Q"'V'1

t5 .‘
I? I
H 4

HM
DH!
WI"!

ll

il‘ "! '
g H | p
i‘ W '_
‘E W

l

1
l

1
l
I. l‘

‘ '1} w
‘ 1‘
rs ’
4 m '

‘ 4

1
il
‘

-—‘ I
.p.

 

OVERT AGGRESSION, FANTASY AGGRESSION,
AND AGGRESSEON ANXIETY
N A POPULATION OF YOUTHFUL OFFENDERS

Thesis for fho Degree of M. A.

MICHQGAN STATE UNIVERSH’Y

Ronald Norman Luehrig
T964

SSSSS

 

MllIllullililllWWIWHIIIHUIIHHIUIIIHIIHLIIMIf

3 1293 10525 8572 M

LIBRA

ichigan State
Universu‘y

3:]

A

.‘g‘ru-w .‘n. ,. . if u: .-
in. A .

 

1",... cum... innit;
.,- u q__«.“_ A. ,. .".r

0.35.?3W".
'*- .-'-'-".‘;f§l'f_"y '

3" 5—9—- q. 7": Eng-1:.“¥ P

WW4”),- ea-h- “'55.. ‘— ATA'C‘“? , _

RGOM USE ONLY

 

 

 

ABSTRACT

OVERT AGGRESSION, FANTASY AGGRESSION,
AND AGGRESSION ANXIETY
IN A POPULATION OF YOUTHFUL OFFENDERS

by Ronald Norman Luehrig

This study addresses itself to the nature of
the relationships among overt aggression (the inde-
pendent variable), "aggressive tendencies," aggres-
sion anxiety, and fantasy aggression (the three
dependent variables). The psychological concepts
of aggression, fantasy, and anxiety are discussed,
followed by a review of the literature bearing on
these relationships. Briefly, the literature sugé
gests that aggression (making no distinction between
overt aggression and "aggressive tendencies") and
aggression anxiety are in an inverse linear rela-
tionship to one another, while aggression and fantasy
aggression have a direct linear relationship.

Three hypotheses are proposed based on the em-
pirical findings reviewed. (1.) If a prison popu-
lation is divided into three groups according to the
degree of overt aggression (as indicated by their
present offences), the higher the overt aggression,
the higher the scores on the aggression measure will

be. (2.) Given these three groups, the higher the

overt aggression, the lower the aggression anxiety

scores. (3.) Again, given these three groups, the

higher the overt aggression, the higher the fantasy
aggression scores.

The tests used to measure the three dependent
variables are (1.) the Zaks and Walters twelve item
scale of aggression, (2.) Sears aggression anxiety
scale, and (3.) a combination of two group-admini-
stered versions of the Thematic Apperception Test:
the Iowa Picture Interpretation Test and the Objec-
tively Scoreable Apperception Test. The IPIT and
the OAT are combined by converting to standard scores,
adding a constant, and taking the arithmetic mean of
the two scores to get the fantasy aggression score.
A product moment correlation is computed between the
hostility scale on the IPIT and the aggression scale
on the OAT and is found to be +.599, indicating that
the two scales are related.

Out of an original population of #23 male prison
inmates, 343 who met the criteria of age and minimum
IQ were retained as subjects in this study. They
were divided into three groups--Violent Personal,
Violent Non-personal, and Non—violent--based on the
classification of their crimes by three independent
raters. The three groups are comparable in terms of

IQ and education, but the Violent Personal group is

significantly older than the other two groups.

The data were analyzed by means of a single
classification analysis of variance, and the indi-
vidual comparisons were made by means of three If
tests for independent means. With the exception of
the aggression scale, where the moderate overt ag-
.gression group showed the highest mean "aggressive
tendency" score, all analyses of the dependent vari-
ables across the three groups are insignificant. The
lack of statistically significant results is attri—
buted primarily to the small n in the middle group.
Despite the generally statistically insignificant
results among the three groups, the data did suggest
some interesting relationships between the independent
and dependent variables. In all cases the relationship
is curvilinear. When a Pearson r was computed for
pairs of the dependent measures, two were found to be
significant. ”Aggressive tendencies" and fantasy
aggression were positively correlated, and aggres-
wion anxiety and fantasy aggression were negatively
correlated. There was no correlation between "ag—
gressive tendencies" and aggression anxiety. Specu—
lations based on these trends lead to suggestions
for the modification of the hypotheses and the
techniques of investigation, to be incorporated into

future research. The significant findings are

discussed in light of the literature in this area.

APPROVED BY

 

Amsakvg‘xw

THESIS CHAIRMAN

OVERT AGGRESSION, FANTASY AGGRESSION
AND AGGRESSION ANXIETY

IN A POPULATION OF YOUTHFUL OFFENDERS

By

Ronald Norman Luehrig

A THESIS

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

MASTER OF ARTS

Department of Psychology

196”

ACKNOWLEDGEMENTS

I wish to express my sincere gratitude and
appreciation to by Thesis Chairman, Dr. Lucy Rau, for
her interest, encouragement, and instruction related
to this investigation. In addition, I am grateful to
Dr. John Hurley and Dr. Bertram Karon for their in-
formative and helpful discussions of the issues in-
volved in this research. I extend my gratitude to
Dr. Shelley, Supervisor of Treatment, Department of
Corrections of the State of Michigan, for his interest
and encouragement and, especially, the assistance he
rendered in speeding the proposal through the neces—
sary red-tape. I also extend my gratitude to Mr.
Harrison, Director of the Department of Corrections,
and his staff for their assistance; to Mr. Hanland,
Supervisor of the Michigan Training Unit and his
staff, for facilitating the collection of the data;
and to the three men-~Roger Olive, John Taliafarro,
and Larry Levey--who arose early on Saturday mornings
to proctor the testing sessions. Last, but by no means
least, I would like to express my deepest gratitude
to my wife, Laurel, for the long hours she spent typing
and proofreading this thesis and for her confidence
and encouragement, without which this thesis might
have been impossible.

ii

TABLE OF CONTENTS

HISTORY . . . . . . . . . . . . . . . .
Hypotheses . . . . . . . . . . . .
PROCEDURE . . . . . . . . . . . . . . .
Tests . . . . . . . . . . . . . .
Index of Overt Aggressive Behavior
Data Collection . . . . . . . . .
Classification of Crimes . . . . .
Subjects . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . .
Summary and Conclusions . . . . .
APPENDICES . . . . . . . . . . . . . .

REFEREIQICES O O O O O O O O O O O O O 0

iii

18
21
21
23
21+
25
27
31
35
no
us

68

TABLE

1.

LIST OF TABLES

PAGE

Means and Variances of Age, IQ, and Edu-
cation for the 3u3 Subjects When Divi-
ded Into the Three Experimental Groups 29

The Means and Variances of the Zaks and
Walters Scores Across the Three Ex-
perimental Groups 31

The Means and Variances of the Sears Ag-
gression Anxiety Scores Across the Three
Experimental Groups 32

The Means and Variances of the Fantasy

Aggression Scores Across the Three
Experimental Groups 3”

iv

LIST OF APPENDICES

APPENDIX PAGE

A. The Four Tests Employed in This Study,
Their Instruction, and Their
Scoring Procedures . . . . . . . . UH

B. Classification of Crimes . . . . . . . 55
C. Summary Tables . . . . . . . . . . . . 61
D. Conversion to "Corrected" Standard

Scores . . . . . . . . . . . . . . 65

HISTORY

The literature dealing with the problem of the
relationship between overt and fantasy aggression and
overt aggression and aggression anxiety is not very
definitive, either empirically or theoretically. For
the sake of clarity, it seems necessary to deal with
the three fundamental aspects of this problem sepa-
rately before attempting to deal with the theoretical
and empirical issues of the problem at hand. That is,
the author will discuss briefly the topics of aggres-
sion, fantasy, and anxiety before focusing on the
problems of the relationship between overt and fantasy
aggression, and overt aggression and anxiety over
aggression.

Sears, et al (25) define aggression as "...be—
havior that is intended to hurt or injure someone...."
We will attempt to refine this definition by making
a distinction between personal aggression, i.e., harm
to another person, and object aggression, i.e., be—
havior that results in damage to property public or
private.

Freud, about the time of the first world war,
became astounded by the aggression that man was ex-

hibiting towards his fellow man, and felt it necessary

I

to attempt to integrate this with psychoanalytic
theory. He posited the concept of the death instinct,
relating this with the biological fact of gradual de-
terioration of the organism with age, and indicated
that along with the sexual drives (Eros) of man there
are also aggressive drives, which he called Thanatos.
Yet, psychoanalytic theory as we know it today is
split on this issue. This split is reflected by
Fenichel (9) when he says, "aggressive tendencies of
all kinds constitute a considerable proportion of all
human drives. In part, they manifest a reactive
character; that is, they are the response to frustra-
tion and have as theirgoal the overcoming of frustra-
tion. In part they appear closely connected with
certain sexual drives, especially with sexual drives
that are predominant in pregenital levels of libido
organization. Other aggressions seem to arise quite
apart from sexuality." In this statement, Fenichel
reflects not only Freud's notion of aggression as a
drive, but also the hypothesis set forth by Dollard,
et al, at Yale University in 1939 (7). The basic \
postulate of this hypothesis is that "...aggression
is always a consequence of frustration. More speci-
fically, the proposition is that the occurrence of
aggressive behavior always presupposes the existence
of frustration, and, contrariwise, that the existence

of frustration always leads to some form of aggression."

Bandura and Walters (2), in a large scale study
of adolescent aggression, formulate a more specific,
theoretically precise statement of the antecedents of
aggression. Briefly,

...the theory of anti-social ag-
gression that is offered...assumes

that such a disorder originates
primarily from the disruption of a
child's dependency relationship to

his parents....the frustration of

the child's dependency needs through

a lack of affectional nurturance on
the part of one or both of his parents
provides the child with continuing in-
stigation to hostility and aggression....
An impaired dependency relationship may
not only be a source of aggressive
feelings, but may also limit a child's
capacity to handle such feelings ade-
quately once they are aroused.... A
child who is severely frustrated and
rejected has little or nothing to gain
through controlling the anger and ag-
gression that the frustration provokes.
This aggression at least brings tem—
porary relief, and at the same time,
forces others to attend to him.

It is evident throughout their book that these authors
follow the conceptual framework set up by the Yale ‘
scholars some twenty years before. Yet, Dollard and
his colleagues (7) posit a different set of frustra-
tion-factors as the stimulator of the aggression.

They are the attainment of sexual maturity which is

in conflict with society's taboo against premarital
sexual gratification and the attainment of functional
skills which would allow for economic autonomy but
which is resisted by the society's economic structure.

It is by now quite apparent that these three

theoretical systems have attempted to deal with the
phenomenon of aggression and have basically come to
agree that frustration is the prime factor underlying
aggressive behavior. There is less agreement as to
the specific source of the frustration, but this is
understandable because there may be many sources of
frustration that can lead to aggressive behavior.
These approaches to the problem have suggested only

a few of the sources of frustration.

Regardless of the sources of frustration, it is
reasonable to assume from the above discussion that
there is frustration throughout childhood. This
being the case, how does the individual learn to deal
with this frustration and the resulting aggressive
impulses? ‘Psychoanalytic theory tells us that part
of the developmental process involves the growth of
the ability for the delay of gratification. In non-
psychoanalytic terms this means we attain the ability
to tolerate frustration. But what if a child does
not learn to tolerate frustration? Sears, et a1 (25)
suggest that aggressive behavior is in part influenced
by the social structure of the society in which one
finds himself. These social limitations aid the in-
dividual, no matter how free from frustration his
childhood was, no matter how great a tolerance for
frustration he has developed, to adOpt subtle and

complex expressions of his aggression which is

approved by his society.

The complexity and subtlety of
adult aggression is the end pro-
duct of two or three decades of
socialization...and bears little
resemblance to the primitive
quality of the infant's action
pattern, from which it developed.
All human societies...have rules
to limit the kinds and direction
of aggression that may be expressed.
The most fundamental of these is the
high degree of prohibition on in-
group fighting [especially] in-
family aggression.... Aggression...
has a strong tendency to evoke
counteraggression or punishment...
[this] means that every child grows
up with the experience of being
punished in some degree for his ag-
gressive behavior.... One signifi-
cant effect of punishment is the
production of anxiety [to be dealt
with later]. If the punishment is
repeated many, many times through
early childhood, situations that
provoke aggressive feelings gradu-
ally come to arouse anxiety.... Even-
tually the aggression itself or the
accompanying feeling of being angry
becomes sufficient to arouse anxiety
[guilt].... The formulation of such
a reaction has two kinds of conse—
quences...one is the uneasiness and
discomfort that becomes connected
with the arousal of aggressive im-
pulses...a second...is the develop-
ment, by the child, of techniques for
avoiding punishment. (25)

Bernabeu (3), taking a more eclectic, yet mar-
kedly Freudian, position, on the one hand reinforces
Bandura and Walters' position—~one of the prime factors
in this issue is the disruption of the dependency rela-
tionship—-and on the other hand supplies some important

additional information. (Bernabeu's article deals with

delinquents, but for the present we have attempted to
keep the discussion more general, although the popu-
lation under investigation is youthful offenders.*)

She suggests, among other things, that fantasy-—espe-
cially fantasy of omnipotence--is an important variable
to be considered. Also, she suggests that aggressive
behavior or "...action has become an evasion of re-
ality rather than an adaptation to it, i.e., the
negative alloplastic reaction. External controls

[as described by Sears, et al, above] are not inter—
nalized.... This acting out, since it does not solve
the fundamental problems of the individual, causes a
continuing lack of socialization and therefore of op-
portunity for development of the various ego functions...,"
primarily control of aggressive urges.

Aside from supporting the Bandura and Walters
notion of aggression stemming from a frustrated de-
pendency relationship and supporting Sears' notion of
socialization of aggression, Bernabeu points to a
factor which might be quite relevant to the study of
aggressive behavior, i.e., fantasy.

Fenichel (9) points out that "as long as thinking

*Youthful offenders refers to offenders sent to
a house of correction by a court other than a juvenile
court. Here we are not dealing with delinquents be-

cause this would pose many legal problems in terms of
accessibility of subjects and their records.

 

_is not followed by action it is called fantasy. There
are two kinds of fantasy: creative fantasy, which pre-
pares some later action, and day—dreaming fantasy, the
refuge for wishes that cannot be fulfilled." Freud (11)
does not make the distinction that Fenichel makes, al-
though he does go into greater theoretic detail on the
subject. "The common origin and normal prototype of
all these phantastic creations are the so-called day-
dreams of adolescence.... These phantasies are wish
fulfillments, products of frustration and desire...."'
Freud sees these fantasy experiences--day-dreams--as
concerned almost entirely with erotic pleasures.

Healy and Bronner (13) also deal, even though
only secondarily, with the phenomenon of fantasy (as
did Bernabeu--the fantasy of delinquents). These
authors define fantasy as "...free wandering of thought[s]
untrammeled by considerations of reality. [They] re-
present for all human beings...the unbidden expressions
of feelings and desires that are potent in forming
conduct tendencies...."

It should be noted that both Fenichel, and Healy
and Bronner, use fantasy as thoughts allowed to wander
free to be later used as a guide for future behavior.
Yet, empirically, one would be hard put to define opera-
tionally the two types of fantasy proposed by Fenichel.
For this reason, the most general definition of fantasy

will be employed for the purposes of this paper;

fantasy is defined as those thoughts that occupy the
mind, which may or may not lead to future behavior,
and that are generally not verbalized and not im—
mediately acted upon. These thoughts must also be
guided more by primary process thinking than by
secondary process thinking. This stipulation is
added to distinguish fantasy from problem solving
behavior; i.e., it is an attempt to de-emphasize

the logical processes in the definition of fantasy.

We now begin to get a feeling that man, in his
bio-social wisdom, has "provided" himself with pos-
sible ways of expressing aggressive urges, arising
from frustrating situations, in socially acceptable
behavioral modes. The one which is of interest to
this author is the use of fantasy to handle one's
aggressive urges.

From the above discussion we might expect, since
fantasy is posited as a means of coping with one's
aggressive feelings, that the following relation—
jship holds: the greater the fantasy aggression,
the less the overt aggression. This expectation
is the same as the one that would be posited by
psychoanalytic theory based on the notion of mental
economics; i.e., "There is a 'mental energy exchange,'
an economic distribution of energy at hand between

intake, consumption, and output (Fenichel, 9). Gordon

and Cohn (12) state this theoretical formulation
very clearly. "According to psychoanalytic theory
instinctual energies may be discharged either mo- i
torically or in fantasy, as alternative discharge
systems. Psychoanalytic theory would predict,lthen,
that the presence of a fantasy discharge in a subject
should reduce the overt expression of the drives that
are expressed by fantasy."

The basic premise of the philosophy of science
is to compare all theoretical statements with em—
pirical findings to see if the theory is substantiated
by fact. It is for this very reason that we turn a-
way from theory momentarily to face the findings of
empirical research designed to test the relationship
between overt and fantasy aggression.

At the outset, it must be noted that the avail-
able literature is sometimes difficult to interpret
for a number of reasons: (a) some of the findings
that are reported are merely interesting by-products
of larger and, more important, almost irrelevant
studies; (b) the measures of overt and of fantasy
aggression are many times subjective and based on
different, incomparable instruments; and (c) the
hypotheses forming the foundation of the studies are
so stated that comparisons are virtually impossible
to make. Generally, the literature in this area,

although fascinating and provocative, does not clarify

10

the nature of the relationship between overt and
fantasy aggression. However, a few studies seem to
agree as to the nature of this relationship, and
others suggest different relationships.

Marquis (22), using several TAT scales ("hypo-
theses") and correlating them with ratings of "many
aggressive behaviors," concludes that less is known
about fantasy behavior and aggressive behavior than
is generally realized. Although Marquis' conclusions
seem almost meaningless, they do express one aspect
of this problem—-we really don't know too much about
the relationship between overt and fantasy aggression.
Let us see what has been done to illuminate the situa-
tion.

Healy and Bronner (13) find as a by-product of
their research that there is some evidence for a
positive relation between overt and fantasy behavior.
They feel that fantasy is some sort of rehearsal be—
havior for future action. The evidence for this,
however, is more anecdotal than empirical. Cohn (6),
in a study employing doll play and subjective measures
of both overt and fantasy aggression, concludes that
"Doll play aggression is comparable to overt aggression
in that [the] characteristic of overt aggression is
likely to show up in doll play aggression also, al-
though not necessarily in the same form." Again, a

positive relationship. Sears (2H), with a firm

ll

foundation of empirical findings, begins to make
things a little clearer. In her study a frequency
count of aggressive acts in a doll play situation is
the index of aggressive fantasy. She compares her
findings with the results of surveys on aggressive
behavior in real-life situations and concludes that
"the fantasy component of doll play...does not pro-
vide many reversals of behavior appearing in more
realistic situations"; i.e., there is a positive
relation between the fantasy and overt aggressive
behavior. Kagan (17) also finds a positive rela-
tionship between overt aggression (ratings of children
as to the likelihood of their getting into fights) and
fantasy aggression (fight themes of TAT stories).

In Lasky's study (18) three questions are asked,
only one of which is relevant to the present study;
i.e., what is the relationship between spontaneous
fantasy and behavioral measures of hostility. Lasky
uses reported day-dreams as an index of fantasy
hostility and finds that there is a significant cur-
vilinear relationship between day-dream fantasy and
self-ratings of hostility (P<.01). Those who rated
themselves high or low in hostility had higher day-
dream productivity scores than the intermediate self-
rating group. This study does not really allow us to
draw any definite conclusions because the index of

"fantasy aggression" is unclear: is the day—dream

l2

productivity an adequate measure of fantasy aggres-
sion; what is the content of the day-dreams being
considered? Jensen (15), on the other hand, finds
himself unable to support the hypothesis of a posi-
tive relationship between overt aggression (in high
school students as rated by their teachers) and fan-
tasy aggression (as measured by TAT stories). The
conclusion reached regarding this relationship is:
"There was very little, if any, relationship between
aggression in fantasy and in overt behavior...."
There are other studies that are related to this
issue. Lesser (20), besides finding a positive pro-
duct moment correlation of .26 between fantasy and
overt aggression (P<.05), also finds a negative pro-
duct moment correlation of —.23 between overt aggres-
sion and anxiety which prevents fantasy aggression.
Sears (2”), as we have seen, finds a positive rela-
tionship between two of the principle variables of
Ithis study (overt and fantasy aggression), but also
finds that sex and presence or absence of father in-
fluence aggressive fantasy. Sears, et a1 (25) find
that aggressive fantasy is not a function of parental
permissiveness but a function of parental punitiveness.
Lesser (19) finds another parent—centered variable
which influences aggressive fantasy; "...under con-
ditions of maternal encouragement of aggression, a

greater degree of correspondence exists between fantasy

l3

and overt aggression...than under conditions of ma-
ternal discouragement of aggression."

With the introduction of important secondary
variables the relationship between overt and fantasy
aggression becomes even more complex, yet these vari-
ables give us clearer understanding of the problem.
The basic theoretical issue remains: is psychoana-
lytic theory's formulation of the relationship be-
tween overt and fantasy aggression supported by em—
pirical evidence, or is it necessary to formulate a
new theory to adequately deal with this phenomenon?
The empirical evidence, although somewhat confused,
suggests that the psychoanalytic formulation is in-
appropriate. We feel it is advisable to follow the
lead of the empirical studies quoted and posit that
the fundamental relationship between overt and fan—
tasy aggression is positive and that the psychoana-
lytic formulation is inappropriate.

Before postulating specific hypotheses, we must
consider the third major variable of this investiga—
tion. "Anxiety," which is less clearly defined than
fantasy and aggression in that it covers a wide range
of conceptual meaning, is a rather broad and nebulous
term. May (23), who expounds the existential view of
anxiety, is at one extreme. "Anxiety is not an affect
among other affects such as pleasure or sadness. It

is rather an ontological characteristic of man, rooted

11:

in his very existence as such. It is not a peri-
pheral threat which I can take or leave...it is
always a threat to the foundation, the center of

my existence." May then continues to make the distinc-
tion between anxiety ("anxiety strikes the center
core of his self—esteem and his sense of value as a
self, which is one important aspect of his experience
of himself as a being") and fear ("a threat to the
periphery of his existence; it can be objectivated,
and the person can stand outside and look at it...").
The source of anxiety, according to May, is "at the
point where some emerging potentiality or possibility
faces the individual...but this very potentiality in-
volves the destroying of present security, which
thereupon gives rise to the tendency to deny the new
potentiality." Therefore, anxiety is the inner con-
flict or choice between the fulfillment of one's own
potentiality and one's present state of security.

The fact of man's ability to choose is synonymous,
for May, with anxiety--a central threat to the core
of the self.

At the other extreme is the position which, un-
like the phenomenological position of the existen-
tialists, observes anxiety from a physio-biological
point of view. Cameron and Magaret (5) define an-
xiety as "the predominately covert skeletal and

visceral reaction which constitutes the unconsummated,

15

preliminary phase of emotional excitement. \Like

most covert reactions, anxiety typically remains

unshared, unverbalized, and often inaccessible to
one's own self reactions...." These authors also
make a distinction between anxiety ("preparatory

flight from danger which is not identified, is no
longer present...") and fear ("preparatory flight
from 'real' danger").

Freud (10), unlike May, views anxiety as an
"affective condition.” He also points out, in con-
trast to Cameron and Magaret, that anxiety when
viewed from the position of "academic medicine" (the
physio—biological position) is not very meaningful,
nor informative. For Freud, anxiety is not a single
entity. He concerns himself with two kinds of anxiety--
objective or real anxiety and neurotic anxiety. The
former is related, it is true, to preparatory flight
but it is not nearly as rational and expedient as the
term "preparatory flight" might suggest. For Freud
"...the development of anxiety is never expedient
[nor rational].... The anxious readiness, seems to me
the expedient element...," but this is not to say that
it is derived from the ability to make rational choices,
as May would lead us to believe.

According to Freud, anxiety "is generally under-
stood to mean the subjective condition arising upon

the perception of what is called 'developed' anxiety."

16

The source and prototype of all anxiety, according

to Freud, is the trauma of birth resulting from the
marked increase in stimulation upon leaving the womb
and the individual's inability to deal with this
stimulation. He continues, "...the development of
(real) anxiety is the reaction of the ego to danger
and the signal preparatory to flight [no anxiety as
preparatory flight, as Cameron and Magaret suggest,
but a signal preparatory :2 flight]; it is then not

a great step to imagine that in neurotic anxiety also
the ego is attempting a flight, from the demands of
its libido, and is treating this internal danger as
if it were an external one." When Freud compares
phobias of children with neurotic anxiety he briefly
points out that the libido "substitutes some other
external object or some situation for the love—object
which it misses."

Arieti (1) makes this last statement of Freud's
much clearer, when he points out that the child re-
quires an atmosphere of warmth, love, respect, and
approval in order to develop a sense of self-esteem
and self-identity. He then points out that "At times,
however, the child is not surrounded by this atmos—
phere of love,acceptance and approval. His strivings
toward self-esteem and self-realization are thwarted
by the destructive influences that the surrounding

adults have upon him. His security is attacked and

l7

anxiety originates." It must be emphasized at this
point that, although Arieti uses the concepts of
"self-esteem" and "self—realization," his view of
anxiety is not at all like that set forth by May.

In his discussion of anxiety, Fenichel (9) re-
iterates Freud's views of anxiety, yet places stronger
emphasis on the concept of primary anxiety (hitherto
not mentioned). He also combines the views of the
psychological and physio-biological positions. Again,
as for Freud, the source and prototype of anxiety is
the trauma incurred at birth. Fenichel points out

The sensations of this 'primary
anxiety' can be looked upon partly
as the way in which the tension
makes itself felt and partly as the
perception of involuntary vegetative
emergency discharges...certainly
this primary anxiety is in no way
created actively by the ego; it is
created by external and internal
stimuli, still unmastered, and in-
sofar as it is experienced as a
conscious painful feeling, it is
experienced passively, as something
that occurs to the ego and has to
be endured.

As Fenichel indicates, anxiety results from both
internal and external stimuli. Among the internal
stimulation that might arouse anxiety are the ag-
gressive impulses. In an attempt to understand ag-
gression, Sears (26) has developed a scale (actually
a sub-scale) which he feels can measure anxiety over

aggression. He says that "this scale measured feelings

of fear, discomfort, and dislike of aggression." Sears'

l8

aggression anxiety scale is primarily based on a
somewhat different concept of anxiety. Dollard and
Miller (8), taking a social learning point of view,
discuss "anxiety" in great detail. In their dis-
cussion they tend to use "anxiety" and "fear" inter—
changeably. They do, however, make the traditional
distinction—-fear is a response to an object or is
associated with a specific stimulus, while anxiety
is a response to some vague, non—specific stimulus;
"...we say that fear is learned because it can be
attached to previously neutral cues." The origins
of this social leaning theory of anxiety are in the
avoidance learning experiments with animals. Dollard
and Miller have extended and refined this "avoidance"
into a precise descriptive concept and have applied
it to such phenomena as ”fear of being alone," "the
masturbation taboo," etc. It seems reasonable there-
fore, that Sears' scale might have the ability to
measure the complex phenomenon of anxiety, since his
definition and description of the scale contains some
of the essential elements of anxiety.

Turning to empirical studies that relate the
phenomena of anxiety and aggressive behavior, we
find that their conclusions are congruent with one
another. Lesser (20), as mentioned above, finds a
correlation of -.23 between overt behavior and ag—

gression anxiety. In another study, Lesser (21) finds

19

that religion plays a major role in the anxiety over
aggression, which in turn affects the quality, not
necessarily the quantity, of aggressive fantasy.
Sears (26), in a rather complex study, finds that
antisocial aggression correlates negatively with
aggression anxiety. Projected and self aggression,
however, correlate positively with aggression anxiety.
From the above discussion of the nature of an-
xiety and the empirical studies relating anxiety

with overt aggression, it seems fairly reasonable to

_ ‘ _- .-.__.—~—_._._.. ‘
~—~~.___ - a-.. - —'

expect that when aggression anxiety is high overtly
aggressive behavior would be low and, conversely, when
aggression anxiety is low, overt aggression would be

high.

Hypotheses

 

1 If a population of prison inmates is subdivided
into three sub-populations (Violent Personal, Violent
Non-personal, and Non-violent) on the basis of the
nature of their offences, then the Zaks and Walters (28)
aggression scale will differentiate these three groups
with the Violent Personal having the highest score,

the Violent Non-personal having the next highest score,
and the Non—violent having the lowest score, i.e., a
direct relationship exists between measured and overt
aggression.

2. Given these three groups, the Sears (26) aggres—

20

sion anxiety scale will differentiate these three
groups as follows: Violent Personal, low aggression
anxiety; Violent Non—personal, higher aggression
anxiety; and Non-violent, highest aggression anxiety,
i.e., an inverse relationship exists between aggression
anxiety and overt aggression.

3. Again, given these three groups, the fantasy
aggression score will differentiate these three

groups in the same manner as in Hypothesis 1, i.e.,

a direct relationship exists between overt and

fantasy aggression.

PROCEDURE

$3.555».

Three tests were selected to obtain the measure-
ments of the two dependent variables, aggressive fan-
tasy and anxiety over aggression. A fourth test,
measuring aggression ("aggressive tendencies") was
also included.

(a) Aggression The Zaks and Walters (28) twelve

 

item scale of aggression was included in the study to
measure "aggressive tendencies" or aggressive impulses.

(b) Aggression anxiety_ Sears' (26) scale of ag-

 

gression anxiety was chosen because of his successful
use of the scale with eleven and twelve year old chil-
dren. Since the subjects under investigation were
primarily still in their teens, it was felt that the
scale would still be appropriate.

(c) Fantasygaggression As was noted above, the

 

majority of the studies concerned with the relation-
ship between overt and fantasy aggression used one

of two measures of the dependent variable. When young
children were studied, doll play was used as the
measure of fantasy aggression. On the other hand,
when older subjects were studied, TAT stories were

used to measure this variable. Two group tests

21

22

derived from the TAT and employing TAT cards were used
in this study. This group technique was decided upon
for two reasons. First, since no unequivocal conclu-
wion could be drawn from the studies using small
samples, it was felt that by employing a large sample
such factors as sampling error and variance would be
reduced. This in turn would permit some conclusive
statement concerning the relationship between overt
and fantasy aggression. Second, these group measures
provide objective scoring criteria for the measurement
of fantasy aggression. But why use two tests for the
measurement of aggressive fantasy?

Three versions of group TAT's are available.
The first one, the Iowa Picture Interpretation Test
(IPIT), was constructed by Hurley (14). Two years
later, in 1957, Johnston (16) came out with an "im-
proved" version of this test, but did not make it
very clear what these improvements were. In 1962,
Stricker (27) brought forth what he considered to be
a far more reliable, and even more valid, test than
the original. He called this test the Objectively
Scoreable Apperception Test (OAT). In the same year
Buss, et a1 (H) used Hurley's original version of the
IPIT in a study concerned with aggression in psychia—
tric patients. Aside from the main results of the
study, these authors found a significant correlation

between the IPIT and two other measures of aggression.

23

For the men there was a correlation of .60 between
the IPIT and the scrambled sentences test and a cor—
relation of .66 between the IPIT and the operant level
test. This independent study by Buss and his associ-
ates gives some evidence as to the validity of the
aggression scale on Hurley's original form of the
test. Since this was the case, it seemed advisable
to investigate the validity of the aggression scale
in Stricker's test and, at the same time, increase
the sensitivity of our measure by making more items
available. (The tests, their instructions, and their

scoring procedures can be found in Appendix A.)

Index of overt aggressive behavior

For the purposes of this study the independent
variable, overt aggressive behavior, is defined in
terms of the present offence (i.e., the offence which
lead to the subject's present prison sentence) and the
rating of three independent judges (see below). The
judges classified the offences into Violent Personal,
Violent Non—personal, and Non-violent, with the first
being equated to high overt aggressive behavior and
the last being equated to low overt aggressive be-
havior. These equalities follow the assumptions found
in the literature.

In order to obtain some support for this assump-

tion, i.e., that as we go from Non-violent crimes

2%

through Violent Non-personal crimes to Violent Personal
crimes, we are in fact following an increasing degree
of overt aggression, five independent judges ranked
the three classes of crimes as to which is most ag-
gressive in terms of overt behavior. The five raters
included one ex-state police captain, two psycholo—
gists, one hospital supervisor, and one attendant.

All five are employed at a state hospital treating

the criminally "insane." All five rated Violent
Personal as the most aggressive classification in
terms of overt aggressive behavior, and all five rated
Non—violent as the least overtly aggressive classifi—
cation. These independent raters support this assump-

tion which is the key to the present study.

Data collection

 

There were two phases involved in the collec-
tion of the data, both occurring during the same time
period. First, all the inmates of the Michigan
Training Unit were tested in four groups. Each
group consisted of the approximately 100—125 men in
each living unit. Two groups were tested in the first
session, and the remaining two groups were tested one
week later, in a second session. During the week in-
tervening, the case history data was collected. The
testing took place in an auditorium which was semi-
darkened. This was to insure that the projected

image of the TAT card could be seen and that there

25

would be enough light for the subjects to read the
printed material in the test booklets. The tests
were presented in the order appearing in Appendix A.
This was to permit the changing of the TAT slides,
without interrupting the on—going testing, from the
IPIT order to the OAT order. This was done while
the subjects were engaged with the Zaks and Walters,
and Sears scales.

Prior to the actual administration of the
tests, a master list of inmates according to living
units and rooms ("locks") was obtained. This list
included the inmates' names and prison numbers.
These numbers were then placed on the test booklets
following the order on the master list. At the time
of testing the inmates entered the auditorium in
single file, each according to his living unit (only
one unit tested at a time) and lock. As each man
entered, his number was called and he responded with
his name. This was to make sure that the right man
received the right test booklet. This procedure was
followed to avoid bias, in that the men had not been
classified into the experimental groups prior to
testing. Another step was taken to assure that bias
would not enter into the study; another list was set
up in the sequence of the inmates' prison numbers.
This list was used to obtain the relevant information——

present offence, date of birth, marital status, race,

26

education, religion, and IQ--from their records. A
third step was employed to avoid bias: the tests
were scored prior to the classification of the crimes
into the three experimental conditions (see below).
Only immediately prior to statistical analysis were
the three types of information--case history data,
test scores, and experimental group-—brought together
and recorded on 5 x 8 index cards, one card for each

subject.

Classification of crimes

 

As originally planned, graduate students clas-
sified the crimes into the three experimental groups--
Violent Personal, Violent Non-personal, and Non-violent.
(For the questionnaire used and the resulting classi-
fications see Appendix B.) When the degree of agree—
ment among the graduate student judges was measured
by means of the proportion of crimes meeting the
criteria of agreement (two out of three judges agreeing
on a classification), it was discovered that only
80.95% of the crimes met this criteria. This 80.95%
is just barely statistically beyond the cance level
of agreement. Of the #2 different crimes classified
only u5.2u% were fully agreed upon by all three judges,
while for 19.05% there was no agreement at all. These
crimes and the subjects associated with them would

have had to been dropped from the study. In order to

27

avoid losing more subjects and, even more important,
in order to avoid the possibility of a type one error,
it was felt that more experienced judges should be
employed. Therefore, three professionals from the
Michigan Department of Corrections classified the
crimes following the same instructions that were

given to the graduate students. When the proportion
of agreement meeting the criteria was computed, 100%
of the crimes were agreed upon by two out of the three
judges, while 76.19% of the crimes had total agreement
(as compared to HS.2U% for the graduate student raters).
Statistically this is significantly greater than
chance well beyond the .001 level (Z=u.58). Due to
the increased agreement among raters, the classifica-
tion provided by the professionals was chosen over

the classification derived from the graduate students.
(See Appendix B for the actual classification and the
frequency of crimes in the population under investi-

gation.)

Subjects

With the exception of five men, the entire in-
mate population (u28 men) of the Michigan Training
Unit at Ionia, Michigan, was tested. Three of these
five were involved in their duties at the unit and
were unable to come for testing; the other two had

visitors from out of town. Of the #23 men who were

28

tested, 51 were lost due to (a) their refusal to take
the test; (b) their inability to follow the instruction
on the IPIT and the OAT, despite the fact that there
were two proctors and one examiner present at all
times during the testing; and (c) minor omissions and
errors encountered on the test forms, such as the
omission of the response to one item or the use of
tied ranks when the instruction specifically indi-
cated that tied ranks were not to be used (see
Appendix A). An additional 29 were lost because

they did not meet the age and IQ requirements esta-
blished prior to the testing. The age requirement was
that the subject not be older than 29 years old. The
IQ requirement was that the subject have at least an
85 IQ. Ten were lost because they were 30 years old
or older, and nineteen were lost because they had an
IQ of under 85. Thus, there was a total of 80 sub—
jects lost who had taken the test leaving a total of
3M3 male subjects. The mean age of all 3H3 subjects
at the time of testing was 19.76 years; and the mean
IQ was 107.19, based on the Army General Classifica-
tion Test.

Before the placement of subjects into experi-
mental groups is discussed, a word about the Training
Unit is necessary. Four factors are involved in
sending a man to the Training Unit: (a) he has to be

young (except when a special skill is needed at the

Q

29

Unit); (b) he has to have had at least one prior
conviction; (c) he must have at least the equivalent
of an eighth grade education; and (d) in conjunction
with or instead of (c), he must be motivated to im—
prove himself. Also, it must be pointed out that
the Michigan Training Unit is not a maximum security
installation.

When the subjects were subdivided into experi-
mental groups, there was a significant difference in
age, but not in education or IQ. (See Table 1. For
the summary tables of the analysis of variance see

Appendix C.)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 1. Means and Variances of Age, IQ, and Education

for the 3H3 Subjects When Divided Into the

Three Experimental Groups

Violent Violent

_Personal Nonfpersonal Non-violent

X 82 X 82 X s2 pp
Age 20.91 6.99 19.00 3.20 19.92 6.83 .01
Educ. 9.60 1.92 8.67 0.67 9.53 1.77 N.S.
IQ 107.86 128.36 101.67 239.87 107.0” 125.H3 N.S.
N 81* 6 156
When individual comparisons were made (see Appendix C)
with respect to the age variable, the Violent Personal

group was significantly older than the Violent Non-

*Except for Education, where there are 80 cases
because one subject's education was not recorded.

30

personal group by 1.91 years (at the .05 level) and
older than the Non—violent group by 1.99 years (at the
.01 level). Unfortunately, the Violent Non—personal
or the moderate overtly aggressive group (see above)
had only six cases. This small group was retained
because there is no reasonable justification for
combining this group with either of the other two
groups since (a) the degree of overt aggression is
different, according to the five independent judges,
and (b) the object of the aggression is different for
this group than for the other two. Thus, the Violent

Non-personal group was left intact.

RESULTS

The data were analysed by means of single classi-
fication analysis of variance. Individual comparisons
were made, when a significant F was obtained, by means
of three separate t—tests for independent means. The
smaller degrees of freedom was used when entering the
3 tables in order to counter-balance the effects of

serial t—tests.

Aggression

 

When the Zaks and Walters scores were subjected
to analysis, the over-all F was significant at the .05
level (see Appendix C). The means and variances for
the three groups are given in Table 2.

TABLE 2. The Means and Variances of the Zaks and
Walters Scores Across the Three Experimental

 

 

 

Groups

Violent Violent

Personal Non—personal Non-violent
Mean 5.04 6.83 5.39
s2 3.88 8.97 3.18

 

When the aggression data were analysed by means of
individual comparisons, the Violent Personal group was
significantly lower on the aggression measure than the
Violent Non-personal. This was the only statistically

31

32

significant difference found (see Appendix C). As

'can be seen in Table 2, the scores in fact tend toward

a curvilinear relationship across the three groups.
Where there is high and low overt aggressive behavior
there is low measured aggression, and where the overt
aggression falls in the middle range the measured aggres—
sion is highest. This is assuming, of course, that the
order of overt aggression from least to most (Non-
violent to Violent Personal) as set forth by the five
independent judges is in fact correct. This ordering

of overt behavior must also be kept in mind when reading

the next two sub-sections.

Aggression anxiety

An analysis of the Sears aggression anxiety scores
(see Appendix C) showed no statistically significant
differences. The meansand variances for the three
groups are given in Table 3.
TABLE 3. The Means and Variances of the Sears Aggres—

sion Anxiety Scores Across the Three Experi—
mental Groups

 

 

Violent Violent

Personal Non—personal Non-violent
Mean 37.58 32.17 36.55
s2 u5.71 uu.56 37.18

 

Despite the statistically insignificant results, the

trend suggests that a curvilinear relationship exists

33

between overt aggression and aggression anxiety. More
specifically, when overt aggression is high and low the
aggression anxiety is high, while the middle range of

overt aggressiveness has a lower level of anxiety over

aggression.

Fantasy aggression

 

Before the data could be analysed, it was neces-
sary to combine the scores from the two fantasy aggres—
sion measures. The raw scores on the IPIT and OAT had
to be converted to standard scores, in order to be
sure that the two sets of scores were comparable. (See
Appendix D.) To facilitate computation all values were
multiplied by 100 to eliminate decimal points in the
new scale. For the purposes of this study this new
value will be called the corrected standard score.

A product moment correlation between the two sets
of standard scores was computed. The resulting corre-
lation was +.599. (This is a slight under-estimation
due to rounding error.) This correlation value is sig—
nificant beyond the .01 level. Thus, the IPIT and OAT
scores could be combined into one fantasy aggression
score. This was achieved by taking the mean value of
the corrected standard score for each test. When
this fantasy aggression score was analysed no statisti—
cally significant results were obtained (see Appendix C).

The means and variances for the three groups are given

38

in Table H.

TABLE H. The Means and Variances of the Fantasy Ag-
Agression Scores Across the Three Experi-
mental Groups

 

 

 

Violent Violent

Personal Non:personal Non-violent
Mean ”12.79 351.17 397.85
s2 6909.89 339.57 888o.u7

 

Again, as with the aggression anxiety scores, no
statistically significant differences were found. Yet
the relationship is interesting and suggestive. Before
looking at the nature of this relationship, we must
understand the nature of the measure more precisely.
Unlike the other measures used in this study, where a
high numerical score means a high variable score, the
IPIT and the OAT variable scores have an inverse rela—
tionship to the numerical scores obtained. Thus, the
higher the numerical score the lower the fantasy ag—
gression score, and the lower the numerical score the
higher the fantasy aggression score.

With this in mind, we note that, as in the case of
the aggression scores, there is a tendency towards a
curvilinear relationship between overt and fantasy ag-
gression. That is, when overt aggressive behavior is
high and low, fantasy aggression is low; and where overt
aggression is in the middle range, fantasy aggression

is high.

35

The relationships among the dependent measures

 

When product moment correlations were computed for
the relationships among the three dependent variables,
it was found that two Pearson r's were significant and
one was not. When aggressive tendencies and aggression
anxiety were correlated, Pearson's r was found to be
+.06, which is not significant. 'Aggressive tendencies"
and fantasy aggression correlated -.32 (significant at
the .01 level). The correlation between aggression
anxiety and fantasy aggression was +.3l (significant at
the .01 level). But these relationships are not what
they seem because of the scoring of the fantasy aggres—
sion measures. As was mentioned above, a high numerical
score is actually a low fantasy aggression score and
vice versa. In other words, there is a direct rela-
tionship between'aggressive tendencies" (or impulses)
and fantasy aggression, while there is an inverse rela-
tionship between anxiety over aggression and aggressive

fantasy.

DISCUSSION

We can now ask two questions: (a) what inter-
pretations can we make from these findings concerning
aggression, and (b) how do the findings relate to
what is already known?

Two brief tangential comments are necessary be—
fore answering these questions. First, it was seen
that a statistically significant difference of age
exists between the Violent Personal groups and the
other two groups; i.e., they are older than the other
two groups by 1.91 and 1.H9 years respectively. Al—
though this age difference is statistically signifir
cant, the psychological significance is questionable
due to the small magnitude of the mean differences.
Second, although it may be superfluous to explain in—
significant results, a brief word is relevant here:
it seems reasonable that the insignifibant results can
be in part explained by the very small n in the Violent
Non-personal group. With an n this small a large mean
difference would be needed for statistical significance.
In terms of future research it would seem appropriate
that close attention be paid to sample size, i.e., an
attempt should be made to keep the groups relatively
equal in size.

36

37

Despite the small sample size in the moderate
overt aggression group, the Zaks and Walters aggres-
sion measure (28) ("aggressive tendencies") does show
a statistically significant difference in favor of this
group (see Table 2). This finding is contrary to the
hypothesis of the study. Nevertheless, it is congruent
with subjective impressions based on the author's
clinical experience. A therapygroup of six male mur-
derers seems far less overtly aggressive than other
groups of men committed for less overtly aggressive
behavior, also conducted by the author. In combining
the clinical experience and the statistical evidence
that the most aggressive group (Violent Personal) in
terms of overt behavior has lower measured"aggressive
tendencies" than a group rated moderately aggressive
(Violent Non-personal) in terms of overt behavior, we
find a novel curvilinear relationship between overt
aggression and aggressive impulses. The generality of
this relationship might well be questioned in terms of
the small n for the middle group. It is, however, an
interesting finding and further research seems in
order to confirm both the clinical experience and the
statistical findings of this study. Such research would
determine whether this curvilinear relationship between
overt aggression and'aggressive tendencies" (impulses)
is in fact descriptive of the relationship. If, after

further research is completed and this relationship is

38

confirmed, an explanation will be necessary, and we
would like to speculate at this point as to what it
might be.

In this study we find a tendency toward a cur—
vilinear relationship between overt aggression, based
on the subject's present offence, and aggression
anxiety. This relationship is opposite in direction
to that found for overt aggression and "aggressive
tendencies." When we look at Table 3, we find that
anxiety over aggression tends to be highest for the
most overtly aggressive group (Violent Personal) and
lowest for the moderately overtly aggressive group
(Violent Non—personal) and high, but not as high as
for the Violent Personal group, for the least overtly
aggressive group (Non—violent). Thus we might specu-
late that high anxiety is a factor which lowers the
measured "aggressive tendencies." Therefore, it seems
reasonable to speculate that anxiety over aggression.
is the key to the relationship between overt aggression
and "aggressive tendencies."

How do the other variables of this investigation
tie in with these speculations? Fantasy aggression
has the same relationship to overt aggression as "ag-
gressive tendencies." This is confirmed by a statisti-
cally significant direct relationship (product moment
correlation) between these two dependent variables (see

Results). This tends to support the notion that fantasy,

39

more specifically fantasy aggression, is a form of
behavior which is preparatory for future action
rather than a mode of control over aggression, as
suggested by Healy and Bronner (13) and Fenichel (9).
That is the fantasy takes the impulses and derives a
"plan of action." But since, as we have speculated,
the impulses are monitored by anxiety when there is
high aggression anxiety, the fantasy aggression should
be lower because the aggressive impulses are lower.
The correlation between aggression anxiety and fantasy
aggression confirms this line of reasoning, since
there is an inverse relationship between these two
variables (aggression anxiety and fantasy aggression).
Turning now to the second question, how do the
findings relate to what is already known, we see that
a number of new hypotheses are suggested by this
research. First, the literature, although equivocal,
tends towards a direct relationship between aggression
and fantasy aggression. But, it makes little or no
distinction between overt aggression and "aggressive
tendencies" This study points out that there is a
direct relationship between aggressive impulses and
fantasy aggression, while there tends to be a curvi—
linear relationship between overt aggression and fan—
tasy aggression (see Table M). This distinction be-
tween overt aggression and aggressive impulses is not

clearly made in the literature. Second, the literature

#0

points out that there is a negative relationship
between aggression and aggression anxiety. Again,
there is no distinction made between overt aggression
and "aggressive tendencies.” In this study there was
no relationship found between "aggressive tendencies"
and aggression anxiety and there was only a tendency
toward a curvilinear relationship between overt aggres-
sion and aggression anxiety.

Basically, then, this study, by making a distinc—
tion between overt aggression and "aggressive tendencies"
(impulses), raises the possibility of novel and hope—
fully meaningful relationships. These relationships
are only speculative and require confirmation in

future research.

Summary and conclusions

 

In short, we see that there is the possibility
of a complex dynamic interrelationship among the three
dependent experimental variables and the independent
variable. For the sake of clarity, conclusions con—
cerning each of these will be stated separately.

Aggression The aggressive impulses tend to get

 

strong as the overt aggression increases, but their
measured value first increases and then decreases.
The decrease may be attributed to the concept of
anxiety over aggression.

Aggression anxiety Anxiety over aggression tends

 

41

to be high in both the high and low overt aggression
groups and low in the middle range of overt aggression.
This tends to indicate, as seen in the body of the
discussion, that anxiety over aggression is one vari-
able that might control aggressive impulses. The
suggested relationship found in this study does not
conform to the relationship discussed in the literature,
i.e., it is not an inverse linear relationship.

Fantasy aggression The suggested relationship

 

found in this study again does not conform to the
direct linear relationship predominately found in the
literature. One study by Lasky (18) found a curvi-
linear relationship between day-dream fantasy and
self-ratings of hostility. He found that when the
hostility variable was high and low the fantasy variable
was high. This is just the opposite of the trends
found in this study.

The parallel between fantasy aggression and ag-
’gression ("aggressive tendencies") suggests that fantasy
is a form of preparation for future action rather than

a mode of control over aggressive impulses.

Implications for further research

 

The discussion of the trends found in the data
suggests hypotheses for future research, although the
significance of the results is limited by failure to
control the size of the groups involved in the study.

This study suggests that a distinction must be made

#2

between overt aggression and aggressive impulses.
Finally, the speculations derived from the results
suggest that the relationships among the four vari-
ables are not simple and clear cut and need more pre-

cise investigation.

APPENDICES

”3

APPENDIX A

The Four Tests Employed in This Study,
Their Instructions,

And Their Scoring Procedures

nu

IOWA PICTURE INTERPRETATION TEST

(Hurley)

INSTRUCTIONS: You will be shown a number of pictures,
each indicated by an arabic numeral (number). On the
paper before you, you will find four (H) descriptions
that are rather frequently given for that particular
picture. Rank the four alternatives in terms of their
correspondence with YOUR idea of what the picture ex-
presses. Make you rankings in the space provided at
the left of each alternative.

Find the interpretation that you would most likely
give. Mark it 1 at the space to the left. Then find the
one that seems Hext most likely. Mark it 2. And so on.
The idea that you would be least likely to_give should
be marked 9.

Here is an example:

 

A. She is listening to her favorite radio program.
(8GF)__—B. She is annoyed because she has to work while
_—_” 'hér'friends go out.

C. She feels that she cannot go to the party be—
cause no one ever asks her to dance.
___D. She is looking forward to her Opening night as
the star of a great show.

If choice B is most like you own interpretation,
you would mark it I. The next most likely interpreta—
tion would be 2. And so on.

Each picture will be shown for one minute. You must
rank each choice. Even if you have difficulty deciding
what the ranks should be, make the best decision you can.
Remember, there are no right or wrong answers. All of
these interpretations are possible and are sometimes
given. Don't spend too much time trying to decide. In—
dicate your first impressions.

:takkztickfikkkékkkkfik'kkfikieé**********:’:****:‘n‘c:’:*3’n’::'::'¢s'¢*:'c:’::'::'::’:*:'cz’:*

1.AI A. He is dreaming of the day when he will become a
great musician.

. He is afraid that he will never be able to play

the violin well.

His violin is on the table and he is waiting for

his music lesson.

He is angry at his mother because she makes him

practice while he'd rather be outside playing.

(1)

l=> l7 M
OED

U

45

#6

She resents the fact that she must do more than
her share of helping with the farm work.

She is looking for a nice quiet place where she
can read and get a little relaxation.

She is rather sad because she feels that she
doesn't fit in at school or on the farm.

She is determined to get an education so that

she can help improve the condition of her people.

He very much wants to stay with her but is a-
fraid of other people's contempt.
He is determined to fight for what he thinks is
right and will win in the end.
He is disgusted with her and trying to get away
as quickly as he can.

He is a patient being helped to his bed.

He is going to look for another room because
her boarding house is full.

He is telling her that she must enter an old-
age home because he refuses to support her any
longer.

He is telling her that he must leave home be-
cause opportunities are greater in the big city.
He is telling her that he has just lost his job
and has little hope of finding another.

The boy is determined to live up to the ideals
and standards of this older man, whom he greatly
admires.

The older man is telling about his childhood
experiences.

The father is telling his son that if he does
not stop his wild ways he will disown him.

The boy is distressed because he feels that he
has let his father down.

The little girl has been left in the care of a
governess and feels that she is less loved by
her parents than other children.

The little girl is resentful because her mother
insists upon drilling her over her homework.
The little girl is listening to a story about
Florence Nightingale and is thinking of the
time when she might achieve so much.

The little girl listens while her mother reads
her stories.

u?

7. B A. He is in a gymnasium dressing room watching the
trainer prepare the players for practice.'

B. He is dreaming of becoming a skilled and famous
surgeon.

C. He would like to go hunting with the others but
is considering the dangers involved.

_A_D. He has shot his cruel stepfather who has mis-

treated and abused him.

father, who won't let him leave the cabin.
. He is wondering why he is unpopular and no one
comes over to play with him.
C. He is just enjoying the warmth of the early spring
sunshine.
AI_D. He is thinking of the great accomplishments of
Abe Lincoln, who was also born in a log cabin.

A. He is thinking of ways of getting back at his
B

9. I A. Things have not been working out for him and he
is wondering if life is worth living.
. He is enjoying the night air just before retiring.
. This is a kidnapper breaking into a child's bed-
room under the cover of darkness.
AI_D. He is looking out into the night dreaming of
great accomplishments.

10. B A He is demonstrating the way to climb a rope.

(178M)

_A_B. He is stealing up to attack a hated enemy.

5::C. He is in a rope climbing contest and is exerting
every effort to win.

_I_D. Although he has tried his best, he sees that the
race is lost.

CODE:
AI = Achievement Imagery B = Blandness
I = Insecurity A = Aggression (Hostility)

SCORING: sum of the ranks given for each item for each

of the four scales.

ZAKS AND WALTERS 12 ITEM SCALE OF AGGRESSION

INSTRUCTIONS: To the right of each item you will find
a bracket [ J, one marked "Agree," the other marked
"Disagree." Please read each of the statements below
and place an "X" in the bracket marked "Agree" if you
agree with the statement, or place an "X" in the bracket
marked "Disagree" if you disagree with the statement.
Please DO NOT put your name anywhere on this sheet,
because we want you to feel free to agree or disagree
as you see fit. Please respond to all items honestly
and openly, since we are interested in your true opinion.
Thank you in advance for your cooperation and
assistance.

AGREE DISAGREE_
[Score 1] [Score 0]

1. There are two kinds of people
in this world: the weak and the
strong. ( ) ( )

2. Dealings with policemen and
government officials are always
unpleasant. ( ) ( )

3. Most people get killed in ac-
cidents because of their own
reckless driving. ( ) ( )

H. Horses that don't pull should
be beaten or kicked. ( ) ( )

5. At times we enjoy being hurt by
those we love. ( ) ( )

6. Many a decent fellow becomes a
crook or a criminal because he can't
stand to be pushed around so much. ( ) ( )

7. I easily lose patience with
people. ( ) ( )

8. I often do things which I regret
afterwards. ( ) ( )

9. It makes me mad when I can't do
things for myself the way I like to. ( ) ( )

u8

H9

10. Occasionally I was in trouble
with the police or law.

11. I almost never dare to express
anger toward people for fear I may
lose their love or approval.

12. As an adolescent (or young
kid) I often mixed with the
wrong crowd.

TOTAL

(Aggression Score)

AGREE

DISAGREE

SEARS AGGRESSION ANXIETY SCALE

INSTRUCTIONS: Here are a number of statements. In
front of each one there are 5 brackets ( ). Read

each statement carefully and then put an "X" in one

of the 5 brackets to show how you feel about that
statement. If you agree strongly, put your "X" in

the left-hand bracket. If you disagree strongly, put
your "X" in the right—hand bracket. If you are pg:
sure how you feel, mark the middle bracket. (The words
at the top of the columns of brackets show what each
box means.)

 

 

z m m >xo

H 51-! (D r-IQ)

w a n cos

c m a) a) w arm

0 m m 8 0rd

fich Q P m Eco

+450 m o a L",

U3< ‘< 2 C1 UJQ
( ) ( ) ( ) ( ) ( ) 1. It makes me uncomfortable

to see two of my friends
fighting.

( ) ( ) ( ) ( ) ( ) 2. If someone gets hurt
in an auto accident, I
usually try to get a good
view of what happened.

( ) ( ) ( ) ( ) ( ) 3. It upsets me to think
that some thoughtless word
or crack of mine might hurt
someone's feelings.

( ) ( ) ( ) ( ) ( ) 9. When I get angry, I us-
ually feel bad afterwards.

( ) ( ) ( ) ( ) ( ) 5. It makes me nervous to
hear a gang of boys getting
argumentative even when I'm
not involved.

( ) ( ) ( ) ( ) ( ) 6. When I get too angry,
I'm liable to get in trouble.

( ) ( ) ( ) ( ) ( ) 7. There are too many vicious
crimes described in the
newspapers.

( ) ( ) ( ) ( ) ( ) 8. I hate to hear people
at a baseball game yelling
"Kill the umpire!"

50

Strongly

VAgree

( )

( )

( )

SCORING:

Agree
Not Sure
Disagree
Strongly
VDisagree

A

9. I usually don't show it
when I get angry, but it
leaves me shaking inside
afterward.

10. Teachers should be
very careful never to let
a class discussion get too
heated or too personal.

11. I like to watch a real
man-sized slugging match
in a movie or on TV.

12. Arguing nearly always
leads to trouble in one
way or another.

When A is indicated at the left, score

"Strongly Agree" 5 and in descending order to

"Strongly Disagree" 1.

When D is indicated, score "

"Strongly Disagree" 5 to "Strongly Agree" 1.

(l)

(2)

(5)

OBJECTIVELY SCOREABLE APPERCEPTION TEST

(Stricker)

INSTRUCTIONS: For each picture that is flashed on the
screen, you will be given five alternative choices. Each
of these choices represents a capsule plot or brief
summary of a story which can be told about the picture.
You will be asked to rank—order these choices, according
to which summary ygg feel is most appropriate for the
picture. Thus, what you are being asked to do is to
place the number 1 next to that option which contains
what ygg feel to be the most appropriate story summary,
the number 2 next to the second most appropriate choice,
and so on, until placing the number 5 next to the least
appropriate choice.

Remember to place one, and only one, number next
to each choice, so that when you are through with each
set of choices the numbers 1-5 will have been assigned
to them, according to their appropriateness in telling
a story about the picture.

1. A The boy is angry at his parents because he would
rather play outside.

_R_ The boy spends many hours alone, as his parents
have little time for him.
_I_ The boy is worrying about whether or not he
passed a test in school that day.
_2_ The boy can only play the violin when the
teacher encourages him.
_N_ The boy is waiting for his violin teacher to arrive.
2._A__She resents the fact that she must do more than
her share of helping with the farmwork.
I She is worried that her homemade clothing is not
——_ as good as her schoolmates'.
_R__Her schoolmates don't invite her to parties be-
cause she is a poor farm-girl.
_N_ She decided to walk to school because it's a nive
day.
D She doesn't like going to school because she has
——— to leave her family.
3. D She is seeking her husband's sympathy after an
——— argument with a neighbor.
N She is going to tidy up the room.
_R— Her husband doesn't like her to be in the room
‘T_ when he's discussing business.
A She is coming into the room to scold her children.
_I: She has always been afraid to be at home alone.

52

(68M)

(6GP)

(78M)

(7GP)

(9GP)

HZIW |Ul>

I” la lip Hz

Ht IH I8

l2 l> lH In l3

l7 l2 |._. l> lb

53

The mother is angry at her son for disobeying her.
Whenever he has a decision to make, he asks his
mother's help.

Her husband has deserted the woman and her son.
She has been visiting with her son for a while.

He is not sure he has the ability to meet his new
responsibility.

He is introducing himself to her.

Her sophisticated manner has him confused and
unsure of himself.

He is blaming her for neglecting some work at
the office.

He wishes that she were as affectionate as his
mother always was to him.

She would like him to ask her out, but he never
has.

The son cannot do anything without first asking
his father.

The young man is not sure how to ask his boss for
a raise.

They are having an argument.

The father refuses to help his son.

The two men are having a discussion about working
conditions.

The little girl is interested in learning so that
her parents will praise her for it.

The mother doesn't seem to care what the little
girl wants to do.

The little girl is afraid that she'll never
understand her lessons.

The little girl wants to play, but her mother is
punishing her and making her stay home.

Mother and daughter are spending a rainy day
together.

One woman has been a little upset, and is running
to the other for comfort and sympathy.

One of the women has deliberately frightened the
other.

One of the women went to a dance, but she was so
uncomfortable that she is running away.

The two women stOpped at the beach on their way
home from work.

One woman is hiding from the other because she
doesn't want anything to do with her.

(19)

(20)

(138)

(8GP)

5M

9. 2_ He would like to go to college but he's afraid
. to leave home.
_3_ He doesn't understand why his schoolmates won't
associate with him.
N He is closing the window before he goes to sleep.
:1: He can't sleep because he's afraid he won't do
well on his tests tommorrow.
_A_ He is mad at his parents for not allowing him
more freedom.
10._I_ The man is afraid that he will not succeed on
his new job.
D The man is new in the city, and is wishing that
—:_ he was back with his parents.
_A_ The man has just been criticized for showing poor
judgment in his dealings.
_R_ The man's wife has left him, and he now feels
all alone.
_N_ The man is standing by the lamppost waiting for
a taxicab.
11. D The little boy doesn't like going to school be—
_—— cause he has to leave his family.
_A_ The little boy is angry at his parents for
punishing him.
_N_ The little boy is waiting for a friend to call
for him.
_R_ None of the other children will play with the
little boy.
_I_ The little boy got a bad report card, and he's
afraid to show it to his parents.
12. N She is waiting for a friend to arrive.
:3: Her husband always leaves her at home alone and
goes out to play golf.
I She is worried that her wardrobe is inadequate
——_ for her husband's social position.
D She is daydreaming about her childhood, when
——_ she had no responsibilities to worry her.
A She is angrily waiting for her husband, who is
__- late coming home.
CODE:
A = Aggression D = Dependence
R = Rejection N = Neutral
I = Insecurity

SCORING: sum the ranks given for each item for each

of the four scales.

APPENDIX B

Classification of Crimes

55

CRIME CLASSIFICATION QUESTIONNAIRE

INSTRUCTIONS: Below you will find a list of crime
titles recently taken from the records of prisoners in
one of Michigan's prisons. You are being asked to
classify these crimes, to the best of your ability and
judgment, in accordance with the following categories:
Violent Personal, Violent Non-personal (or Violent
Object), Non-violent, and Unclassifiable. Try to
avoid using the last category (Unclassifiable) as much
as possible. The following brief definitions will aid
you in your classification:

Violent Personal — those offences that do direct
(or indirectI’harm or injury to another person. (V.P.)

Violent Non-personal — those offences that do ‘
damage to or destroy property in some way, whether
public or private, but do not injure another person.
(V.N-P.)

Non-violent - those offences that do not damage
property or injure another person. (N-V.)

Unclassifiable - those offences that do not
clearly fall into the above three classes. (UC.)

Check the appropriate column next to each crime
listed.

 

 

CRIMES V.P. V.N-P. N-V. UC

 

Breaking and entering, night

Forgery

Assault with intent to rob,
armed

Perjury

Breaking and entering, day

Robbery, unarmed

Malicious destruction of
property

Uttering and publishing

Robbery, armed

Unlawful driving away of
an automobile

Felonious assault

Larceny form a person

Larceny from an automobile

Gross indecency

Larceny from a building

Assault with intent to rob,
unarmed

Arson

AA
UV

AA
V

AAAA
VVVV
AAA/K
VVVV
AAA“
VVVV
AAA“

AAA
VV

V V
AAA
VVV
A A

AAA/NINA
VVVVVV
AAA/\AA
VVVVVV
AAAAAA
VVVVVV
AAAAAA

AA

V
AA
UV

56

VVV VVVV VV

VVVVVV

VV

57

-v
54

CRIME

<
"U

V. N-P. n-V. UC.

 

Assault less than murder
Violation of check law
Indecent liberties

Inciting perjury

Unlawful use of an automobile
Grand larceny

Statutory rape

Assault with intent to rape
Manslaughter

Kidnapping

Pandering

Larceny by conversion

Safe breaking

Possession of burglary tools
Carrying a concealed weapon
Embezzelment

Second degree murder

Rape

Receiving stolen property
Larceny from a boat

Larceny from a store
Larceny from a church
Possession of stolen car
Violation of the drug law
Negligent homicide

“AAAAAA/\AAAAAAAAAAAAAAAAA
VVVVVVVVVVVVVVVVVVVVVVVVV
AAAAAAAAAAAAAAAAAAAAAAAAA
VVVVVVVVVVVVVVVVVVVVVVVVV
AAAAAAAAAAAAAAAAAAAAAAAAA
vvvvvvvvvvvvvvvvvVVVVVVVV
AAA“AAAAAAAAAAAAA/‘AAAAAAA

vvvvvvvvvvvvvvvvvvvvvvvvV

CLASSIFICATION OF CRIMES
USING GRADUATE STUDENT RATINGS

CRIME

CLASSIFICATION

NUMBER
AGREEING

 

Arson
Assault less than murder
Assault to rape
Assault to rob, armed
Assault to rob, unarmed
Breaking and entering, day
Breaking and entering, night
Carrying a concealed weapon
Embezzlement
Felonious assault
Forgery
Grand larceny
Gross indecency
Inciting perjury
Indecent liberties
Kidnapping
Larceny by conversion
Larceny from an automobile
Larceny from a boat
Larceny from a building
Larceny from a church
Larceny from a person
Larceny from a store
Malicious destruction

of property
Manslaughter
Negligent homicide
Pandering
Perjury
Possession of a stolen car
Possession of burglary tools
Rape “
Receiving stolen property
Robbery, armed
Robbery, unarmed
Safe breaking
Second degree murder
Statutory rape
Unlawful driving away

an automobile
Unlawful use of

an automobile
Uttering and publishing
Violation of check law
Violation of drug law

58

Viol. N-personal
Viol. Personal
Viol. Personal
Viol. Personal
Viol. Personal
(No Agreement)
Viol. N-personal
(No Agreement)
Non-violent
Viol. Personal
Non—violent
Viol. N-personal
Non-violent

(No Agreement)
Viol. Personal
Viol. Personal
Unclassifiable
Viol. N-personal
Viol. N-personal
Viol. N-personal
Viol. N-personal
(No Agreement)
Viol. N-personal

Viol. N-personal
Viol. Personal
Viol. Personal
Non-violent
Non-violent
Non-violent
Unclassifiable
Viols‘Personal
Non—violent
Viol. Personal
Viol. N—personal
Viol. N-personal
Viol. Personal
Viol. Personal

(No Agreement)

(No Agreement)
(No Agreement)
Non-violent

(No Agreement)

wowmwmmmwomwwwmomowwwww

wwwmwmwmmwmwww

C)

00000

CLASSIFICATION OF CRIMES
USING PROFESSIONAL RATINGS

CRIME

Arson

Assault less than murder
Assault to rape

Assault to rob, armed
Assault to rob, unarmed
Breaking and entering, day
Breaking and entering, night
Carrying a concealed weapon
Embezzlement

Felonious assault

Forgery

Grand larceny

Gross indecencies

Inciting perjury

Indecent liberties
Kidnapping

Larceny by conversion
Larceny from an automobile

Larceny from a boat
Larceny from a building
Larceny from a church
Larceny from a person
Larceny from a store

Malicious destruction

of property
Manslaughter
Negligent homicide
Pandering
Perjury
Possession of stolen car
Possession of burglary tools
Rape
Receiving stolen property
Robbery, armed
Robbery, unarmed
Safe breaking
Second degree murder
Statutory rape
Unlawful driving away

an automobile
Unlawful use of

an automobile
Uttering and publishing
Violation of check law
Violation of drug law

59

CLASSIFICATION

Viol. N-personal

Viol. Personal
Viol. Personal
Viol. Personal
Viol. Personal
Non—violent
Non-violent
Non—violent
Non-violent
Viol. Personal
Non-violent
Non-violent
Viol. Personal
Non—violent
Viol. Personal
Viol. Personal
Non-violent
Non—violent
Non—violent
Non-violent
Non-violent
Non-violent
Non-violent

Viol. N-personal

Viol. Personal
Viol. Personal
Non—violent
Non-violent
Non-violent
Non-violent
Viol. Personal
Non-violent
Viol. Personal
Viol. Personal

Viol. N—personal

Viol. Personal
Viol. Personal

Non—violent

Non-violent
Non-violent
Non-violent
Non—violent

NUMBER
AGREEING

wmwwwwmwwmmmwwwwwwwwwww

wwmwwwwwwmmwww

N

(DNCQOJ

H}

N

H
OOHLOHmemHNmLOQONOQOQ-CNM

i—Im
mNHO-PONl-‘OI-‘wF-‘U'Iw

()0
C7

I—‘oomco

MEASURED AGREEMENT
AMONG RATERS

For Graduate Student Raters:

N

o\°

 

 

All three agree 19 ”5.24
Only two agree 15 35.71
No agreement 8 19.05
% meeting criterion = 80.95
For Professional Raters:
N %

All three agree 32 76.19
Only two agree 10 23.81
No agreement 0 0.00

O

6 meeting criterion = 100.00

60

APPENDIX C

Summary Tables

61

 

ANALYSES OF VARIANCE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Age

Source df 88 MS F p

Between 2 l”0.50 70.25 10.2” *

Within 3”0 2330.8” 6.86

Total 3”2 2”71.3”

*Significant beyond .01 level.

Education

Source df SS MS F p

Between 2 ”.87 2.”” 1.36 N.S
Within 339 610.3” 1.80

Total 3”l 615.21

I9.

Source df SS MS F pp

Between 2 225.0” 112.52 N.S
Within 3”0 ”3707.52 128.55

Total 3”2 ”3732.56

Aggression

Source df SS MS F p_
Between 2 21.33 10.67 3.09 **
Within 3”0 1173.”” 3.”5

Total 3”2 119”.77

**Significant beyond .05 level.

62

63

Aggression anxiety

 

 

 

 

 

 

 

 

Source df SS MS F p
Between 2 191.33 95.67 2.”0 N.S.
Within 3”0 13525.80 39.7”

Total 3”2 13717.13

Fantasy aggression

Source df SS MS F p
Between 2 28637.20 l”318.60 1.72 N.S.
Within 3”0 283”775.32 8337.57

Total 3”2 2863”12.52

(Variance not homogeneous)

 

 

INDIVIDUAL COMPARISONS

 

 

 

 

 

 

Age

3 Significance
V.P. VS. VON—P. 2.14:5 Beyond 005
VON—P. VS. N-Vo lou8 N.S.
V.P. vs. N-V. 7.93 Beyond .01
Aggression

3 Significance
V.P. VS. V.N‘Po 2009 Beyond 005
V. N—P. vs. N—V. l.”6 N.S.
V.P. VS. N-Vo long N.S.

 

 

 

6”

APPENDIX D

Conversion to "Corrected" Standard Scores

65

 

. .
n

CONVERSION FOR THE IPIT

 

 

 

Mean = 28.”3 S.D. = ”.27
Corrected
Deviation Standard Standard
Raw Score Score Score Score
15 -13.”3 —3.15 85
16 -12.”3 -2.91 109
17 ~11.”3 -2.68 132
18 ~10.”3 —2.”” 156
19 - 9.”3 —2.21 179
20 - 8.”3 —1.97 203
21 - 7.”3 -1.7” 226
22 - 6.”3 -1.51 2”9
23 - 5.”3 —l.27 273
2” - ”.”3 -1.0” 296
25 - 3.”3 -0.80 320
26 - 2.”3 -0.57 3”3
27 — 1.”3 -0.33 367
28 — 0.”3 —0.10 390
29 0.57 +0.13 ”13
30 + 1.57 +0.37 ”37
31 + 2.57 +0.60 ”60
32 + 3.57 +0.8” ”8”
33 + ”.57 +1.07 507
3” + 5.57 +1.30 530
35 + 6.57 +1.5” 55”
36 + 7.57 +1.77 577
37 + 8.57 +2.01 601
38 + 9.57 +2.2” 62”
39 +10.57 +2.”8 6”8

66

 

CONVERSION FOR THE OAT

 

 

 

Mean = 37.96 S.D. = 7.05
Corrected
Deviation Standard Standard
Raw Score Score Score Score
16 -21.96 -3.11 89
20 ~17.96 —2.55 1”5
21 -16.96 -2.”1 159
22 -15.96 -2.26 17”
23 —1”.96 —2.12 188
2” —13.96 -1.98 202
25 -12.96 -l.8” 216
26 —11.96 -1.70 230
27 -10.96 -1.55 2”5
28 - 9.96 -1.”1 259
29 - 8.96 -1.27 273
30 - 7.96 -1.13 287
31 — 6.96 -0.99 301
32 - 5.96 -0.85 315
33 - ”.96 -0.70 330
3” - 3.96 -0.56 3””
35 - 2.96 -0.”2 358
36 - 1.96 -0.28 372
37 — 0.96 -0.1” 386
38 + 0.0” +0.01 ”01
39 + 1.0” +0.15 ”15
”0 + 2.0” +0.29 ”29
”1 + 3.0” +0.”3 ””3
”2 + ”.0” +0.57 ”57
”3 + 5.0” +0.71 ”71
”” + 6.0” +0.86 ”86
”5 + 7.0” +0.99 ”99
”6 + 8.0” +1.1” 51”
”7 + 9.0” +1.28 528
”8 +10.0” +1.”2 5”2
”9 +11.0” +1.57 557
50 +12.0” +1.71 571
51 +13.0” +1.85 585
52 +1”.0” +1.99 599
53 +15.0” +2.13 613
5” +16.0” +2.28 628
55 +17.0” +2.”2 6”2
56 +18.0” +2.56 656

REFERENCES

l. Arieti, 8., Interpretation of schizophrenia, New York:
Robert Brunner, 1955, pp ”3-52.

 

2. Bandura, A. and Walters, r., Adolescent aggression:
a study of the influence of child training practices
and family interrelationships, New York: Ronald Press,
1959.

 

3. Bernabeu, E.P., Underlying ego mechanisms in delin-
quency, Psychoanal. Quart., 1958, 27, 383-396.

 

”. Buss, A.H., Fischer, H., and Simmons, A.J., Aggres-
sion an hostility in psychiatric patients, I.Consult.

5. Cameron, N. and Magaret, Ann, Behavior pathology,
Houghton Mifflin, 1951, pp278-279.

 

6. Cohn, F.S., Fantasy aggression in children as studied
by the doll play technique, Child Deve1., 1962, 33, 325ff.

 

7. Dollard, J., Doob, L.W., Miller,N.E., Mowrer, O.H.,
and Sears, R.R., Frustration and aggression, New Haven:
Yale Univ. Press, 1939.

8. Dollard, J., and Miller, N.E., Personality and psycho-
thera , McGraw Hill, 1950.

9. Fenichel, Otto, The pgychoanalytic theory_of neurosis,
New York: Norton 8 Co., 19”5.;

10. Freud, S., A general introduction to psychoanalysis,
Twenty-fifth lecture: Anxiety, pp ”00-”18, Permabook,
1953.

ll. Freud, S. Hysterical phantasies and their relation
to bisexuality (1908) in Sigmund Freud, Collected Papers,
Vol. II, New York: Basic Books, 1959.

 

12. Gordon, J.E. and Cohn, F.S., Effects of fantasy arousal
of affiliation drive in doll play aggression, J. Abnor.
Soc. Psychol, 1963, 66, 301-307.

 

 

13. Healy, W. and Bronner, Augusta, New lights on delin-
guency_and its treatment, New Haven: Yale Univ. Press,
1936, pp 67-70.

 

1”. Hurley, J.R., The Iowa Picture Interpretation Test:
A multiple-choice variation of the TAT, J. Consult.
Psychol., 1955, 373—376.

 

68

69

15. Jensen, A.R., Aggression in fantasy and overt
behavior, Psychol. Monogr., 1957, 71, #HAS.

 

16. Johnston, R.A., A methodological analysis of
several revised forms of the Iowa Picture Inter-
pretation Test, J. Pers., 1957, 25, 283-299.

17. Kagan, J., The measurement of overt aggression
from fantasy, J. Abnor. Soc. Psychol., 1956, 52, 390—
393.

 

18. Laskey, D.I., The relationship between spontaneous
fantasy, induced fantasy, and behavioral measures

of hostility, Ph.D. Dis., Temple Unive., 1961, Dis.
Abstr., 1962, 22 (12), #908.

19. Lesser, G.S., The relationship between overt and
fantasy aggression as a function of maternal response
to aggression, J. Abnor. Psychol, 1957, 55, 218-221.

 

20. Lesser, G.S., Conflict analysis of fantasy aggres—
Sion’ J. Pers., 1958, 26, Zg-L‘lo

21. Lesser, G.S., Religion and the defensive response
in children's fantasy, J. Proj. Tech., 1959, 23, 69-68.

 

22. Marquis, J.N., Fantasy measures of aggressive be—
havior, Ph.D. Dis., Univ. of Michigan, 1961, Dis.
Abstr., 1961, 21 (12), sesu-aess.

23. May, R., Existence, New York: Basic Books, 1958, 50-55.

 

2H. Sears, P.S., Doll play aggression in normal children;
influence of sex, age, sibling status, and father's
absence, Psychol. Monogr., 1951, 65, #323.

 

25. Sears, R.R., Maccobby, E.E., Levin, H., The sociali-
zation of aggression. In Maccobby, B.E., Newcomb, T.M.,
and Hartley, E.L., Readings in social_psychology, Holt,
Rinehart, 8 Winston, 1958, 350-359.

 

. ...H 1"“? -__‘

26. Sears, R.R., Relation of early socialization experi—
ences to aggression in middle childhood, J. Abnorm.
Psychol., 1961, 63, ussff.

 

27. Stricker, G., The construction and partial validation
of an objectively scoreable apperception test, J. Pers.
1962, 30’ 51-620

28. Zaks, N.S., Walters, R.H., First steps in the con-
struction of a scale for the measurement of aggression,
J. Psychol., 1959, #7, 199-208.

 

HICHIGRN STATE UNIV. LIBRQRIES

ll llll
9

312 3105258572