A COMPARISON OF." AMERKCAN AND HLIPINO
CHILDREN BY MEANS OF THE
DRAW-A.PERSON TEST

Thesis for “10 Degree 0‘ M. A.
MICHISAN STATE UNWERSITY

Josefina Averilla Limuaco

3.959

J

”’9

‘_. -n u. .4-‘—_'_“'" ' ‘. _.

L [B R A R Y

'Michigan State

University a.
5?

 

A COMPARISON OF AMERICAN AND FILIPINO CHILDREN

BY MEANS OF THE DRAW-A—PERSON TEST

By

Josefina Averilla Limuaco

A THESIS

Submitted to the College of Science and Arts

Michigan State University of Agriculture and '

Applied.8cience in partial fulfillment of
the requirements for the degree of

MASTER OF ARTS

Department of Psychology

1959

Dedication

To My Fami 1y

ACKNOWLEDGMENTS

The author wishes to express her utmost and sincere gratitude
to Dr. Albert I. Rabin, who in his capacity as chairman of this thesis
committee, has been very patient and helpful in guiding the author in
this undertaking.

Further gratitude is also expressed to Dr. Mary Haworth for her
helpful suggestions and to Dr. Terrence Allen for his guidance on the
statistical analysis.

Also a note of thanks to Mrs. Concepcion.Mangona, Mrs. Isabel
Fajardo, Mrs. Crisanta Limuaco Fabella and to the principals of the
Michigan Avenue School and the Kellogg Elementary School who assisted

in the collection of data.

A COMPARISON OF AMERICAN AND FILIPINO CHILDREN

BY MEANS or THE DRAW-A—PEFSON TEST

By

Josefina Averilla Limuaco

AN ABSTRACT

Submitted to the College of Science and Arts
Michigan State University of Agriculture and
Applied Science in partial fulfillment of
the requirements for the degree of

MASTER OF ARTS

Department of Psychology

1959

 

Approved ngm

1
Josefina Averilla Limuaco
ABSTRACT

The DrawaA—Person test was administered to 129 American children
and 159 Filipino children to compare them in their degree of sexual dif-
ferentiation. Of the American subjects, 69 were boys and 60 were girls.
Of the Filipino subjects, there were 76 boys and 83 girls. The ages
of the subjects ranged from.nine years to twelve years. Three predic-
tions were formulated on the basis of the different family structures

in the two cultures.

1. The American children will Show a lesser degree of sexual dif-
ferentiation than the Filipino children.

2. The Filipino boys will tend to identify more with the male
figure than the American boys.

3. The American and Filipino girls will show no differences in

their identification model.

A revision of Swensen's 9-point Rating Scale for Sexihl Different-
iation was used as an index for sexual differentiation. The revised
scale contains S~points i.e. l, 2, 3, b, 5, which correspond to the odd
numbers 1, 3, 5, 7, 9, of Swensen’s scale. The data were analyzed by
means of the chi—square to determine the significance of the differences
between the two groups on sexual differentiation. For the self-sex
identification the Goodenough Scale was utilized as the measure. .A t-
test was employed on the scores obtained from the Goodenough scale
(male and female figure drawings) to find out the difference between
the two groups on their self-sex identification. Since there was doubt

as to whether the drawing ability of the children was reSponsible for

2

Josefina Averilla Limuaco

the difference between the two groups, the analysis of covariance was
used to show the relationship between the degree of sexual differentia—
tion and the scores obtained from the Goodenough Scale.

The results supported the above predictions. The relationships be-
tween the results of the study and the parental roles in the two cul-
tures were also d13cussed. It was concluded that the Filipino children
Show a higher degree of sexual differentiation as compared to the Amer-
ican children and the Draw-A—Person test is an adequate and useful in—

strument in a cross-cultural study.

I. Introduction . . . . . . .

TABLE OF CONTENTS

I O O O O O O O O O O I O I O l

A. Importance of child-rearing practices for person-

ality development . . . . . . . . . . . . . . . . . . l
B. Child-rearing practices in the American culture . . . l
C. Child-rearing practices in the Philippine culture . . 3
D. Feeding and toilet training . . . . . . . . . . . . h
E. Socialization process . . . . . . . . . . . . . . . S
F. Identification . . . . . . . . . . . . . . . . . . 5
G. Sex typing . . .-. . . . . . . . . . . . . . . . . 6
H. Differences between the American and Filipino

children in their sexual identification . . . . . . . 7

I. Measure used in the study of the degree of the
children's sexual identification . . . . . . . .

II. Statement of hypotheses . . . . . . . . .

III. Application of drawings in the assessment of sexual ident-

ification O O O O I O O O O O I I O O O I O O O O I O

A. Historical background in the use of drawings

with children . . . . . . . . . . . . . . . . . . . ll
B. The Goodenough Scale . . . . . . . . . . . . . . . 12
C. Drawings used as projective techniques . . . . . . 12
D. The Draw-A-Person test . . . . . . . . . . . . . . 1h
E. The Scale of Sexual Differentiation . . . . . . . . 16
F. .Adaptation of Swensen's Scale to children's
drawings ._. . . . . . . . . . . . . . . . . . . 16
IV. Subjects and Procedure . . . . . . . . . . . . . . . . 18
V. Results . . . . . . . . . . . . . . . . . . . . . . . . 22

PAGE
VI. DiSCUSSion O O C O O O O O C O C C O O O O O O O O O O O 27
VII. Summary and Conclusion . . . . . . . . . . . . . . . . . 30

VIII. Bibliography 0 o o o o o o o o o o o o o o o o o o o o o 32

TABLE

I Some Characteristics of the American and Filipino Samples

LIST OF TABLES

II Correlation between judges scores on Goodenough Scale. .

III Correlation between judges rating on sexual

differentiation . .

IV" A comparison between groups on sexual differentiation

scores I O O O O O O

V A comparison between groups on the total scores obtained

from the Goodenough Scale (male figure drawing) . . . .

VI A comparison between groups on the total scores obtained

from the Goodenough Scale (female figure drawing) . . .

VII A comparison.within the culture groups on the male and

female figure drawings

0 O O O O O O O O O O C O O O O 0

VIII A comparison between groups on their degree of sexual
differentiation and scores obtained from the Goodenough

Scale (male figure)

PAGE
19
2O

21

22

2h

25

25

26

I INTRODUCTION
A. Importance of Child-Rearing Practices for Personality Development

Child—rearing practices within various cultures play an important
role in the development of the child's personality and in the parent-
child relationship. It is in the family, Seward (36) points out, that
the child first learns to love and to be loved. The success of his
early relationships with both his parents determines to a large extent
his later relationships. The mother is practically all the child knows
of society in his early infancy. The amount of time Spent by the mother
with the child gives her more opportunities to influence the child's be-
havior and social growth. Symonds (A1) is of the opinion that the
mother's behavior and attitudes are more important in the development
of the child's personality than the father's attitudes and behavior.

Yet the father, although he may Spend less time with the child, influ-
ences the child's behavior indirectly. He influences the mother's at-
titudes toward the home and the care of the child. He may have the

reSponsibility of maintaining discipline in the home.
B. Child-Rearing Practices in the American Culture

In.American society child—rearing practices, Masters (28) states,
"have the essential characteristics of a fad: they reverse themselves
every few years, destrqying any real continuity from one generation of
parents to the next. (p. 531)." Gorer (21) maintains that "the idio-
syncratic feature of the American conscience is that it is predominantly
feminine owing to the major role played by the mother (p. 56)." Child-

rearing is the mother's responsibility and privilege. Her influence on

2
the child, Gorer further states, is reinforced from the age of six in
the schools by the fact that most of the teachers are women. The role
of the father in mattensof discipline has been so diminished that these
functions have been delegated to the mother. The father, Masters (28)
claims, Spends so much time earning a living that he has very little
time to spend with the children. Thompson (A2) points out that the
father sees his children only occasionally "before bedtime, on Sundays,
and during an annual vacation (p. Slb)."

Davis and Havighurst (9) and Ericson (12) found significant dif-
_ferences in child-rearing practices of the middle and lower-class par-
ents. The middle—class parents were considerably more rigorous than
the lower-class parents in training their children for feeding and
cleanliness habits. They expected their children to take responsibil—
itdes for themselves at an earlier age than the lower-class parents.
Permissive feeding and weaning treatment was more characteristic of the
lower-class than of the middle-class mother. .A staff of investigators
at the Laboratory of Human Development at Harvard university (25) found
few differences between the social classes in infant feeding practices.
The major differences of these studies, according to Mussen and Conger”
(31) are that, at present, middle-class American mothers have adopted

more permissive feeding practices through education.
C. Child-Rearing Practices in the Philippine Culture

Studies regarding child-rearing practices in the Philippines are
not available at present. The following discussion is based on the

writer's personal experience.

The Filipino mother has the sole responsibility of taking care of
the children. The fathers are in most cases responsible for the disci-
pline of the children over the age of six. In the middle and upper—class
families the maids and "Chinese amahs" take care of the children and
Spend more time with them than do either the mother or the father.

The middle-class Filipino mothers are quite rigid in training
their children in feeding and cleanliness habits. The children are fed
on schedule and are weaned at the age of one. They are toilet-trained
before they reach the age of one; sometimes coercive methods are used
to achieve this. .A child may be punished or scolded for bedwetting.

The middle-class parents are overprotective and seem to train their
children to be dependent on them.

The lower-class Filipino mothers, on the other hand, are less con-
cerned about the feeding habits and toilet training of their children.
The children are fed on demand and are nursed for a long time. Some—
times the children are not weaned until they are at least two years old.
Toilet training of the children is neglected and the children are neither
punished or scolded for bedwetting and defecating anywhere. They may
learn cleanliness habits when they reach the age of six or seven, either
through imitation from the elder members of the family, or at school.
These parents also give more work responsibility to their children at
an earlier age.

- Filipino parents, in general, are strict and authoritarian in deal—
ing with their children. The r83ponsibility of discipline belongs not
'only to the father, but also to the older brothers and sisters. The
grandparents, aunts and.uncles who stay in the home also take part in

the rearing of the children. The only peOple the American child has

contact with ordinarily at home and possibly identifies himself with
are usually his parents, brothers and sisters. The Filipino child may
therefore identify himself with more people with whom he comes in con—

tact at home than the American child.
D. Feeding and Toilet Training

Feeding constitutes the child's earliest experience in social
participation. In accordance with the theory of Dollard and Miller
(10), the kind and degree of satisfaction the child gets from the feed—
ing situation may determine his future reactions to people. If the
child is fed in a relaxing and comfortable way and if he is cuddled and
properly held, the feeding situation becomes more pleasant and reward-
ing to the child. The relaxing quality of this experience will be
attached to the mother who cares for the child. Thus the child devel-
ops a positive attitude toward the mother. But if the feeding situation
is unsatisfactory - that is, the child is given food when not hungry or
scolded while crying when hungry - " a hungry—anxiety conflict" results
and a negative attitude toward the mother may develop.

Toilet training or cleanliness is another aSpect of the child's
experience which may have an important bearing on his subsequent in-
terpersonal realationships. Wolfenstein(A3) points Out that this ié
one phase of the mother-child relationship that is most laden with
emotional risks. Severe punishment or coercive methods used may lead
to undesirable consequences for the child's future development. Ac-
cording to Dollard and Miller (10), infliction of punishment may arouse
the child's anger toward the parent administering it. Furthermore,

the anxiety and hostility produced by the harsh training may become

attached to the parent who enforces it. Consequently, the parent's
presence may become capable of evoking anxiety, and the child in order
to escape the anxiety may avoid interaction with the parent. The
child's reactions to his parents developed during the toilet training

may then generalize and affect the child's future social relationships.
E. Socialization Process

Aside from the feeding situation and toilet training the child
experiences, his social growth may also be affected by the attitudes
of his parents who may be rejecting, overprotective, or democratic.
Thompson (A2) states that parents are changeable and.often.unpredict—
able. Their interactions with the children are often colored with
"traditions, personal prejudices, emotional tone and rule-of-thumb
procedures (p. 50h)."

According to Shoben (37) the parents have the duty of teaching
the child certain social norms and good conduct. They often resort
to punishment when these rules are violated. "Since they are asso-
ciated with both the experience of drive arousal and the experience
of drive reduction, they have become the objects of love and hatred
(p. 51)." The parents tend to be loved by the child to the extent
that they are associated with feeding, love, warmth and protection.
They tend to be feared to the extent that they are associated with
pain and anxiety. It is this situation that gives rise to the ambi—
valent attitude which the child develops for his father and mother.

F. Identification

Further complication in the parent-child relationship is found in

the development of the Oedipus Complex. Freud (17) describes the

Oedipus Complex as follows: The child's first object is the mother
whom he has associated with his sucking activity, comprising both the
function of obtaining nourishment and that of giving pleasure. The
idea of mother is not.present at the beginning, and the impression he
has is that the source of pleasure and satisfaction are his mother's
breast and love. This impression is later differentiated into the
person of the mother who in the course of feeding and care of the child
arouses in the child pleasurable and unpleasurable sensations. At
about the age of 2 or 3 the child develops an interest in his genitals.
At this stage the boy desires his mother and wants to get rid of his
father who is his rival for the mother's love. He becomes "his mother's
lover." The girl, on the other hand, takes her father as her love ob-
ject because of her wish for the penis which has been denied to her by
her mother and which she now expects from her father. With the increase
in the family the Oedipus Complex expands into what Mullahy (30) calls
the "Family Complex." The older brothers and sisters may serve as
father and mother substitutes.

In view of the danger of castration the boy represses his desire
for his mother and identifies with his father. The girl, because of
the fear of losing her mother's love, turns to her mother and identi-
fies with her mother. With the resolution of the Oedipus Complex the
child's primary identifications with his parents are greatly intensi—
fied. Besides the parents, the child's siblings and parent surrogates
may also influence the development of the child's super—ego and become

the child's ideal models.

G. Sex—Typing

Sex-typing begins at home with the male and female names, clothes,

7

play patterns and toys. Seward (35) points out that the growing child
adopts his sex role as early as he accepts the facts of genital sex
differences. His acceptance of his social sex role is facilitated by
his identification with the same—sex parent. Davis (8) claims that
the emotional interaction between the child and his parents contributes
to the child’s learning of his appropriate sex role. As the child
grows older the school plays an important role. His peers help further
the achievement of his proper sex identification. Mowrer (29) is of
the opinion that sex-typing is established by the parent's use of re-
ward and punishment for appropriate and inappropriate sex responses.
H. Differences Between American and Filipino Children in their
Sexual Identification

It would not be surprising to find a difference in the kind of
sexual identification between the American and Filipino children due
to the differences in the child-rearing practices in the two cultures.
This is especially true of the roles of the parents. In the Filipino
family both parents have somewhat specific roles. The mother is re-
sponsible for the caring of the children and the father is in charge of
the discipline. The father also assumes the major role in the family
and is often the source of frustration for the children.) On the other
hand, we find a diminishing role of the father in matter of discipline
in the American family. This function is taken over by the mother who
is also reSponsible for the care of the children. Thus the mother as-
sumes the major role in the family. As for the training habits, there
seems to be more permissiveness among the American parents as compared

to the Filipino parents. “The Filipino parents are more rigid. These

8
differences in parental roles in both cultures may therefore influence
the development of the children's sexual identification.

I. Measures used in the Study of the Degree of Children's Sexual

Identification

Various methods were used to find out the extent of the children's
awareness of sex differences. Smith (39) used a list of desirable and
undesirable traits to which the boys and girls gave their opinions con-
cerning themselves. By using the Terman and.Miles "Attitude-Interest
(M-F) Test", Ferguson (13) sought to determine some of the cultural
origins of masculinity and femininity. The use of tqys to determine
the child's identification with his sex-role was found to be a feasible
medium of investigation by Benjamin- (2) and Rabban (33). One satis-
factory approach that may be used in this connection is that of children's
drawings. The present study will attempt to show how the "DraweA-Person"
test can be used as a valid measure in comparing American and Filipino

children in their degree of sexual identification.

II STATEMENT OF HYPOTHESES

In the process of identification the child tends to identify with the
person.who is the source of frustration. This has been hypothesized by
Anna Freud (16) who says:

In the identification with the aggressor the children who identi-
fied themselves with the elder's threat of punishment were internal-
izing other people's criticism of their behavior. When the child con—
stantly repeats this process of internalization and introjects the
qualities of those responsible for his upbringing, making their char—
acteristics and opinions his own, he is all the time providing material
from which the super-ego may take shape (p. 125).”

Thus the child's perception of his sex role may be influenced by his

identification with the same—sex parent.

Hypothesis I: Fewer American children, as compared to the Filipino child-

ren,ishow a high degree of sexua1 differentiation.

As has been stated earlier, the mother in the American family assumes a
dominant role while the father, whom we might expect to assume this role,
maintains a diminished role eSpecially in matter of discipline. These par-
ental roles may be perceived by the child as confusing thus accounting for

the above hypothesis.

Hypothesis II: More Filipino boys, as compared to the American boys, tend

to identigy'with the male figure.

With the resolution of the Oedipus Complex, the boy identifies with the
father. This is further reinforced if the father is seen as a threat to
the child. If the father is the source of frustration for the Filipino boy,

the father will be serving as an appropriate model for the boy's sex role.

10
Hypothesis III: The American and the Filipino girls will Show no difference

with respect to identification with the female figure.

Because of the fear of losing the mothers' love, the girls in both cul-
tures tend to identify with their mothers. This identification with the
mother is further strengthened if the mother is the child's constant compan-

ion at home and serves as the feminine model for the girl.

III. THE APPLICATION OF DRAWINGS IN THE ASSESSMENT OF
SEXUAL DIFFERENTIATION

A. Historical Background in the use of Drawings with Children

Early interest in children's drawings was directed mainly at the
study of the mental development of the child. This interest, Gooden-
ough (18) points out, dates back to Cooke in 1885 and that of Ricci in
1887. Goodenough developed a scale for the measurement of children's
intelligence. The drawings were later utilized as indices of children's
interests and also in a number of genetic studies in the field of per—
ception. Lamprecht, Rouma, and other investigators, as cited by Gooden—
ough (18), conducted a series of studies to Show the relationships be-
tween the drawings of modern children and those of the primitive races
Comparative studies of national and racial groups were undertaken with
the use of drawings. Manuel and Hughes (27) gave the Draw-A-Man test
to compare the Mexicans and non-Mexicans in Texas in their ability to
draw. Papavassilion (32) used this test to evaluate its suitability
for Greek children. Aside from these two investigators, several other
researchers used the drawings in their studies in different cultures.

Recent interest in drawings has shifted to their use as projective
devices. According to Goodenough and Harris (19) earlier studies con—
cerning children's drawings were based on the belief that "a child
draws what he knows, rather than what he sees." .At present the studies
are conducted on the hypothesis that "a child draws what he feels, rath-
er than what he sees." Present day psychologists feel that drawings
and other overt behavior of children reveal the children's feelings

and desires. The psychologists further claim that the drawings ex-

press the children's personality.

12

B. The Goodenough Scale

As has been mentioned earlier, Goodenough devised a scale to
measure the intellectual factors involved in the drawings of children.
This scale is based on the drawing of the human figure. It consists
of fifty-one points. These points were obtained by means of "(a) the
observation of differences which appeared to be characteristic of the
performances of children at successive ages or school grades; (b) the
formulation of objective definitions or descriptions of these differ-
ences; and (0) their statistical validation based on a comparison be-
tween the performances of children of different Ages, and also between
the performances of children who were accelerated in school and those
who were retarded (18, p. 81)." Goodenough claims that this scale is
useful for making comparisons between groups and is applicable for in—

vestigating the mentality of children from other cultures.
C.. Drawings used as Projective Techniques

The present trend in the use of drawings is their utilization as
projective devices. .A number of studies have been undertaken using
drawings for this purpose. Appel (1) used a child's drawing of his
home and family in the course of psychiatric interview with the child.
He claimed that drawings can be used as aids in the free verbal ex-
pression of the child. To distinguish between the socially adjusted
and the socially maladjusted mentally deficient subjects, Brill (3)
utilized the Draw-A-Man test. He concluded that there was a strong
probability that the adjusted group would score higher than the mal-
adjusted group, but the differences was not satisfactory enough to be

of diagnostic value in individual cases.

13

Fingert _e_t. a_1_. (1h) gave the Draw-A-Man test to schizophrenics to
measure their improvement following insulin and metrazol therapy. They
felt that by this method they could get a deeper insight of the mental
states of the patients. They found a steady increase in score during
the period of recovery from the treatment.

England (11) investigated the manifest expression of parent selec—
tion and identification by the use of sexless stick figures. A total
of 13h children (8h boys and 50 girls) were subjected to the experiment.
The experiment consisted of three exercises. In the first exercise, a
drawing of a woman and a man both seated on a bench with a tree between
them was presented to the subject. .A sexless figure was seated beside
the woman labeled "mother" and the man labeled "father." The subject
was asked.whether the stick figure was a boy or a girl. In the second
exercise, a drawing of a man and a woman walking down the street with
joined hands was shown to the subject. In this drawing the stick fig—
ure held hands of each parent. The subject was again asked whether the
figure was a boy or a girl. The third exercise was a drawing of a
woman seated on a bench. Immediately at her left was a tree. The sub-
ject was told to draw a boy and a girl near the woman. Since there was
no Space left for him to draw at the woman's left, he was forced to
draw the boy and the girl at the mother's right. In this experiment
England found that the boys and girls identified the girls with the
mother and the boys with the father. The boys, however, were more con-
fused in their feelings in identifying the sex of thestick figure with
either the father or the mother. He felt that the exercises were well

framed to obtain the true projective feelings.

lb
The above experiments are just a few of the studies which have

tried to use children's drawings as projective devices. Since these
drawings uncover the deep-seated feelings and unconscious motives of

the person who draws them, their use as indices of identification

seem very feasible. This has been demonstrated by England. But the
"DrawsArPerson" test is more suited for this purpose. Since the drawing
of each sex will enable the examiner to get an insight into the subject's

conception of sexual roles and attitudes toward parental roles.
D. The Draw-A-Person Test

.After the publication of Goodenough's scale in 1926, Machover (26)
in the course of her work with psychiatric patients found that although
the intellectual development of any two children was the same, their
drawings of the man were not identical. The drawings produced valu-
able materials which she believed could be used in the study of the
adult's personality. These findings prompted her to devise the Draw-
A-Person test. Instead of utilizing just the male figure as used by
Goodenough in her scale, she used both the male and female figures in
her test. She also formulated some assumptions for the analysis and
interpretation of the figure drawings. Her basic assumption is that
"the human figure drawn by an individual who is directed to draw a
person relates intimately to the impulses, conflicts, and compensations
characteristic of that individual (p. 35)."

In recent years increasing interest has been shown in the use of
this test as a projective technique. It has become an instrument used
routinely by many clinical psychologists. King (2h) showed by means

of a single case the usefulness of the test as an adjunct to therapy.

15
She found an interesting feature of the test was the subject's remark
that the drawings seemed sexless. This brought into focus of therapy
the subject's confusion of sex identity.

Fisher and Fisher (15) in their study concluded that the general
sex role structure which the subject projects in the figure drawings
have important meaning in relation to the subject's past sexual con-
duct. The relationship between the sexual adjustment and the nature of
the human figure drawings of psychotherapy patients was explored by
Siprelle and Swensen (38). The drawings were rated on sexual differ-
entiation using Swensen's (9—point scale (h0)f. The patients were
also rated on a sexual adjustment rating scale. The patients were com-
pared on the basis of these two scales. The investigators found that
women who drew the female figure first tend to be dissatisfied with
their sexual partners, and women who drew more feminine figures tend to
be better adjusted sexually.

Cutter (7) used the DAP to verify the hypothesis that poor sexual
differentiation in the figure drawing may reflect general deficit or
specific sexual disturbances. He used as subjects sexual psych0paths.
He found that the sexual offenders did not differentiate between the
sexes on the DAP any worse than the normals.

The DAP and the M—F scale of the MMPI were administered by
Granick and Smith (20) to 185 male and female students for the purpose
of determining the relationship between the self-sex drawing as the
first reSponse to the DAP and the results obtained from the M—F scale
of the MMPI. They found that the tendency to draw the self-sex first
was stronger in men than in women. No relationship was found between

the order of the human figures and the Scores of the M—F scale of the

 

v
n .u, Jun—er

 

_ .
, r—“- '— — .
-_,.__.— mag—_—

16

MMPI. According to them, cultural factors might be the cause for the

tendency of most of the subjects to draw their own sex.

E. The Scale of Sexual Differentiation

Using the Draw—A-Person test Swensen (NO) devised a scale for
rating sexual differentiation. The Object of his experiment was to
investigate the relationship between sexual differentiation on the DAP
and various behavioral characteristics. In developing this scale he
used the drawings of the patients of the veterans Administration Mental
Hygiene Clinic at Knoxville, Tennessee. The scale contains nine points,

ranging from little or no sexual differentiation (point 1) to excellent

sexual differentiation (point 9). It is arranged in ascending numerical

order with the poorest sexual differentiation coming first and the best
sexual differentiation coming last. The five odd numbers i.e. l, 3, 5,
7, 9, contain descriptions and three examples of drawings and the four

even numbers i.e. 2, h, 6, 8, are used for rating drawings that may be

considered to fall between the defined points on the scale.
F. Adaptation of Swensen's Scale to Children's Drawings

Swensen's scale was developed on the basis of the drawings of
adult subjects. Using the scale for children's drawings which are
more simple and less.complicated than the adult drawings, might pre-
sent some difficulties. To overcome these difficulties the Scale was
reduced to a 5-point scale containing numbers 1, 2, 3, h, 5 corres-

ponding to the odd numbers i.e. l, 3, 5, 7, 9, of Swensen's Scale. The

scale is as follows:

17

Little or no sexual differentiation

There is little or no difference between the two figures,
and what difference exists between them does not particularly
suggest sexual differentiation.

Poor sexual differentiation

Longer hair on the female than on the male. There may be
a slight suggestion of difference in body contour and/or
clothing.

Fair sexual differentiation

The female definitely has longer hair than the male. The
female clearly has different body contour, with either rounded
hips or breasts, or both present. There may be a suggestion of
difference in the clothing of the pair.

Good sexual differentiation

The female has longer hair than the male. The female has
rounded body contour, the male has angular contour. Breasts
and/or rounded hips present, with both usually present. There
is a clear difference in clothing, with the female wearing
feminine apparel, although the apparel may be copied after
that of the male, e.g. slacks. There may be the suggestion of
differentiation in minor details, such as eyelashes or fuller
lips on the female.

Excellent sexual differentiation

Female hair is longer than male hair, with definite feminine
hair styling in the female. The male body has angular contour,
the female body has rounded contour with both breasts and round-
ed.hips present. The male wearing clothing that is definitely
masculine, the female wearing clothing that is clearly feminine.
Minor details, such as eyes, mouth, earrings, bracelets, etc.,
clearly appropriate for the sex of the figure on which they are
drawn, (NO, p. 38).

IV SUBJECTS AND PROCEDURE

A. Subjects

The subjects used in the present study were 129 American children
and 159 Filipino children. The American subjects are currently enroll-
ed in the Michigan Avenue School in Lansing and in the Kellogg Element-
ary School at Hickory Corners, Michigan. The Filipino subjects are in
the Francisco Benitez Memorial School in.Pagsanjan, Laguna and in the
Juan Luna Elementary School and in the J. Zamora Elementary School in
Manila. These subjects were made up of two groups - one comprising the
tenryear olds and the other, the eleven-year olds.

The school children in the Philippines are classified according
to their scholastic achievement. There are seven sections in the
fourth and fifth grades. The upper sections represent the bright stu-
dents and the lower sections the Slow students. The subjects for this
study were selected at random from all these sections. The American
subjects were chosen from the fifth and sixth grades. The drawings of
some of the fourth grade subjects from the Perrin—Palmer School were
obtained from the files of the Psychological Clinic, Michigan State
University.

The subjects were equated both for age and social status. They
were classified into the middle-class and the lower-class groups, ac—
cording to their parents' occupations. The middle—class group comprise
subjects whose parents are doctors, engineers, lawyers and those en—
gaged in other professions. The lower-class group is made up of sub-

jects whose parents are laborers, factory workers and those engaged in

19
the unskilled work. Of the American subjects, 21 per cent and 79 per
cent were classified into the middle-class and lower-class groups re-
Spectively. Of the Filipino subjects, 22 per cent were in the middle-
class group and 78 per cent in the lower-class group. Table I shows

the classification and ages of the subjects.

TABLE I

Some Characteristics of the.American and Filipino Samples

 

 

 

 

Boys Girls Age

N M Age N M Age Range Total
Ten-year olds
Americans bl ‘ 10.3 37 10.1 10-11 78
Filipinos Ml ' 10.2 1.6 10.0 9-11 90
Elevenryear olds
Americans 28 11.1 23 11.1 11-12 51
Filipinos 32 11.1 37 11.0 10-12 .69
Total N lh5 1h3 288

 

B. Instructions

The subjects were given plain white paper (8 1/2" x 11") and
pencils and were told to "Draw—a person." After they had completed
their drawings they were given the following directions: "0n the back
of your paper draw the opposite sex of the first figure you drew. If

you drew a man, draw a woman; and if you drew a woman, draw a man."

20

C. Scoring

In sCoring the drawings Swensen's Rating Scale for Sexual Differ-
entiation was used as an index for sexual differentiation and the Good—
enough Scale as a measure of the self-sex identification. All the
drawings were first scOred using Goodenough's scale. Based on this
scale a drawing could receive a maximum score of 51 points. To estab—
lish the scorer reliability two independent judges1 scored every fourth
drawing. The pearson.product-moment correlation between the two sets

of scores is presented in Table II.

TABLE II

Correlation Between Judges Scores on.Goodenough Scale

 

 

 

 

Male Figure Female Figure
Ten-year olds
Americans .79 .86
Filipinos .91. ’ .93
Eleven-year olds
Americans ' .9h .92
Filipinos ‘ .91 .96

 

The drawings were further rated for sexual differentiation. The
5-point rating scale for sexual differentiation described above was
used. TwO independent judges2 rated the drawings. Complete agreement
was reached between the two judges after a discussion of the major dif-

ferences in the rating of the drawings. The Rulon formula (22, p.h97)

 

EMrs. Mary M. Boroughs in addition to author.

2Mr. Gilbreth DeRath in addition to author.

21
was used to find the reliability of judgments. Agreement between the

two judges is shown in Table III.

TABLE III

Correlation Between Judges Rating on Sexual Differentiation

 

 

 

Boys Girls
Tenryear olds
Americans .90 .96
Filipinos .87 .8h

 

Elevenryear olds
Americans .92 .88
Filipinos .96 .87

 

V RESULTS

An analysis of the data by means of the chi-square (see table IV)
Shows the results of the test to be significant in relation to hypothesis
I. According to the hypothesis, the American children will Show a les-
ser degree of sexual differentiation as compared to the Filipino child—
ren. The data were based on the 5-point Rating Scale for Sexual Differ-
entiation. In computing for the chi-square the data were dichotomized
because of the small frequencies in the 5 cells. Points 1 and 2 of the
scale were combined into one category and points 3 to 5 in another cate-

gory. The analysis indicates a significant difference at the 0.01 level.

TABLE IV

A Comparison Between Groups on Sexual Differentiation Scores

 

 

Ten4Year Olds

 

Americans Filipinos Chiz p.*
A. Boys
Categories 1 - 2 31 18 10.58 0.01
Categories 3 — 5 10 26
B. Girls
Categories 1 - 2 2h 17 6.93 0.01
Categories 3 - 5 13 29
C. All Children
Categories 1 - 2 55 35 16.8h 0.01

Categories 3 - 5 23 55

 

*TWo-sided test

23
TABLE IV (Cont.)

 

ElevenéYear Olds

 

 

 

Americans Filipinos Chi2 p.*
A. Boys >
Categories 1 - 2 l9 5 17.88 0.01
Categories 3 - 5 9 27
B. Girls
Categories 1 - 2 11 8 h.h6 o.o§**
Categories 3 - 5 12 29
C. All Children
Categories 1 - 2 , 3o 13 20.0h 0.01
Categories 3 - 5 21 56
Teneand E1even4Year Olds Combined
Americans Filipinos Chi2 p.
Categories 1 - 2 85 LB 33.h7 0.01

Categories 3 - 5 AA 111

 

*Two-sided test.

**Yates correction applied.

For hypotheses II and III the t—test was used to find out the mean
difference between the two groups on the total scores obtained from the
Goodenough Scale. Hypothesis II states that the Filipino boys will
tend to identify more with the male figure than will the American boys.
The results show that the Filipino boys drew more details in their
drawings indicating identification with the self-sex figure. There is
a significant difference at the 0.01 level among the boys on the scores

obtained from the male figure drawing as shown in Table V.

2A

TABLE V

A comparison Between.Groups on the Total Scores Obtained from
the Goodenough Scale (Male Figure Drawing)

 

 

Boys N M SD t p

 

Ten-year olds Americans A1 2A.A 6.1' 3.33 0.01
Filipinos AA 29.5 7.6

 

Eleveneyear olds Americans 28 27.2 6.3 3.57 0.01
Filipinos 32 33.5 7.A

 

Girls

 

Ten-year olds Americans 37 2A.3 6.9 0.27 N.S.
Filipinos A6 2A.6 6.5

 

Elevensyear olds Americans 23 27.9 5.7 2.11 0.05
Filipinos 37 23.9 6.3

 

The results of the test of significance of hypothesis III are pre—
sented in Table VI. Based on this hypothesis we will expect to find no
difference between the American and Filipino girls in their degree of
identification with the female figure. The t—tests of 1.67 and 0.25 are
are not significant.

In the comparison within cultures, which is shown in Table VII, we
find a significant difference at the 0.01 level among the American
children.based on their scores obtained from the female figure drawing.
The girls drew more details than did the boys. 0n the other hand, a
significant difference at the 0.01 level was found among the Filipino
children on the basis of their scores obtained from the male figure draw—

ing. Here we find the boys drew more details than did the girls.

TABLE VI

A Comparison Between Groups on the Total Scores Obtained from
the Goodenough Scale (Female Figure Drawing)

 

 

Boys N M SD t p

 

Ten—year olds Americans A1 20.8 6.A 3.A9 ' 0.01
Filipinos AA .25.9 6.9

 

Elevenryear olds Americans 28 2A.1 5.8 2.61 0.05
Filipinos 32 29.A 9.6

 

'Girls

 

Tenryear olds Americans 37 23:9 6.7 1.67 N.S.
Filipinos A6 26.A 6.7

 

Elevenryear olds ’ Americans 23 27.A 5.9 0.25 N.S.
Filipinos 37 27.8 6.1

 

TABLE VII

A Comparison Within the Culture Groups on the Male and Female
Figure Drawings (Goodenough Scale)

 

 

 

 

Male Figure A Female Figure
N M - SD t p M 'SD t p
Americans
Boys 60 25.5 6.3 ‘N.S. 22.0 6.A 2.67 0.01
Girls 69 25.7 6.7 25.0 6.6
Filipinos
Boys 76 31.2 7.8 5.71 0.01“ 27.A 8.A N.S.

Girls 83 2t.u 6.A 27.0 6.A

 

26

Since the results so far have shown a significant difference be-
tween the two groups, we might ask whether this difference is due to
the drawing ability or intelligence of the children. .An analysis of
covariance was employed to show the relationship between the degree
of sexual differentiation and the scores obtained from the Goodenough
Scale (male figure drawing). Scores obtained from the female figure
drawing were not utilized because there was no significant difference
between the two groups based on these scores. By eliminating drawing
ability as a factor, a difference between the two groups was still
found. The F-ratio of 1A.20 which is significant at the 0.01 level
is shown in Table VIII. The results indicate that the difference
between the two groups on their degree of sexual differentiation can
not be accounted for by the mean difference in their scores obtained

from the Goodenough Scale.

TABLE VIII

A Comparison Between Groups on their Degree of Sexual Differentiation
and Scores Obtained from the Goodenough Scale (male figure)

 

Total Within Between

 

 

Sum of products 79A-9 699.2 95.7
Sum of squares: X 153AA.A 13202.9 21Al.5
Sum of squares: Y 201.8 176.9 2A.9
df 287 28A 3
Adjusted sum of squares 160.7 139.8 20.9
df 286 283 3
Note: N - 288 F = 1A.20 p. 0.01

X - scores obtained from the Goodenough Scale..
Y - rating on Sexual Differentiation.

VI DISCUSSION

The present study is an attempt to compare the American and Fili—
pino children in their degree of sexual differentiation and self—sex
identification as projected in the DrawnA-Person test.

In the first hypothesis, fewer of the American children show a
high degree of sexual differentiation as compared to the Filipino
children. The major role of the mother at home and the diminishing
role of the father eSpecially in the matter of discipline may have ac—
countedfku‘the confusion on the part of the American children in the
perception of their proper sex roles.

The American parents are also more permissive in the training of
their children in the sense that the children do not have rigid.patterns
to follow. .As Seward (35) states, "Today in the post'Wor1d4War United
States there is a good deal less self-conciousness about the sex role
and probably more freedom of choice for the individual than ever before
(p. 175)." Brown (6) further claims that the feminine and masculine
roles in the American society are "becoming broader, less rigidly de-
fined, less sex typed and more overlapping with each other (p. 239)."
There seems, according to him, a developing convergence of the two sex
roles. This trend is seen in the parental roles which may in turn af-
fect the children's perception of their appropriate sex roles.

In the Filipino family the parents assume specific roles - the
mother cares for the children and the father takes charge of the dis-
cipline of the children. The father also assumes the dominant role in
the family. This family structure may have helped the children in

perceiving their apprOpriate sex roles. Furthermore, the parents are

28
strict and rather rigid in the training of their children. In this kind
of family system, the children tend to conform to some standards such
as, the girl should stay at home and the boy is given the privilege to
go out with his friends. These standards which have been set up by the
parents may enable the children to identify more clearly with their
proper sex roles.

We may also advance Britton's hypothesis (A) that the children
from the higher socio-economic status do better in the Draw-ArMan test
because of the attention given by the parents to their children in the
details of dressing and personal neatness. This, he Claims, may have
influenced the children’s awareness of specific details and thus af—
fected the scores in the test. In this connection, we may say that in
the course of training the Filipino parents might give more attention
to the dressing and personal appearance of their children than do the
American parents.

The second hypothesis deals with the greater degree of identifica-
tion of the Filipino boy with the male figure. Since he sees the father
as the source of threat, he tends to identify with the father. Besides
the father, the elder brothers who take part in the discipline of the
boy may also serve as identification models for the boy. The American
boy tends to identify less clearly withthe father than does the Fili-
pino boy. The mother, rather than the father, is seen not only as the
source of threat, but she also plays the principal role in the home.
Brown (5) points out that a great number of boys are more exposed to
the feminine patterns when the mother is seen as the prominent figure
than the father. This situation is further complicated by the fact

that the father is away most of the time while the mother is around

29
serving continually as the model. This results in the complication in
the identification of the boy with the same—sex parents.

In the third hypothesis, the American and Filipino girls show no dif-
ference with reSpect to identification with the female figure. As has
been mentioned earlier, because of the fear of losing their mother's love
the girls of both cultures tend to identify with their mothers. This
identification is facilitated because the mothers are the girls' constant
companion at home and serve as the feminine models for the girls.

Another example which may Show why the girls tend to identify with
the female figure is Brown's findings (5) which showed that girls of
the ages from 9 to 11 exhibit a change in their preference. In contrast
to their masculine preference during their earlier years, they demon-
strate a greater degree of identification with the female figure.

In the comparison within cultures, we can clearly see the parental
roles in the child-rearing process in the two cultures. In the American
family the mother plays a major role while it is the father who assumes
this role in the Filipino family. These parental roles seem to have
affected the identification of the children with the same-sex parent.

The pattern in the Filipino family seems to give the boy an opportunity
to identify more with the father. The girl in the American family
seems to have more advantage than the boy in identifying with the same-
sex parent. The dominant role of the mother results in the confusion
of the boy's sex identity.

We can conclude from the above study that the Filipino children
show a greater degree of sexual differentiation than the American
children and that the DrawnA—Person test is an adequate and valid in-
strument in comparing two cultural groups in the degree of sexual dif~

ferentiation.

VII SUMMARY.AND CONCLUSION

The DrawnA-Person test was administered to 129 American children
and 159 Filipino children to compare them in their degree of sexual
differentiation. 0f the American subjects, 69 were boys and 60 were
girls. 0f the Filipino subjects, there were 76 boys and 83 girls. The
ages of the subjects ranged from nine years to twelve years. Three
predictions were formulated on the basis of the different family struc—

tures in the two cultures.

1. The American children will show a lesser degree of sexual dif-
ferentiation than the Filipino children.

2. The Filipino boys will tend to identify more with the male
figure than the American boys.

3. The American and Filipino girls will Show no differences in

their identification model.

A revision of Swensen's 9-point Rating Scale for Sexual Different-
iation was used as an index for sexual differentiation. The revised
scale contains 5 points i.e. 1, 2, 3, A, 5, which correSpond to the odd
numbers 1, 3, 5, 7, 9, of Swensen's scale. The data were analyzed by
means of the chi—square to determine the significance of the differences
between the two groups on sexual differentiation. For the self-sex
identification the Goodenough Scale was utilized as the measure. A t—
test was employed on the scores obtained from the Goodenough scale
(male and female figure drawings) to find out the difference between
the two groups on their self-sex identification. Since there was doubt

as to whether the drawing ability of the children was reSponsible for

31
the difference between the two groups, the analysis of covariance was
used to show the relationship between the degree of sexual differentia-
tion and the scores obtained from the Goodenough Scale.

The results supported the above predictions. The relationships be-
tween the results of the study and the parental roles in the two cul—
tures were also discussed. It was concluded that the Filipino children
show a higher degree of sexual differentiation as compared to the Amer—
ican children and the Draw—AePerson test is an adequate and useful in—

strument in a cross—cultural study.

10.

ll.

12.

13.

1A.

15.

16.

VIII BIBLIOGRAPHY

Appel, K. Drawings by children as aids to personality studies.

Amer. J, Orthopsychiat., 1931, 1, 129—1AA.

 

Benjamin, H. Age and sex differences in the toy preferences of

young children. 'J. Genet. Psychol., 1932, A1, A17

 

-A29

Brill, M. A study of instability using the Goodenough Scale.

‘J. Abnorm.‘Soc. Psychol., 1937, 32, 288-302

 

Britton, J. Influence of social class upon performance on the
DrawnA-Man test. 'J. Educ. ngchol., 195A, A5, AA-5l

Brown, D. Masculinity and femininity development in
‘1. Consul. ngchol., 1957, 21, 197-202

children.

Brown, D. Sex-role development in a changing culture. Psychol.

Bull., 1958, SS, 323—3A1

Cutter, F. Sexual differentiation in figure drawings and overt

deviation. ‘1. Clin. P§ychol., 1956, 12, 369-372

 

Davis, A. American status system and the sooidlizati
child. Amer. Soc. Rev., 19A1, 6, 3A5- 356

on of the

Davis, A. and Havighurst, R. J. Social classes and color differ-

ences. Amer. Soc. Rev., 19A6, 11, 698-710

 

Dollard,J. and Miller, N. Personality and Psychotherapy. New

 

York: MeGrawsHill, 1950

England,.A. Cultural milieu and parental identification. Nervous

Child, 19A7, 6, 301-305

Ericson, Martha Child rearing and social status. Amer. J. Socio.,

19A6, 52, 190-192

Ferguson, L. The cultural genesis of masculinity - femininity.

Psychol. Bull., 19A1, 38, 58A-585

 

Fingert, H.3 Kagan, Julia and Schilder, P. The Goodenough test

in insulin and metrazol treatment of schizophrenia.

ngchol., 1939, 121, 3h9-365.

i- 922-

Fisher, S. and Fisher, Rhoda Lee Style of sexual adjustment in
disturbed women and its expression in figure drawings.

g, Psychol., 1952, 3A, 169-179

Freud, Anna The Ego and the Mechanisms of Defense.
Interna'1.Univer. Press, 19A6

 

 

New‘York:

17.

18.

19.

20.

21.

22.

23.

BA.

25.

26.

27o

28.

29.

30.

31.

32.

33.

33
Freud, S. New Introductory Lectures on Psychoanalysis. New

York: InternaTl Univer. Press, '1' 9116‘

 

 

Goodenough, Florence Measurement of Intelligence blerawings.
Yonkers, New York: World Book, 1926

 

 

Goodenough, Florence and Harris, Dale Studies in the psychology
of children's drawings. Psychol. Bull., 1950, A7, 369-A33

Granick, S. and Smith, L. Sex sequence in the DAP test and its
relation to the MMPI M—F scale. J, Consul. Psychol., 1953, 17,,
71-73

 

Gorer, G. :The American People. NeW'York: W. W. Norton, 19A8

 

Guilford, J. P. Fundamental Statistics 13 Psychology and Educa-
tion. New York: McGraw—Hill, 1950

 

 

Havighurst, R., Gunther, Minna Korol, and Pratt, Inez Ellis
Environment and the DrawnA-Man test: The performance of Indian
children. J, Abnorm. Soc. Psychol., 19A6, Al, 50-62

 

King, Frances The use of drawings of the human figure as an ad-
junct in psychotherapy. J, Clin. Psychol., 195A, 10, 65—69

Maccoby, Eleanor; Gibbs, Patricia and the Staff of the Laboratory
of Human Development - Harvard University. In Martin and Stendler
(Ed.) Readings 12 Child.Development. New Y0?R: Harcourt and
Brace, 195A

 

 

 

Machover, Karen Personality Projection 12 the Drawings 2£.EEE
Human Figure. Illinois: Charles C. Thomas, 19A9

 

 

Manuel, H. T. and Hughes, Lois S. The Intelligence and drawing
ability of young Mexican children. J, Applied ngchol., 1932,
16, 383-387 -

 

Master, E. E. Modern Courtship and Marriage. New York: McMillan,
1957

Mowrer, 0. H. Learning Theory and Personality Dynamic . New York:
Ronald, 1950

 

 

Mullahy, P. Oedipus Myth and Complex. New York: Grove Press, 1955

 

Mussen, P. and Conger, J. Child Development and Personality, New
York: Harper, 1956 '

 

Papavassilion, Th. The validity of the Goodenou h DrawaA—Man test
in.Greece. .£° Educ. ngchol., 1953, AA, 2AA? 8

Rabban, M. Sexarole identification in.young children in two diverse
social groups. Genet. Psychol. Monogr., 1950, A2, 81-158

 

3A.
35.
36.

37.
38.
39.
A0.
A1.

A2.

A3.

3A

Sears, R., Maccoby, Eleanor and Levin, H. Patterns of Child Rear-
Jgg. New York; Row, Peterson, 1957

 

Seward, G. Sex roles in postwar planning. ‘J. Soc. Psychol., 19AA,
19, 163-185

 

Seward, Georgene Sex and the Social Order. New York: McGraw-
Hill, 19A6

 

Shoben, E. Theoretical Frames of Reference in Clinical Psychology.
‘In Pennington and Berg (Ed.) Introduction to Clinical ngchology.
New York: Ronald Press, 195A

Sipprelle, C. and Swensen, C. Relationship of sexual adjustment
to certain sexual characteristics of human figure drawings.

IJ. Consul. Psychol., 1956, 20, 197-198

Smith, 5. Age and sex differences in childrens’ opinion concern-
7 ing sex differences. J, Genet. Psychol., 1939, 5A, 17-25

 

Swensen, C. Sexual differentiation on the DrawaA—Person. ‘J. Clin.

ngchol., 1955, 11, 37-Al

Symonds, P. Some basic concepts in parent-child relationships.
Amer. J. Psychol., 1937, 50, 195-206

Thompson, G. Child Psychology. New York: Riverside Press,
Cambridge, 1952

Wolfenstein, Martha Trends in infant care. Amer. J, 0rthopsychiat.,
1953, 33, 120-130

 

anrggn a” *-' ,3 C‘
Til 4U A ' ;~~ 1‘ ’
’56 ‘1I Lui-‘u Lidia-i;

 
   

AUG 11.81980 3“

A08 31718801.

 

  
    

 

 

 

 

 

MICHIGAN STATE UNIVERSITY LIBRA

H l’ Illlllllllllllllll
’ 1 1

31293 03 45 6 18

IES

'3
‘l
l