods and sexual abuse) and children with Developmental
Disorders (Mental Retardation, Autistic Disorder,
Participants were part of a prospective study of the Language disorders and Learning Disorders). We
Psychiatry Developmental Section at the Department have decided to exclude children with Developmental
of Child ad Adolescent Psychiatry, University of Disorders because these children usually show draw-
Rome, ‘‘La Sapienza’’. ing difficulties.
36 subjetcs (18 female, 18 male), 5–10-years-old
were included in the study.
All partecipants were causasian and have the same j Measures
social–economic background.
The total sample was subdivided in three groups: The distributions of sex, checked with the v2-test,
have resulted homogeneous in the three assessed
• GROUP M: 12 physically abused children (4 girls groups. The distribution of the age, I.Q., Bender test,
and 8 boys; between 5 and 9 years old); checked with T-Test for independent groups, have not
• GROUP A: 12 sexually abused children (7 girls and shown statistically significant results.
5 boys; between 5-years-old and 10-years-old); In analyzing the FDs we have developed an Specific
• GROUP C: 12 children from a matched control Screening Inventory (Table 1: FDI-Family Drawings
group (7 girls and 5 boys; between 6-years-old and Inventory).
9-years-old). This Inventory takes into consideration such quali-
tative and quantitative variables:
A sample of 24 case of abused children (physically
and sexually) were recruited at the department of • Graphically-expressive maturity: the aim is to
child and adolescent psychiatry at the University of measure, using Piaget’s parameters [16, 26–30]
Rome ‘‘la Sapienza’’, Italy. whether the graphic level is compatible with the
These cases were at random selected from abused child basic cognitive organization;
(physically and sexually) children diagnosed in the • Omitted subjects: the omissions of some family
past 2 years (2001–2003) at the department of child members reflect a modality that the child may use
and adolescent Psychiatry and were compared with 12 to belittle a subject by removing an imaginary
non-abused children, between 5–10-years-old, at person and thus communicate something either
random selected from two different primary schools unfulfilled or unaccepted by its family;
of the Rome district. The control children were se- • Body distortions: This variable gives us a measure
lected by the child’s class teacher for age, sex, socio- of the care and attention that the child puts into
economic background. Exclusion criteria included drawing human figures and of his capability to
Family violence, neglect, divorce, bereavement of the portray the body in a reasonable way in relation to
mother/father, Developmental Disorders (Mental shape, size and details;
Retardation, Autistic Disorder, Language disorders • Identification roles: this concerns the choice of
and Learning Disorders). ‘‘persona’’ or the character that the child will
The comparison of three groups was made between choose when questioned: ‘‘who would you like to
the variables of the Screening Inventory (FDI-Family be?’’;
392 European Child & Adolescent Psychiatry (2007) Vol. 16, No. 6
 Steinkopff Verlag 2007

• Emotional proximity: the distance the child draws 12

between itself and the other family members. 10
The screening categories for coding the FD were 8
derived from the research of different authors [5, 6,
32, 40]. 6
Family Drawing, according to different researches
4
[39] can be used to identify children in distress by
focusing upon the family dynamics drawn within the 2
drawing.
The study utilized the convergence of the indica- 0
GROUP M GROUP A GROUP C
tors to obtain diagnostic evidences concerning the
quality of feelings, either of the depressive or of the ADEQUATE INADEQUATE
relational anxiety/anguish type.
The categories about the quality of the feeling Graph 1 Representation maturity
(depression, anxiety) were derived from the study of
Luigi Castellazzi [5, 6]. views. Assessors were supported by clinical team and
decisions were reviewed and confirmed through
j Depressive feelings consensus of the team.
The distribution of each variable has been tested
At least three of the following criteria must be met: by comparing the groups with all possible coupling
(M–A; M–C; A–C), using the v2-test. In case of charts
• Thin lines;
with void columns, the v2-test has been applied only
• The drawing situated in the lower part of the paper;
on those columns that showed some remarks.
• Small dimensions;
• Static-schematic family or drawn with basic ‘‘stick-
like’’ lines. Results
• Menacing objects and/or clouds, dark sun.
j Graphic-representative maturity (Graph 1)
j Feelings of relationship anxiety and/or anguish
The FD graphic level of children who have been
At least three of the following criteria must be met: abused (physically and/or sexually) results below the
basic cognitive organization.
• Family drawn in a basic-schematic way; Children who have been physically abused show a
• Family drawn in simple schematic way with graphic-representative maturity significantly inferior
essential marks (stick-like); compared to children who have been sexually abused
• Lines or objects (birds, clouds, various markings) (v2 = 5.042; P = 0.023), and compared to the non-
drawn above the heads; abused matched control group (v2 = 20.007; P =
• Caricatural/grotesque or completely absent family; 0.001).
• One or more subjects outlined or enhanced. Children who have been sexually abused: half of
the subjects produce drawings suitable to their
The FD test was administered in a standardised chronological age; even though they show a superior
manner: graphic-pictorial capacity compared to the physically
• Individually session without family members; abused children, their differences with the children of
• Drawing material: white papers,one black-lead the matched control group are statistically significant
pencil, coloured pencils; (v2 = 8; P = 0.005).
• The only instructions given to che subjects was: Children from the matched control group produce
‘‘draw a not real family’’. in fact drawings adequate to their chronological age in
12 cases out of 12.
The FD test was part of a larger assessment protocol Graphic 1 shows the graphic level of the drawings
for children who have been abused (physically and/or in the three samples.
sexually).
All participants provided written informed consent
before taking part in this study. Assessments were j Omission of subjects (Graph 2)
carried out by mental-health professionals (child and
adolescent psychiatrists, psychologists) who were Parents are the most absent figures in the drawings of
trained and experienced in conducting clinical inter- both physically and sexually abused minors: 67% of
 F. Piperno et al. 393
Drawings of the family

12 8

7
10
6
8
5

6 4

3
4
2
2
1

0 0
PARENTS SELF NO OMISSIONS THE FAMILY PARENTS SELF OTHERS NO IDENTIFICATION
GROUP M GROUP A GROUP C
GROUP M GROUP A GROUP C

Graph 2 Omissions of figures

Graph 4 Identification

12

10
Research has proved the evidence of a statistically
significant difference between group M and group C
8
(v2 = 16.666; P = 0.001) and between group A and
group C (v2 = 10.971; P = 0.001).
6 The experience of abuse (physical and/or sexual)
results to be correlated in a statistically significant
4 way to a modified and distorted graphic representa-
tion of the body image.
2 Graph 3 shows how the children (physically–sex-
ually and normal children) distorted the graphic ra-
0 presentation of the body image.
GROUP M GROUP A GROUP C

DISTORTION NO DISTORTION j Identification (Graph 4)

Graph 3 Body distortion
A statistically significant difference results if we
compare children of Group M with those of Group A
the physically abused children and 42% of the sexu- (v2 = 6.476; P = 0.038).
ally abused ones. Physically abused minors tend to identify in the
No statistically significant difference results be- 67% of cases with themselves, 33% with others (ani-
tween the two clinical groups (v2 = 2.692; P = 0.556). mals, friends, brothers and sisters, etc.) and only 0%
A statistically significant difference does show be- of the assessed cases identifies with their parents or
tween children who have been physically abused and with nobody; 42% of the sexually abused children will
the matched control group (v2 = 12.571; P = 0.002) identify with nobody, 33% with themselves, 25% with
and between the sexually abused children and control others, and 0% with their parents.
group (v2 = 11.571; P = 0.009). The study furthermore reveals a statistically sig-
Graph 2 shows how the children of the three nificant difference between Group M and Group C
samples omitted some family members. (v2 = 12.8; P = 0.005). Children of the matched con-
trol group identify in 50% of cases with one of their
j Body distortion (Graph 3) parents, 42% with others (brothers, sisters), 8% with
themselves, and 0% with nobody. Graph 4 shows the
On a graphic level we have reported the immaturity in choice of identification in the three samples.
the drawings of 17 of the children in our sample
groups. j Emotional proximity (Graph 5)
Distribution of the variable tested by v2 has not
proved statistically significant in comparing physi- Comparison between group M and group A does not
cally abused and sexually abused children (v2 = 1.2; bring to our attention any statistically significant
P = 0.273). difference (v2 = 1.8; P = 0.591).
394 European Child & Adolescent Psychiatry (2007) Vol. 16, No. 6
 Steinkopff Verlag 2007

12 absolutely no details, by distorted bodies and by their

drawing themselves small and isolated.
10

Discussion
6

The findings from the recent literature suggest that

4
drawing techniques are largely concerned with
evaluation of emotional states of children, because
2
drawings are useful for the identification of emo-
tional problems in children. Their use is all the more
0
GROUP M GROUP A GROUP C valid when one considers that children are able to
convey in their drawings thoughts and feelings they
PARENTS SLIBLINGS NOBODY
cannot express in speech or writing. This may be
Graph 5 Affective proximity particularly true of abused children (physically and
sexually abused children) who feel fear to talk about
the abuse. Only two drawing techniques specifically
In Group C the totality of children, 12 out of 12, convinced for the assessment of physically and sex-
graphically locate themselves near, or between, the ually abused children have been identified: The
parent couple. Favourite Kind of Day Drawing [38] and the Kinetic
Assessment and comparisons of the drawings have FDs [40].
highlighted a statistically significant difference be- We have decided to use the FD Test [8] because it
tween group M and group C (v2 = 14.4; P = 0.001) is widely employed as a test in psychodiagnostic
and between group A and group C (v2 = 20.307; assessment and it aims to investigate the relationship
P = 0.001). that children have established with their parents and
Graph 5 shows the distance that the children draw with other family members.
between themselves and other family members. The FD is not a test to determine if a child has been
abused or not, but it’s a good ‘‘method of allowing
j Depressive feelings abused children to communicate more freely’’ and to
express their feelings about their family [37, 39]. The
No statistically significant differences emerge between FD helps professionals to recognise emotional distress
the clinical sample groups (v2 = 0.75; P = 0.609). A in children, but it cannot be used as an indicator of
significant difference does show in comparing groups abuse (physical or sexual abuse).
M and C (v2 = 14.4; P = < 0.001), and groups B and C Carpenter [4] has found that physically children
(v2 = 9.882; P = 0.002). were more likely to distort the bodies they draw. In
Both groups express their emotional distress in our sample, physically abused children draw sche-
similar ways and both differ from the matched control matic bodies, often barely outlined or sketched. The
group. Their mental affliction encompass feelings of drawings are poor of details and the use of perspec-
sadness, loneliness and anguish: the threat is repre- tives is missing even in children older than seven
sented hanging over their heads, like sharp pointed years old. The most disturbing fact concerns the
shapes, black clouds or enormous menacing birds. Or distortions and schematization of the body, the ab-
a motionless figure that stands out alone on the white sence or deformity of the face. Distortion suggests a
paper. poor image of its own body.
Children who have been sexually abused don’t
show homogeneous characteristics, only half of them
j Relationship anxiety/anguish feelings appear to be inadequate in the graphic representation.
A significant number of the assessed sexually
No statistically significant difference emerges between abused children draw distorted bodies. Physical con-
the clinical standards (v2 = 0; P = 1.00). There is a tact with adult sexuality generates an altered percep-
significant difference between group A and group C tion of their own body model. Puppets are devitalized
(v2 = 6.75; P = 0.009) and between group A and by drawing faces void of any expression.
group C (v2 = 6.75; P = 0.009). The are no substancial We think it noteworthy to emphasize how the
diversities between the two clinical groups but they ‘‘eroticization of the picture,’’ one of the aspects
are both significantly different from the matched generally considered important in the sexual abuse
control group. Their suffering is shown by faces with literature [39] in our study has emerged on two
 F. Piperno et al. 395
Drawings of the family

children only out of a total of 12. We think that sexual

abuse creates a deformation much more complex and
involving the whole body.
In the control group both distortion and schema-
tization are absent. These children accomplish artic-
ulate drawings, the body is rich in details and the
human figures are complete.
We may consider the graphic poverty and the
immaturity of the drawings as a first indicator of the
distress. The distress involves the mind functions in
the symbolic-representative domain. Fig. 1 Family drawing of a physically abused child
Intelligent children who draw in an immature way
are seeking the specialist’s attention, but only a
comprehensive assessment will indicate the diagnos-
tic and the therapeutic direction that needs to be
undertaken.
The other important indicators of distress in our
study are the omission of parents and the identifica-
tion.
In the drawings of Physically abused children the
parents are the most omitted figures. Outside figures
not pertaining to the family group are added to the
picture, mostly friends or children of the same age,
animals and/or objects.
The omission of the parents shows problems in the
identification process. To remove the caregiver is an
important indicator of emotional distress (D., (Fig. 1),
instead of his family, draws a little fish that calls for
‘‘HELP,’’ while a bigger fish is about to swallow it).
Fig. 2 Family drawing of a sexually abused child
All physically abused children exclude their par-
ents from their drawings. Their wish to be someone
else is surmounted by a reality choice: to be them-
selves. The literature on the subject shows that it is all- confirms that sexual abuse involves the whole family.
important, in the years of growth, the choice of an The wish to be themselves confirms the children’s
‘‘engram’’ revolving around the idea of a family with a attempt to rely on their own resources, but the wish
father, mother and child, and that 75% of the children not to be like anybody else appears to be the extreme
will choose this particular ‘‘engram’’ [31]. resource of a child crushed by reality who cannot
(C. draws the picture of a little house in a corner and even find a way to escape.
comments ‘‘the house of a puppy... father breaks down It is significant that almost the totality of the as-
the house... they find a kitten, but the dog kills her’’). sessed sexually abused children portray themselves in
Half of the assessed children draw themselves total isolation. The wish of isolation from the family,
alone or isolated. Only three of them are standing and more specifically from the parents, may put these
by their mother and three of them are near a children at risk to miss the all important investments
brother. that are essential requisites for the identity process.
Sexually abused children will either erase them- Castellazzi et al. [5, 6] have shown that usually
selves or their parents. In one instance the whole normal/non-abused children draw not real and com-
family is removed (G., Fig. 2, draws only a stylized plete family, with all the family members. Our control
house in the midst of a very rough sea; S. Draws three group confirms these dates: children from the normal
figures suspended in mid air, with no hands, and a sample group never omit one of their parents. The
male figure, but she leaves herself out of the picture). Parent couple is eloquently enhanced. All normal
About the identification, all sexually abused chil- children draw themselves next or in between their
dren exclude their parents from their choices. Both parents.
the wish to be themselves and the wish to be ‘‘no- (G.,, draws her parents that walk arm in arm while
body’’ will count in equal measure. she smiles at them; L., Fig. 3, draws a family group
It is interesting to note that both abusive and non- rich in details and the parents couple surrounds three
abusive parents are perceived as undesirable, this differently aged children).
396 European Child & Adolescent Psychiatry (2007) Vol. 16, No. 6
 Steinkopff Verlag 2007

lege an interpretation that would reflect the emotional

point of view the child has of itself and its family [1,
25, 26].
Our data, in concordance with literature [4–6, 12,
18], show that FD characterizes the normal from the
emotionally stressed populations. All the variables
that have been taken into consideration distinguish in
a statistically significant percentage clinical cases
from the control group.
Children who have suffered any form of abuse,
(either physical or sexual) tend to project their inner
drama in their drawings: their past experiences of
loneliness, esclusion, anxiety, and inadequacy emerge
both in the graphic and in the contentual analysis.
Our results and the last researches cited confirm
that the FD is a good method to identify the child
emotional distress in physically and sexually children.
It seems that the most significant indicators of emo-
tional distress are the body distortion, the omission of
the parents and the identification.
In Sexually abused children the answer ‘‘nobody’’
gives us a clear signal of a rising disinvestments from
Fig. 3 Family drawing of a non-abused child
emotional relations.
All children who have been exposed to violence
About identification, normal children are equally and to the abuse of their bodies, but specially of their
divided between the wish to be one of their parents or minds, reflect a pain, a distress, in the Ôgrand theatre
one of their brothers, these conclusions are confirmed of affectivity’’ [21].
by the literature on the subject [2] where it is shown Family Drawing cannot single out individual cases
the need of models but also the tensions of competi- by traumatic events but can describe their effects and
tion between both brothers and sisters, a compatible the psychopathological points of view and thus help
process with the normal family dynamics. the clinicians in their understanding of children
The Fig. 1 shows a FD of a physically abused child, mental distress.
the Fig. 2 of a sexually abused child and the Fig. 3 of a One major limitation of this study is the small size
normal child (control group). that limits the interpretations of the results. A more
comprehensive study involving more participants is
required in order to evaluate the real significance of
Conclusion the FD Test with abused children.
The findings of this study confirm that FD is not
The limited number of assessed cases does not allow suitable for the identification of abused children, but
us to draw conclusions so we limit ourselves to sug- it is a good method to recognize the emotional dis-
gest cues for reflections on the problem. tress in children. It should be used as a part of a larger
According to the most recent literature on the assessment to help professionals (child and adoles-
subject we have chosen to distance our study from cent psychiatrists and psychologists) to understand
Corman’s traditional interpretation [8] and to privi- better suffering children.

References
1. Bender LA (1938) A visual Motor Ge- 2. Bruening CC, Wagner WG, Johnson JT 3. Burgess AW, Hartman CR (1993)
stalt Test and its clinical use research (1997) Impact of rater knowledge on Children’s drawings. Child Abuse Negl
monograph. Am Ortopsychiatric As- sexually abused and nonabused girls’ 17:161–168
sess, New York scores on the Draw-a-Person: screening
procedure for emotional disturbance. J
Pers Assess 68:665–677
 F. Piperno et al. 397
Drawings of the family

4. Carpenter M, Kennedy M, Amstrong 17. Kerig PK, Fedorowics AE, Brown CA, 31. Pratt C (2001) The predictive impact of
AL, Moore E (1997) Indicators of abuse Warren M (2000) Assessment and domestic violence on three types of
or neglect in preschool children’s intervention for PTSD in children ex- child maltreatment. Child Abuse Negl
drawings. J Psychosoc Nurs Ment posed to violence. The Haworth Mal- 25:869–883
Health Serv 35:10–17 treatment & Trauma Press 3:161–184 32. Tambelli R, Zavattini GC, Mossi P
5. Castellazzi VL (1996) Il test del disegno 18. Koch K (1949) Der Baumtest. Hans (1995) Il senso della famiglia. Le rela-
della famiglia, LAS, Roma Hube, Bern zioni affettive del bambino nel ‘‘Dise-
6. Castellazzi VL, Nannini MF (1992) Il 19. Koverola C, Pound J, Heger A, Lytle C gno della Famiglia’’. La Nuova Italia,
disegno della figura umana come tec- (1993) Relationship of child sexual Roma
nica proiettiva. LAS, Roma abuse to depression. Child Abuse Negl 33. Thomas GV, Gray R (1992) Children’s
7. Cohen-Liebman M (1995) Drawings as 17:393–400 drawings of topics differing in emo-
judiciary aids in child sexual abuse 20. Leo J (1973) I disegni dei bambini tional significance: effects on place-
litigation: a composite list of indicators. come aiuto diagnostico. Giusti Barbera, ment relative to a self drawing. J Child
Arts Psychother 22:475–483 Firenze Psychol Psychiatry 33:1097–1104
8. Corman L (1967) Le test du dessin de 21. Levi G (1993) Trauma, rappresentazi- 34. Thomas GV, Jolley RP (1998) Drawing
famille dans la pratique medico-peda- one e violenza mentale. Imago. 1:2–3 conclusions: a re-examination of
gogique. Presses Universitaires de 22. Liberman A, Van Horn P (1998) empirical and conceptual bases for
France, Paris Attachment, trauma, and domestic psychological evaluation of children
9. Cox M (1993) Children’s drawings of violence: Implication for child custody. from their drawings. J Clin Psychol
the human figure. LEA, Hove Child Adolesc Psychiatr Clin N Am 37:127–139
10. Czenner Z (1986) The reliability of 7:423–443 35. Thomas GV, Silk AMJ (1990) An
information gained by a child’s draw- 23. Luquet GH (1913) Les Dessin D’un introduction to the psychology of
ings. Acta Med Leg Soc (Liege) 36:119– Enfant: Etude Psychologique. New children’s drawings. Harvester Wheas-
207 York, Alcan theaf, New York
11. Di Leo J (1970) Young children and 24. Luquet GH (1927) Il disegno infantile. 36. Veltman MW, Browne KD (2000) Pic-
their drawings. Brunner/Mazel, New Armando, Roma tures in the classroom: can teachers
York 25. Machover K (1949) Personality projec- and mental health professionals iden-
12. Ferro A (1992) La tecnica nella Psico- tion in the drawings of the human fig- tify maltreated children’s drawings?
analisi Infantile. Edizioni Scientifiche ure. C.C. Thomas, Springfield Child Abuse Rev 9:328–336
Magi, Cortina 26. Mortone N, Browne K (1998) Theory 37. Veltman MW, Browne KD (2001) Three
13. Freeman NH (1980) Strategies of rep- and observation of attachment and its decades of child maltreatment research
resentation in young children. Analysis relation to child maltreatment: a re- implications for school years. Trauma,
of spatial skills and drawing processes. view. Child Abuse Negl 22:1093–1104 Violence Abuse 2(3):215–239
Academic Press, New York 27. Piaget J (1927) La causalitè phisique 38. Veltman MW, Browne KD (2001)
14. Hammer E (1997) Advances in projec- chez l’enfant. PUF, Paris Identifying childhood abuse through
tive drawing interpretation. Charles C. 28. Piaget J, Inhelder B (1956) The child favourite kind of day and kinetic family
Thomas, Springfield, IL conception of space. Routledge & Ke- drawings. Arts Psychother 28(4):251–
15. Kaufman J, Birmaher B, Brent D (1997) gan Paul, London 259
Schedule for affective disorders and 29. Piaget J, Inhelder B (1969) The psy- 39. Veltman MW, Browne KD (2002) The
schizofrenia for school-age children- chology of the child. Routledge & Ke- assessment of drawings from children
present and lifetime version (K-SADS- gan Paul, New York. Basic Books, who have been maltreated: a systematic
PL): initial reability and validity data. J London review. Child Abuse Rev 11(1):19–37
Am Acad Child Adolesc Psychiatry 30. Piaget J, Inhelder B (1971) Mental 40. Veltman MW, Browne KD (2002)
36:980–988 imagery in the child. Basic Books, New Trained raters’ evaluation of Kinetic
16. Kellogg R (1970) Analysing chil- York Family Drawings of physically abused
dren’art. National Press Books, Palo children. Arts Psychother 625:1–10
Alto, CA