University of Southern California Dissertations and Theses

Behavioral Seriousness And Impulse-Control Balance In Delinquency

(USC Thesis Other)

Behavioral Seriousness And Impulse-Control Balance In Delinquency

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been
microfilmed exactly as received 67-427
ROCHLIN, Martin, 1928-
BEHAVIORAL SERIOUSNESS AND IMPULSE-CONTROL
BALANCE IN DELINQUENCY.
University of Southern California, Fh.D., 1966
Psychology, clinical
University Microfilms, Inc., Ann Arbor, Michigan
BEHAVIORAL SERIOUSNESS AND IMPULSE-CONTROL
BALANCE IN DELINQUENCY
by
Martin Rochlin
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Psychology)
June 1966
UNIVERSITY O F SOUTHERN CALIFORNIA
THE GRADUATE SCHOO L
U NIVERSITY PARK
LOS ANGELES, CALIFO RNIA 9 0 0 0 7
This dissertation, written by
............................M axtin.Jlachlija............................
under the direction of Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by the Graduate
School, in partial fulfillment of requirements
for the degree of
D O C T O R O F P H I L O S O P H Y
.......
Dean
D ate ................
ERTATION COMMI
TABLE OF CONTENTS
LIST OF
Chapter
I.
II.
III.
IV.
Page
TABLES........................................ iv
PROBLEM
Delinquent tendencies
Petty theft
Burglary
Grand theft, auto (GTA)
Defining Delinquency
Measuring and Predicting Delinquency
THEORETICAL FORMULATIONS AND HYPOTHESES . . . 11
Theoretical Formulations
Experimental Hypotheses
METHOD........................................ 25
Operational Definitions of the Variables
Behavioral Seriousness
Offense category 1
Offense category 2
Offense category 3
Offense category 4
Psychometric Asocialization
Intelligence
Social position
Family status
Procedure
The Sample
Statistical Techniques
RESULTS...................................... 42
Experimental Results in Terms of Hypotheses
Unhypothesized Findings
Race and Jesness scales
Social position and Jesness scales
Age and Jesness scales
IQ and Jesness scales
Validity of the Cattell test
Intercorrelations of Jesness scales
DISCUSSION OF RESULTS....................... 52a.
ii
iii
Chapter Page
VI. SUMMARY........................................ 71
Results
APPENDIXES
Appendix A. Intercorrelations of Jesness Inventory
Scales............................. 76
Appendix B. Significant Correlations Between
Variables Within OC I ............. 77
Appendix C. Significant Correlations Within
OC I I ........................ 78
Appendix D. Significant Correlations Within
OC III........................ 79
Appendix E. Significant Correlations Within
OC I V ........................ 80
Appendix F. Significant Correlations Within OC I
& II (Less Serious Offenses) .... 81
Appendix G. Significant Correlations Within OC III
& IV (More Serious Offenses) .... 82
Appendix H. Significant Correlations Between
Variables Within the Higher Social
Position Group ..................... 83
Appendix I. Significant Correlations Between
Variables Within the Lower Social
Position Group ................. 84
Appendix J. Hollingshead's Two-Factor Index of
Social Position ................... 85
REFERENCES....................................... 87
LIST OF TABLES
Table Page
1. General Characteristics of the Total
N of 200 ...................................... 37
2. Group Characteristics Within Offense
Categories.................................... 38
3. Significant Mean Differences Between
Subgroups.................................... 43
4. Partial Correlations Between Social Position
(ISP) and Motivational and Inhibitory In
dices (Sum^ and Sum2) , with Asocialisation
(ASI) Held Constant, Within Offense
Categories.................................... 45
5. Significant Correlations Between Variables in
the Total Sample of 200, Excluding Inter
correlations Between Jesness Scales ......... 47
6. Means of Major Variables Within Offense
Categories.................................... 54
CHAPTER I
|
| PROBLEM
I Defining Delinquency
"O remember not the sins and offences of my youth."
! (The Book of Common Prayer: Psalms XXV.6)
The content of this simple entreaty attests the
;perennial prevalence of antisocial behavior in youth. It
!is only in recent years, however, that there have been num
erous systematic attempts to increase our understanding of
the dynamics of juvenile delinquency. The work of August
Aichhorn (1935) helped stimulate a dynamic approach to a
problem that previously had been considered largely one of
iconstitutional inheritance, and had been treated mainly as
a penal or custodial issue. The years since the publica
tion of Aichhorn's Wayward Youth (19 35) have seen the accu-
;mulation of an increasing literature in the field of delin
quency. Unfortunately, however, the value of the growing
I literature has been curtailed by a concomitant increase,
■ rather than reduction, of dissension and confusion regard-
iing the problem.
| One of the most fundamental steps in the study of a
problem is its definition in terms of the indentification
and specification of its relevant aspects or variables.
The study of delinquency thus far has failed to yield ade
quate and commonly acceptable classification systems or j
|
measurements by which youthful offenders can be differen- i
I
tiated from one another in terms either of personality |
characteristics, psychodynamics, or even the severity of
the symptomatology.
Perhaps the knottiest problem in the behavioral
definition and classification of delinquent acts is the
assessment of "seriousness." From Roscoe Pound's "social
interests" (Pound, 1914) to the Model Penal Code (American
I
Law Institute, 1954; 1959), attempts at classification,
using various criteria such as social injury, psychic in
jury, type of victimization, and diverse moral precepts,
have failed to yield a satisfactory seriousness index.
The most recent, and perhaps most estimable, scheme
for the definition and measurement of seriousness is that
of Sellin and Wolfgang (1964), which takes into account
weighted scores on a number of elements of delinquent
!
events, such as the number of victims of bodily harm, in
timidation, the number of premises forcibly entered, and
the value of property stolen, damaged, or destroyed. This
system, however, is inadequate for research purposes be
cause its applicability is limited to only certain types
of delinquent behavior, and the detailed information re
quired for scoring is unlikely to be readily accessible in
most police records.
i Etiologic definitions and classifications, focusing
| attention on the delinquent child rather than on the delin-
!
| quent act, have been favored by social scientists, whose
i !
j concern is not limited to those who have been officially j
i apprehended. They generally consider all behavior, ap-
i - '
j proved or not, symptomatic of individual needs, and feel
I
j that official labels may obscure rather than clarify the
I real problems, the discovery and treatment of which are
!
1 their goals. In practice, however, definitions based on
! underlying characteristics of offenders have resulted in
i
I
I even more ambiguity and controversy than behavioral defini
tions, primarily because they depend on theoretical assump
tions which remain largely undemonstrated. The literature
on the etiology of delinquency reveals, as does the litera
ture on its behavioral manifestations, a plethora of con-
fused, contradictory, and ambiguous concepts, and a mass of
unsystematic, fortuitous, and often irrelevant or trivial
i data, which may be classified according to their relative
emphasis on either constitutional, sociogenic, or psycho
genic factors. Although the symptomatic and etiologic ap
proaches have often been treated as inimical, it is not
i
;only possible but perhaps essential to think of the study
i
;of the delinquent child and the delinquent act as comple
mentary, each contributing to a fuller understanding of the
i other.
Measuring and Predicting Delinquency |
I
"A boy's will is the wind's will. ..."
i
: (Longfellow: My lost youth)
i j
The inadequacy of testing techniques has constitu- |
i ted a major problem in delinquency research. Schuessler
I " ;
: and Cressey (1950) reviewed studies in which 30 of the best:
| known objective personality tests developed before 1945
i
: were used, and found personality traits to be distributed
I among offenders in much the same way as among law abiding
peers. More recent data suggest that these early tests,
: rarely designed for delinquency research, may not have been;
tapping relevant dimensions.
Projective techniques, notably the Rorschach and
the Thematic Apperception Test, have been used extensively
I and have yielded interesting data about individual cases,
but suffer the persistent difficulty of adequate delinea-
tion and validation of interpretive dimensions.
Many attempts have been made to predict moral-
ethical behavior on the basis of measurement techniques
designed to tap selected personality dimensions judged rel—
j evant to such behavior. Among such attempts, those of
|
| Hartshorne and May (1930), and Havighurst and Taba (1949)
i
found "reputation scores" (estimates of teachers and peers)
I
! somewhat useful in predicting external behavior, but not
! in providing indices of individual character dimensions.
■ The Gluecks (1950) differentiated delinquents from
I
nondelinquents on the basis of five social factors involv
ing the quality of interpersonal relationships and of dis
cipline in the family. Using the Glueck Social Prediction
Scale , several independent investigators found the scores
derived from ratings on the five social factors effective
in predicting future delinquency years before the children
became involved with authorities (Glueck & Glueck, 1959).
These studies have been criticized with regard to their de
sign and method, however, and Robison (1960) pointed out
that the efforts of the New York City Youth Board to vali
date the Glueck prediction scales pose many questions re
garding the comparability of populations, the method of se
lecting samples, and the adequacy of the information assem
bled. Furthermore, this instrument is limited to a unidi
mensional rather than diagnostic use, and is extremely cum
bersome in the extensive amount of personal historical data
required for scoring.
The Gough Socialization Scale (Gough, 1954) is a
true-false test designed to measure the ability to place
oneself in the role of another. This single dimension is
given paramount importance by Gough as the determinant of
antisocial behavior, a position stemming from the earlier
theory of Mead (1922) . Validation studies by the author of
the scale (Gough, 1954) and others (Reed, 1957) have sup
ported the instrument's ability to differentiate between
nondelinquents and offenders characterized by different
6
degrees of antisocial involvement. However, the unidimen
sionality of the scale, as with the Glueck scale and Doc
tor's use of the Porteus Maze for measuring impulsiveness
(Doctor, 1954), limits its connotative range to a narrow
dimension which implies a closer correspondense with the
larger condition of delinquency than is warranted either by
experimental or theoretical evidence. Gough (1954) found
a significant relationship between scores on the delin
quency scale of the California Personality Inventory and a
socialization dimension defined by a sample of 3,2 85 males
ranging from those nominated "best citizens," through "men
in general," through "disciplinary problems," to young de
linquents and prisoners. The single score, however, again
suffers the same connotative limitations as the other uni
dimensional measures.
The problem of unidimensionality is inherent in the
use of the Rosenzweig Picture-Frustration Study, which
measures characteristics of aggression, certainly a rele
vant, but not exclusive, dimension in delinquency research.
The transparency of this test, furthermore, embodies anoth-
! er serious flaw that constitutes a serious consideration in
:work with antisocial youngsters. Several studies using
this test with delinquents (e.g., Silverstein, 1951; Vane,
j
1954) resulted in significant findings in direct opposition
to what could reasonably be expected on theorectical
grounds.
Vane's delinquent subjects showed significantly
less than average extrapunitiveness, less concentration on
frustrating objects, and more interest in solving problems
than nondelinquent controls. The meaning of these sur
prising results was illuminated by Silverstein1s finding
that delinquents differed from norms in the same ways as
did normal subjects who were instructed to try to make a
favorable impression. He found, furthermore, that the
ability to fake on the Rosenzweig P-F is independent of in
telligence, and advised against the use of the test with
delinquent populations.
The MMPI has been used with some success in dis
criminating between delinquent and nondelinquent groups
(Wirt & Briggs, 1959), in predicting future delinquent be
havior in children several years before the behavior comes
to public attention (Hathaway, 1952) , and in differentia
ting the severity and chronicity of psychopathology among
offenders (Gynther, 1962) . The usefulness of the MMPI,
however, is limited by the theoretical irrelevance of its
dimensions to the problems of delinquency. As an empiri
cally derived, pragmatic instrument, based on psychopatho-
logical symptoms and nosological categories, its results
are meaningful only at a descriptive level, and it can re
veal no more about an individual than is known about those
in the original criterion groups.
An empirically derived predictive measure may make
8
use of nonstructured stimuli, for, as reasoned by Berg
(1961), the particular form of stimulus content is rela
tively unimportant in the pragmatic approach to test de
sign. Berg found considerable evidence that deviant re
sponses tend to be general, and that responses made in a
noncritical area of behavior can be used to predict behav
ior in critical areas. In various researches, the stimuli
employed have included lists of foods, the autokinetic phe
nomenon, preference for abstract designs, meaningless
sounds, etc. Deviant and criterion groups are identified
by means of behavioral characteristics, and deviant re
sponses identified and scaled, if desired, by purely sta
tistical methods. It is essential in this approach, how
ever, that the establishment of deviant and criterion
groups rests on carefully validated operational referents.
Berg concluded, from his failure to distinguish delinquents
from matched controls on the basis of Perceptual Reaction
Test deviant responses, that delinquency as such is too
heterogeneous a concept to constitute a valid behavioral
category. If delinquents are not identifiable by external
criteria, they do not form a valid deviant group in terms
of this "deviation hypothesis."
The importance of identifying deviant groups by
valid behavioral characteristics has led to a variety of
attempts to establish an objective behavioral classifica
tion of juvenile delinquents. Mueller (1959) found that
ratings from case histories on six behavioral scales (ag-
j
i
gression— nonaggression, maladaptive dominance— adaptive |
submission, acceptance— rejection of gang codes, accep
tance— rejection of conformity codes, desurgency— surgency,:
and withdrawal— participation with people) could be used to
classify 204 delinquent boys reliably in four categories,
but had practically no prognostic value. In a study of the
reliability and validity of the psychiatric assessment of
100 delinquent boys, based on ratings of interviews, obser
vations, and background data by psychiatric teams, Hurwitz,;
Hutcheson, and Cooper (1961) found a positive correlation
between the level of reliability and the level of inference
involved. They found also that observations and background
data together produced more reliable ratings than did back
ground data alone, and that knowledge of previous difficul
ty with the law influenced diagnostic judgment more than
did social and historical data.
In contrast to the psychiatric-rating approach,
Nye and Short (1957) advocated the self-report for measur- ,
ing delinquent behavior in non-institutional populations,
with the aim of eliminating or minimizing the socioeconomic
and cultural biases involved in differential arrest, deten
tion, and penal practices. Dentler and Monroe (1961) found
that survey and retest data from 912 junior high school
youths in three types of communities provided a highly
reliable self-report scale of theft, which correlated
i ' ' 10
I I
I significantly with demographic factors, parent-child rela-
I !
i tionships, and leisure activities. Because the theft scale!
i _ i
i was associated with but not predictive of truancy, vandal- i
3 j
j ism, and antisocial misconduct, the investigators recom
mended that self-report scales be devised for each type of
I misconduct to permit research on the interrelations among
i
j them. Like Nye and Short (1957), they felt that the inci-
; dence of certain types of behavior in terms of configura
tions of offenses has etiologic significance. Rating
I scales, generally, are a precarious method of measurement,
requiring highly trained judges with qualities of objectiv
ity, precision, and quantitative perception rarely found in
human observers.
The Jesness Inventory (Jesness, 1963) was designed
for use in the classification, treatment, and prediction of!
delinquency. Its development was especially intended to
correct the shortcomings of instruments previously used forj
these purposes, and to fulfill the need for a structured
test, suitable for use with young offenders with low read- i
ing ability and short interest spans, which would be sensi-
; tive to change, yet stable enough to provide a measure of
personality types. Its multidimensionality permits diag-
| nostic profile analysis, and its single index of asocial
tendencies allows for the prediction of delinquent
behavior.
CHAPTER II
THEORETICAL FORMULATIONS AND HYPOTHESES
Theoretical Formulations
The theoretical orientation of the present study
stems largely from, but is not identical to, the "struc
tural" conception of psychoanalytic theory, first formu
lated explicitly in 1923 by Sigmund Freud (1927). The
structural determinants of behavior (id, ego, and superego)
were introduced as intervening variables to account for the
observation that drives do not unequivocally determine ei
ther general behavior or symptom formation. In 1915,
Freud published two papers (Freud, 1953a; 1953c) in which
he introduced the concept of "countercathexes" as interfer
ing factors which tend to delay the discharge of "drive
cathexes." Within the psychoanalytic framework, Redl and
Wineman (1957) have pointed to the generality with which
the "ego" processes have been discussed, compared to the
detail and specificity with which the manifestations of im
pulses, emotions, and drives have been studied, in the lit
erature on delinquency. To counteract this general empha
sis on the study of the "impulsive system," Redl and
Wineman have concentrated their efforts on a careful explo
ration of the "control system," emphasizing the specific
12
: characteristics of the "delinquent ego" and the "sick con-
j
I science." Among other investigators who have emphasized
I "control" factors in delinquency are the McCords (1956) ,
i
jwho attribute delinquency largely to "lack of inhibition,"
j
land Cleekley (1955), who spoke of "defective brakes."
" t
Virtually nonexistent in the literature are studies of the
| dynamic interaction of specific impulses with specific ways
I of controlling or dealing with them.
The major theoretical hypothesis of the present
study, simply stated, is that the seriousness of overt de-
: iinquent behavior is a function of the degree of imbalance
between the individual's antisocial impulses and his re
sources for controlling those impulses. The dynamic inter
action of antisocial impulses and controls is viewed here
as the psychological modifying process or mediating mechan
ism underlying the behavioral manifestations of delinquen
cy. The etiologic value of this formulation lies less in
its explanatory cogency than in its integrative and heuris
tic utility. It is not intended to supply answers to ques-
| tions of why individuals come to have various kinds and de-
| grees of antisocial motivation or inhibitory mechanisms,
| but rather to provide a general framework which may serve
jto better synthesize the existent mass of fragmentary and
| inconsistent etiologic hypotheses and research data, and to
| stimulate and give direction to further efforts to improve
|
| understanding, explanation, and prediction. The present
formulation differs from most similar views mainly in its
greater generality and interdisciplinary connotation. In
contrast to a strict psychoanalytic interpretation, the
"impulse" component is not limited to a concept of instinc
tual "id" drives or frustration-determined aggressive re
sponses (although these are included) , nor is the "control"
component limited to the concepts of "ego" and "superego"
restraints (although these too are included). The factors
that contribute to an individual's antisocial impulsivity
as well as to his control capacity may have their sources
in a wide variety of constitutional, sociologic, and psy
chological areas, all of which may be treated as complemen
tary rather than antithetical in the present scheme. This
can be illustrated by the following outline.
ANTISOCIAL IMPULSE COMPONENT CONTROL COMPONENT
CONSTI-
| TUTIONAL
! FACTORS
SOCIO
GENIC
FACTORS
Male sexuality
(Schwarz, 1939)
Morphological anomolies
(Loinbroso, 1911)
Mental deficiency (Goddard,
1915; Bennett, 1960)
Hereditary predisposition
(Burt, 1925)
Neurological defects
(Thompson, 1953; Strauss &
Lehtinen, 1947)
Cerebral dysfunction, as
indiqated by abnormal EEG
patterns (Fontanesi & Zilli,
1961; Kilch & Osselton,
1961; Yoshii et al., 1961)
Unconditionability—
extraversion (Eysenck, 1947
1952; Hildebrand, 1953;
Payne, 1961)
Perceptual defects (Eskin,
I960; Petrie et al.. 1962;
Pesetsky, 1961)
Social disorganization
(Shaw & McKay, 19 31)
Culture conflict (Shaw &
McKay, 1931)
Female sexuality
(Schwarz, 19 39)
Intellectual
competence
Favorable heredity
Conditionability—
introversion
(Eysenck, 1947,
1952; Hildebrand,
1953; Payne, 1961)
Perceptual compe
tence
Social organiza
tion, stability,
solidarity
Cultural unity,
authoritative stan
dards , codes, and
models
Anomie and nonconformity
(Merton, 19 49)
Clear norms,
conformity
ANTISOCIAL IMPULSE COMPONENT CONTROL COMPONENT
i SOCIO- Depressed living conditions Comfortable living
j GENIC (Shaw & McKay, 1931, 1942; conditions
| FACTORS Carr-Saunders et al., 1942;
(Cont'd.) Mannheim, 194 8, 1949;
j Schwartz, 1945; Murphy et al.
i 1946; Falk, 1959)
Differential association
(Sutherland, 1947)
| Lower-class definition of
the male role (Parsons,
1947)
Association with
lawabiding persons
Middle or upper-
class definition of
the male role
Gang membership and identi
fication with the culture
of the gang (Cohen, 1955)
Horizontal mobility (Falk,
1959)
Differential opportunity
systems— "illicit means"
(Cloward & Ohlin, 1963)
Value infringements
(Eissler, 1949)
Environmental sta
bility
Lack of opportuni
ties or means of at
tainment by illicit
activity
Value conformity
Optimal supervision
concern, and disci
pline in the family,
the school, and the
community
PSYCHO- Need for punishment, from
GENIC a sense of guilt (Freud,
FACTORS 1953b)
Alloplastic aggression
(Eissler, 1949)
Superego lacunae
(Johnson, 1959)
Normal unconscious crimi
nality (Alexander & Staub,
1956)
Self-acceptance
Autoplastic aggres
sion
Normal superego
functioning
Outlets for antiso
cial drives in sub
limated forms
16
PSYCHO
GENIC
FACTORS
(Cont'd.)
ANTISOCIAL IMPULSE COMPONENT CONTROL COMPONENT
Unrewarded needs for excite
ment, variety, and self-
actualization (Freyham,
1951)
Identification with parent
of the opposite sex, with
reactive pseudo-masculinity
(Fenichel, 1945)
Identification with hostile
or criminal parental figures
(Aichhorn, 1935; A. Freud,
1949; Glover, 1960) or with
antisocial aspects of social
ized parental figures (Bandura
& Walters, 1959)
Implicit, covert reinforce
ment from superficially so
cialized parental figures
who seek vicarious gratifi
cation of repressed antiso
cial drives (Johnson, 1959)
Repressed dependency needs Sense of security
(Bandura & Walters, 1959)
Sociopathic impulsivity,
associated with underdevel
oped socialization, deriving
from early deprivation of
nurturance through parental
absence, neglect, or rejec
tion (Reich, 1925; Ferenczi,
1929; Lippman, 19 37; Lander,
1941; Lindner, 1944; Bender,
19 43; Goldfarb, 1947; Healy
& Bronner, 1936)
Drive to act out suppressed
phallic masturbation impul
ses (A. Freud, 1949)
Insecurity and identity dif
fusion (Er'ikson, 1956)
Ample outlets for
energy release, ex
pression of inter
ests, and social
ized forms of grat
ification
Normal identifica
tion with a "good
object" (Fenichel,
1945)
ANTISOCIAL IMPULSE COMPONENT CONTROL COMPONENT
PSYCHO
GENIC
FACTORS
(Cont'd.)
Techniques of "neutraliza
tion" (Sykes & Matza, 1957)
and ego functioning in the
service on impulse defense
(Redl & Wineman, 195 7)
Failure of norm contain
ment— erosion of norms
(Reckless & Shoham, 1963)
Psychotic impulsivity
(Bandura & Walters, 1959)
Need to counteract a psy
chotic fear of being help
lessly manipulated
(Bergler, 1961)
Positive strivings for ego
development and communica
tion; attempts to compen
sate for loss of feelings
of weakness and dependency
(Allchin, 1962)
Unfavorable attitudes of
self-reference (Purcell,
1961; Lively et al., 1962)
Self-insight
(Rogers et al.,
1948; Trent, 1957)
Norm containment—
retention of norms
(Reckless & Shoham,
1963)
Adequate reality
testing and ego
integration
Capacity for inhi
bition (McCord &
McCord, 1956) Ef
fective "brakes"
(Cleckley, 1955)
Favorable self-
image (Purcell,
1961; Lively et al.
1962)
Socioempathic capa
city and role-play
ing ability (Mead,
1922; Aichhorn,
1935; Gough, 1954;
Sarbin, 1954)
Ego functioning di
rected toward syn
thesis between de
sires , reality de
mands , and social
values (Redl &
Wineman, 195 7)
A striking characteristic of the etiologic hypoth
eses outlined above is the fact that most of the control
components are derived logically as opposites of impulse
components, reflecting the heavy emphasis on consideration
of the motivational aspects of delinquency in the litera
ture. This emphasis has constituted a deterrent to theory
and research in other areas of psychopathology as well as
in delinquency, and appears to persist despite the repeated
finding that a knowledge of drives alone is insufficient
for the prediction of specific behavior. On the other
hand, it is obvious that a knowledge of control resources
and coping techniques alone is equally insufficient for the
explanation and prediction of behavior. Defective con
trols, for example, tell us little or nothing about the na
ture of the particular impulses that are likely to pene
trate them. Another notable characteristic of the control
components is their frequent vagueness and non-specificity.
An individual who displays "conditionability"value con
formity," and "ego integration," for example, is as unlike-
;ly to display neurotic or psychotic symptomatology as he is
unlikely to display character disorders or to be over-
i
iwhelmed by his antisocial impulses. It seems clear that
ietiologic enlightenment and predictive efficacy depend on
the consideration of both motivational and inhibitory vari-
; ables, and require investigation of their specific quali-
| ties and the ways in which they interact in specific
iindividuals and specific circumstances. The outline pre-
; sented above is not intended to represent a complete list
!
| of the factors contributing to antisocial motivation or
| control, nor are the factors included purported to be ne-
i cessarily significant. It is intended rather to illustrate
! the manner in which a wide variety of disparate etiologic
|hypotheses may be incorporated and integrated in the theo-
i retical framework presented here, and how this interdisci-
I plinary scheme may stimulate and suggest possibilities for
further study of the interrelationships of diverse relevant
! variables within its theoretical scope.
By regarding the impulse-control balance as the me
diating variable which relates motivational stimuli of var
ious kinds to the behavioral manifestations of antisocial
responses, many of the shortcomings of delinquency research
may be remedied, and the possibility of meeting the cri
teria of effective scientific investigation may be en
hanced. In proposing criteria for effective research in
the etiology of delinquency, Herzog (1959) stressed the im
portance of studying the relative roles of the physical en-
I vironment, the subculture, and the family pattern, as well
j
; as the special needs of the individual designated delin-
: quent. She urged the study of the dynamics of delinquent
i
: behavior as experienced in interrelationships, both psycho-
; logical and structural-functional, between these various
i
I influences. The impulse-control concept, because it is
20
general enough to encompass any relevant variable, and yet
specific enough in its dynamic connotation to give direc
tion to the research hypotheses it may generate, seems pe
culiarly well adapted to the kind of study proposed by
Herzog.
The second and third hypotheses of the present study
were developed directly from the concept of "impulse-control
balance" as the mediating psychodynamic process underlying
delinquent behavior.
Experimental Hypotheses
I. The sociologically defined "seriousness" of de
linquent behavior is positively related to a psychometric
index of "asocialization."
This hypothesis is intended to examine the three
major assumptions on which it is based. Its first implica
tion is that "seriousness" constitutes a valid dimension of
delinquent behavior, and that the dimension can be identi
fied, defined, and measured by social conduct norms. These
assumptions are open to question. Although it is known
that delinquent behavior varies in many ways, the question
of whether or not the qualitative diversity of its manifest
forms is subject to quantification on a value dimension is
debatable. Furthermore, even if the concept of "serious
ness" is accepted as applicable to delinquent behavior, the
validity of its definition in terms of social conduct
; norms is questionable. i
| i
The second implication of this hypothesis xs that aj
|
psychometric index may be found that constitutes a continu-l
j i
| ous, linear, quantitative measure of asocialization. Even ;
|
if the conceptual validity of such a measure and its abil-
i ity to differentiate delinquents from nondelinquents are
i acceptable, the question inherent in the hypothesis con-
! cerns its ability to discriminate among finer gradations of
: degrees of asocialization among delinquents.
The third assumption implicit in the first hypoth- .
esis is the notion that, granting the validity of the be- !
havioral and psychometric variables, there is a meaningful ;
relationship between them. Even if the dimensions are in-
! dividually conceptually valid and accurately defined, the
nature of the relationship between tested attitudes or per
sonality traits and overt behavior remains questionable in
general, and even more uncertain with regard to the speci- j
fic variables under discussion here.
II. There is a closer correspondence between the
"seriousness" of delinquent behavior and the "asocializa-
' tion" index than between behavioral seriousness and either :
| :
i the motivational or inhibitory components of the asociali-
i zation index.
i
This hypothesis derives from the theoretical orien-
; tation of the present study, that it is the impulse—
| control balance rather than the impulses or controls alone
22
; which constitutes the mediating variable underlying delin- j
| quent behavior. j
| i
III. With behavioral seriousness and psychometric |
I asocialization held constant, "lower-class" and "middle-
! and upper-class" delinquents can be differentiated by the
: relative magnitudes of the motivational and inhibitory com-
; ponents of the asocialization index. Lower class delin-
iquents are expected to attain lower scores on these compo-
I nents than middle and upper class delinquents, with equal
degrees of asocialization or behavioral seriousness.
The third hypothesis is intended to examine the ;
two major sources of its derivation. Its first implication
pertains to the theoretical orientation of the study, which
attributes greater etiologic significance to the impulse-
control balance than to impulse or control factors alone.
In this orientation, the absolute magnitude or intensity of
antisocial impulses or of controls is viewed as an unsound |
I basis for predicting the degree or severity of delinquency.1
Intense aggressive drive, for example, need not be asso
ciated with delinquent attitudes or behavior if it is bal-
| anced by intense social interest, and, conversely, if lit-
| tie social interest is present, a lesser degree of aggres
sive drive is required to result in delinquent behavior.
! Furthermore, not only are absolute degrees of impulse or
control lacking in predictive cogency, but absolute degrees
| of the resultant attitudes, inclinations, and behavior are
| similarly wanting in connotative power. Just as a single
j
Wechsler IQ score sheds little light on the particular
! scatter patterns that may produce it, a single measure of
j delinquency provides little information about the nature of
its components. An asocialization score, for example, may
i be the result of high motivational and inhibitory compo-
i
|nents or of low motivational and inhibitory components.
! The same principle applj.es to any given degree of behavior-
:al severity, which may reflect either high or low impulse
level balanced by either high or low control.
The second major implication of the third hypothe
sis pertains to the frequent assumption that delinquency is
linked with social position. Lower class culture is com
monly viewed as a generator of delinquent attitudes and be
havior (e.g., W.B. Miller, 1958). If the assumption that
delinquency is a more typical adjunct of lower class than
of middle and upper class conditions and values is a valid
j one, lower class delinquents would be expected to require
| less psychological motivation, balanced by fewer psycholog
ical inhibitory tendencies, than middle and upper class de-
j
linquents to reach any given level of asocial inclination
j or behavioral seriousness. As cited previously, Levy
i
j
| (1932) held that in the lower classes the enviornment may
I create delinquency in the child, while in the middle and
upper classes the child creates the delinquency in the en-
; viornment as an attempted solution to a neurotic problem.
24
Thus, it would seem to take a considerably stronger indi
vidual antisocial drive for a middle class boy to become as
delinquent as a lower class boy whose delinquency is fos
tered and facilitated by the environment.
CHAPTER III
METHOD
Operational Definitions of the Variables
Behavioral Seriousness
An important implication of the accumulated data in
the study of delinquency is that an "either-or" theory,
postulating a simple dichotomy between delinquent and non
delinquent personalities, clearly cannot provide adequate
conceptualizations for dealing with the phenomenon. Grada
tions and degrees of variation among delinquents must be
recognized, and appropriately related to the realities of
the sociocultural environment. Without such differential
tion, delinquency research must remain as restricted as med
ical research would be if it were limited to comparisons of
"healthy" populations with those described simply as "ill,"
with no consideration of the nature and severity of the
illness. Gough (1954) suggested that the systematic vali
dation of a measure of delinquency requires the derivation
of a sociologically defined continuum of delinquency, and
the checking of a psychometric index on samples representing
various positions along this continuum.
In the present study, an attempt was made to estab
lish a sociologically defined continuum of delinquency, as
25
suggested by Gough (1954), by means of an operational defi
nition of the "seriousness" of antisocial behavior. When
police contact with a youngster is made and recorded, the
police authorities have the legal and formal alternative of
either simply leaving it at that stage, "counselling and
releasing" the child, or of requesting the probation depart
ment to submit a petition to the juvenile court, implying
that the child needs sanctioning or help according to the
legal provisions of the state. The proportion of petitions
requested for particular offenses is the basis for the pres
ent index of the seriousness of those offenses. This meth
od of quantifying a "seriousness index" was suggested by
data compiled by McEachern (1963) in a study of the factors
related to disposition in juvenile police contacts in Los
Angeles County. The results of this study offer justifica
tion to certain limitations that should be considered in
its application. The technique makes use of what Robison
called a "conduct norm," which she described as a "rule"
supported by sanctions which reflect the value attached to
the norm by the normative group. (Robison, 1960) The
"resistance potential" of such a norm, i.e., the extent to
which the normative group resists its violation, if meas
urable, can provide a basis for offense classification.
Sellin (1938) suggested that "scales of penalties," one
way of measuring "resistance potential," would afford a
sounder theoretical basis for offense categorization than
27
the labels found in criminal codes. Similarly, Waller
(1955) advocated a classification system based on the type
of social value protected by the relevant conduct norm.
Robison (1960), in applying this concept to delinquency,
spoke of studying the proportion of cases brought to the
court in various communities in which the disposition was
dismissal or commitment, the two extremes in penalties.
A "conduct norm" definition of "seriousness," as
used in the present study, is established, in effect, by
those persons most thoroughly familiar with, and most real
istically representing, the cultural values and conduct
norms of the state or jurisdiction in question. The data
provided by McEachern (1963, p. 9) on the proportion of pe
titions requested for various offenses, based on a random
sample of 1,010 records drawn from the Central Juvenile
Index of the Los Angeles County Sheriff's Department, con
stitutes the norm for assessing the seriousness of offenses
in the present study. McEachern*s classification of of
fenses divided them into two major legal categories, juve
nile and adult offenses, and subdivided those into logical
categories, least through most serious, giving the number,
of cases and the proportion of petitions requested for each
offense. Certain modifications were made in adapting this
classification to the needs of the present study, in which
it was decided to select four offense categories as repre
sentative of positions on a seriousness continuum.
n' 2 81
: ]
Offense category 1: Delinquent tendencies.— This
i
;constitutes the least serious category in the present stu-
i
! j
dy, consisting of a variety of juvenile offenses, includingj
I truancy and other school violations, disorderly conduct,
runaway, curfew violation, improper companions, liquor law
I violations, glue inhalation, transience, and incorrigibili-
ity. While there is considerable variability in the speci-
ific activities involved in the violations included in this
|category, they all have a lower proportion of petitions re-
|quested than the next offense category, except for truancy
i j
I and incorrigibility. The high proportion of petitions re
quested for these two offenses is misleading, however, be- I
cause of the high recidivism factor necessary in order for ;
these violations to come to public attention. Truancy cas
es are usually referred by the school, and are not referred!
until or unless the school authorities are insistent that
action be taken. Incorrigibility cases are usually re
ferred by the parents, and generally require custody by the;
juvenile court because of the inability or unwillingness of;
the parents to deal with the youngster at home. Therefore,
jthe high proportion of petitions requested of the juvenile :
I court in these cases does not necessarily reflect an offi-
.
t i
icial judgment of the seriousness of the offense, but is
more likely to represent the inadequacies of school author-:
;ities and parents in dealing with the children. Delinquent
:tendencies are rated 1-3 by McEachern.
Offense category 2; Petty theft,— This category is
given a rating of 5 by McEachern, under adult offenses
which range from 4 to 7.
Offense category 3: Burglary.— Although this of
fense is given the same rating of 5, in McEachern's classi
fication, as petty theft, the proportion of petitions re
quested for burglary arrests, in the sample of 122, was
.36, appreciably higher than the proportion of .25 for
petty theft.
Offense category 4; Grand theft, auto (GTA).— This
offense is given a rating of 6 (more serious adult offens
es) in McEachern's classification, with a proportion of .61
petitions requested by the probation department.
Several considerations influenced the choice of the
four offense categories used in the present study, among
which the number of cases to be found in the probation de
partment's case movement was quite important, from a prac
tical point of view. For example, it might have been more
desirable to choose for the most serious offense category
one of the offenses rated 7 (most serious adult offenses)
by McEachern. However, his finding of only one case of
forcible rape, three cases of grand theft (other than auto
theft), twelve cases of narcotic possession and use, and
eighteen cases of aggravated assault among 1,010 records
made it apparent that there was little likelihood of
obtaining enough cases in this category to form an adequate
I sample for statistical analysis.
Psychometric Asocialization
i .
The psychometric index, against which the sociolog-!
i ically defined continuum of seriousness was checked, is the;
'Asocialization Index (ASI) of the Jesness Inventory. The
inventory consists of 155 true-false items, and yields
scores on 11 scales.
Three of the 11 scales were derived empirically
from item analysis, using criterion groups: Social Malad- |
justment (labeled "Delinquency Orientation" in the initial !
form of the inventory), Immaturity, and Value Orientation.
Seven scales were defined by means of a statistical cluster
analysis: Autism, Alienation, Manifest Aggression, With
drawal, Social Anxiety, Repression, and Denial. The final
scale— Asocialization— is based on a regression equation
which combines the information on attitude syndromes and
i
personality traits from the ten other scales into a single
index most predictive of delinquency. Since this index is
: determined by a motivational component modified by an in-
j hibitory component, the general score is interpreted as a
; measure of "impulse-control balance," and the separate com-;
ponents may be studied in their interrelationships.
The "impulse" component referred to in the present
study consists of the sum of the Social Maladjustment (de-
! linquent orientation) weighted raw score and the Social
Maladjustment (delinquent orientation) weighted raw score
and the Social Maladjustment special weight score, referred
: to as Sum^. The "control" component consists of the sum ofj
| weighted scores on scales 2,4,5,6,7,8, and 9, referred to
| ;
' as Surr^. Scales 3 (Immaturity) and 10 (Denial) are omitted
in the derivation of the Asocialization Index (ASI), which
I consists simply of Sum^ minus Sum2 . The improved predic-
' tive power resulting from a consideration of the interac
tion of motivational and inhibitory components over motiva-;
tional components alone fits the present theoretical orien-;
: I
tation, and makes the test particularly suitable for the
purposes of the present study. The score on the Asociali- ■
zation Index is interpreted by Jesness as reflecting "a
generalized disposition to resolve problems in social and
personal adjustment in ways ordinarily regarded as showing
a disregard for social customs or rules. ..." (1963, p.9X
The other 10 scales of the Jesness Inventory are
briefly described by Jesness as follows:
1. Social Maladjustment (6 3 items) Social Malad
justment refers to a set of attitudes associated with
unfulfilled needs, as defined by the extent to which an
individual shares the attitudes of persons who demon
strate inability to meet, in socially approved ways,
the demands of their environment.
2. Value Orientation (39. items) Value Orientation
refers to a tendency to hold values characteristic of
persons in the lower social classes.
3. Immaturity (45 items) By Immaturity is meant
the tendency to display attitudes and perceptions of
self and others which are usual for persons of a young-i
er age.
32
4. Autism {2 8 items) Autism refers to a tendency
in thinking and perceiving to distort reality according|
to one's personal desires or needs. j
5. Alienation (26 items) Alienation refers to the
presence of distrust and estrangement in a person's
attitudes toward others, especially toward persons re- \
presenting authority.
6. Manifest Aggression (31 items) Manifest Aggres
sion refers to an awareness of unpleasant feelings, es
pecially of anger and frustration, a tendency to react
readily with emotion, and perceived discomfort concern
ing the presence and control of these feelings.
7. Withdrawal (24 items) Withdrawal involves a j
perceived lack of satisfaction with self and others and:
a tendency toward passive escape or isolation from oth-;
ers.
8. Social Anxiety (24 items) Social Anxiety is de
fined as the perceived emotional discomfort associated
with interpersonal relationships.
9. Repression (15 items) Repression refers to the
exclusion from conscious awareness of feelings and emo
tions which the individual normally would be expected
to experience, or his failure to label these emotions.
10. Denial (20 items) The term Denial refers to the:
failure to acknowledge unpleasant events or aspects of :
reality normally encountered in daily living. (196 3,
pp. 8-9)
^Scale intercorrelations, in the present study as
well as in Jesness' data, are presented in Appendix I.
There is a growing interest in and use of the Jesness
;Inventory among workers in the field, and it has been used
successfully in differentiating delinquents from nondelin-
Iquents, in showing sensitivity to change under treatment,
i
and in predicting parole performance.
Intelligence
The choice of the Culture Fair Intelligence Test !
(Cattell & Cattell, 1960) as the measure of mental ability
in the current study was determined by several considera
tions, the most important of which were its practicality
and suitability for use with delinquent adolescents.'" The
size of the sample and the economic limitations of the stu
dy precluded the use of an individual test such as the
Wechsler, particularly since intelligence was not one of
the principle variables under investigation. Once it was
decided to use a group test, the choice was narrowed cons id-!
erably by the scarcity of group tests with adequate relia
bility and validity that are not either designed for older
age groups or highly verbal. The particular value of the
Cattell test is that it appears to be as saturated with "g"
as any standard verbal test, and yet as free of scholastic
and cultural influences as any standard performance test.
Thus, it would seem more likely to separate the individ
ual1 s general mental ability from the influences of educa
tional opportunity and socioeconomic status than most stand
ard tests, a vital consideration in a delinquent population
composed largely of lower-class boys with school problems.
The subtitle of the test, "A Measure of 'g'," appears to be
justified by the results of a number of cross-cultural stu-:
dies, as reviewed by Smith (Buros, 1959).
Scale 2, Form A, used in the present study, has
four subtests, all nonverbal and each involving a differentj
type of perceptual reasoning. The first three involve fam-j
iliar types of spatial relationships, series, classifica
tions, and matrices, and the fourth, called "conditions,"
involves a novel type of "topological" reasoning. The
testing time is 14 minutes. The manual contains extensive
information regarding reliability and validity, tables of
norms suitable to the requirements of different types of
subjects, and detailed administration and scoring instruc
tions. The test was standardized on 4,328 pupils, sampled
from various regions of the United States and Great
Britain.
Social position
The father's or guardian's occupation and educa
tional level were used to determine the socioeconomic sta
tus of each subject, according to Hollingshead's Two-Factor
Index of Social Position (ISP). Hollingshead's educational
scale is based upon the degree of skill required for the
job, and the prestige value assigned to the job by the
community.
The criteria and examples for educational and occu
pational scale ratings are provided in detail in a mimeo
graphed bulletin by Hollingshead, and summarized in Appen
dix J. In the present study, social position was dichoto
mized by combining Hollingshead1s classes 1, 2, and 3, here
designated as "higher" class, and combining Hollingshead's >
|
classes 4 and 5, designated "lower" class. j
Family status (FS)
In the present study, family status was a dichoto-
mous classification, designating either a united or a bro
ken home. The home was considered united only if the
child's natural parents were still alive and living togeth
er. The dichotomous classification is in agreement with
Monahan (1957), who concluded from research data concerning
the relationship between family status and delinquency that;
the fact of a break in his home is of more importance to
the child than the nature of the break.
The data recorded for each subject included the of
fense classification, test scores, social position, family
status, age, race, and previous IQ scores if any were
available.
Procedure
Two hundred boys, aged 13-17, confined in one of
two facilities of the Los Angeles County Probation Depart
ment, were selected on the basis of their offense records,
to be divided into the four offense categories chosen to
represent four points on the seriousness continuum. They
were selected for testing if they were between the ages of
13 and 17, if the most serious offense in their records
was in one of the four predefined categories, and if the
36
record contained the information necessary for classifying
social position and family status. If these criteria were
metf the full file was examined and the relevant data re
corded. The boys were tested in groups of 8 to 17, most
frequently 12 or 13, in sessions of approximately an hour.
Subjects were eliminated as candidates for testing if their
history included offenses of a nature so discrepant with
the research categories that their inclusion in the sample
would preclude the possibility of meaningful definition of
experimental groups. Examples of such offenses were narco
tics violations, armed robbery, forcible rape, and other
sex offenses.
The Sample
In an age range of 13 to 17, Table 1 reveals that
14 and 15 year olds constitute the bulk of the sample.
Table 1 reveals further that the total sample contains a
preponderance of boys from broken homes in the lower socio
economic strata.
Table 2 shows that the distribution of races in the
petty theft and burglary categories is similar to their
distribution in the total sample. The delinquent tenden-
dies category, however, appears to contain a disproportion
ate percentage of Caucasians over Negroes, while the auto
theft group shows a converse preponderance of Negroes over
Caucasians. Age distributions in offense categories
37
TABLE 1
GENERAL CHARACTERISTICS OF THE TOTAL N OF 200
Variable N Per cent of Total N
Race:
Caucasian 105 52.5
Negro 63 31.5
Mexican-American 30 15.0
Other (Japanese-American) 2 1.0
Age:
13 24 12.0
14 52 26.0
15 81 40.5
16 22 11.0
17 21 10.5
ISP:
Higher 49 24.5
Lower 151 75.5
FS:
United 69 34.5
Broken 131 65.5
38
TABLE 2
GROUP CHARACTERISTICS WITHIN OFFENSE CATEGORIES
Variable
OC I
(N:41)
N %
OC
(N:
N
II
55)
%
OC
(N:
N
III
52)
%
O.C
(N:
N
iva
52)
tb
Race:
Caucasian 27 6 6 28 51 29 56 21 40
Negro 6 15 20 36 13 25 24 46
Mex-Arner 8 19 6 11 10 19 6 12
Jap-Amer 0 0 1 2 0 0 1 2
Age:
13 3 7 10 18 7 13 4 8
14 8 20 16 29 15 29 13 25
15 18 44 14 26 24 46 25 47
16 3 7 9 16 4 8 6 12
17 9 22 6 11 2 4 4 8
ISP:
Higher 16 39 9 16 11 21 13 25
Lower 25 61 46 84 41 79 39 75
FS:
United 12 29 14 25 22 42 21 40
Broken 29 71 41 75 30 58 31 60
^ C is an abbreviation for Offense Category: OC I-
Delinquent Tendencies; OC II-Petty Theft; OC III-Burglary;
OC IV-GTA (Grand Theft, Auto).
t>The symbol H%" designates the percentage of the
subjects in that particular offense category who are, e.g.,
Caucasian, 13 years old, etc.
(Table 2) are similar to that of the total sample, except
for an apparent preponderance of 17 year olds in the delin
quent tendencies group (22% as opposed to 10.5% in the total
sample) and a scant representation of 17 year olds in the
burglary group (4% as opposed to 10.5% in the total sample).
Inspection of Table 2 reveals that the distribution of so
cial classes in the more serious offense categories shows
little deviation from that in the total sample,
but shows a disproportionate percentage of middle-class
boys in the delinquent tendencies group, and of lower-class
boys in the petty theft group.
Simple inspection of the descriptive sample charac
teristics and their interrelationships, in summary suggests
the possibility that certain sociological characteristics
may be associated with each of the offense categories in
the present sample:
1. Among the least serious delinquents, those boys
whose most serious recorded offense was in the delinquent
tendencies category, there appears to be a preponderance of
older, middle class Caucasians from united homes.
2. Among those boys whose most serious recorded
offense was petty theft, there appears to be a preponder
ance of younger, lower class minority group members from
broken homes.
3. Among those whose most serious recorded offense
was burglary, there is an apparent preponderance of younger
40
boys from united homes.
4. Among the most serious delinquents, according to
the present classification, those whose most serious offense
was auto theft, there appears to be preponderance of Negroes
from united homes in a lower socioeconomic class setting.
Statistical Techniques
Pearson product-moment coefficients of correlation
between each of the variables and all the others resulted
in the correlation matrix for the total sample summarized in
Table 5. Correlation matrices were calculated within each
of the 4 offense categories and within offense categories I
and II combined, and III and IV combined. Similar matrices
were calculated within each of the social position classifi
cations .
Each of the 4 offense groups was compared to each of
the others with respect to differences between them in means
of each variable, and t-tests were calculated for the signi
ficance of mean differences.
First-order partial correlations were computed be
tween social position and Sum^, and between social position
and Sum2 , with ASI held constant, within each of the 4 of
fense categories, and within categories I and II combined,
and III and IV combined.
The biserial coefficient of correlation was
^The author acknowledges the contribution of the
University of Southern California Computer Sciences Labora
tory and the assistance of its staff.
calculated between ASI and OC, as a check on the suitabili
ty of the product-moment coefficient with an artificially
dichotomized variable, but yielded a lower coefficient than
did the Pearson r, supporting the use of the product-moment
coefficient in a sample of this size.
CHAPTER IV
RESULTS
Experimental Results in
Terms"'of ""Hypotheses'
Hypothesis I, that the degree of sociologically de
fined seriousness of delinquent behavior is positively re
lated to the psychometric index of asocialization, can be
neither rejected nor supported by the experimental results.
Table 3 shows that the difference between OC I and
OC II, in ASI means, is significant beyond a .2 level of
probability, and the difference between OC I and OC IV (the
least and most serious classifications) is significant be
yond a .1 probability level. Because the limited range of
the sample on both the behavioral and psychometric serious
ness dimensions restricted the magnitude of differences to
be expected, a .05 level of confidence was not deemed nec
essary to report these results. The facts that the differ
ences are in the predicted direction, and that the differ
ence between the least and the most serious offense groups
is greater than between groups I and II in the predicted
direction, offer some support for the first hypothesis,
with the recognition that there is a substantial margin of
error in the relationship between the two variables.
43
TABLE 3
SIGNIFICANT MEAN DIFFERENCES BETWEEN SUBGROUPS
V t
*1
°Xi
x2 °x2
P
OC I vs,
OC II
IQ
3.1650 84.5120 4.0850 75.5370 3.9560 .01
AS I 1.6509 22.5850 .7130 24.5370 .6850 .20
OC I V S.
OC IV
AS I 1.8085 22.5850 .7130 24.7310 .6950 .10
OC II V S .
OC III
IQ
1.9681 75.5370 3.9560 80.8080 3.2170 .10
OC II vs.
OC IV
IQ
4.3633 75.5370 3.9560 86.5000 2.3570 .01
OC III vs.
OC IV
IQ
i
2.4109 80.8080 3.2170 86.5000 2.3570 .05
OC I & II
vs .
OC III & IV
IQ
1.9201 79.4110 2.8790 83.6540 2.0040 .10
ISP 1, 2, 3
vs .
ISP 4 & 5
IQ
2.2400 92.9590 20.3620 77.9270 24.5350 .05
44
Hypothesis II, that there is a closer correspond
ence between behavioral seriousness and psychometric asoci
alization than between behavioral seriousness and either
the motivational or inhibitory components of the Asociali
zation Index, likewise cannot be rejected by the experimen
tal results. Neither Sum^ nor Sum2 showed a relationship of
any significance to offense classification, in contrast to
the small but intelligible relationship to offense classifi
cation evinced by the ASI derived from the subtraction of
Sum2 from Sum-^.
The experimental results pertinent to Hypothesis
III are presented in Table 4. The prediction that, with
behavioral seriousness and psychometric asocialization held
constant, lower-class and middle- and upper-class delin
quents would differ in the relative magnitudes of the moti
vational and inhibitory components of the asocialization
index, was supported by the results. As can be seen in
Table 18, the relationship was found to be significant in
all but one of the offense classifications. However, the
second part of the hypothesis, the expectation that the
magnitude of motivational and inhibitory factors would de
crease with descending social position, at any level of
behavioral seriousness or psychometric asocialization, was
contradicted by the results. The positive sign of the sig
nificant coefficients indicates that the direction of the
relationship is in direct opposition to the predicted di
rection. With ASI partialled out statistically, and of
fense seriousness held constant experimentally, scores on
45
TABLE 4
PARTIAL CORRELATIONS BETWEEN SOCIAL POSITION (ISP) AND
MOTIVATIONAL AND INHIBITORY INDICES (SUMx AND SUM2),
WITH ASOCIALIZATION (ASI) HELD CONSTANT,
WITHIN OFFENSE CATEGORIES
oc ISP and Sumj ISP and Suni2
I
.415**a
.466**
II .179*k .179*
III -.094 (Non-Sign.) -.095
IV .182* .183*
Less Serious
(I & II) .282** .281**
More Serious
(III & IV) .149* .148*
a. ♦♦Significant at .01 level
b. ♦Significant at .05 level
46 !
Suxnj and Sum2 were found to rise rather than diminish with ,
lower social position.
Analysis of the results revealed a number of signi
ficant relationships worthy of mention, although they do
not bear directly on the experimental hypotheses. Most of
the statistical results referred to in the following sec
tion are presented in tabular form in Tables 5 and 6, and
in Appendices A through J .
Unhypothesized Findings
Offense Seriousness appears to be related to minor
ity group membership. The significant correlation coeffi
cients between race and offense category in the total sam
ple (Table 5) and in the lower classes (Appendix I) suggest
a tendency for reported offenses to increase in seriousness
with minority status.
Offense Seriousness appears to show a curvilinear
relationship with IQ, with the highest IQ means in the
least and most serious offense groups. Table. 3 shows that
the mean IQ is significantly higher in the delinquent ten
dencies group than in the petty theft group, and higher in
the GTA group than in the petty theft group beyond the .01
level of confidence, and higher in the GTA group than in
the burglary group beyond the .05 level of confidence.
Offense Seriousness appears to be related to Mani
fest Aggression scores in the higher class group (Appendix
H), and to Denial scores in the lower class group (Appendix
47
TABLE 5
SIGNIFICANT CORRELATIONS BETWEEN VARIABLES IN THE
TOTAL SAMPLE OF 200, EXCLUDING JESNESS
SCALE INTERCORRELATIONS
Age Race ISP OC IQ
Age -.223** -.181*
Race .451** .174* -.282**
ISP -.223** .451** -.278**
OC .174*
IQ .181* -.282** -.278**
SM .171* -.211**
VO -.176* .241** -.296**
Im -.161* .296** .317** -.329**
Au .169*
A1 -.191* .269** .233** -.276**
Wi -.220**
Re .287** .279** -.184*
De .159*
Sum j .191* -.247**
Sum2 .177* .215** -.258**
I), both at the .05 level of confidence, suggesting a ten
dency for MA to increase with seriousness of offense in the
higher class, and for De to increase with behavioral ser
iousness in the lower class group. Offense Seriousness is
related to Repression scores among less serious offenders
(Appendix F) at a .05 level of confidence, indicating more
repression among those in the petty theft group than among
those in the delinquent tendencies group.
Offense Seriousness shows a relationship to social
position among the lower classes (Appendix I), suggesting
that seriousness of delinquent behavior increases from low
to lowest social status. This relationship is significant
beyond the .05 level of confidence.
The relationship between race and social position
is significant beyond a .01 level of confidence, suggesting
that minority group membership is associated with low soci
al position among delinquents, and that the association is
more prominent among those committing adult offenses (OC
II, OC III, & OC IV) in the lower social classes.
The relationship between race and IQ is significant
beyond a .01 level of confidence, suggesting that IQ scores
tend to diminish with minority status. This relationship,
as that between race and social position, appears to stand
out more prominently among the more serious offenders in
the lower social strata.
49
Race and Jesness scales.— Coefficients significant
at the .01 level of confidence indicate a tendency for
scale scores to increase (high scores on the Jesness scales
are unfavorable) with minority status on Value Orientation,
Immaturity, Alienation, Repression, and Sun^. Coefficients
significant at the .05 level indicate a tendency for scores
to rise with minority status on the scales Social Malad
justment, Autism, and Sum^ (Table 5). Relationships be
tween race and Jesness scales appear to differ with social
position. Within the higher class, a coefficient signifi
cant at the .01 level indicates a relationship between race
and Autism score, not found among the lower classes, sug
gesting that among delinquent boys, higher class minority
group members show more autistic tendencies than do higher
class Caucasians.
Social position appears to be related to age among
delinquent boys in the present sample. Highly significant
coefficients suggest a decrease in age with descending so
cioeconomic status, particularly among higher class minor
offenders.
Highly significant coefficients suggest a rise in
IQ scores with ascending social position, particularly
within the more serious offense categories.
Social position and Jesness scales.— In the total
sample (Table 5), coefficients at the .01 level indicate
50
a relationship between social position and the Immaturity,
Alienation, and Repression scales, with these scale scores
rising (becoming less favorable) with descending socioeco
nomic status, particularly among the lower classes. On the
other hand, two relationships appear among the higher clas
ses that are not found at a significant level either in the
lower class group or in the total sample. In the higher
classes, coefficients significant at a .05 level indicate a
rise in Social Anxiety and a decrease in the ASI with de
scending social position. It is only in this group of
higher class delinquent boys, that ASI shows a significant
correlation with any other variable.
The relationship between age and social position
has already been described. Age shows some relationship
with family status as well, in the present sample. Among
lower class offenders, a coefficient significant at the .05
level shows a tendency for decreasing age to be associated
with increasing incidence of broken homes.
The relationship between age and IQ in the present
sample is similar to that between race and IQ. It is sig
nificant at the -05 level in the total sample, suggesting a
rise in IQ with increasing age, particularly among more
serious offenders at lower socioeconomic levels.
Age and Jesness scales.— Table 5 shows a tendency,
at the .05 level of confidence, for increasing age to be
51
associated with lower (more favorable) scores on Value
Orientation, Immaturity, and Alienation in the total sample.
The negative relationship between age and Immaturity scores
supports the validity of the Immaturity scale.
Family status is related to Manifest Aggression
scores, at a .05 level, only in two subgroups, and in oppo
site directions in those groups. In OC I, MA tends to in
crease with a greater incidence of united homes, while an
increase in MA appears to be associated with a greater in
cidence of broken homes in OC II.
IQ. and Jesness scales.— IQ is significantly related
to scores on all Jesness scales except Autism, Manifest Ag
gression, Social Anxiety, and the predictive ASI. All the
coefficients are significant at a .01 level, except those
between IQ and Repression and Denial, which reach only a
.05 significance level. All but one of the significant
correlations are negative, indicating a tendency for rising
IQ to be associated with more favorable scale scores. Only
Denial shows a positive relationship to IQ.
Validity of the Cattell test.— Among the 87 boys
whose files included previous IQ scores the correlation be
tween those scores and the present IQ scores on the Culture
Fair Intelligence Test was .456, a coefficient highly sig
nificant beyond the .01 level of confidence. As a validity
coefficient, this was higher than expected in view of the
nature of the population, the test, and the criterion
52
measures, which included a diversity of mental tests of
various kinds, given at various times, in various situa
tions, under unknown circumstances. In this context, the
validity coefficient of .456 may be taken as evidence of
the usefulness of this test in the present study as well as
in similar subsequent studies.
Intercorrelations of Jesness scales.— Appendix A
shows the scale intercorrelations obtained in the present
sample, alongside those obtained by Jesness in a sample of
1888 male delinquents and nondelinquents aged 10 through 18.
In view of the great differences in the size and nature of
the two samples, the present correlation coefficients are
remarkably similar to those obtained by Jesness (1963). As
in the Jesness sample, there is considerable relatedness
among''the scales SM, VO, Au, Al, and MA, and somewhat les
ser correlation among the scales SA, Wi, and MA. Re tends
to correlate with Im, but with nothing else. The fact that
VO tends to correlate rather highly with several scales,
but negatively with De, suggested to Jesness an association
with a yes-saying or plus-getting attitude. The intercor
relations shown in the inner box in Appendix A are those
for the scales derived from cluster analysis, which seem to
show considerable independence in both samples.
CHAPTER V
DISCUSSION OF RESULTS
The relationship between the behavioral dimension
of offense seriousness and the psychometric index of asoci-
alization is suggestive, but somewhat problematical. It is
not so substantial as to warrant the prediction of behavi
oral seriousness on the basis of Jesness ASI scores without
a great deal of error. The finding of this positive but
weak relationship raises several questions regarding the
inferences that may be drawn from it. First, it could sig
nify that the weakness of the relationship is inherent in
the nature of the variables. This interpretation would
take the experimental results at face value, as implying
that delinquency potential or antisocial inclination is
positively but tenuously related to the seriousness of its
behavioral manifestations. This interpretation, while log
ically reasonable, is unwarranted in view of the imperfec
tions and rudimentary stage of the methods of defining and
measurbing the variables. Indeed, the finding of even a
. \
slight positive relationship of some significance despite
those imperfections, suggests the possibility of a stronger
potential correspondence between the variables than is ap
parent here.
52
53
The second possibility, that the techniques of
identification and measurement of the variables are insuf
ficiently refined or sensitive to ascertain accurately the
degree of the relationship, is more tenable. The next
question, then, involves an evaluation of the techniques to
decide whether the major errors lie in the behavioral di
mension, the psychometric dimension, or in both. Analysis
of the data discloses greater shortcomings and sources of
error in the behavioral dimension.
Inspection of Table 6, which shows the means of ma
jor variables for each offense category, suggests that the
groups may not represent points on a psychological continu
um. Even when the differences between offense group means
are nonsignificant, their direction on several variables
contradicts the assumption of a linear ordinal scale. On
the Jesness indices, there appears to be greater congruity
between offense groups I and III, and between groups II and
IV, than between adjacent groups. This is due largely to
the interruption of continuity by the second group. The IQ
means, on the other hand, show a curvilinear pattern, with
the highest scores in the extreme offense categories. In
this case, it is the first group which breaks the linearity.
Only the social position (ISP) means reveal an apparent
consistency with the assumption of a continuous dimension,
social position descending regularly with offense serious
ness. These patterns suggest the possiblity that the
\
TABLE 6
MEANS OF MAJOR VARIABLES WITHIN OFFENSE CATEGORIES
OC I OC II OC III OC IV OC I & II OC III & IV
Age 5.171 4.596 4.596 4.865 4.905 4.731
ISP 3.805 4.037 4.154 4.154 3.937 4.154
FS 1.707 1.741 1.577 1.596 1.726 1.587
IQ
84.512 75.537 80.808 86.500 79.411 83.654
Si
51.537 53.630 52.750 53.731 52.736 53.240
s2
28.927 29.093 28.981 29.000 29.031 28.990
ASI 22.585 24.537 23.769 24.731 23.695 24.250
ui
iC^
55
present behavioral seriousness index may constitute a valid
sociological continuum, but not a psychological one. This
is not surprising, in that the index is based on a purely
sociological criterion, that of police disposition of juve
nile cases, which depends more on extrinsic conditions than
on intrinsic characteristics of individual offenders. The
implication is that police disposition, while perhaps ade
quately reflecting general social values, may not adequately
represent values meaningful in terms of the psychology of
delinquency.
The results of this study do not warrant the aban
donment of this type of sociological behavior index, but
indicate a need for its refinement. Further research may
profitably be aimed at investigating the relationship be
tween a behavioral seriousness index based on official so
cial judgment, as represented by police disposition in the
present study, and, for example, the judgments of delin
quent youngsters themselves, and/or the judgments of pro
fessional workers in correctional facilities, regarding the
seriousness of various offenses. Evaluations of seriousness
from these and other sources might be compared in terms of
their relative correspondence with each other and with the
Asocialization Index. The information derived from such
research would be useful for the further development and
refinement of a behavioral seriousness index in which
56
empirical, logical, and legal factors are taken into ac
count in making the appropriate modifications to bring so
ciological and psychological dimensions into closer corre
spondence .
The theoretical expectation that behavioral serious
ness would be more closely related to the Asocialization
Index score, representing impulse-control balance, than to
either the Sum^ {impulse component) or Suir^ (control com
ponent) score alone, is supported by the experimental re
sults . The value of taking both motivational and inhibi
tory components into consideration in arriving at a pre
dictive index is manifested in several ways. Perhaps the
most striking achievement of the subtractive score is its
apparent invulnerability to influence by intelligence and
the social desirability of responses. Despite the highly
significant relationship between intelligence and each of
the components of the ASI (Sum^ and Suir^) , IQ shows no re
lationship with ASI itself. Although greater intelligence
apparently enables subjects to present more desirable pic
tures of themselves on a number of test scales, it does not
appear to permit faking on the predictive index.
Another value of the subtractive index is that it
appears to represent more than do any of its separate com
ponents a quality more unique to the individual subject.
In a sense, the single score represents a configuration
rather than a point on a single dimension. In that sense,
57
It becomes more personally descriptive in the same way that
a profile on any multidimensional test presents a more
unique picture of an individual than does the score on any
single scale. While the relationship between ASI and of
fense seriousness is minimized by the shortcomings of the
seriousness index, it is greater than the relationship be
tween ASI and any other variable, showing no correlation
with age, race, ISP, or family status. As noted above,
single scale scores, including scores on the motivational
and inhibitory components of the ASI, are significantly re
lated to these other variables, but not to offense serious
ness. It may be that the ASI is relatively immune to in
fluence by sociological and ethnic factors, correcting for
them as it does for intellectual ability to "fake good,"
but reflects only what it purports to measure, the antiso
cial inclination of the individual child as reflected in
his antisocial behavior, regardless of his environmental
setting. If this is borne out by similar subsequent stu
dies with other delinquent populations, it points to the
usefulness of the index for assessing delinquency disposi
tion in youngsters of any socioeconomic, cultural, ethnic,
or environmental background, and of a wide range of age and
intellectual ability.
The theoretical expectation of the third experimen
tal hypothesis, that with behavioral seriousness and psy
chometric asocialization held constant, lower-class and
middle- and upper-class delinquents could be differentiated
by the relative magnitudes of the motivational and inhibi
tory components contributing to the asocialization index,
was supported by the experimental results. The fact that
the direction of the difference was contrary to expecta
tions however, requires re-examination of the theoretical
premises involved in the prediction. The direction of the
hypothetical prediction rested on the assumption that the
motivational and inhibitory indices of the Jesness Inven
tory represent intrinsic rather than extrinsic qualities of
the subject. It was reasoned that if the frequently made
assumption that delinquency is a more typical adjunct of
lower-class conditions and values is valid, the inner pro
pulsions and prohibitions of lower-class delinquents need
not be as intense as those of middle- and upper-class de
linquents to result in a disposition toward or in actual
delinquent behavior. If cultural and environmental factors
in the lower classes facilitate antisocial behavior, such
behavior need not be energized by inner promptings in these
classes as intensely as in the middle and upper classes,
where there are more stringent cultural and environmental
deterrents to such behavior. The question of the nature
of "inner" and "outer" determinants, and the degree to
which they are represented by the test scores, is the cru
cial issue here. The experimental results contradict the
assumption that the motivational and inhibitory test
59
indices represent inner determinants, but point rather to a
more overt interpretation of the scores. If Sum^ and Sum2
are understood as reflecting overt propensities resulting
from a variety of influences, rather than as representing
the covert motivational influences themselves, the experi
mental results are consistent with the theoretical hypoth
esis. The higher motivational and inhibitory components
found in the lower-class subjects, with equal degrees of
behavioral seriousness and psychometric asocialization, may
be interpreted as merely reflecting proclivities resulting
from a combination of endogenous and exogenous influences.
This appears to be the sounder interpretation of the
scores, and is in fact more congruent with the nature of
the test. The investigator was perhaps too easily misled
by the terms employed by Jesness. "Motivational" and "in
hibitory," like "impulse" and "control," are terms with
greater psychodynamic than sociogenic connotation, and
therefore inappropriately applied to scores derived from an
objective, highly structured test consisting of stimulus
content designed to be highly specific and relevant to the
social behavior to be predicted. More neutral terms, per
haps "positive" and "negative" components, referring to
their empirical relationships to the incidence of delin
quent and nondelinquent behavior, might more clearly de
scribe Sums 1 and 2.
The experimental results offer striking support for
60
the preponderance of current evidence against the proposi
tion that low intelligence is causally related to delin
quency. Most theorists find the relationship between IQ and
delinquency a spurious one, determined mainly by social fac
tors. This view is exemplified by Robison (1960), who rea
soned that IQ scores are related to educational opportuni
ties, educational opportunities to socioeconomic and cultur
al factors, and socioeconomic and cultural factors to the
likelihood of official apprehension as a delinquent, which
is greatest in the most deprived social groups. The pres
ent results are congruent with, and add another facet to,
those that contradict a direct relationship between IQ and
delinquency. Most of the studies referred to are designed
to differentiate delinquents from nondelinquents, frequent
ly finding significantly higher IQ scores among nondelin
quents, but refuting the causal implication of the differ
ence. The present study was concerned exclusively with the
differentiation of delinquent subgroups. Since the experi
ment involved manipulation of the variable by testing it at
several points, and because it was not feasible to assemble
an adequately matched group of nondelinquent boys that
could be considered a "zero point" on a delinquency scale,
the inclusion of a nondelinquent group was deemed unneces
sary. With this focus on differentiation among delin
quents, the relationships between IQ, delinquent behavior,
and sociocultural factors appeared to stand out more
61
clearly in some respects. While IQ was significantly re
lated to social position, to ethnic identification, and to
every scale (except Au) of the Jesness, it showed no rela
tionship either to the behavioral seriousness dimension or
to the ASI, intended to predict antisocial inclination. In
essence, the present results revealed IQ to be significant
ly and linearly related to almost every variable except
those pertaining directly to delinquency, either the behav
ioral or the psychometric dimension. The significant dif
ferences in IQ found among offense categories, as noted
previously, suggest the possibility, however, that although
there is no linear relationship between IQ and offense ser
iousness, there may be a curvilinear one. The fact that
the highest IQ means were found in the least serious and in
the most serious offense categories is a provocative source
for speculation, but is not adequately accounted for either
by the present data or by other research findings.
The relationship between age and social position
found in the present study is consistent with and suppor
tive of similar findings in previous studies. The fact
that conflicts with the law occur at an earlier age in the
lower social strata has been noted by Glaser and Rice
(1959), who asserted that economic conditions have markedly
contrasting relationships to crime for different age groups,
owing largely to the importance of age role differences in
American society. Cloward and Ohlin (1961) interpreted the
62
greater youthfulness of lower-class offenders in terms of
their theory of differential opportunity systems. Their
studies of "illicit means" in lower class culture led to
the observ ation that
delinquent role preparation and role performance may be
integrated even at the "play-group" stage of illegiti
mate learning. The child has an opportunity to actual
ly perform illegitimate roles because such activity
finds support in his immediate neighborhood milieu. The
rewards— monetary and other— of successful learning and
performance are immediate and gratifying at each age
level.
Another observation of Cloward and Ohlin, regarding the
prestige ordering of criminal activities, is confirmed by
and may account for the present finding that the relation
ship between age and social position in the total sample
stands out more significantly in the petty theft group than
in the other offense categories. Cloward and Ohlin singled
out petty theft, in concurrence with the earlier observa
tion of Shaw (1933), as an example of a particular type of
criminal activity looked upon with contempt by older delin
quents, among whom there is a definite stigma attached to
any form of petty stealing.
The significant relationship between age and family
status among lower-class youths in the present study is
consistent with previous findings, but cannot be interpre
ted etiologically because of unavoidable sampling biases of
an unknown degree. The tendency for the incidence of bro
ken homes to increase with decreasing age in the present
63
sample has been noted also by Toby (1957), who attributed
the greater observed incidence of broken homes among pre
adolescent than among adolescent delinquents to a stronger
relationship between family disorganization and delinquency
for the younger children. Despite the logical feasibility
of a causal interpretation of this data, however, it is im
possible to ascertain the degree to which the data reflect
an artifact, associated with the common police policy of
refraining from taking legal action against younger chil
dren unless the family situation is regarded as unfavorable
enough to require the supervision of the juvenile court.
Three scales of the Jesness showed significant re
lationships to offense classification, two of which ap
peared to be influenced by social position (OC with De and
MA), and one (OC with Re) by the specific offense classifi
cation, none appearing significantly in the total sample.
All three relationships have etiologic implications that
warrant discussion.
The tendency for Denial scores to increase with of
fense seriousness only in the lower classes, together with
the finding that Denial increases with IQ only in the low
er classes, is consistent with several sociological studies
of lower-class values. W.B. Miller (1958), for example,
discussed "smartness," conceptualized as "capacity to out-
. smart, outfox, outwit, dupe, 'take,' or 'con' another or
others," as a long standing and highly valued tradition in
64
lower-class culture. The present results offer support for
Miller's observation, suggesting that Denial, which is as
sociated in the Jesness Inventory with a plus-getting or
"conning" attitude, is a more striking attribute of lower-
class than of higher-class delinquents, and more apt to be
used among lower-class offenders who are both brighter and
more serious in their delinquent behavior.
The tendencies for Manifest Aggression scores to
increase with offense seriousness only in the higher-class
group, and for Social Anxiety to rise with social position
only in the higher-class offenders, are consistent with so
ciological studies of differences between lower and middle-
class values. Most studies find that middle-class values
place great emphasis on being lawabiding, while lower-class
culture tends to foster delinquency. If this difference is
valid, middle-class delinquents should be expected to suf
fer more discomfort than do lower-class delinquents, either
as part of the propulsion toward delinquency, a reaction to
it, or both. Cohen (1955), in discussing the negativistic
quality of the delinquent subculture, made the following
observation, which is relevant here:
The delinquent's conduct is right, by the stand
ards of his subculture, precisely because it is wrong
by the norms of the larger culture. . .he will often
assure us. . .sometimes with a touch of glee or even
pride, that he is "just plain mean."
The pride in negativism or antisocial behavior
among lower-class offenders, as opposed to the guilt likely
to be associated with illicit activities in the middle and
upper classes, may account for the fact that among delin
quents in the present sample, social anxiety rises with so
cial position only in the higher classes. It may similarly
elucidate the increase in Manifest Aggression scores with
increasing offense seriousness only among the higher clas
ses, if the Manifest Aggression scale is understood to refer
to awareness of anger and frustration, lability, and per
ceived discomfort concerning the presence and control of
these feelings. Thus, the increase in Manifest Aggression
with behavioral seriousness among higher class offenders
may be viewed in terms of the causative role of MA in gen
erating delinquent behavior of increasing severity, in
terms of MA as a reaction to antisocial behavior among
middle-class delinquents, or both.
The difference between the delinquent tendencies
and petty theft groups in the degree to which normally ex
perienced feelings are excluded from awareness, as re
flected by the higher Repression scores in the petty theft
group, is explicable in terms both of the nature of the of
fenses, and the characteristics of the offenders. Offen
ses in the delinquent tendencies category differ from of
fenses in the other categories more than the other three
offense categories differ from one another in quite an im
portant respect. That is, the offenses comprising the de
linquent tendencies classification are for the most part
66
not crimes per se, in that they are largely dependent on
the youth of the offenders, and not considered criminal be
havior in adults (e.g., truancy, runaway, curfew violation,
incorrigibility), while petty theft, burglary, and grand
theft are considered criminal activities per se, at any
age. Therefore, the meanings of delinquent tendencies of
fenses to the individual offenders are likely to vary even
more than do the personal meanings of the other offenses.
While theft and burglary have in common a manifest, univer
sally acknowledged criminal character, delinquent tenden
cies violations are frequently expressions of various kinds
of disturbance (often involving family and school con
flicts) in which criminal intent need not be present. This
is supported in the data by the tendency for boys in the
delinquent tendencies group to score lower on ASI (delin
quency inclination) than those in the petty theft group.
Differences in the extrinsic composition of the of
fense groups, as well as in the intrinsic nature of the of
fenses, are apparent in the present sample. The mean IQ in
the delinquent tendencies group, for example, is signifi
cantly higher than the mean IQ in the petty theft group.
It will be recalled that inspection of group characteris
tics revealed variations between the two offense groups in
age, ethnic identification, and social position. General
ly, the delinquent tendencies group, as compared with boys
in the petty theft group, appear to be older and more
67
intelligent, have lower antisocial inclination as measured
by the ASI, higher social position, and a higher proportion
of Caucasians, all attributes which might tend to be con
ducive to greater discomfort and conflict in connection
with delinquent behavior than might be experienced by mem
bers of the petty theft group, even though the offenses con
cerned may be of a less serious nature from a social stand
point. These may be the factors reflected by the greater
awareness of unpleasant feelings indicated by the lower Re
pression scores in the delinquent tendencies group, in the
same way that similar factors may be reflected by the
greater awareness of discomfort indicated by the rise in
Manifest Aggression scores with offense seriousness in the
middle and upper social levels.
A general aspect of the experimental results that
merits discussion is the pervasive tendency for relation
ships between variables to occur either exclusively or more
prominently among lower class offenders. The significant
relationships found exclusively among the lower-class boys
are: IQ with age, IQ with race, and IQ with Denial, Imma
turity, Social Maladjustment, and Repression; age with fam
ily status, Value Orientation, Social Maladjustment, Aliena
tion, Manifest Aggression, Denial, and Sums 1 and 2; social
position with offense seriousness. Alienation, and Repres
sion; and offense seriousness with Denial. None of these
relationships appear at significant levels in the higher
class segment of the sample. The general question engen
dered from this difference between the social class groups
in the number of significant findings they manifest is
whether the source of the difference lies mainly in the na
ture of the variables or in the nature of the samples. The
former alternative, that certain characteristics of lower-
class boys tend to enhance the interrelationships among a
number of factors or influences pertinent to them, has in
teresting theoretical implications. Such an interpretation
would be consistent with and supportive of the school of
thought typified by Levy (19 32) and Johnson (1959), which
distinguishes between "individual" or "neurotic" delinquency
and "sociologic" or "milieu" delinquency. In the former,
said to be more common in the middle class, individual psy
chodynamics are given primary importance, while cultural
and environmental factors are imputed to be the major deter
minants of the latter type of delinquency, more prevalent in
the lower social strata. The present finding could be in
corporated in this theoretical framework, by attributing the
scarcity of significant general relationships between vari
ables in the middle and upper-class offenders to the greater
individuality of their psychodynamics. Statistical analysis
of the data, however, failed to support this interpretation,
since standard deviations showed less variability among the
middle and upper-class youngsters both in IQ and in person
ality scales. The alternative interpretation, that the
differences between class groups in the number of signifi
cant relationships found between variables stems from the
nature of the sample, rather than from the nature of the
variables, appears more tenable. The most obvious differ
ence in the nature of the samples is their size. The higher
class group, with 49 subjects, is less than one third the
size of the lower-class sample, with 151 subjects. The dif
ference in sample sizes is reflected in the larger standard
errors of means in the higher-class group, pointing to lower
reliability and greater error of mean scores despite the lo
wer population dispersion. This increased margin of error
in the sample statistics would tend to diminish the signifi
cance of correlation coefficients or mean differences in the
middle and upper-class group.
The difference in variability between the samples,
as indicated by a comparison of standard deviations of
means, is another possible source of the discrepancy at is
sue. It has been noted by Block (1960) that the likelihood
of significant findings is sharply reduced if, as in the
middle and upper-class sample here, many of the measures em
ployed show relatively little dispersion because of sample
homogeneity. He reasoned that when the dispersion of a mea
sure decreases because of the homogeneity of the sample to
which it is applied, its reliability within the sample de
creases, and the likelihood is lessened of a significant
association between the measure and an independent variable.
70
It may be concluded from the general results of the
study that further research aimed at establishing and im
proving meaningful, quantitative dimensions, and unifying
theoretical concepts, in the study of delinquency, may prof
itably be continued along similar lines. At the present
stage of conceptual chaos and diagnostic and prognostic im
potence, serious efforts to refine measurement and to better
organize etiologic thinking may offer potentially rich
rewards.
CHAPTER VI
SUMMARY
An important implication of the accumulated data in
the study of delinquency is that an "either-or" theory,
postulating a simple dichotomy between delinquent and non
delinquent personalities, cannot provide adequate concepts
for dealing with the phenomenon. Gradations and degrees of
variations among delinquents must be recognized and related
to the realities of the socio-cultural environment. The
major goals of the present study were the validation of a
measure of delinquency, by checking a psychometric index on
samples representing positions along a sociologically de
fined continuum of the seriousness of delinquent behavior,
and the investigation of relationships between these dimen
sions within the framework of a unifying theoretical formu
lation.
Seriousness was predefined operationally, by the
sociological criterion of the proportion of petitions sub
mitted to the juvenile court for specific offenses in Los
Angeles County. The psychometric dimension was the Asoci-
alization Index (ASI) of the Jesness Inventory, an objec
tive test designed for use with adolescent delinquents,
consisting of 155 true-false items, and 10 personality and
71
attitude scales. The predictive ASI, intended to assess
antisocial inclination, was derived from the Jesness Inven
tory by the subtraction of an inhibitory component (Sun^)
from a motivational component (Sum^), each of which is
based on weightings of the other scales derived from re
gression equations.
The properties of the ASI and its motivational and
inhibitory components are appropriate to the theoretical
orientation of the study, which conceptualizes "impulse-
con trol balance" as the mediating variable relating motiva
tional stimuli of all kinds (constitutional, sociogenic,
and psychogenic) to the behavioral manifestations of anti
social responses.
The hypotheses of the present study stated that (a)
the sociologically defined "seriousness" of delinquent be
havior is positively related to the psychometric index of
asocialization, (b) there is a closer correspondence be
tween seriousness and asocialization than between serious
ness and either the motivational or inhibitory components
of asocialization, and (c) with equal degrees of behavioral
seriousness and psychometric asocialization, "lower-class"
and "higher-class" delinquents can be differentiated by the
relative magnitudes of motivational and inhibitory compo
nents. The third hypothesis was intended not only to test
the theory that "impulse-control balance," rather than the
degree of impulse or of control alone, is the mediating
73
process underlying delinquency, but also to test the fre
quent assumption that lower class culture and milieu gener
ates delinquent attitudes and behavior.
Subjects were 200 delinquent boys, aged 13-17, con
fined in two facilities of the Los Angeles County Probation
Department. They were selected on the basis of offense re
cords, and divided into four categories intended to repre
sent four positions on a continuum of behavioral serious
ness. Data regarding social position and family cohesive
ness was recorded, and each subject was tested on the
Jesness Inventory and the Cattell Culture Fair Intelligence
Test. The first two hypotheses were tested by correlation
al techniques and tests of significance of mean differences
between offense groups, and the third hypothesis was tested
by correlational techniques and tests of significance of
mean differences between offense groups, and the third hy
pothesis was tested by first-order partial correlations
within each offense category.
Results
(a) A positive but weak relationship was found be
tween behavioral seriousness and ASI. The major source of
error minimizing the significance of the relationship was
attributed to the limited range of the offense classifica
tion, and refinements of the seriousness index were sug
gested. (b) Behavioral seriousness showed closer corre
spondence to ASI than to either Sumi or Sum2, which was
74
interpreted as supportive of the theoretical formulation.
(c) With behavioral seriousness held constant by experi
mental design, and ASI partialled out statistically, lower-
class and higher-class delinquents differed significantly
in the relative magnitudes of Sum^ and Sum2 , which was in
terpreted as supportive of the third hypothesis. The di
rection of the difference indicated that the motivational
and inhibitory components of the Jesness Inventory repre
sent extrinsic propensities rather than intrinsic motiva
tional influences. (d) A number of unhypothesized findings
emerged, regarding relationships between social position,
ethnic identification, age, family status, IQ, Jesness
scales, and delinquent behavior, which were for the most
part consistent with other relevant research data.
Problems for future research were suggested, and it
was generally concluded that the present line of endeavor
to improve measurement and develop the concept of "impulse-
control balance" as the mediating process in delinquency
constitutes a potentially rewarding direction for further
study.
APPENDIXES
.A
75
APPENDIX A
INTERCORRELATIONS OF JESNESS INVENTORY SCALES
SM VO IM Au Al MA Wi SA Re De
SM: J [esness] .78 .38 .71 .66 .65 .48 .18 .13 -.51
Rfochlin] .82** .37** .73** .65** . 70** .41** .22** .14 -.53**
VO: J .78 . 30 .72 .81 . 83 .43 .14 -.02 -.53**
R .82** .33** .70** .78** . 79** .39** .18* . 06 -.70
Im: J . 38 .30 .48 .35 .14 .25 .06 .59 -.14
R
.37**
.33** .44** .35** .16* .28** .04 .61** -.10
Au: J .71 .72 .48 .58 .63 .38 .14 .09 -.54
R .73** . 70** . 44 ** .52** .62** .34** .11 .13 -.56**
Al: J .66 .81 .35 .58 .58 .24 -.15 .06 -.57
R .65** .78** .35** .51** .48** .23** -.18** .11 -.55*
MA: J .65 .83 .14 .63 .58 .38 .25 -.18 -.66
R . 70** .79** .16* .62** .48** .30** .27** -.04 -.60**
Wi: J .48 .43 .25 . 38 .24 .38 .51 .03 -.59
R .41** . 39** .28** .34** .23** .30** .40** .09 -.54**
SA: J .18 .14 .06 .14 -.15 .25 .51 -.16 -.33
R .22** .18* .04 .11 -.18* .27* .40** -.03 -.20**
Re: J .13 -.02 .59 .09 .06 -.18 .03 -.16 .18
R .14 .06 .61** .13 .11 -.04 .09 -.03 .16*
De: J -.51 -.70 -.14 -.54 -.57 -.66 -.59 -.33 .18
R -.53** -.70** -.10 -.56** -.55** —; 60** -.54** -.20** .16*
'j
-------- C T i
APPENDIX B
SIGNIFICANT CORRELATIONS BETWEEN VARIABLES WITHIN OC I
Age Race ISP FS IQ
SM .320*
VO .322* .422** -.345*
Im .416** -.313*
Au .444***
Al .525**
MA -.361*
Re .423**
De .330*
S± .366* -.326*
S2 .362* .396**
APPENDIX C
SIGNIFICANT CORRELATIONS BETWEEN VARIABLES WITHIN OC II
Age Race ISP FS IQ
Age -.330*
Race .477**
ISP -.330* .477** -.387**
IQ -.387**
VO .380**
Im .497** -.292*
Al .382**
MA .282*
Re .396**
s2
.307*
o o
APPENDIX D
SIGNIFICANT CORRELATIONS BETWEEN VARIABLES WITHIN OC III
Age Race ISP FS
IQ
Age -.283*
Race .608**
SM -.350*
VO -.350*
Im .297* -.319*
Al -.311*
Wi -.289*
SA -.314*
S1
-.374**
S2
-.330*
vo
APPENDIX E
SIGNIFICANT CORRELATIONS BETWEEN VARIABLES WITHIN OC IV
Age' Race ISP
IQ
Age .370**
Race .471** -.422**
ISP .471** -.432**
VO .307* -.303*
Im .323* -.373**
Al .383** -.417**
Re .377** .277* -.293*
S1
.276* -.283*
S2
.291* -.328*
oo
o
APPENDIX F
SIGNIFICANT CORRELATIONS BETWEEN VARIABLES WITHIN OC I & II
(LESS SERIOUS OFFENSES)
Age Race ISP OC IQ
Race .358** -.330**
ISP -.304** .358** -.294**
SM .258*
VO -.206* .353** .310** -.278**
Im .318** .424** -.320**
Au .24 8*
Al .285** .372** -.226*
Wi -.264*
Re .323** .338** .242*
51 .284** .230*
52 .323** .280** -212*
APPENDIX G
SIGNIFICANT CORRELATIONS BETWEEN
(MORE SERIOUS
VARIABLES WITHIN OC
OFFENSES)
III & IV
Age Race ISP
IQ
Age .232*
Race .520** -.257**
ISP .520** -.296**
SM -.299**
VO -.328**
Im .276** .206* -.350**
Au -.236*
Al -.227* .254** -.354**
Re .270** .243*
De .207*
S1
-.324**
S 2
-.325**
00
ro
APPENDIX H
SIGNIFICANT CORRELATIONS BETWEEN VARIABLES WITHIN THE
HIGHER SOCIAL POSITION GROUP
Race ISP OC
IQ
Age -.398**
VO -.343*
Im .304*
Au .430** -.298*
Al -.347*
MA .289*
Wi -.294*
SA .358*
S1
-.297*
S2
\
-.365*
ASI -.305*
tkt
i * >
APPENDIX I
SIGNIFICANT CORRELATIONS BETWEEN VARIABLES WITHIN THE
LOWER SOCIAL POSITION GROUP
Age Race ISP OC
IQ
Age .176*
Race .414** .232** -.253**
ISP .414** .192*
FS -.189*
SM -.168* -.183*
VO -.214** .229** -.259**
Im .193* -.273**
Al -.209* .274** .209* -.212**
MA -.168*
Wi -.209*
Re .216** .194* -.193*
De .168* .176* .163*
s . i
-.193* -.212**
s2
-.211** .198* -.213**
00
. ■ . ■
APPENDIX J
HOLLINGSHEAD'S TWO-VACTOR INDEX
OF SOCIAL POSITION (ISP)
Educational Scale
Years of School Completed
Graduate professional training (M.A.:
Ph. D.; LL. B.)
Four-year college graduation (A.B.: B.S.)
Partial college or business school training
(1-3 years)
High school graduation
Partial high school (10-11 years of school)
Junior high school (7-9 years of school)
Less than seven years of school
Scale Value
1
2
3
4
5
6
7
Occupational Scale
Occupation Scale Value
Executives and proprietors of large
concerns, and major professionals.
Managers and proprietors of medium-sized
business and lesser professionals.
Administrative personnel of large concerns,
owners of small independent businesses,
and semi-professionals.
Owners of little business, clerical and sales
workers, and technicians.
Skilled workers.
Semi-skilled workers.
Unskilled workers.
4
5
6
7
86
The scale value for education is multiplied by a
factor weight of 4, and the scale value of occupation is
multiplied by a factor weight of 7. The sum of these two
figures is the ISP score. For example, John Doe is the man
ager of a Safeway store; he completed high school and one
year of business college. He is scored as follows:
Factor Scale Score Factor Weight Score x Weight
Occupation 3 7 21
Education 3 4 12
Index of Social Position Score 33
When the ISP score is calculated, the individual may
be stratified either on the continuum of scores or into a
"class," as follows:
Scores Social Class
11-17 I (Highest)
18-31 II
32-47 III
48-63 IV
64-77 V (Lowest)
REFERENCES
» .
87
REFERENCES
Aichhorn, A. Verwahrloste jugend. Vienna: International
Psychoanalytischer Verlag, 1925. English translation;
Wayward youth. New York: Viking Press, 19 35.
Alexander, F., & Staub, H. The criminal, the judge, and the
public. Glencoe, 111.: Free Press, 1956.
Allchin, W. H. Some positive aspects of delinquent behav
ior. Brit. J. Criminol., 1962, 3^ 38-46.
American Law Institute. Model penal code. Washington,
D.C.: American Law Institute, 1954 & 1959.
Bandura, A., & Walters, R. H. Adolescent aggression. New
York: Ronald Press, 1959.
Bender, Lauretta. Aggression in childhood. In Round table
discussion. Airier. J. Orthopsychiat., 1943, 14, 384-389.
Bennett, Ivy. Delinquent and neurotic children. New York:
Basic Books, 1960.
Berg, I. A. Measuring deviant behavior by means of deviant
response sets. In Berg, I. A. & Bass, B. M. (Eds.),
Conformity and deviation. New York: Harper, 1961.
Pp. 328-379.
Bergler, E. Psychopathic personalities are unconsciously
propelled by a defense against a specific type of psy
chic masochism— malignant masochism. Arch, crim.
Psychodyn. , 1961, 4_, 416-434.
Block, J. On the number of significant findings to be ex
pected by chance. Psychometrika, 1960, 2j^, 369-380.
Buros, 0. K. (Ed.). The fifth mental measurements yearbook.
Highland Park, N.J.: Gryphon, 1959, 473-474.
Burt, C. The young delinquent. London, 1925.
Carr-Saunders, A. M., Mannheim, A. M., & Rhodes, E. C.
Young offenders: an enquiry into juvenile delinquency.
Cambridge: Cambridge Univ. Press, 1942.
Cattell, R. B., & Cattell, A. K. S. Culture fair intelli
gence test: a measure of "g": Scale 2. Champaign,
88
89
111.: Institute for Personality and Ability Testing,
1960.
Cleckley, H. The mask of sanity. St. Louis: C. V. Mosby
Co. , 1955.
Cloward, R. A. & Ohlin, L. E. Delinquency and opportunity.
Glencoe, 111.: Free Press, 1955.
Cohen, A. K. Delinquent boys: the culture of the gang.
Glencoe, 111.: Free Press, 1955.
Dentler, R. A. & Monroe, L. J. Social correlates of early
adolescent theft. Amer. sociol. Rev., 1961, 26, 733-
743.
Doctor, R. F. & Winder, C. L. Delinquent versus non
delinquent performance on the Porteus qualitative maze
test. J. consult. Psychol., 1954, 1J3, 71-73.
Eissler, K. R. Some problems of delinquency. In K. R.
Eissler (Ed.), Searchlights on delinquency. New York:
Internat. Univ. Press, 1949.
Epps, P., & Parnell, R. W. Physique and temperament of wo
men delinquents compared with women undergraduates.
Brit. H. med. Psychol., 1952, 25_, 249-255.
Erikson, E. H. The problem of ego identity. J. Amer.
psychoanal. Assn., 1956, £, 56-121.
Eskin, L. D. A study of some possible connections between
criminal behavior and perceptual behavior. Unpublished
doctoral dissertation, New York University, 1960.
Eysenck, H. J. The dimensions of personality. London:
Routledge & Kegan Paul, 1947.
Eysenck, H. J. The scientific study of personality. London
Routledge & Kegan Paul, 1952.
Falk, G. J. The role of social class differences and hori
zontal mobility in the etiology of aggression. J. educ.
Sociol., 1959, 33, 1-10.
Fenichel, 0. The psychoanalytic theory of neurosis. New
York: Norton, 19 45.
Ferenczi, S. The unwelcome child and his death instinct.
Internat. J. Psychoanal., 1929, 10, 125-130. Also in
Ferenczi, S. Final contributions to the problems and
90
methods of psycho-analysis. New York: Basic Books,
1955. Pp. 102-107.
Fontanesi, M. & Zilli, E. Les 17-chetosteroids urinaires
dans un groupe de delinquants auteurs de viol avec vio
lence. Arch. Psicol. Neurol. Psichiat., 1961, 22, 147-
153.
Freud, Anna. Certain types and stages of social maladjust
ment. In K. R. Eissler {Ed.), Searchlights on delin
quency . New York: Internat. Univer. Press, 1949.
Freud, S. Repression. (1915) In S. Freud, Collected pa
pers . London: Hogarth Press, 1953 (a), £, 84-97.
Freud, S. Some character-types met with in psycho-analytic
work. (1915) In S. Freud, Collected papers. London:
Hogarth Press, 1953 (b) , jJ, 318-344.
Freud, S. The unconscious. (1915) In S. Freud, Collected
papers. London: Hogarth Press, 1953 (c), £, 98-136.
Freud, S. The ego and the id. (192 3) Translated by Joan
RiviereT London: Hogarth Press, 1927.
Freyham, F. A. Psychopathology of personality functions in
psychopathic personalities. Psychiat. Quart., 1951, 25,
469-474.
Glaser, D. & Rice, K. Crime, age, and employment. Amer.
sociol. Rev., 1959, 2A.t 679-686.
Glover, E. H. The roots of crime. New York: Internat.
Univ. Press, 1960.
Glueck, S. & Glueck, Eleanor. Unravelling juvenile delin
quency. New York: Commonwealth Fund, 1950.
Glueck, S. & Glueck, Eleanor. Predicting delinquency and
crime. Cambridge, Mass.: Harvard Univer. Press, 1959.
Goddard, H. H. The criminal imbecile. New York: Macmillan
1915.
Goldfarb, W. Variations in adolescent adjustment of insti
tutionally-reared children. Amer. J. Orthopsychiat.,
1947, 17, 449-457.
Gough, H. G. Systematic validation of a test for delin
quency. Amer. Psychologist, 1954, 9, 381. (Abstract of
paper presented at 62nd annual convention of the Amer.
Psychol. Assn.)
91
Gynther, M. D. Crime and psychopathology. J. adnorm. soc.
Psychol. , 1962, 6j[, 378-380.
Hartshorne, H. , & May, J. A. Studies in the nature of char
acter and deceit. New Yorkl Macmillan, 1930.
Hathaway, S., & Monachesi, E. D. The MMPI in the study of
juvenile delinquency. Amer. sociol. Rev., 1952, 17,
Havighurst, R. J., & Taba, H. Adolescent character and per
sonality . New York: John Wiley & Sons, 1949.
Healy, W., & Bronner, A. F. New light on delinquency and
its treatment. New Haven: Yale Univer. Press, 19 36.
Herzog, Elizabeth. Some guide lines for evaluative re
search assessing psycho-social change in individuals.
Washington, D.C.: U. S. Children*s Bureau, 1959.
Hildebrand, H. P. A factorial study of introversion-
extraversion by means of objective tests. Unpublished
doctoral dissertation, Univer. of London Libr. , 1953.
Hurwitz, J. I., Hutcheson, B. R., & Cooper, S. Problems in
refining the psychiatric assessment of juvenile delin
quents. J. Hlth. hum. Behav. , 1961, 2_, 276-283.
Jesness, C. F. The Jesness Inventory: redevelopment and
revalidation. Calif. Youth Authority Research Report
No. 35, Nov. 30, 1963.
Johnson, Adelaide M., Juvenile delinquency. In S. Arieti
(Ed.), American handbook of psychiatry., New York:
Basic Books, 1959, Vol. I, Pp. 840-856.
Kilch, L. G. & Osselton, J. W. Clinical encephalography.
London: Butterworth, 1961.
Lander, J. Traumatic factors in the background of 116 de
linquent boys. Amer. J. Orthopsychiat. , 1941, 11, ISO-
159.
Lederman, D. G. Delinquency and the concept of identifica
tion. Unpublished doctoral dissertation, Washington
State Univer., 1961.
Levy, D. M. On the problem of delinquency. Amer. J.
Orthopsychiat., 1932, 2, 148-154.
Lindner, R. A rebel without a causae. New York: Grune &
Stratton, 194 4.
92
Lippman, H. S* The neurotic delinquent. Amer. J. Ortho
psychiat. , 1937, J 114-119.
Lively, E. L., Dinitz, S., & Reckless, W. C. Self-concept
as a predictor of juvenile delinquency. Amer. J. Ortho
psychiat. , 1962, 32^ 159-168.
Lombroso, C. Crime, its causes and remedies. Boston:
Little, Brown, 1911.
Mannheim, H. Juvenile delinquency in an English Middletown.
London: Kegan Paul, 194 8.
Mannheim, H. The limits of present knowledge. In Why de
linquency? The case for operational research. Report
of a conference on the scientific study of juvenile de
linquency. London: N.A.M.H,, 1949.
McCord, W., & McCord, Jean. Psychopathy and delinquency.
New York: Grune & Stratton, 1956.
McEachern, A. W. Factors related to disposition in juve
nile police contacts. Paper presented at the S.S.S.P.
Meeting, Los Angeles, Calif., August 25, 1963.
Mead, G. H. A behavioristic account of the significant
symbol. J. Phil., 1922, 19, 157-16 3.
Merton, R. K. Social theory and social structure. In
Social structure and anomie. Glencoe, 111.: Free
Press, -1949.
Miller, R. C, Personality patterns among delinquent behav
ior types. Washington, D.C.: Catholic Univer. Amer.
Press, 195 8.
Miller, W. B. Lower class culture as a generating milieu of
gang delinquency. J. soc. Issues, 195 8, 14, 5-19.
Monahan, T. P. Family status and the delinquent child: a
reappraisal and some new findings. Social Forces, 1957,
35, 250-258.
Mueller, P. F. C. An objective approach to a behavioral
classification of juvenile delinquents and its relation
ship to treatment assignment. Dissert. Abstr., 1959,
20, 383.
Murphy, F. J., Shirley, M. M., & Witmer, H. L. The inci
dence of hidden delinquency. Amer. J. Orthopsychiat.,
1946, 16, 686-696.
Nye, F. I., & Short, J. F., Jr. Scaling delinquent behav
ior. Amer. sociol. Rev., 1957, 22_, 326-331.
Parsons, T. Certain primary sources and patterns of aggres
sion in the social structure of the western world.
Psychiat., 1947, 1£, 167-181.
Payne, R. W. Cognitive abnormalities. In H. J. Eysenck
(Ed.), Handbook of abnormal psychology. New York:
Basic Books, 1961. Pp. 193-261.
Pesetsky, F. J. Variability in the meaning of deviant be
havior. Unpublished doctoral dissertation, Vanderbilt
Univer., 1961.
Petrie, A., McCulloch, R., & Kazdin, P. The perceptual
characteristics of juvenile delinquents. J. nerv. ment.
Disord., 1962, 134, 415-421.
Pound, R. A theory of social interests. In Outlines of
lectures in jurisprudence. (2nd ed.) Cambridge, Mass.:
Harvard Univer. Press, T5"14.
Purcell, J. F. Expressed self-concept and adjustment in
sexually delinquent and non-delinquent adolescent girls.
Unpublished doctoral dissertation, Fordham Univer.,
1961.
Reckless, W. C., & Shoham, S. Norm containment theory as
applied to delinquency and crime. Excerpta Criminolo-
gica, 1963, 3, 637-645.
Redl, F., & Wineman, D. The aggressive child. Glencoe,
111.: Free Press, 1957.
Reed, C. F., & Curdra, C. A. The role taking hypothesis in
delinquency. J. Phil., 1957, 21, 254-261.
Reich, W. Per Triebhafte character. Vienna: Intemation-
aler Psychoanalytischer Verlag, 1925.
Robison, Sophia M. Juvenile delinquency: its nature and
control. New York: Jenry Holt, 1960.
Rogers, C. R., Kell, B. L., & McNeil, Helen. The role of
self-understanding in the prediction of behavior. J.
consult. Psychol., 1948, 12^, 174-186.
Sarbin, T. R. Role theory. In G. Lindzey (Ed.), Handbook
of social psychology. Vol. I. Reading, Mass.:
Addison-Wesley, 1$54. Pp. 223-258.
94
Schuessler, K. F. & Cressey, D. R. Personality characteris
tics of criminals. Amer. J. Psychol., 1950, 55, 476-
484.
Schwartz, E. E. Statistics of juvenile delinquency in the
United States. Annals Amer, Acad, polit. soc. Sci.,
1949, 261, 9-20.
Schwartz, E. E. A community experiment in the measurement
of juvenile delinquency. In Yearbook of the National
Probation Association. New York: National Progation
Association, 1945. (Reprint by U.S.G.P.O., 1947)
Sellin, T., & Wolfgang, M. E. The measurement of delin
quency. New York: Wiley, 1964.
Shaw, C. R., & McKay, H. D. Social factors in juvenile de
linquency. Vol. II of National Commission on Law Ob
servance and Enforcement, Report on the causes of crime.
Washington, D.C.: U. S . G. P. 0., 1931.
Shaw, C. R., & McKay, H. D. Juvenile delinquency and urban
areas. Chicago: Univer. of Chicago Press, 1942.
Sheldon, W. H., Hartl, E. M., & McDermott, E. Varieties of
delinquent youth: an introduction to constitutional
psychiatry. New York: Harper, 1949.
Silverstein, A. B. Changes in performance on the Rosen-
zweig Picture-Frustration^Study. Unpublished Master's
Thesis, New York Univer., 1951.
Strauss, A. A., & Lehtinen, Laura E. Psychopathology and
education of the brain-injured child. New York: Grune
& Stratton, 1947.
Sutherland, E. H. Principles of criminology. Philadelphia:
Lippincott, 1947.
Sykes, G. M., & Matza, D. Techniques of neutralization: a
theory of delinquency. Amer. sociol. Rev., 1957^ 22,
664-670.
Thompson, G. The psychopathic delinquent and criminal.
Springfield^ 111.: Charles G. Thomas, 1953.
Toby, J. The differential impact of family disorganization.
Amer. sociol. Rev., 1957, 22, 505-512.
Trent, R. D. Socioempathic ability in a group of institu
tionalized delinquent boys. J. genet. Psychol., 1957.
91, 99-108. -------- -----
95:
Vane, Julia R. Implications of the performance of delin
quent girls on the Rosenzweig Picture-Frustration Study.
J. consult. Psychol., 1954, lji, 414.
Waller, W. Some sociological aspects of the definition and
classification of crimes. Paper read at the Criminology
Section of the American Sociological Society, December,
1955.
Wirt, R. D., & Briggs, P. F. Personality and environmental
factors in the development of delinquency. Psychol.
Monogr., 1959, 73^ 47 p.
Yoshii, N., Shimokochi, M., & Tani, K. The electroencephal
ograms in juvenile delinquents. Folia Psychiat. Neurol.
Jap., 1961, 15, 85-91.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Delinquency As A Function Of Intrafamily Relationships

PDF

Prediction Of Therapeutic And Intellectual Potential In Mentally Retardedchildren

PDF

Perceptual Defense And Somatization: A Comparison Of The Perceptual Thresholds Of Obese And Peptic Ulcer Patients

PDF

Positive And Negative Verbal Reinforcement In The Conditioning Of The Verbal Behavior Of Adolescent Underachieving Delinquents With Dependent, Neutral, And Aggressive Stimuli

PDF

A Study Of Affective Sets: The Effects Of Family And Non-Family Verbal Contexts On Word-Need Stimuli In A Word Association Experiment With Reference To Pleasant And Emotional Tones Of Associated...

PDF

An Experimental Study Of Aggression Habit Patterns And Sex Role

PDF

Improvement Rate In Tuberculosis As A Function Of Accessibility To Repressed Emotional Disturbance

PDF

Ego Diffusion In Women With Behavioral Disorders And The Integrating Effects Of Psychodrama In Identity Consolidation

PDF

An Investigation Of The Response To Stress Of Patients Hospitalized For Anxiety State And Peptic Ulcer Patients

PDF

Acculturation And Value Change

PDF

An Investigation Of Overt Versus Covert Responding In Programmed Instruction

PDF

A Semantic Differential Investigation Of Critical Factors Related To Achievement And Underachievement Of High School Students

PDF

Measuring Thought Process As An Ego Function In Schizophrenic, Mentally Retarded And Normal Adolescents By Means Of The Rorschach

PDF

The Effects Of Delay In Knowledge Of Results On The Amount Learned In Teaching Machine Programs Of Differing Cue Content

PDF

Behavioral (Social) Intelligence: A Factor Analysis

PDF

Validity Concomitants Of Various Scoring Procedures Which Attenuate The Effects Of Response Sets And Chance

PDF

An Investigation Into The Communication Style Of Suicidal Individuals

PDF

A Problem-Centered Approach To Introductory Psychology

PDF

The Effects Of Prior Part-Experiences On Visual Form Perception In The Albino Rat

PDF

Predicting Success In The Study Of Descriptive Linguistics

Asset Metadata

Creator Rochlin, Martin (author)

Core Title Behavioral Seriousness And Impulse-Control Balance In Delinquency

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,Psychology, clinical

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Seward, Georgene H. (committee chair), Jacobs, Alfred (committee member), Lefever, David Welty (committee member), Metfessel, Newton S. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-104656

Unique identifier UC11360197

Identifier 6700427.pdf (filename),usctheses-c18-104656 (legacy record id)

Legacy Identifier 6700427.pdf

Dmrecord 104656

Document Type Dissertation

Rights Rochlin, Martin

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA