AGGRESSION IN BLACK AND WHITE
CHILDREN AS PROJECTED IN
THE DRAW~A.-'PERSON TEST

Thesis for the Degree of M. A..
MICHIGAN STATE UNIVERSITY.
HUGH E. JONES
1972

«r!!! '

 

 

 

 

 

 

 

v I":
‘5 5 h
i :-
'l' v ’ {— ' ‘

 

y "a. ‘. 3 f“ .
’ ‘x
‘f ' .. f
. 1‘.

f 2

’-

3 JUN“? B fiffi

1 CC;

 

 

 

ABSTRACT

AGGRESSION IN BLACK AND WHITE CHILDREN AS
PROJECTED IN THE DRAW-A-PERSON TEST

BY

Hugh E. Jones

The present study, which is based on the assumption
that frustration leads to aggression but accepts the tena-
bility of other theories of aggression, examines the effect
of such frustrating conditions as racism and poverty upon
the degree of aggressiveness in children as reflected in
the Draw—A-Person test. It was hypothesized that black
subjects would show more aggression than white subjects;
that lower-class subjects would show more aggression than
middle—class subjects, and that the groups would assume
the following rank of least aggression to greatest aggres-
sion: white middle-class, black middle-class, white lower-
class, black lower-class.

The subjects included seventy-two fourth and fifth
grade male students--thirty-two black and forty white.
Socio—economically, there were twenty white middle-class
subjects; twenty white lower-class subjects; sixteen

l

Hugh E. Jones

black middle-class subjects, and sixteen black lower-class
subjects. A measure of the relative degree of aggression
of each subject was obtained through the use of eleven
graphic indices of aggression in the drawing of the human
figure identified by Machover (1949), Buck (1948), and
Koppitz (1968). Each drawing was scored for the number of
indicators present.

Statistical analysis demonstrated that neither
race, social class, nor the interaction of race and social
class were significantly related to the subjects' aggres-
siveness as projected in the D-A-P test. Trends in the
data suggest that race may be related to the level of
aggressiveness in that blacks consistently received greater
mean scores of aggression than whites. The effect of
social class is equivocal.

Serious questions are raised about the reliability
of utilizing specific graphic indices of the Draw-A-Person
test as a measure of aggression. A global or "intuitive"
approach is recommended.

Further investigation of the causes of aggression,
techniques for measuring it, and methods of controlling it

are suggested. Specifically, the need for further study

Hugh E. Jones

of the functional relationship between aggression and
assertiveness (Feshbach, 1970) and the determination of

aggression behavior as a group influence are indicated.

AGGRESSION IN BLACK AND WHITE CHILDREN AS

PROJECTED IN THE DRAW-A-PERSON TEST

BY

x
F.

Hugh E: Jones

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS
Department of Psychology

1972

ACKNOWLEDGEMENTS

I would like to express my appreciation to Dr. Mary
Leichty, chairman of my thesis committee, for her guidance
and encouragement during each phase of this research. I
would also like to thank Dr. Albert Rabin, Dr. Martha
Karson, and Mr. Lennard Leighton for their participation
as members of my thesis committee.

In addition, I would like to thank my wife,
Vernell, who served as coder, editor, and counselor, and
my friend, Marilyn Golisek, who also served as a coder.
Lastly, I would like to thank my youngsters, Gilbert and
Derrick, for their love and great patience during my

academic pursuits.

ii

TABLE OF

LIST OF TABLES. . . . . . . .

INTRODUCTION. . . . . . . . .
REVIEW OF LITERATURE. . . . .
Psychoanalytic Theory . .

Learning Theory . . . . .

CONTENTS

Page

. . . . . . . . . . . V
. . . . . . . . . . . l

. . . . . . . . . . 2
. . . . . . . . . . . 2
. . . . . . . . . . . 4

Reinforcement and Punishment. . . . . . . . 7

Identification (Modeling) . . . . . . . . . 10

Socio-Cultural Variables.
Poverty . . . . . . .

Racism. . . . . . . .
Hammer's Study. . . .
Hypotheses. . . . . . . .
METHOD. . . . . . . . . . . .
Sample. . . . . . . . . .
Instrument. . . . . . . .
Procedure . . . . . . . .

RESULTS 0 O O O O O O O O O O

. . . . . . . . . . . 11

iii

TABLE OF CONTENTS (cont.)

Page

DISCUSSION. . . . . . . . . . . . . . . . . . . . . 37

SUMMARY 0 O O O O O O O O O O O O O O O O O O O O O 52

REFERENCES. 0 O O O O O O O O O O O O O O O O 0 O O 54

iv

LIST OF TABLES

Table Page

I. SUMMARY OF ANALYSIS OF VARIANCE RACE AND

SOCIO-ECONOMIC STATUS . . . . . . . . . . . 32

2. DISTRIBUTION OF S's WHO HAD AT LEAST ONE
INDICATOR IN DRAWING. . . . . . . . . . . . 33

3. MEAN AGGRESSION SCORES--RACE AND SOCIO-
ECONOMIC STATUS . . . . . . . . . . . . . . 33

4. FREQUENCY OF AGGRESSION INDICATORS. . . . . . 36

INTRODUCTION

Man's record of aggression and violence spans
human history and holds compelling prominence in our
twentieth century culture. Aggression has been considered
from a number of viewpoints-~some of which will be dis-
cussed below. These various approaches may be included
under the rubrics of psychoanalytic theory and learning
theory--two theoretical notions about the cause and char-
acter of aggressive behavior that have dominated twen-

tieth century psychology.

REVIEW OF LITERATURE

Psychoanalytic Theory

 

.Freud (1922), in his early writings, considered
aggression to be of sexual origin and directed towards
"the overpowering of the sexual object in so far as the
carrying out of the sexual act demands it" (Ibid., p. 69).
He equated aggression with hate and with the destructive
impulse-~as well as a normal component of the instinct of
sex. Recognizing the difficulty of the foregoing assump-
tion that the sadistic impulse (which aims to injure)
could derive from life-sustaining Eros, Freud developed
his hypothesis of a Death instinct, which represents the
organism's wish to return to the state of nothingness from
which it came. According to this view, aggression is an
expression of the death instinct, and its primary aim is
self-destruction. In each individual's life, there is a
struggle between life and death instincts. The stronger
the death instinct in a person, the more necessary it is

for him to direct aggression outward against objects and

 

 

'l.|l'llll‘llll|

 

people. Aggression that is not directed against others is
turned against oneself.

The treatment of aggression among major psycho-
analytic theorists is variable. Many question the death
instinct construct, but most accept the reactive character
of human aggression. Waelder (1956) takes a position
typical among Freudian analysts in accepting the validity
of both reactive and instinctive sources of aggression.
The neo—Freudian position elaborated by Lowenstein (1948)
and Hartman, Kris, and Lowenstein (1949) concurs with
Freud's assumptions regarding the instinctual origin of
aggression. They differ from Freud who interpreted out-
ward aggression as a manifestation of the death instinct,
however, for they view self—destructive behavior as a
deflection of aggressive impulses. They also suggest that
the destructive function of aggressive energy can become
modified through neutralization and sublimation.

Among non-Freudian psychoanalysts, the status of
aggression is equally variable. For Fromm (1941), the
sadomasochistic personality pattern represents a means of
reducing feelings of powerlessness and alienation from
the social structure. Horney (1945) rejects both aggres-

sive and death instincts. She considered aggression as

“wk

39.737

one fundamental response used by the child to cope with
"basic anxiety," the feeling a child has of being isolated
and helpless in a potentially hostile world. Horney dif-
ferentiates reactive aggression which intends to injure

and reactive aggression which is basically instrumental.

Learning Theory

 

Learning theory, while not denying the possible
biological underpinnings of aggression, stresses the impor-
tance of experience (i.e., learning) in the causation and
mode of expression of aggression. This is exemplified by
the Yale studies of aggression (Dollard et 31., 1939).
The Yale group postulated that aggression is always a
consequence of frustration. More specifically, they in-
dicated that the occurrence of aggressive behavior always
presupposed the existence of frustration and, conversely,
that the existence of frustration always led to some kind
of aggression.

The Yale group offered the following operational

definitions of aggression and frustration:

l)

2)

l)

2)

3)

Aggression is "a sequence of behavior whose goal
response is the injury of the person to whom it

is directed";

Frustration is "an interference or blocking of a
behavioral sequence that prevents the goal response

from occurring" (Ibid., pp. 6-11).

These theorists also hypothesized the following:

The stronger the instigation to the frustrated
response, the more the frustration is an instiga-

tion to aggression.

The greater the interference with a frustrated
response, the stronger the instigation to aggres-
sion. A minor interference is less frustrating

than a complete blocking.

Frustration accumulates. Minor frustrations add
together to produce an aggressive response of

greater strength than would normally be expected
from the frustrating situation that appears to be

the immediate antecedent of aggression.

The immediate reaction to Dollard's assertion that
aggressive behavior always presupposes the existence of
frustration and that the existence of frustration always
leads to aggression was negative. Miller (1941), one of
the Yale group, considered the assertion that the occur-
rence of aggression always presupposes frustration as
defensible, but he recognized the following difficulties
with the assertion that the existence of frustration always
leads to some form of aggression: 1) It suggests that
frustration has no consequences other than aggression;

2) It fails to distinguish between instigation to aggres—
sion and the actual occurrence of aggression. Miller
attempted to remedy this situation by making the following
reformulation: "Frustration produces instigations to a
number of different types of responses, one of which is an
instigation to some form of aggression" (Ibid., p. 342).

It has been demonstrated that frustration often
does increase the probability of an aggressive response.
Otis and McCandless (1955) report increased aggression
following mild repeated frustration of nursery school
children. Increases in doll play aggression in both boys
and girls were found by Hartup and Himeno (1959) in pre-

school children following social isolation. A number of

other studies utilizing diverse techniques for assessing
aggression have found increases in aggressive tendencies
following frustration (Miller & Buhelski, 1948; Moore,
1964; Mallick & McCandless, 1966). Negative evidence of
the frustration-aggression hypothesis is reported by
FrederickSon (1942), Burton (1942), and Yarrow (1948).
Jegard and Walters (1960) failed to find an increase in
aggression following frustration in nursery school

children.

Reinforcement and Punishment

 

A number of behavior theories question the inevi—
tability of aggression as the sole response to frustration,
but others have gone even further, questioning frustration
as the necessary antecedent condition to aggression.
Reinforcement and modeling have been presented as alter-
natives.

Buss (1961) who exemplifies the reinforcement
position, argues that frustration is not a very potent
antecedent to aggression. According to his view, aggres-
sion occurs primarily if it has instrumental value--i.e.,

if the individual gets what he wants by an aggressive act.

Berkowitz, a current proponent of the frustration-
aggression hypothesis (1962), counters that the depriva-
tion of some object or goal which the individual attempts
to obtain by aggressive behavior is in itself a frustra-
tion. Reinforcement theorists hypothesize that the child
who is rewarded for aggressive behavior in a given situa-
tion will tend to respond aggressively in similar situa-
tions, and the child who is punished for such behavior
will develop anxiety about its expression and will tend
not to react with aggression in similar situations
(Mowrer, 1939). These formulations have been empirically
tested. The effect of reward for aggression on subsequent
aggressive behavior has been demonstrated in a number of
laboratory studies with children. Davitz (1952) found
that an aggressively trained group behaved more aggres-
sively after frustration than a constructively (non-
aggressively) trained group. A number of studies have
demonstrated that it is possible to shape aggressive re-
sponses using a variety of reinforcers (Lovaas, Baer, &
Bijou, 1963; Walters & Brown, 1963).

The effect of punishment as an inhibitor of aggres-
sion has been demonstrated in the laboratory with animals

(Kahn, 1951; Scott and Marston, 1953; Seward, 1946). Also,

Chasdi and Lawrence (1955) demonstrated that nursery school
children who were punished for aggressiveness during ini-
tial play sessions exhibited significantly less aggression
during later sessions than children who received no punish-
ment. Punishment is defined here as "a reduction of the
future probability of a specific response as a result of
the immediate delivery of a (punishing) stimulus for that
response" (Azrin & Holz, 1966, p. 381).

Survey studies of the effect of punishment as an
inhibitor of aggression have varied in their results.
Lesser (1952) found some indication that punishment for
aggression as reported by parents was associated with the
inhibition of aggression, whereas reward for aggression
was associated with increased aggression. Other survey
studies of child-rearing practices have failed to find the
predicted relation between increased punishment and lowered
aggression (Bandura & Walters, 1959; Glueck & Glueck, 1950;
Sears et 31., 1957). In terms of their findings, Sears
and his colleagues concluded that punishment is "ineffec-
tual over the long term as a technique for eliminating the
kind of behavior toward which it is directed" (Ibid.,

p. 484).

10

Identification (Modeling)

 

The modeling theorists stress the importance of
observational learning in the acquisition of aggressive
behavior. They postulate that a situation in which an
individual observes another perform an aggressive response
has a strong influence on the learning of aggressive re-
sponses. The response may be an exact imitation of a non-
identical behavior in the same class as that observed.
Modeling is more effective when the observer has been
previously instigated to aggress (Berkowitz, 1965) and
when the observer sees the model being reinforced (Ban-
dura, Ross, & Ross, 1963; Rosenbaum & Tucker, 1962; Wal-
ters, 1966). The power status and other stimulus qual-
ities of the model have also been found to be important
conditions of the effectiveness of modeling (Epstein,
1966; Hicks, 1965; Lefkowitz, Blake, and Mouton, 1955).

Observational learning (imitating a model) is con-
sidered to be relevant to Freud's concept of identifica-
tion. Freud (1923) developed this concept to help account
for the continuity in children of the values, desires, and
manners of their same-sex parent. Freud designated two

processes--anaclitic identification in which a child

ll

acquires attributes of the loved parent and identification
with the aggressor, in which a child defends against
anxiety by taking on the attributes of a punitive parent
(A. Freud, 1937). According to Sears, Rau, and Alpert
(1965), the relevance of modeling to identification is
that the child, "by performing certain acts which, in the
mother's behavior repertoire, have become secondary re-
wards or reinforcers for the child, now has a mechanism
by which he can reward himself. By imitating his mother,
he can provide a substitute for her when she begins with-

drawing affectionate interaction and nurturance from him"

(p. 4).

Socio-Oultural Variables

A number of investigators have produced empirical
evidence supporting the position that aggression is a re-
sult of specific types of social milieu. Parental influ-
ence has been identified as an important variable. McCord
25 31. (1961) attempted to evaluate a large array of
familial factors in the development of aggressiveness

among lower-class nondelinquent boys. They found that

12

aggressive boys were most likely to be raised by parents
who (a) treated the boy in a rejecting, punitive fashion;
(b) failed to impose direct controls on his behavior;

(c) offered him an example of deviance; and (d) were often
involved in intense conflicts. Non-aggressive boys came
from a decidedly different environment in which they were
(a) treated in an affectionate, non-punitive manner;

(b) guided by a consistent set of controls; (c) exposed

to examples of social conformity; and (d) reared by affec-
tionate, satisfied parents.

McCord's findings were similar to those of Sear,
Maccoby, and Lewin (1957) who, from carefully executed
interviews with 379 New England mothers, concluded that
aggression in small children (defined as "behavior intended
to hurt or injure someone") was associated with such en—
vironmental factors as parental permissiveness for aggres-
sion, the use of physically punitive discipline, maternal
lack of self—esteem, disagreement between the parents,
and dissatisfaction on the mother's part with her role in
life.

Chasdi and Lawrence (1955) investigated the effects
of home frustration, home and experimental punishment, and

experimental permissiveness on the frequency and intensity

13

of aggressive doll play behavior of fifty-three nursery
school children. Their findings include the following:

1) Children who are severely frustrated by their mothers
at home tend to be more aggressive in their doll play than
children who are mildly frustrated. 2) Children who are
highly punished for aggression at home tend to be more
aggressive in doll play than non-punished children. Also,
the punished group decreased in doll play aggression and

the permissive group increased in doll play aggression.

Poverty

Michael Harrington (1963) defines the poor as
"those who for reasons beyond their control cannot help
themselves." The door to opportunity and advancement is
closed to them. They are victims in a "vicious circle of
poverty" (Ibid., p. 22).

Hess (1970) described the following circumstances

of lower working-class life in metropolitan society:

1) The economic poverty of families in urban slums
is the most visible and pervasive feature of life.
In 1964, roughly one-sixth of the population of

the United States had family incomes of $3000 or

2)

3)

14

less or individual incomes of $1500 or less (Key-

serling, 1964).

Powerlessness is one of the central problems of

the poor. They are more likely to be arrested
without justification and detained without adequate
regard for individual rights. Health services are
often inferior although not substantially less
expensive. In many areas of their lives, the poor
have difficulty defending themselves against inva-

sions of privacy, i.e. welfare agents.

Related to these conditions of poverty and power-
lessness is a vulnerability to disaster. Typically
the poor are not only without financial reserves

of their own, but they are most likely to be given
little advance notice when laid off from work, to
be victims of legislative and bureaucratic delay

or interruptions of welfare service. They typi-
cally possess little credit or borrowing power

and are less likely to have friends with resources.
Life is lived on the edge of incipient tragedy

which they are powerless to avert.

15

4) The circumstances of life of the poor combine to
restrict the range of alternatives of action avail-
able to them. Lack of economic resources, lack of
power, lack of education, and lack of prestige
greatly reduce physical mobility and the oppor-
tunity for different options concerning areas of
residence, housing, employment, and most areas of

their lives.

5) The poor are without status and prestige in the
society. They are aware of their position in
society, and this awareness is probably transmitted
to children directly through their own observations
and indirectly through adults who define for their

children their relative position in the community.

6) The poor have few experiences in common with the
upper working-class or the middle-class. Although
there is sufficient contact and visibility to make
the poor aware of the values of the middle-class,

their range of experience is quite different.

A number of studies have examined the socialization
practices directly related to aggressive responses of

middle— and lower-class parents. Bandura and Walters (1959)

16

and Sears 33 31. (1957) found a relationship between par-
ental permission of aggression and the child's aggressive
behavior. In each case, mothers of aggressive boys were
found to be significantly more permissive of aggression.
Many lower-class mothers encourage aggression against
other children, believing that a child must learn to take
care of himself. Kenneth Clark (1963) observed that
lower-class children are more likely to react aggressively
since aggressive patterns of behavior are part of the
struggle for survival within the lower-class pattern of
living.

Allinsmith (1954) found that middle-class parents
typically used psychological punishment while the punish-
ment of lower-class parents was predominantly physical.
Physical punishment led to more aggression than psycholog-
ical punishment. Furthermore, physical punishment led to
direct aggression against authority while psychological

punishment led to indirect aggression against authority.

Racism

Racist practices continue to be a source of con-
stant frustration to the black man in America. Harrington

described the black man's predicament in the following:

17

The Negro is poor because he is black; that is
obvious enough. But perhaps more importantly,
the Negro is black because he is poor. The
laws against color can be removed, but that
will leave the poverty that is the historic
and institutionalized consequence of color.

As long as this is the case, being born a
Negro will continue to be the most profound
disability that the United States imposes upon
a citizen (1963, p. 73).

Unlike the young white immigrant of several genera-
tions ago, the black child does not find society as open
to him. Instead, White America keeps the black man down.
"It forces him into slums and the dirtiest, lowest paying
jobs; it theorizes about him and stereotypes him as lazy,
shiftless, and irresponsible. The consequence is heart-
breaking frustration" (Ibid.).

Kenneth Clark indicates that blacks as a group are
subject to more general conditions of social isolation,
rejection, and frustration than any other group in America.

Robert Coles (1964) reports that the late 1960's
brought improved self-esteem and constructive channeling
outward in black children. The writer's own experience
support this view. There seems to be more encouragement

among blacks to express aggression openly--not to squelch

it as the writer, like most blacks, was taught as a child.

18

Coles quotes a black mother who stands in awe of
her nine-year—old son's angry denunciation of white people
and her five-year—old daughter's equally vehement words
and feelings. She says that her children tell her:

We've got to keep pushing, and we can't let "the
white" beat on us no more, and we should go
across town (Boston) and tell them to give us
what we should have been having three hundred
years ago:(p. 7).
The son called his mother a "Black coward" when she tried
to silence him. The mother describes her own Southern
upbringing—-a striking contrast to her children's words:
My mother, she used to tell us kids we shouldn't
even let ourselves know what we are thinking and
wanting. And even in Church, she'd say we
should be careful what we pray for, because you
never can tell, you can get in trouble for being
"uppity" even if you look "uppity" when you come
out of Church. The white man, he'll see you
and tell you, "Nigger, you look too smart"
(£313., p. 50).

Clark also observes the tendency of young blacks
to refuse to assume the meek, unaggressive role which in
the past had been stereotypic of blacks in their relation—
ships with whites. Young blacks have accepted the same
goals as whites and are demanding the immediate attainment

of these goals rather than in the future. Their methods

for attaining these goals are typically American—-based

19

on competition and aggressive assertion of individual
worth rather than meekness and subservience.

Clark contends that middle-class black children
are trained to control their impulses, to adhere to de-
mands of respectability, and to avoid negative contact
with whites. Direct expression of aggression, overt vio-
lence, or anti-social behavior is discouraged.

Margaret Brenman (1940) investigated the effects
of social class on the personal adjustment of Negro girls.
She found that lower-class black girls gave more open
expression to aggressive feelings in their relationships
with whites. Middle- and upper-class black girls resembled
white girls in that they accepted the same restraints and
social demands which they thought would give them higher

social status.

Hammer's Study

 

Hammer (1953) extended the frustration-aggression
to socio-racial areas by testing the relative degree of
aggression in 148 black children and 252 white children
attending grades one through eight in Virginia. No attempt

was made to control socio—economic level since few blacks

20

in Virginia were of comparable socio-economic level as
whites. In this study, frustration was defined as the
thwarting or blocking of an individual's significant
drives, motives, and needs. Also, Maslow's (1941) theory
that frustration consists of deprivation and personality
threat was used. Hammer assumed that the black child suf-
fered more frustration than the white child as defined by
the criteria of deprivation and personality threat. He
used the House-Tree-Person Test, a projective technique,
to tap the presence of aggression. A number of qualita—
tive indicators of aggression were developed. Based on
the number of indicators present in a drawing and the
judge's clinical impression, the drawings were rated on
an aggression scale of zero to two. A rating of zero
represented no apparent aggression or hostility; a rating
of one represented mild aggression and hostility, while a
rating of two represented severe aggression and hostility.
The mean aggression and hostility rating for white child-
ren was .308 while that of black children was .823. Thus,
white children proved to be closer to the point of no
apparent aggression and hostility while black children

scored closer to the point of mild aggression.

21

Hypgtheses

 

The present study is a modified replication and
expansion of Hammer's study. Like Hammer's study, the
primary assumption underlying this study is the notion that
frustration leads to aggression. This emphasis upon the
frustration-aggression hypothesis, however, is not to
negate the tenability of the various other antecedents of
aggression that have been discussed above. Regardless of
the tenability of any of these approaches, there is plenty
of evidence that a number of factors related to poverty
and racism act as strong instigators to aggression. The
present project attempts to assess groups where such
alleged conditions exist and to use the Draw-A-Person
test, a projective technique, to determine the relative
degree of aggression as projected in the DAP of children
of differing racial and socio-economic backgrounds who
are subjects of racial discrimination and poverty. The
present study expands Hammer's by examining social class
differences in the degree of aggressiveness as projected

in the DAP and by controlling for age, sex, and grade.

I III: II

22

Hypothesis 1

 

A black child growing up in white America finds
the world to be hostile and threatening~~full of disap-
pointment and frustration. The discrimination and inequal-
ities he experiences often cause him lifelong, inescapable
frustration. One reaction to this frustration is aggres-
sion. Because they experience strong instigators to
aggression associated with racial discrimination:

1) Black children will show significantly more
aggression than white children.

a. Black middle-class children will show
significantly more aggression than white
middle-class children.

b. Black lower-class children will show
significantly more aggression than white
lower-class children.

Hypothesis 2

 

Poor housing, inadequate health services, economic
deprivation, feelings of helplessness and hopelessness,
and other conditions associated with poverty serve to pro-
duce considerable frustration. Children growing up amidst
such conditions are highly frustrated. Because they exper-
ience strong instigators to aggression associated with

poverty:

23

2) White lower-class and black lower-class children
will show significantly more aggression than
white middle-class and black middle-class
children.

a. White lower-class children will show signif-

icantly more aggression than white middle—
class children.

b. Black lower-class children will show signif—

icantly more aggression than black middle-
class children.

Hypothesis 3

 

The group mean scores of aggression will assume
the following rank in order of least aggression to most

aggression:

3) a. white middle-class-—least aggression
b. black middle-class
c. white lower-class

d. black lower-class--most aggression

Black lower-class children will show the most aggression
because they are subjected not only to instigators to
aggression related to racial discrimination but also to

instigators associated with poverty.

METHOD

Sample

The sample consisted of a total of seventy-two
4th and 5th grade male students in a large midwestern
school district. Forty of the subjects were white and
thirty-two were black. Socio-economically, there Were
twenty white middle-class and twenty white lower-class;
sixteen black middle-class and sixteen black lower-class
subjects. Socio—economic status was determined by par-
ental occupational information provided by classroom
teachers. For the most part, the middle-class sample
attended a school on the city's northwest side. This
school is fairly well racially integrated (about 25-30%
black) and is considered a highly desirable educational
setting by both black and white parents. Children of a
number of local and state governmental employees attend
this school including the children of the city's mayor.
The parents of the middle-class subjects held professional
and technical jobs. Typical parental occupations of these
subjects include the following: teaching--elementary,

24

25

secondary, and university; medicine-—physicians and
nurses; social work; management; architecture; pharmacy,
computer programming, map making; government; engineering;
educational administration; insurance, and personnel.
White middle—class parents tended to hold higher—level,
more prestigious jobs than parents of black middle—class
subjects.

The lower—class subjects attended a school located
on the south side of the city. The school is located in
an area recently annexed to the city and serves a popula-
tion of blacks, whites, and Spanish-Americans, many of
whom moved to the area after being displaced by urban
renewal projects. This school was formerly a small
"country" school, but its enrollment has increased four-
fold as a result of several housing projects built recently
in the area. Typical parental occupations of these sub-
jects include such semi-skilled and unskilled jobs as the
following: custodial, cooking, security, service station
attending, manual labor, domestic work, and factory pro-
duction work. A sizable number of these subjects were
from one-parent homes where the family received support

from public funds.

26

These schools were chosen for the study because
both were racially integrated and thus had a significant
number of blacks as well as white students and because
both schools served rather homogeneous socio-economic
groups. All of the fourth and fifth grade classes in each

of the schools participated in the study.

Instrument

 

The Draw-A—Person or Human Figure Drawing, a pro-
jective technique described by Machover (1949) and others
was used in the study because of its similarity to the
House-Tree-Person projective technique utilized by Hammer.
The Draw-A-Person requires the subject to make a drawing
of a person. It is generally accepted that in drawing the
human figure, a person projects his own physical aspects
as well as his psychological self. The adaptability of
this technique for group administration made it quite
appropriate for the present study. Specific graphic
indices of the DAP have been found to be predictors of

aggression.

27

Procedure

 

The Draw-A-Person was administered by the writer
to four entire fourth and fifth grade classes at the west
side school containing a total of approximately 120 stu-
dents and three intact fourth and fifth grade classes at
the south side school containing approximately 90 students.
Each student in the seven classrooms involved was given a
sheet of paper size 8-1/2" x 11" and a number two pencil
with an eraser. The following instructions suggested by
Koppitz (1968) were given: "On this piece of paper, I
would like you to draw a WHOLE person. It can be any kind
of a person you want to draw, just make sure that it is a
whole person and not a stick figure or a cartoon figure."
Subjects were permitted to ask questions in order to clar-
ify the directions. They were also instructed to put
their name, age, and grade on the back of the drawing.

While the subjects were drawing, teachers made
lists of all the children in their classroom present that
day and provided information regarding the racial identity
and parental occupation or source of income of each of
their students. Drawings of children for whom economic

status could not be determined were not used in the study.

28

Upon receiving the drawings, the writer assigned
codes, identifying the subjects racially and socio-
economically. The codes used were devised by the writer
and were not known to any other person.

A total of 210 drawings were collected. Of this
number, approximately 110 of them were drawn by female
students and therefore were not used in the study. An
additional twenty-five drawings could not be used for
various reasons, e.g. child did not follow directions.
Thus, a total of seventy-two drawings were suitable for
use in the study.

The following graphic indices of aggression and
hostility, some of which were utilized by Hammer and
others which were identified by Koppitz (1968), Machover
(1949), and Buck (1948) were used to assess the presence
of aggression in the subjects' drawings.

(1) Poor integration of parts—~one or more parts not
joined to the rest of the figure, part only connected
by a single line or barely touching. Poor integra-
tion appears to be associated with impulsivity.

(2) Gross asymmetry of limbs——one arm or leg differs
markedly in shape from the other arm or leg. This
item is not scored if arms or legs are similar in
shape but just a bit uneven in size. Koppitz found

this item to occur frequently in drawings of aggres-
sive children.

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(ll)

29

Teeth--any representation of one or more teeth. Kop-
pitz, Machover, and Hammer consider the presence of
teeth to be related to overtly aggressive and possibly
sadistic behavior.

Long arms--arms excessively long, arms long enough to
reach below the knee or where the knee should be.
Koppitz and Buck found that long arms reflected exter-
nally directed, aggressive needs, an aggressive reach-
ing out into the environment.

Big hands--hands as big or bigger than the face of the
figure. Big hands seem to be associated with aggres-
sive and acting-out behavior involving the hands,
either directly as in overt aggression, or indirectly,
as in stealing.

Genitals--realistic or unmistakably symbolic represen-
tation of genitals. Koppitz found this item to be
rare in children's drawings, but when it did appear,
it was invariably in the drawings of overtly aggres-
sive children. The presence of genitals or symbols
suggest poor impulse control.

Sharply pointed fingers or toes--spear or talon-like
fingers. Buck found this item to reflect aggressive
tendencies.

Sharply squared shoulders--shoulder line a straight
line or at least one angular shoulder. According to
Buck, this item connotes over-defensive, hostile
attitudes.

Arms drawn folded across the chest--suggests attitudes
of suspicion and hostility.

Person carrying weapons such as guns, swords, black-
jacks, etc. Buck and Hammer found that the presence
of weapons in the drawing reflected aggressive and
hostile tendencies.

Drawings made conspicuously too large for the page--
without adequate page space framing them (particularly
when they touch or almost touch the page's side

3O

margins), tend to indicate a feeling of frustration
produced by a restraining environment, with concomi-
tant feelings of hostility and a desire to react
aggressively either against the environment, the
self, or both.

Two coders trained by the examiner scored the
drawings for the presence or absence of each of the above-
described indicators of aggression. They were not aware
of the race or socio-economic status of the subjects whose
drawings they scored. The drawings were distributed ran-
domly between them. On two pretesting trials in which
each coder scored five drawings of children not included
in the study, there was a 80 and a 100 percent of agree-
ment in Trials I and II, respectively. If an indicator
were present in a drawing, a score of one (1) was assigned;
if the indicator was not present, a score of zero (0) was

assigned. Each subject, therefore, received a score of

zero to eleven.

RESULTS

Since there were eleven (11) indicators of aggres-
sion, each subject could potentially receive a total
aggression score of zero to eleven. However, in the data,
the scores ranged only from zero to three with an overall
mean score of .51. No indicators of aggression were found
in 57% of the drawings.

A two-way analysis of variance was used to deter-
mine the effect of race, socio-economic status, and the
interaction of race and socio-economic status upon the
relative degree of aggression of each subject. The failure
of the indicators of aggression to appear in the drawings
and the lack of variance among the total aggression scores
of the subjects caused the differences among them to be
statistically non—significant (Table 1). No statisti-
cally significant differences were found among any of the

variables.

31

32

TABLE 1

SUMMARY OF ANALYSIS OF VARIANCE RACE AND
SOCIO-ECONOMIC STATUS

 

 

 

Source . .. SS df MS F
Race .89 1 .89 .4263
Social Class (SES) .356 1 .356 .17
Race & Social Class .178 l .178 .085
Subj./Race & Social Class 8.36 26 2.09

Total 9.784 29

 

Hypothesis One

Hypothesis One that black subjects would show
significantly more aggression than white subjects was not
supported. However, visual examination of mean differ-
ences in Table 2 show that the data were in the direction
of the hypothesis. The mean score for blacks is .63 while
that of whites is .425. Fifty percent (50%) of all black
subjects had at least one indicator in their drawings
while thirty-five percent (35%) of all white subjects had
at least one indicator (Table 3). The data are also in

the direction of the two sub-hypotheses that black

33

TABLE 2

MEAN AGGRESSION SCORES--RACE AND
SOCIO-ECONOMIC STATUS

 

 

 

 

Race Middle Lower Total Race
Black .70 .56 .63
White .40 .45 .425
Total Social Class .53 .50

TABLE 3

DISTRIBUTION OF 3's WHO HAD AT LEAST ONE
INDICATOR IN DRAWING

 

 

 

Lower Middle Total Race

Pos.* Neg.** Pos. Neg. Pos. Neg.

Black 7 9 9 7 16 16

White 7 l3 7 13 14 26
Total SES 14 22 16 20

Total 30 42

 

*Pos. indicates number of subjects who had indicators in
drawings.

**Neg. indicates number of subjects who did not have
indicators in drawings.

34

middle-class subjects (mean .7) would show more aggression
than white middle-class subjects (.4), and that black
lower-class subjects (.56) would show more aggression than

white lower-class subjects (.45).

Hypothesis Two

 

Hypothesis Two that lower—class subjects would
show significantly more aggression than middle-class sub-
jects was not supported. In fact, an examination of mean
scores in Table 2 reflects almost equal means for both
the middle- and lower-classes, .53 and .50, respectively.
Forty-four percent (44%) of all middle—class subjects had
at least one indicator of aggression in their drawings
while thirty-nine (39%) of all lower-class subjects had
at least one indicator. The data were in the direction
of the sub—hypothesis that white lower-class subjects (.45)
would show more aggression than white middle-class sub-
jects (.40), but mean scores were not in the direction of
the hypothesis that black lower-class subjects (.56) would

show more aggression than black middle—class subjects (.70).

35

Hypothesis Three

 

Hypothesis Three that the group mean scores would
assume the following rank order of least aggression to
most aggression: white middle, black middle, white lower,
and black lower--was not supported. An inspection of
means in Table 2 reveals the following order of least to
greatest aggression: white middle, white lower, black

lower, and black middle.

Additional Findings

 

One of eight of the eleven indicators of aggres-
sion was present in at least one drawing (Table 4).
Three of the indicators did not appear in any drawings.
A total of 37 indicators appeared in the 72 drawings
(mean .51). Indicators appeared in 31 or 53% of the
drawings. Teeth and sharply squared shoulders were the
most frequently appearing indicators. Teeth appeared in
eleven drawings and sharply squared shoulders were pres-

ent in ten drawings.

36

 

 

 

TABLE 4
FREQUENCY OF AGGRESSION INDICATORS
Indicator BM BL WM WL Total

Poor Integration of Parts 0 l 0 0 1
Gross Asymmetry of Limbs 2 0 0 0 2
Teeth 2 3 4 2 11
Long Arms 0 0 0 0 0
Big Hands 0 0 O 0 0
Genitals 0 0 0 0 0
Sharply Pointed Fingers

and Toes 0 2 0 2 4
Sharply Squared Shoulders 4 l 2 3 10
Arms Folded Across Chest 1 0 0 l 2
Person Carrying Weapon 2 0 l 0 3
Person Conspicuously

too Large 0 3 l 0 4

 

DISCUSSION

The lack of significant results in the present
study may be attributed largely to inadequacies of the
instrument used to assess the presence of aggression.
Considerable controversy surrounds the use of specific
graphic indices in human figure drawings as was done in
the present study. Granick and Scheflen (1958) pointed
out the difficulty of devising acceptable objective scor-
ing procedures for projective tests. They suggest that
it is not feasible to develop an overall statistical re-
liability of certain aspects. Most investigations of the
reliability of specific graphic indices of human figure
drawings (Gunderson and Lehner, 1958; Bradshaw, 1952) have
used a "percent of agreement" method. Swenson (1957)
criticizes this approach, maintaining that percent of
agreement and reliability are not the same. The research
on the use of specific graphic indices is inconclusive
with a number of studies supporting the validity of var-
ious graphic indices (Schmidt and McGown, 1959; Budwin
and Bruch, 1960; Kamana, 1960; Goldstein and Rawn, 1957)

37

38

and other similar studies failing to find significant re-
sults (Starr and Marcuse, 1959; Strumpfer and Nichols,
1962).

Guinan and Hurley (1965) and others (Zimmer, 1956;
Rosenzweig, 1960; Laane, 1960) criticize the practice of
making judgments and interpretations of the DAP based on
specific elements. Guinan and Hurley report that research
on the DAP has focused on specific variables in the draw-
ings while in clinical settings the test is used to obtain
a general impression of personality organization. They
conducted a reliability study of the DAP utilizing global
impressions as criteria for judgments and found it to be
quite reliable when used in this way. They therefore
suggest a global or "intuitive" approach to the DAP in
research rather than the traditional atomistic approach.
Such an approach may have been valuable in the present
study since many drawings which gave a very aggressive
impression clinically failed to reveal any of the desig-
nated indicators. Hammer used the wholistic approach in
his study in that he provided judges with criteria of
aggression to be used as guidelines, but they exercised
clinical judgment to rate the drawings in terms of the

degree of aggressiveness.

39

The most notable finding is the failure of the
eleven indicators of aggression to appear in the drawings.
Of the possible eleven, the largest number of indicators
to appear in a single drawing was three. Most drawings
had one or less (total mean--.51). No indicators of
aggression were revealed in 57% of the drawings.

The expectation that a greater number of indicators
would appear in the drawings may have been somewhat un-
realistic. Koppitz (1969) in a comparison of 30 Emotional
Indicators in the Human Figure Drawings of 468 4th and 5th
grade boys and girls from lower- and middle-class back-
grounds found that a little more than one-fourth of them
revealed two or more Emotional Indicators in their draw-
ings. Approximately half of the 72 subjects in the pres-
ent study revealed one or more of the possible eleven
indicators of aggression.

The data were in the direction of the hypothesis
that blacks would show more aggression than whites. The
lack of significance of this hypothesis may be related to
the inadequacy of the instrument as a measure of aggres-
sion. However, other factors may be operative. It is
possible that blacks and whites have equally aggressive

fantasy lives, but they may differ in the way they express

40

it. Blacks, because they have been discouraged in ver-
bally expressing aggression, may act out their aggression
in a physical manner whereas whites--especially middle-
class whites--may tend to express their aggression more
verbally because they are socialized to "talk about their
problems." This avenue traditionally has not been an
option for blacks because black parents quite early in
the child's life, began squelching the expression of
verbal aggression which could cause the child to come into
conflict with the white majority. He had to learn early
to inhibit verbal aggression and to appear passive and
docile. This pattern is changing somewhat. There is an
increasing recognition among blacks, especially middle-
class blacks, of the value of verbally expressing aggres-
sion, and such expression is being encouraged.

Another consideration which may account partially
for the lack of significant differences in the aggressive-
ness of blacks and whites is the possibility that there
may be in the environments of whites instigators to
aggression that are equally as powerful as the racism
experienced by the black child.

The data were not in the direction of the hypoth-

esis that lower—class subjects would show more aggression

41

than middle-class subjects. The lack of significant class
differences in the degree of aggressiveness may be due to
the sole use of economic circumstances and resources as
represented by occupational level as the determinant of
socio-economic status. The use of more systematically
defined indices of social status as occupation, education,
and ecological variables would have been appropriate as
well as possible emyhasis upon social and psychological
features of socio-economic status. The fact that middle-
class blacks, in contradiction to the prediction, showed
considerably more aggression than lower-class blacks may
be related to the fact that the middle-class black sub-
jects, while they were economically more advantaged than
the lower-class blacks, lived on the fringe of the west
side black ghetto while the lower—class blacks lived about
five miles away from the innercity in a recently developed,
multi-ethnic area with diverse income levels.

Securing adequate numbers of middle-class black
subjects and differentiating them in terms of socio-economic
status from the lower class, has been a problem experienced
by most investigators of race-related studies. This has
been the experience in the present study. There were a

number of "borderline" blacks in terms of socio—economic

42

status. It would have been easier to differentiate blacks
socio—economically if the lower-class subjects had been
chosen from a more hard-core, inner-city setting. The

fact that the data were in the direction of the hypothesis
that white lower-class subjects would show more aggression
than white middle-class subjects may be due to the more
clearly defined class differences between whites than be-
tween blacks. The middle-class whites were very obviously
at a higher level socially and economically than the lower-
class whites.

Social classes, like racial and ethnic groups, may
not differ in their aggressiveness but may differ consider-
ably in their expression of aggression. Koppitz (1969), in
the study of Emotional Indicators in the drawings of middle-
and lower-class children, found that when two groups of
students were matched for age, intelligence, and sex,
there was "not a single significant difference" between
the frequency of occurrence of any of the 30 Emotional
Indicators nor for the total number of indicators on each
drawing.

According to McNeil (1957), a middle-class child
and a lower-class child with equally aggressive fantasy

lives may differ considerably in behavior, with the

43

lower-class child's actions matching his fantasy while the
middle-class child inhibits expression of his. Feshbach
(1970) in his comprehensive review of the literature on
aggression discovered that the data bearing on the rela-
tionship between social class background of the child and
his aggressiveness were contradictory. He suggested that
psychological factors linking social class to aggression
such as socialization techniques employed and the degree
of frustration and other instigators to aggression exper-
ienced by children from different social strata--need to
be more systematically investigated. Feshbach contends
that there are important differences in the psychological
environment of children of each social class. They are
not homogeneous groups.

The groups ranked themselves as predicted in order
of least aggression to most aggression with the exception
of black middle-class who, contrary to prediction, showed
more aggression than white and black lower classes. This
occurrence may be explained by the problems related to the
black middle-class sample previously discussed. Another
possible explanation is that there may be a greater inci-
dence of apathy and corresponding less aggression among

lower-class whites and blacks while middle-class blacks

44

have greater expectancies, which when unrealized, create
increased frustration and resultant aggression.

The most frequently occurring indicator of aggress-
sion was the presence of teeth. This sign was revealed
in eleven drawings and was equitably distributed among
blacks and whites and middle- and lower-classes. Machover
considers mouth detailing with teeth to indicate oral
aggression. Griffith and Lemley (1967) found that two
signs-~teeth and a threatening look--were necessary to
permit inferrence of verbal aggression and that even the
presence of both signs was not enough to connote physical
aggression. Koppitz (1968), when she determined that teeth
occurred most often in the drawings of overtly aggressive
children, concluded that teeth are a sign of aggressive-
ness and not only of oral aggression. Goldstein and Rawn
(1957) found that teeth along with seven other specific
drawing details were related to aggression. K0ppitz, too,
discourages the use of one indicator as conclusive. She
considers the occurrence of teeth in a human figure draw-
ing to be diagnostically meaningful only if it occurs with
other indicators. .In the present study, teeth occurred
as the only indicator in nine of the eleven drawings in

which it appeared. Therefore, the teeth in most drawings

45

may not necessarily represent aggression, or it may only
represent a normal amount of aggression in children neces-
sary for achievement and leadership.

The second most frequently occurring indicator was
sharply squared shoulders which according to Buck (1948)
connotes overdefensive, hostile attitudes. The emphasis
on shoulders may be a way of expressing masculine strivings.
Though pre-adolescent, the subjects are concerned about
body image. They are approaching that awkward age when
they are neither youngsters nor grde-ups. Needless to
say, in our society, masculinity is viewed as synonymous
with aggression since boys are socialized to be aggressive
and girls are socialized to be passive. It would be of
interest to compare the occurrence of "sharply squared
shoulders" in the drawings of male and female subjects.
Because of the favored position of masculinity and the
increasing assertiveness of women, differences may be
minimal.

Sharply pointed fingers appeared only in the draw-
ings of lower-class children. This may be related to our
earlier observation that lower-class children tend to act
out their aggression physically while middle-class children

may be more verbally aggressive. In line with this thinking,

46

one would have expected "teeth" to appear more in the
drawings of middle-class subjects, but this did not occur.

"Persons carrying weapons" appeared in only three
drawings-—two black middle and one white middle. With the
excessive use of weapons on television, the Vietnam war,
and the increasing use of weapons by private citizens for
security, one would expect a greater occurrence of this
indicator. A common stereotype is that of the lower-class
black carrying weapons. In the present study, there were
not significant differences betWeen whites and blacks in
terms of the presence of weapons in the drawings, and all
of the drawings containing weapons were those of middle-
class subjects.

In Hammer's study a significantly larger number of
black subjects drew persons conspicuously too large for
the page than white subjects. Since this indicator had
been defined by Buck (1948) as representative of feelings
of great frustration produced by a restraining environment
with concomitant feelings of hostility and a desire to
react aggressively, Hammer interpreted this finding as
support for his judgment that blackIsubjects harbor greater
feelings of frustration with a desire to react aggressively,

than do whites. The data in the present study is in the

47

direction of Hammer's finding; This sign occurred four
times—~three times in the drawings of black lower-class
subjects and once in the drawing of white middle-class
subjects.

Fish and Larr (1972) compared the drawings made
by black youngsters during the 1950's with those made by
a comparable group of blacks during the 1970's and found
that the recent drawings revealed more identifiably black
characteristics than those of the 1950's. This phenomenon
was attributed to increased black awareness and pride.
In the present study, one of the black middle-class (N = 16)
subjects and eight (50%) of the black lower-class subjects
made identifiably black drawings. The most obvious dis-
tinguishing feature in the drawings was the Afro hair
style; however, in some cases treatment of the nose and
mouth features were also indicative. None of the white
subjects made identifiably black drawings. The large dis-
crepancy between middle- and lower-class blacks in the
drawing of black features could be related to the possi-
bility that lower-class blacks are more likely to adopt
Afro hair styles and to be more overt in their identifica-
tion with blackness while middle-class blacks may remain

more accepting and imitative of white styles and values.

48

It is generally expected that a person will draw
his same-sex first when instructed to draw a person. Of
the 72 subjects, 71 of them drew males, and only one drew
a female.

An issue of great concern to social scientists,
educators, and parents is the control of aggression. Eron
(1971) has suggested the following procedures for those
persons who want to reduce the aggressive behavior of
children. (1) Reduction of the amount of instigation to
aggression, i.e. bickering between the parents; criticizing
the child; not nurturing the child; (2) Making aggressive
models less available to the child--reduce the violence
observed by the child on TV, in the parents' disciplinary
practices, and in the parents' social and occupational
behavior in general. (3) Making non-aggressive models
more available to the child—-on TV, in disciplinary prac-
tices, and in social and occupational behaviors, and
(4) Programming the consequences of aggression and of
non-aggression so that aggressive behavior will decrease
and non-aggressive behavior will increase.

Eron (1971) and Feshbach (1970) raise questions
about the ramifications of shaping a child's behavior in

such a way as to minimize aggressiveness. In a capitalistic

49

society such as ours, low aggressiveness may handicap an
individual in coping with the socio-economic environment.
Our country places a high value on aggressiveness, revering
and adulating those who have evidenced this behavior. In
this society, low aggressiveness could be economically and
psychologically crippling. A non-aggressive male would
appear almost as a devrant in our society. On the other
hand, questions may also be raised about the consequences
of shaping a child's behavior toward aggressiveness. The
positive consequence would be the ability to cope in a
competitive society. The negative would be a continuation
of the pattern of violence that has existed from the incep-
tion of the country and continues to exrst today as evi-
denced by rising crime rates and our continued military
involvement in Southeast Asia. Eron considers the issue
of control to be the major problem of systematically train-
ing for aggression. Aggression that is learned, expressed,
and rewarded in one setting (like the military) is con-
demned, considered criminal, and punished in another con—
text.

Feshbach (1970) suggests the need for further
study of the functional relationship between aggression

and assertiveness. These are generally treated

50

synonymously during pre-school years but become differen-
tiated by adolescence. Assertiveness is seen as "a behav-
ioral pattern of active coping with environment in the
pursuit of one's own needs when confronted with counter-
pressure and is distinguished from aggression in that it
lacks the painful and destructive consequences of aggres-
sion" (1313., p. 243). Further exploration is necessary
to establish how closely related these two behavior pat-
terns are and to determine how difficult it would be for
a child to discriminate reinforcement for assertiveness
and aggression.

According to Feshback, insufficient attention has
been given to the social context as it contributes to the
determination of aggressive behavior as a group influence.
Studies to date have focused mainly on individual aggres-
sion even though much of human aggression of children and
adults occur in the context of a group activity. There is
little empirical data that bear on differences in aggres-
sion when a child is part of a group compared to when he
is acting as an individual. In light of the increased
group violence in our society-—race riots, violent student
demonstrations-—it is important to better understand the

influence of the group upon aggressive behavior.

51

Although some very extensive and fruitful empir-
ical studies utilizing a variety of instruments have been
done on aggression, there is a need for continued efforts
to identify those forces which contribute to aggressive
behavior and to formulate methodology for the effective
control of aggression. Further study should focus more
upon utilizing measures of aggressive behavior rather than
predicted or fantasy aggression based on projective tech-

niques.

 

III-Ill

S UMMARY

The present study, which is based on the assumption
that frustration leads to aggression but accepts the tena-
bility of other theories of aggression, examines the effect
of such frustrating conditions as racism and poverty upon
the degree of aggressiveness in children as reflected in
the Draw-A-Person test. It was hypothesized that black
subjects would show more aggression than white subjects;
that lower-class subjects would show more aggression than
middle-class subjects, and that the groups would assume
the following rank of least aggression to greatest aggres—
sion: white middle-class, black middle-class, white lower-
class, black lower-class.

The subjects included seventy—two fourth and fifth
grade male students--thirty-two black and forty white.
Socio-economically, there were twenty white middle—class
subjects; twenty white lower-class subjects; sixteen black
middle-class subjects, and sixteen black lower—class sub-
jects. A measure of the relative degree of aggression of
each subject was obtained through the use of eleven

52

53

graphic indices of aggression in the drawing of the human
figure identified by Machover (1949), Buck (1948), and
Koppitz (1968). Each drawing was scored for the number of
indicators present.

Statistical analysis demonstrated that neither
race, social class, nor the interaction of race and social
class were significantly related to the subjects‘ aggres-
siveness as projected in the DrA-P test. Trends in the
data suggest that race may be related to the level of
aggressiveness in that blacks consistently received greater
mean scores of aggression than whites. The effect of
social class is equivocal.

Serious questions are raised about the realiability
of utilizing specific graphic indices of the Draw-A-Person
test as a measure of aggression. A global or "intuitive"
approach is recommended.

Further investigation of the causes of aggression,
techniques for measuring it, and methods of controlling it
are suggested. Specifically, the need for further study
of the functional relationship between aggression and
assertiveness (Feshbach, 1970) and the determination of

aggression behavior as a group influence are indicated.

REFERENCES

REFERENCES

Abt, L. and Bellak, L. (Eds.). Projective Psychology.
New York: Grove Press, 1950.

 

Allinsmith, B. B. "Parental Discipline and Children's
Aggression in Two Social Classes." Dissertation
Abstracts, 1954, 14, 708.

 

 

Azrin, N. H. and H012, W. C. "Punishment." In W. K.
Honig (Ed). Qperant Behavior: Areas of Research
and Application. New York: Appleton-Century-
Crofts, 1966, 380-447.

 

 

Bandura, A., Ross, 0., and Ross, S. A. "Vicarious Rein-
forcement and Imitative Learning." Journal of
Abnormal and Social Psyghology, 1963, 67, 601-607.

 

 

Bandura, A., and Walters,R. H. Adolescent Aggression.
New York: Ronald Press, 1959.

 

Bandura, A., and Walters,R. H. Social Learning and Per-
sonality Development. New York: Holt, 1963.

 

 

Berkowitz, L. Aggression: A Social Psychological Analysis.
New York: McGraw-Hill, 1962.

Berkowitz, L. "Some Aspects of Observed Aggression."
Journal of Personality and Social Psychology, 1965,
2, 359-369.

Berkowitz, L. Roots of Aggression. New York: Atherton
Press, 1969.

 

Bodwin, R., and Bruck, M. "The Adaptation and Validation
of the DAP As a Measure of Self-Concept." Journal
of Clinical Psychology, 1960, 16, 427-429.

 

54

55

Bradshaw, D. "A Study of Group Consistencies on the DAP
Test in Relation to Personality Projection."
Unpublished M.A. Thesis, Catholic University,
1952.

Brenam, Margaret. "The Relationship between Minority
Group Identification in a Group of Urban, Middle-
Class Negro Girls." Journal of Social Psychology,
1940, 72, 179—196.

 

Buck, J. N. "The H-T-P Technique: A Qualitative and
Quantitative Scoring Manual." Journal of Clinical
Psychology, 1948, 4, 317-396.

 

 

Burton, A. "The Aggression of Young Children Following
Satiation." American Journal of Orthopsychiatry,
1942, 12, 262-268.

 

Buss, A. H. The Psychology of Aggression. New York:
John Wiley and Sons, 1961.

 

Chasdi, E. H. and Lawrence, M. S. "Some Antecedents of
Aggression and Effects of Frustration in Doll
Play." In D. McClelland (Ed.), Studies in Motiva-
tion. New York: Appleton-Century-Crofts, 1955.

 

Clark, K. Prejudice and Your Child. Boston, Beacon Press,
1963.

 

Coles, R. Children of Crisis. New York: Dell Books, 1964.

 

Davitz, J. "The Effects of Previous Training on Post-
frustration Behavior." Journal of Abnormal and
Social Psychology, 1952! 47, 309-315.

 

 

Dollard, J. 33 31. Frustration and Aggression. New Haven:
Yale University Press, 1939.

 

Epstein, R. "Aggression Toward Out-Groups as a Function of
Authoritarianism and Imitation of Aggressive Models."
Journal of Personality and Social Psychology, 1966,
3, 574-579.

56

Eron, L., Walder, L., and Lefkowitz, M. Learning of V

Aggression in Children. Boston: Little Brown
and Company, 1971.

 

 

Feshbach, S. "Aggression." In P. Mussen (Ed.), Manual
of Child Psychology. New York: John Wiley and
Sons, 1970,,159_259.

 

Fish, J. and Larr, C. "A Decade of Change in Drawings -
by Black Children." Unpublished. “

Frederickson, N. "The Effects of Frustration on Negati-
vistic Behavior of Young Children." Journal of
Genetic Psychology, 1942, 61, 203-226.

 

 

Freud, A. The Ego and the Mechanisms of Defense. London:
Hogarth Press, 1937.

 

Freud, S. The Ego and the Id. London: Hogarth Press,
1923.

 

Freud, S. Beyond the Pleasure Principle. London: Ho-
garth Press, 1922.

 

Fromm, E. Escape From Freedom. New York: Farrar & Rine—
hart, 1941.

 

Glueck, S. and Glueck, E. Unravelling Juvenile Delinquency.
Cambridge: Harvard University Press, 1950.

 

Goldstein, A. and Rawn, M. "The Validity of Interpretive
Signs of Aggression in the Drawing of the Human
Figure." Journal of Clinical Psychology, 1957,
13, 169-171.

 

Granick, S. and Scheflen, N. "Approaches to Reliability
of Projective Tests with Special Reference to the
Blacky Pictures Test." Journal of Consulting

 

Psychology: 1958, 22, 137-141.

 

Griffith, A. and Lemley, D. "Teeth and Threatening LookV’
in the D-A-P Test as Indicating Aggression."
Journal of Clinical Psychology, 1969, 23, 83-86.

 

57

Guinan, J. and Hurley, J. "An Investigation of the Relia-
bility of HFD's." Journal of Proj. Techniques &
Pers. Assessment, 1965, 300-304.

 

 

Hammer, E. "Frustration-Aggression Hypothesis Extended to
Socio-Racial Areas: Comparison of Negro and White
Children's H-T-P's." Psychiatric Quarterly, 1953,

 

27, 597-607.

Harrington, M. The Other America. Baltimore: Penguin
Books, 1963.

 

Hartmann, H., Kris, E., and Lowenstein, R. "Notes on the
Theory of Aggression." In A. Freud (Ed.), 133
Psychoanalytic Study of the Child, Vol. 3. New
York: International Universities Press, 1949,
9-36.

 

Hartup, W. and Himeno, Y. "Social Isolation Versus Inter-
action with Adults in Relation to Aggression in
Preschool Children." Journal of Abnormal and
Social Psychology, 1959, 59, 17—22.

 

 

Hays, W. L. Statistics. New York: Holt, Rinehart, &
Winston, 1963.

 

Hess, R. "Social Class and Ethnic Influences Upon Social-
ization." In P. Mussen (Ed.), Manual of Child
Psychology. New York: John Wiley and Sons, 1970,
457-557.

 

 

Hicks, D. "Imitation and Retention of Film Mediated Ag-
gressive Peer and Adult Models." Journal of
Personality and Social Psychology, 1964, 2, 97-100.

 

 

Horney, Karen. Our Inner Conflicts. New York: Norton,
1945.

 

Jackson, L. Aggression and Its Interpretation. London:
Methuen and Company, 1954.

 

Jegard, S. and Walters, R. "A Study of Some Determinants
of Aggression in Young Children." Child Develop—

 

58

Kahn, M. W. "The Effect of Severe Defeat at Various Age
Levels on the Aggressive Behavior of Mice."
Journal of Genetic Psychology; 1951, 79, 117-131.

 

Kamana, D. "An Investigation on the Meaning of Human «/
Figure Drawings," Journal of Clinical Psychology,
1960, 16, 429-430.

 

Kaufmann, "Definitions and Methodology in the Study
of Aggression." Psychological Bulletin, 1965,
351-364.

 

Keyserling, L. Progress on Poverty. Washington, D.C.:
Conference on Economic Progress, 1964.

 

Kirk, R. Experimental Design: Procedures for the Behav-
ioral Sciences. Belmont, California: Brooks/Cole,
1968.

 

 

Koppitz, E. M. Psychological Evaluation of Children's
Human Figure Drawings. New York: Grune and Strat-
ton, 1968.

 

 

/

Koppitz, E. M. "Emotional Indicators on Human Figure
Drawings of Boys and Girls from Lower and Middle-
Class Backgrounds." Journal of Clinical Psychology,
1969, 431-434.

 

Korner, A. Some Aspects of Hostility in Young Children.
New York: Grune and Stratton, 1949.

 

Laane, C. "Clinical Experience with the Figure Drawing “I
Test," Journal of Clinical and Experimental Psy-
chopathology, 1960, 21, 129—141.

 

 

Lawson, R. Frustration. New York: Macmillan Company,
1965.

 

Lefkowitz, M., Blake, R., and Mouton, J. "Status Factors
in Pedestrian Violation of Traffic Signals."
Journal of Abnormal and Social Psychology, 1955,
51, 704-706.

 

59

Lehner, G., and Gunderson, E. "Reliability of Graphic
Indices in A Projective Test (the D-A-P)."
Journal of Clinical Psychology, 8, 125-128.

 

Lesser, G. "Maternal Attitudes and Practices and the
Aggressive Behavior of Children.“ Unpublished
doctoral dissertation, Yale University, 1952.

Loovaas, 0., Baer, D., and Bijou, S. "Experimental Pro-
cedure for Analyzing the Interaction of Social
Stimuli and Children's'Behavior." Paper read at
S.R.C.D. Convention, Berkeley, 1963.

Machover, K. Personality Projection in the Drawing of the
Human Figurs. Springfield, Illinois: Charles C.
Thomas, 1949.

 

 

Mallick, S. and McCandless, B. "A Study of Catharsis of
Aggression." Journal of Personality and Social
Psychology, 1966, 4, 591-596.

 

 

Maslow, A. "Deprivation, Threat, and Frustration." Psy-
chological Review, 1941, 48, 364-366.

 

McCord, W., MrCord, J., and Howard, A. "Familial Corre-
lates of Aggression in Non-Delinquent Male Child-
ren." Journal of Abnormal and Social Psychology,
1961, 62, 79-93.

 

Miller, Neal E. "The Frustration-Aggression Hypothesis."
In R. Lawson (Ed.), Frustration. New York: Mac-
millan Company, 1965, 72—76.

 

Miller, N. E. and Bugelski, R. "Minor Studies in Aggres-
sion: The Influence of Frustrations Imposed by
the In-Group on Attitudes Expressed Towards Out-
Groups." Journal of Psychology, 1948, 25, 437-432.

 

McNeil, E. "Psychology and Aggression." Journal of Con-

flict Resolution, 1959, 3, 195-293.
./
Moore, S. "Displaced Aggression in Young Children."
Journal of Abnormal and Social Psychology, 1964,
68, 200-204.

 

 

 

60

Mowrer, O. H. "A Stimulus-Response Analysis of Anxiety
and Its Role as a Reinforcing Agent." Psycholog-
ical Review, 1939, 46, 553-565.

 

 

Nunally, J. Psychometric Theory. New York: McGraw-Hill,
1967.

 

Otis, N. and McCandless, B. "Responses to Repeated Frus-
tration of Young Children Differentiated According
to Need Area." Journal of Abnormal and Social
Psychology, 1955, 50, 349-353.

 

 

Palmer, J. O. The Psychological Assessment of Children.
New York: John Wiley and Sons, 1970.

 

Rabin, A. I. and Haworth, M. R. (Eds.). Projective Tech- V
niques with Children. New York: Grune and
Stratton, 1960.

 

 

Rosenbaum, M. and Tucker, I. "The Competence of the Model
and the Learning of Imitation and Non-Imitation."
Journal of Experimental Psychology, 1962, 63,

 

183-190.
Rosenzweig, S. "The Rosenzweig Picture Frustration Studyrv
Children's Form." In Rabin and Haworth (Eds.),

Projective Techniques with Children. New York:
Grune and Stratton, 1960, 149-176.

 

\/
Schmidt, L. and McGown, J. "Differentiation of Human

Figure Drawings." Journal of Consultinngsychology,
1959, 23, 129-133.

 

Scott, J. and Marston, M. "Non-Adaptive Behavior Resulting
from a Series of Defeats in Fighting Mice." Jour-
nal of Abnormal and Social Psychology, 1953, 48,
417-428.

 

Sears, R. R., Macoby, E. E., and Levin, H. Patterns of
Child Rearing. Boston: Row Peterson, 1957.

 

 

Sears, R. R., Rau, L., and Alpert, R. Identification and
Child Rearing. Stanford: Stanford University
Press, 1965.

 

 

61

Selltiz, C. 33 31. Research Methods in Social Relations.
New York: Holt, Rinehart, and Winston, 1951.

 

Seward, J. P. "Aggressive Behavior in the Rat: IV. Sub-
mission as Determined by Conditioning, Extinction,
and Disuse. Journal of Comparative Psychology,
1946, 39, 51-76.

 

Singer, J. E. (Ed.). The Control of Aggression and Vio-
lence. New York: Academic Press, 1970.

 

Starr, S. and Marcuse, F. "Reliability in the 'DAP' Test."v/
Journal of Projective Techniques, 1969, 23, 83-86.

 

Strumpfer, D. and Nichols, R. "A Study of Some Communi-
cable Measures for the Evaluation of Human Figure
Drawings." Journal of Projective Techniques,
1962, 26, 342-353.

I/

 

Swenson, C. "Empirical Evaluations of HFD's." Psycholog-
ical Bulletin, 1957, 431-466.

 

 

Waelder, R. "Critical Discussion of the Concept of an
Instinct of Destruction." Bulletin of the Phila-
delphia Association of Psychoanalysis, 1958, 6,
97-109.

 

 

Walters, R. H. "Implications of Laboratory Studies of
Aggression for the Control and Regulation of Vio-
lence." Annals of the American Academy of Polit-
ical and Social Science, 1966, 364, 60-72.

 

 

Walters, R. H., and Brown, M. "Studies of Reinforcement
of Aggression: III. Transfer of Responses to an
Interpersonal Situation." Child Development,
1963, 34, 563-571.

 

 

Yarrow, L. "The Effect of Antecedent Frustration on Pro-
jective Play." Psychology Monograph, 1948, 62,
293.

Young, P. Scientific Social Surveys and Research. Engle-
wood Cliff: Prentice-Hall, 1966.

 

62

Zimmer, H. "Validity of Sentence Completion Tests and
Human Figure Drawings." In D. Brower and L. Abt
(Eds.), Progress in Clinical Psychology. New
York: Grune and Stratton, 1956.

 

HICHIGQN STQTE UNIV. LIBRQRIES

31293103106104