A correlation study leading to the development of a scale useful in the prediction of potential juvenile fire setters

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content A CORRELATION STUDY LEADING TO THE
DEVELOPMENT OF A SCALE USEFUL IN THE
PREDICTION OF POTENTIAL JUVENILE FIRE SETTERS
by
Suzanne Robins Dudek
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
(Education)
July 1982
UMI Number: DP24883
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
Dissertation Publishing
UMI DP24883
Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106- 1346
UNIVERSITY OF SOUTHERN CALIFORNIA
T H E G R A D U A T E S C H O O L
U N IV E R S IT Y P A R K
LO S A N G E L E S . C A L IF O R N IA 9 0 0 0 7
This dissertation, written by
Suzanne Robins Dudek
under the direction of h.e.r... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Graduate
School, in partial fulfillm ent of requirements of
the degree of
D O C T O R O F P H I L O S O P H Y
I
Dean
~ , SEPTEMBER 2, 1982
D ate................
DISSERTATION CO
Chairman
Ph-O-
t A
D345
£275/ 3,
7 iA /
/
ACKNOWLEDGEMENTS
The author wishes to thank the following persons
for their assistance and encouragement during this
study:
My Dissertation Committee:
Eddie H. Williams, Ph.D., Chairman
James F. Magary, Ph.D.
Gayla Margolin, Ph.D.
Members of the Fire Services and Arson Prevention
Committee, Division I of the California State
Psychological Association:
Kenneth Fineman, Ph.D.,
Clinical Psychologist
Lynne Michaelis,
Fountain Valley Fire Department
Captain Joe B. Day,
Los Angeles County Fire Department
Employees of the Los Angeles County Probation
Department:
Alfred P. Parsell, Ph.D.,
Research Director
Edward J. Pawlowski,
Data Systems Supervisor
A friend and supporter:
Richard Charles Flaten, Ph.D.,
Kaiser-Permanente Medical Group
TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS . . . . . . . . . ii
LIST OF TABLES . . . . . . . . . . . . . . . . . . v
Chapter
I. THE PROBLEM. . ................... 1
Introduction
Background of the Problem
Statement of the Problem
Purpose of the Study
Importance of the Study
Questions to be Answered
Hypotheses
Conceptual Assumptions
Definition of Terms
Limitations
Delimitations
Organization of the Remainder of
the Study
II. REVIEW OF THE LITERATURE ............ 24
Introduction
Conceptual Background
The Sociopath
Etiology in Juvenile Delinquency
The Juvenile Fire setter
Fire Setting and Emotional Disturbance
Environmental Theories
Behavioral Programs
Organic Causes
Summary
Hypothesis Generation
ii i
Chapter
III. METHODOLOGY
Page
64
Overview
Research Methodology
Research Design
Selection of Subjects
Instrumentation
Data Collection, Processing and Analysis
Methodological Assumptions
Limitations of the Study
Summary
IV. ANALYSIS AND INTERPRETATION OF FINDINGS .... 83
Descriptive Analysis
Evaluation of Initial Equivalence of Groups
Hypothesis Testing of the Parent Questionnaire
(7-12)
Hypothesis Testing of the Parent Questionnaire
(13-17)
Effects of Factor Analysis Upon, the
Children's Questionnaire
Summary
Hypothesis Testing of the Children's
Questionnaire (7-12)
Summary
Hypothesis Testing of the Children's
Questionnaire (13-17)
Summary
Summary
V. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS . . . 135
Summary
Problem t
Review of the Literature
Methodology
Conclusions
Recommendations
BIBLIOGRAPHY .......... 165
APPENDIXES . ................ 178
A. PARENT QUESTIONNAIRE IN ENGLISH . . ... . . . . . . 179
PARENT QUESTIONNAIRE IN SPANISH . . . . . . . . 184
B. CHILDREN'S QUESTIONNAIRE....................... . 189
iv
LIST OF TABLES
Table Page
1. Means, Medians, Modes, and Standard
Deviations of Categorical Variables ..... 84
2. Analysis of Variance for Age of Respondents . 87
3. Analysis of Variance for Respondent's
Grade in School ......................... 88
4. Analysis of Variance for Age of
Respondent (7 to 12) 89
5. Analysis of Variance for Grade of
Respondent (7 to 12) 90
6. Analysis of Variance for Age of
Respondent (13 to 1 7 ) ........................... 91
7. Analysis of Variance for Grade of
Respondent (13 to 17) . . . . . . . . . . . . 92
8. Analysis of Variance for the Number of
Little, Definite, and Extreme Concern
Items for Children and Parents (7-12) .... 96
9. Arralysis of Means to Determine Differential
Group Responses (7-12) . . 97
10. Analysis of Variance for Individual
Questions (7-12). ................... 99
11. Analysis of Covariance for the Number of
Little, Definite and Extreme Concern Items
(13-17) 102
12. Analysis of Covariance for Individual
Questions on the Parent's Questionnaire
(13-17).............. . . . . . . . . . . . . . 105
v
Table
Page
13.
14.
15.
16.
17.
Factors and Questions Belonging to
that Factor . . . . . . . . . . . . . . . . .
One-way Analysis of Variance with
19 Children's Factors (7-12) .. . ....
One-way Analysis of Covariance with
19 Children's Factors (13-17) .. . . . . .
Analysis of Variance of the Questions
Which Discriminated Between Normal and
Fire Setting Subjects . . . . . . . . . . .
Analysis of Covariance of the Questions
Which Discriminated Between Fire Setter,
Sociopath and Normal Subjects (13-17) . . .
Ill
113
114
118
123
vi
CHAPTER I
THE PROBLEM
Introduction
This correlational study was directed toward the
development of an instrument that would be helpful to
teachers and counselors in the screening of potential juve
nile fire setters. In addition, a parental questionnaire,
previously developed by the Fire Services and Arson Preven
tion Committee, Division I of the California State Psycho-
locial Association, was validated through statistical
analysis. Much research has been generated on the topic of
firesetting. However, little has been done to relate the
concepts in the manner prescribed in this study. No cur
rent checklist is presently being used in an effort to
screen potential fire setters. In order to develop such an
instrument, subjects were chosen from referrals to the
Los Angeles City, Los Angeles County, and Orange County
Fire Departments. Subjects were also chosen from the
Los Angeles County Probation Department and from special
education classes in the Los Angeles City Schools. The
subjects were administered an initial test consisting of
1
questions in a variety of areas (Appendix B), including
impulsivity, locus of control, parental permissiveness,
peer relationships, school success and emotional variables.
A control group of subjects, chosen from Los Angeles Unified
School District was also administered these questions.
Parents in both groups were tested using the parental
questionnaire form (Appendix A) already developed by the
Arson Committee. Analyses by computer determined if the
parental and child's questionnaires were successful in dis
criminating among the normal, fire setting, and socio-
pathic subjects. In future research, a questionnaire could
be developed to be used by educators, counselors and fire
department personnel in an effort to detect potential
students with fire setting tendencies. By assembling those
questions found significant in this investigation into a
new questionnaire, school and fire department personnel
could apply appropriate remedial, educational or psycho
logical techniques to prevent the problem from reaching its
full potential.
Background of the Problem
According to arson investigators, arson is the
fastest growing crime in this country. (Barracato, 1979)
It is a crime that effects everyone, although only recently
receiving the national attention it deserves. The cost in
terms of property losses is estimated to be in the billions
2
of dollars each year (Deukmejian, 1980). The cost in terms
of lives lost is immeasurable; that arson is one of the
nation's most serious human-made disasters. The motto of
the Crime Prevention Center of the Office of the Attorney
General for the 1980's is Remember, Arson Burns Everyone.
In 197 9, the Law Enforcement Assistance Administra
tion convened a workshop of nationally-known experts on the
arson problem. From this came a national arson control
assistance strategy, combining the investigative and
prosecutorial expertise of the federal criminal justice
agencies with the financial assistance of the Law Enforce
ment Administration. An arson unit was also established
within the Administration Office of Criminal Justice.
Research is sponsored by the National Institute of Law
Enforcement and Criminal Justice. Other agencies which
have been directly involved in combating the nationwide
problem of arson include The United States Fire Administra
tion; The Federal Bureau of Investigation; The Bureau of
Alcohol, Tobacco, and Firearms; The United States Postal
Service; The Organized Crimes Strike Forces of the Depart
ment of Justice; The United States Attorney Offices; and
The Federal Insurance Administration. These agencies work
with state and local officials in order to provide a
comprehensive criminal justice strategy. Through combined
efforts, a major step to action will be
3
initiated in an attempt to diagnose and assess the problem
as well as to combat it.
Statement of the Problem
A great deal of research has been generated in this
area. Lewis and Yarnell (1951) classified fire setters
into five major groups:
1. Accidental or Unintentional,
2. Psychotics,
3. Erotics,
4. Revenge,
5. Children.
It is the last group with which this research was concerned.
According to Pat McGuiness (personal interview,
1981), Captain, Los Angeles City Fire Department, Arson
Control Unit, of 7,023 arson fires set in the City of
Los Angeles last year and of 1,084 others which were of
suspicious origin, approximately twenty-two percent were
set by juveniles. The Sheriff's Department, in that same
period of time had 1,450 fires. Juveniles were reported to
be responsible for more than half of those.
In a Los Angeles Times article (December 10, 1980),
Kenneth Fineman, Chairman of the California Psychological
Association's Committee on Fire Services and Arson Preven
tion, stated that various reasons have been entertained as
to why there are so many juveniles involved in this problem.
4
Out of one hundred given fires, at least thirty-five of
them have been set by children. He cited curiosity about
fire as one of the major reasons, along with emotional dis
turbance, as to why so many children are involved. Certain
children may be predisposed to fire setting behavior. These:
children generally come from a disrupted family with poor
communication, an over-protective mother, an absent or
uninvolved father, or an environment not supportive of
normal development. These factors alone are not considered
enough by Fineman, however, since the child also needs to
have the tools and the know-how for making the fire. He
needs a parent who is not properly supervising the child
and he needs a motive. Anger or revenge are two motives
that have appeared in the literature as often underlying
fire-setting behavior.
Among older children a need to feel powerful may
provide the motive. Fire setting is often seen as part of
a (M_inquency pattern that includes drugs, sex, truancy,
shoplifting and vandalism.
In the same Los Angeles Times article L. J. West,
Chairman, Psychiatry Department, University of California
at Los Angeles, stated that pyromania resembles an impulse
disorder, much like compulsive gambling or kleptomania in
which a sense of inner tension builds? which is accompanied
by a desire to see fire. No other mental or personality
5
disorder exists in those juveniles who are identified as
pyromaniacs. Fineman (1979), however, does not agree with
this assessment. Since so many juveniles are involved in
this nationwide problem, this study was instituted in an
effort to provide more information as to why children set
fires. This investigation was also designed to fill a
need in the area of prevention. Instead of simply concen
trating on reasons for fire setting, this study was also
devoted to finding solutions, to deal with the juvenile
fire setter before he has begun his deadly act.
Purpose of the Study
There is much controversy in the area of arson
research. Much of this revolves around speculation as to
the causes of fire-setting behavior. Techniques to handle
and prevent the problem are also involved.
It was the goal of the Fire Services and Arson
Prevention Committee to specifically address the issue as
it related to causes and prevention. By developing manuals
which included background information, interview sheets and
treatment procedures, it was hoped that fire department
personnel could become better adept at distinguishing
between juveniles in need of simple educational techniques
to solve their curiosity fire-setting problems and those in
need of referrals to mental health professionals for further
evaluation and indepth interventions.
6
This dissertation went even further into this area
with the following special purposes:
1. To determine if fire setters are any different
from other juveniles committing other types of
crimes.
2. To determine if fire setters had any particular
and distinguishing personality traits.
3. To determine if the questionnaire (Appendix A
and Appendix B) developed by the Fire Services
and Arson Committee would significantly dis
tinguish the fire setter from other juveniles.
4. To determine if fire setters could be
distinguished from other juveniles based upon
their answers to specific questions.
5. To determine if it might be possible to develop
a questionnaire that could be used in the pre
diction of potential juvenile fire setters.
Importance of the Study
The field of arson research is in disarray.
Although much information has been generated in past years,
it is lacking in general acceptance by those who are prac
ticing in the area. in addition, little attention has been
focused upon the practical problem of distinguishing the
curiosity fire setter from the potentially more dangerous
emotionally disturbed individual. This was the task of
the Fire Services and Arson Prevention Committee, and this
study sought to complete that task.
In the area of juvenile fire setters, many theories
have been generated as to why children engage in this type
of activity. Some researchers have viewed the problem as
an isolated behavior without any additional personality
defects. Others have listed a host of other behaviors that
often accompany fire setting.
It was a goal of this investigation to further
clarify the issue. By grouping together a variety of
questions covering various behaviors, often described as
part of the fire setters personality, it was hoped that a
separate profile would emerge: the profile of the fire
setter.
Additionally, it was the goal of this study to
determine if the fire setter would present himself as any
different from juveniles who commit other types of crimes,
specifically adjudicated sociopaths. This particular
questions has not been thoroughly addressed in the litera
ture. If an isolated group of juveniles, known as fire
setters, could be found, police and fire officials, as well
as school personnel, could refine their intervention
strategies and deal specifically with the behaviors pre
sented. Once officials know what the problem is, it will
be much easier to find the solution.
8
A real milestone.in the area of arson prevention
will be passed if children could be identified as high
risk before the problem has a chance to develop, and
appropriate measures applied. If this could be done, the
first step in the control of the nationwide problem of
arson will have been taken.
It has already been stated that arson is one of
this nation's fastest growing crimes. To solve this
problem, the public must be better educated. Surprisingly,
another group must also be better educated; police and
fire personnel. Rank-and-file police and fire personnel
who come in contact daily with the tragedy of arson-induced
fires, know very little about fire investigation, and even
less about fire prevention. Additionally, there is conflict
between these two groups of public servants, each claiming
the field of fire investigation as part of its own
territory. Consequently, instead of working together to
solve crime, there is often bickering and little attempt to
combine resources. This competition not only exists
between police and fire personnel, but exists among fire
departments as well, including the local agencies of
Los Angeles City and Los Angeles County Fire Services. To
complicate the situation even further, there is additional
difficulty surrounding the issue of female participation in
9
a male dominated field. Given all these factors, plus the
ever present danger of budget cuts in the area of fire
education, there is little wonder that the state of fire
prevention is in complete disarray.
Nationwide, situations exist wherein one researcher
may be working and developing an idea in the' area of fire
prevention while another department in another city has
already implemented this same idea in different form.
The situation is something like reinventing the wheel.
Many attempts have been made for a nationwide program in
the area of fire education. This study was directly
involved with the program developed by Fineman, Brudo,
Brudo, Morris, Michaelis and Day (1979) (Kenneth Fineman,
Charles S. Brudo, Esther S. Brudo, Connie Morris of City
of Fountain Valley, California, Lynne Michaelis, Fountain
Valley, California, Fire Department, and Captain Joe B. Day
of the Los Angeles County Fire Department). Working
together as members of the Fire Services and Arson Preven
tion Committee and funded by the United States Fire Admini
stration, Office of Planning and Education of the Federal
Emergency Management Agency, this group developed a set of
training manuals and video tapes as part of a program de
signed to teach fire personnel the difference between a
curiosity and emotionally disturbed fire setter. The pro*
gram includes an instruction manual, interview sheets,
directions on scoring, and strategies that can be used as
10
part of a fire prevention program. This program has con
tinued to receive funds, making it possible for Captain
Day to travel throughout the United States instructing
other fire departments in the use of the procedure which
deals with juvenile fire setters from the early years of
life until they are teenagers. As part of this package,
fire department personnel present parents with a question
naire which they are to complete (Appendixes A and B).
This dissertation sought to validate that questionnaire.
Fire department officials are using the results to deter
mine, in part, if the juvenile should be referred for addi
tional educational and psychological help. As an off
shoot to the original program, this study was also interes
ted in determining if fire setters would respond any dif
ferently to a set of questions than either normals or
adjudicated offenders. As previously noted, the field of
fire investigation is wide open. Much research has been
done but little of it is actually being used by the fire
professional in the field. This study was designed to
fill that need. Many types of motives and personality
attributes have been assigned to the juvenile fire setter.
Much of the research is contradictory in nature.
Little of it is presented in a form that can readily be
used. This investigation sought to clear some of that con
fusion: do fire setters fit a certain profile?
11
One of the major goals of this dissertation was to
provide some new light and information to the men and
women in the field, who daily put their lives on the line
to save others. If this investigation provided one new
fact or concept useful to fire officials, one of the goals
of this research project will have been met.
Questions to be Answered
This study sought to answer the following major
questions:
1. Would juvenile fire setters respond any differently
to a series of questions about personality and mo
tivational factors than either juveniles convicted
of other crimes or normals?
2. Would there be qualitative differences between
younger and older fire setters?
3. Would parents of juvenile fire setters respond any
differently to a series of questions about person
ality and motivational factors than either parents
of juveniles convicted of other crimes or parents
of normals?
4. Are juvenile fire setters a distinct group or do
they belong to the larger category defined as
sociopathic?
12
5. Would juvenile fire setters respond any differently
to a variety of scales of behavior than either
juveniles convicted of other crimes or normals?
Hypotheses
The following research hypotheses were tested in
this study:
1. Juvenile fire setters are similar to other types
of juveniles defined as sociopathic. Therefore,
they would score highly on a scale designed to
measure sociopathic tendencies.
2. Juvenile fire setters should differ from other
groups of sociopaths by their tendencies toward
pyromania. Therefore, juvenile fire setters should
score highly on a scale designed to measure fire
setting interests and activities.
3. Juvenile fire setters and sociopaths are basically
unhappy people. Therefore, fire setters and
sociopaths should score poorly on a scale designed
to measure good positive adjustment.
4. Juvenile fire setters have been defined in the
literature as having hyperactive tendencies.
Therefore, they should score highly on a scale
designed to measure the speed with which activities
are done.
13
5. Juvenile fire setters as well as sociopaths have
been described as having many personality problems.
Therefore, these two groups should score highly
on a scale designed to measure degree of psychotic
tendencies. Normals, by definition, would score
quite low on this same scale.
6. Fire setters and sociopaths are immature in their
level of psychological development. Therefore,
they should score highly on a scale designed to
measure degree of interest in childlike activities.
7. Sociopaths have been reported to be accident prone.
If sociopaths and fire setters are similar, they
will both score highly on a scale designed to
measure a high degree of accident proneness.
8. Juvenile fire setters, as well as sociopaths,
have a poorly developed super ego and are not
above telling lies to get out of a situation.
Therefore, both groups should score low on a scale
designed to measure truth-telling.
9. Sociopaths have been defined in the literature as
exhibiting no other personality problems, such as
neurotic behavior. Therefore, if juvenile fire
■ e.
setters are similar to sociopaths, both groups, as
well as normals, should score poorly on a scale
_____________________________________ 14
designed to measure a high degree of fears and
frightened behavior patterns.
10. Juvenile fire setters and sociopaths are exter
nally motivated. They do not feel they have con
trol over their actions. Therefore, they will
score poorly on a scale designed to measure degree
of responsibility-taking.
11. Juvenile fire setters and sociopaths are
basically loners with the possible exception
of their own subculture group. Therefore, these
groups should score poorly on a scale designed
to measure degree of positive social interactions.
12. Sociopaths and fire setters are likely to be
interested in activities traditionally viewed
as male-oriented. Therefore, they should score
highly on a scale designed to measure male
interests or machismo.
13. Fire setters and sociopaths should score poorly
on a scale designed to measure positive-self
concept, since they are involved in so many
negative interactions with parents, teachers,
and significant others.
14. Since there are so many negative interactions
with others, fire setters should score low on
a scale designed to measure positive socialization
skills.
15. Sociopaths are described in the literature as
having no personality defects, other than a
history of continuous and chronic antisocial
behavior. Therefore, both sociopaths and fire
setters should score similarly with normals on
questions designed to measure degree of depres
sion .
16. Fire setters and sociopaths, by nature of the
activities in which they are involved, should
have many negative contacts with parents.
Therefore, on questions concerning positive
parental interactions, these groups should score
poorly, as compared to normals.
17. Fire setters and sociopaths may have some dif
ferences. By nature of their crimes, sociopaths
may be more inclined toward violence, especially
toward other people than fire setters. Therefore,
sociopaths will score higher on questions of
violence than fire setters or normals. Of the
three groups, fire setters should fall in the
middle in terms of degree toward violent behavior.
18. A high degree of anxiety often accompanies
neurotic behavior patterns. People with highly
developed super egos will have high degrees of
anxiety. Therefore, sociopaths and fire setters
16
should score low on questions designed to measure
degree of anxiety whereas normals should
demonstrate the opposite response.
19. Younger fire setters often set fires out of
curiosity. Older fire setters are propelled by
different motivations, primarily pathological
in origin. Therefore, older fire setters will
demonstrate more signs of emotional disturbance
in their response to questions than younger fire
setters.
20. Parents of fire setters and sociopaths will
respond with a higher incidence of maladaptive
behaviors on a questionnaire than parents of
normal subjects.
21. Parents of fire setters and sociopaths will
rate their own lives as more unhappy and disturbed
than parents of normal juveniles.
Conceptual Assumptions
The following conceptual assumptions were implicit
in this investigation:
1. The locus of control model as proposed by Rotter
(1977) constituted a valid and viable framework
within which the behavior of fire setters and
sociopaths could be conceptualized.
17
2. The sociopathic antisocial personality syndrome
as defined by the American Psychiatric Associa
tion's Diagnostic and Statistical Manual (1980)
was a valid framework within which the behavior
of sociopaths could be described.
3. The symptoms and behaviors of sociopathic
personalities as defined by Robins (1966) were
accurate and were accepted in this investiga
tion as given.
4. The adjudicated subjects in this study were
assumed to fit the category of sociopathic
antisocial personality and to display the
behaviors as outlined by Robins (1966).
Definition of Terms
The following terms are defined as used in this
study:
Arson: The crime of maliciously or fraudulently
setting fires. It is a felony and a person who causes
death while committing arson is guilty of murder. This
definition is in keeping with the State of California's
Crime Prevention Center.
Other definitions of arson exist. Barracato (1979)
defined an arsonist as a fire setter who was neither a
juvenile nor a pyromaniac. In this study, such a distinc
tion was not made.
18
Fire setter; Someone known to school officials,
fire department personnel, Los Angeles County probation
officers, or local mental health officials, as having a
history of fire setting behavior.
Normal; A normal juvenile was a youth who had
never set a fire. During the screening phase of the study
when subjects were being chosen, a question was asked
of parents and children as to whether the youngster had
ever set a fire. If the answer was in the affirmative, the
juvenile was not chosen for participation in the study as
a normal subject. This was deemed necessary to avoid con
tamination of subjects across groups.
Pyromaniac: A person with a compulsion to set
fires for no apparent reason. Such a person is often
known as a pyromaniac. Although this term is commonly used
in everyday life, it has become controversial among
psychiatric experts as being worn-out and a catch-all
phrase. While some acknowledge its existence, other
experts feel that such a condition does not exist. For
this study, the term was used interchangeably with fire
setter and arsonist.
Sociopathic: Adjudicated juveniles composed a
single group known as sociopaths. They were considered
to be individuals displaying such characteristics as is
19
described in the American Psychiatric Association's
Dianostic and Statistical Manual (1980; p . 11) as:
truancy, expulsion or suspension from school,
delinquency, running away from home, persistent lying,
repeated sexual intercourse in casual relationships,
drunkenness, substance abuse, thefts, vandalism,
school grades below expectation.
Limitations
The following limitations existed which served to
limit the scope and breadth of this study and the gener-
alizibility of its conclusions:
1. The number of subjects available for study was
small.
2. This study was accomplished during the summer,
the season of the year defined by fire officials
as the slowest for fire setting activity.
3. Cooperation was needed from the Orange County,
Los Angeles County, and Los Angeles City Fire
Departments in order to get subjects defined as
fire setters.
4. Cooperation was needed from local mental health
agencies in order to obtain a broad sample of
fire setters and sociopaths. This was limited
due to restrictions imposed on community agencies
by the privacy act, and fears of breaking
confidentiality rules.
20
5. Cooperation was needed from the parents of the
subjects to return the questionnaires. This
varied from group to group with the normal
parents being the least cooperative and the
parents of the sociopaths the most cooperative.
Parents of the last group often asked for final
study results.
6. The limited size of the sample did not allow
for the control of all relevant variables.
7. The questionnaires that were finally developed
were not standardized on an appropriate national
sample.
D e 1 i m i t ai t ion s
The study was subject to the following delimita
tions which served to narrow its focus and generalizi-
bility:
1. The age limit for the subjects participating
was limited from seven to seventeen. The younger
limit was chosen to insure that the subjects would
understand the nature of the question. The older
limit was chosen since older subjects would be
beyond the age limit for juvenile court and would
be liable for adult proceedings.
2. Only boys were chosen for participation. This
was decided on the basis of a review of the
21
literature suggesting that the overwhelming
majority of fire setters were boys.
3. Sociopaths were restricted to subject incarcerated
in certain facilities run by the Los Angeles County
Probation Department. Other probation facilities
in other parts of the state were not available for
study. This limited the generalizibility of the
sociopathic sample.
4. All normal subjects were either enrolled in two
Los Angeles Unified School District schools in
middle and lower income areas or were the children
of faculty members or school psychologists.
5. All juvenile subjects were required to speak and
understand some English since the children's
questionnaire was not available in Spanish.
Organization of the Remainder
of the Study
Chapter II presents a theoretical framework as well
as a review of the relevant literature in the field. From
the reported articles, inferences will be drawn and explana
tions will be given as to relationships with the hypotheses
used in this particular investigation.
Chapter III reviews the research methodology,
covering such areas as design, hypotheses, selection of
22
subjects, instrumentation, data collection and procedures,
study weaknesses, and methodological assumptions.
Chapter IV reviews the results of the investiga
tion, both in terms of the child’s as well as the parent's
questionnaire. Individual questions and scales were
analyzed and conclusions were discussed. Data was presentee
in the form of tables, both within the chapter and at the
conclusion of the paper in the Appendixes.
A Summary of the study, including Findings, Con
clusions and Recommendations are presented in Chapter V.
23
CHAPTER II
REVIEW OF THE LITERATURE
Introduction
This chapter presents the theoretical and experi
mental framework upon which this investigation was de
veloped. Studies are reviewed in the areas of juvenile
delinquency and fire setting behavior. Background informa
tion in the areas of the nature-nurture controversy and the
locus of control model are also presented.
The results of these studies will establish the
need for this particular investigation and will afford a
basis upon which the hypotheses presented in Chapter One
were developed.
Conceptual Background
The nature-nurture question is one of the oldest
controversies in the history of psychology. It has been
argued since time immemorial as to whether heredity or en
vironment is responsible for a host of difficulties. The
controversy is no stranger to the area of juvenile delin
quency or to fire setting in particular. Studies in these
24
areas have looked into genetic and biological causes to
explain criminal behavior as well as the act of arson.
Environmental causes have also been proposed and apparently
have the edge in the current literature, although a promise
of future breakthroughs in the area of genetic research may
be imminent.
From Freud's (1932) thesis of fire setting as a
sexual behavior to Fineman's (1980) exposition of a social-
learning determinant, the literature in the area of
juvenile arson has run the gamut. Another predominant
theme in the environmental area is the current emphasis
upon family background. According to a recent study,
juvenile delinquency, in fifteen Tennessee counties sur
rounding Knoxville, was attributed to problems in the home
environment, including family breakups, financial difficul
ties, weak parental guidance and control, as well as a
lack of recreational facilities, limited job outlets,
changing social values, and peer pressures (Bolton, 1978).
In order to solve the problem, professional coun
seling, improved probation services, and recreational
opportunities were suggested. Themes such as this, as well
as theories about biological causes, will be reviewed.
In this investigation, juvenile fire setting was
viewed as one type of sociopathic behavior, falling within
the continuum of behavior broadly known as juvenile
25
delinquency. Conceptually, . this study looked at fire
setting as the result of environmental determinants. One
of these determinants was based upon the locus of control
model developed by Rotter (1977). It was assumed in this
study that the sociopaths were externally motivated or had
an external locus of control. With this orientation, it is
assumed that the subject has been conditioned to believe
that he has no control over his life or the things that
happen to him. Using Rotter's interpretation, a person's
locus of control orientation is a function of expectancy of
the situation and a generalized expectancy for all situa
tions. Specifically as it relates to juvenile delinquency,
it is suggested that the youth comes from an environmental
situation best described as chaotic. With this condition
ing, the youth comes to expect that he has no control over
his life. He learns that what happens to him is totally
unrelated to his own particular effort. Life's experiences
are related to luck, fate, or to just one of those things.
This type of environment is often found in ghetto
or very poor areas, the breeding grounds, it turns out, for
gang violence and juvenile delinquency. With wide-spread
unemployment, budget cuts in medical services, a lack of
adequate food, and a deteriorating public school system,
juveniles are faced with an uphill battle just to receive
essential services. These children, under such conditions,
26
are ripe to develop an' attitude often known as learned
helplessness. In this review of the literature, articles
will be presented indicating just such a developmental
pattern. At the conclusion of this chaper, some remedia
tion for the problems will be presented. Using the locus
of control model, Lawrence (1975) suggested techniques to
change from an external to an internal orientation. With
this approach children are able to perceive the influence
of their actions on events occurring around them. When
people feel that they have some control over their lives,
they are more willing to assume responsibility. After all,
why bother being responsible or caring about something that
cannot be changed. Perhaps, through a restructuring of the
environment, the locus of control orientation can be
changed. With this may come a decreasing occurrence in
juvenile delinquency in general, and fire setting in parti
cular .
The Sociopath
A tremendous amount of work has been generated in
this area. The sociopath has been defined by the American
Psychiatric Association (1980, p. 11) as a
chronically antisocial individual who is often involved
in activities which violate the rights of others...
failure to accept social norms and to observe law
ful behavior...a disregard for the truth...
failure to plan ahead or to honor financial obli
gations ... inability to maintain an enduring attachment
27
to a sexual partner...a lack of ability to function
as a responsible parent... lack of concern for the
feelings of others...manifested by the absence of
feelings of guilt....
The age of onset when considering this diagnosis is
usually before fifteen years, and is often characterized by
such behaviors as theft, incorrigibility, school truancy,
running away from home, having associates with bad reputa
tions and physical aggressiveness. Much of the research in
this area was the result of a longitudinal study done by
Robins (1966). Other behaviors most often found in the
sociopath include a poor employment record, impulsiveness,
recklessness, irresponsible behavior, slovenly appearance,
enuresis, lack of guilt, lying, homosexuality, and premari
tal sexual intercourse.
Forty-three percent of the subjects exhibiting ten
of these sociopathic symptoms in childhood were likely to
be described as antisocial adults.
Juveniles with three or fewer of the antisocial
behaviors made up only about four percent of adults de
scribed as antisocial.
Robins concluded from his study that the more
symptoms exhibited in younger years, the more life areas
would be effected and the resulting prognosis was poor.
Areas that were ultimately effected by antisocial behaviors
included parental relationships, peer relationships, sexual
and school behavior.
Although forty-three percent of the juveniles
exhibiting ten or more behaviors were described as anti
social as adults, the remaining subjects could not be de
scribed by Robins as clinically well.
Many other psychiatric symptoms were evident among
the remaining adults, chief of which is alcoholism. Of
those subjects exhibiting three or fewer of the antisocial
behaviors, thirty-four percent were described as clinically
well as adults.
One of the study's final conclusions was that a
great number of sociopathic symptoms in childhood is a
serious prognostic sign for adult psychiatric illness.
The incidence of antisocial reaction in the past
has been preponderantly male, both with adults as well as
children. Psychiatric journals often give numbers of five
or ten males to one involved female.
Etiology in Juvenile Delinquency
Studies in the etiology of juvenile delinquency
have been inconclusive, ranging from hypotheses in the area
of sexual and physical factors to totally environmental
causes. No proof of a direct link between brain damage and
criminal behavior has yet been ascertained, although genetic
factors have been found to correlate rather highly with the
incidence of antisocial behavior (Katz, 1972). Much of the
recent research has been with environmental determinants.
29
Further into this chapter it will be seen that
essentially the same reasons have been given for fire set
ting behavior, with hypotheses made in physical as well as
in environmental areas. In both juvenile delinquency and
in juvenile fire setting, studies have continued to
indicate a high incidence of criminal behavior when the
youth emanates from an environment of broken homes,
alcoholism and low socioeconomic status.
Research in the area of juvenile delinquency has
concentrated upon causes, types, personality characteris
tics and preventive programs. It must be understood,
however, that violence, upon which juvenile delinquency is
based, is culturally defined. What may constitute acts of
delinquency in one environment may not necessarily be so-
defined in another. Violence is also institutionally
sanctioned. What goes on in a person*s home is considered
fair game, but not so in other environments.
This situation-specific aspect is a central concept
in the understanding of violent behavior. Hammond (1978)
has determined from her research that violence is now part
of the basic fabric of everyday life, especially when
viewed in terms of domestic or family violence. It is an
act that occurs among people who know each other intimately,
and which is seldom, if ever, punished. In. this type of
environment, children grow up believing that such acts are
30
normal, rather than examples of deviant behavior. In
addition to violence in the home, Hammond (1978) found that
alcoholism, bad rearing, often total neglect of the child,
and criminality combine to produce negative conditioning
and ultimately possible social maladjustment or juvenile
delinquents.
Zabcynska (1979) found that the factors already
mentioned in the previous paragraph combine with school
failure to produce an environment that is high risk for the
development of maladaptive behavior. School failure was
found to be especially significant if it occurred before
the age of ten or eleven. The risk of juvenile delinquency
is greatly intensified in the child who becomes involved
with alcohol, stealing, or running away from home.
The theme of the home environment was studied by
Kulich (1975). In a visit to a juvenile prison, he found
that early childhood deprivation caused disturbances in
motivation for achievement. This most often occurred in an
environment lacking in good occupational training programs
and a total lack of good educational opportunities. In
order to combat the problem of delinquency, Kulich (1975)
suggested a restructuring of the environment, with specific
emphases in education and vocational opportunities.
Cemef (1973) investigated delinquency and drug use.
He also found that poor education, low social status,
31
unstable family conditions, intellectual deficiency and
frequent depressions were among the common variables in
adjudicated delinquents' personalities. This type of
background would provide the backdrop for youths to begin
experimenting with drugs, usually from curiosity at the
beginning, and ultimately ending as an act designed to ap
pease peer pressure.
Schuberth (1974) correlated early home environment
and delinquency. While looking at aggression, he deter
mined that incarcerated youths had higher scores in the
area of uninhibited aggression than a matching control
group. In addition, the threat of punishment did not act
as an inhibitory agent for these youths as it did for the
control group. In conclusion, Shuberth suggested that the
motivational factors underlying aggressive behavior were
related to the social disorders engendered during childhood
resulting in unsolved problems in self-realization and life
orientation.
Koznar (1978) reviewed family rearing practices and
their relationship to dissocial behavior. He found that
internal disharmony in the family, insufficient relation
ship of the parents to the child, including indifference,
emotional negligence, and hostility, plus inadequate child-
rearing practices were the strongest factors in the
development of the disturbed personality.
32
Schindler (1975) also worked in the field of family
relationships and delinquency. He found that the fewer the
number of children in the family, the lower the percentage
of aggressive behaviors used as a solution for conflict
situations. The fewer the number of children, the fewer the
number of convictions for assault and battery. In conclu
sion, the author suggested that with fewer children, parents
had more time to spend with their offspring reinforcing ap
propriate behavior patterns.
Katz (1972) looked at patterns in the development of
juvenile delinquency and found four major groups as possible
causative agents in its development. These included
environmental factors, superego factors, definite
psychiatric syndromes, and a brain functioning factor. In
the first category, the upbringing of a child in a family
that is openly criminal or conditions which exist that pro
duce great anger in the child would be considered. Also in
this heading would be deprivation, neglects and cruelty.
Superego factors include the development of a conscience
that was either too weak or too rigid. The psychiatric
syndromes identified in the article included neurotic,
characterological, or psychotic bases. The last factor
would include juveniles who are brain damaged, epileptic,
or mentally deficient with low frustration tolerances and
poor controls.
33
Anthony also looked at psychiatric syndromes and
their relationship to juvenile delinquency in a research
completed in 1969. He specifically reviewed the relation
ship of depression to acts of violence and concluded that
it was especially associated with it.
Another study concentrating upon the psychosocial
characteristics of young offenders was done by Pospiszyl
(1973). He found that such criminal categories as thieves,
prostitutes, rapists and hooligans demonstrated a higher
instance of neuroticism than control groups. Another of
the study's conclusions was the extroverted nature of the
hooligans, prostitutes and rapists, while the thieves were
more perseverative. Also, prostitutes had more positive
sociometric ratings toward others while the hooligans and
rapists demonstrated more negative ratings.
Pospiszyl (1973) tested Eysenck's (1964) assertion
that criminals are predominantly extroverts. He concluded
that seriously delinquent youths do indeed demonstrate this
trait.
Sorrells (1977) looked at psychiatric syndromes and
children who have killed in Alameda County. In his research
he found that juvenile murderers were not more disturbed
than any other delinquent sample. The families of the
children involved were chaotic. He noted that in addition
to disruptive families, the popular media also had a
34
substantial impact in these cases, by teaching the value of
excitement in committing violent acts. No family values
opposing this emphasis on violence were evident.
Schmideberg (1973) also studied the role of tele
vision and violence. He concluded that it is a contribu
ting factor toward the act of murder when committed by a
juvenile. He also found that another contributing factor
included the knowledge by juveniles that they would not be
dealt with harshly. Exposure to gang wars was a final fac
tor identified by Schmideberg (1973) in his investigation
of contributions toward violence in juveniles.
Heacock (1976), in a study that looked again into
the environment as a possible cause of delinquency, as well
as considering contributing psychiatric syndromes, reviewed
the causes of aggression in the Black child. Looking at
antecedent conditions existing in urban ghettos, Heacock
(1976) identified such factors as frustration and violence
at home, decreased living space, the experience of racial
antagonism and discrimination, abuse at school, constant
exposure to danger in the streets, and the high incidence
of deaths caused by fires, crime and violence as contribu
ting causes. It was suggested that much of the aggression
expressed by the Black slum child in the Bronx can be
viewed as an adaptive response to an unfavorable environ
ment. Thus, the idea of viewing behavior often described
as delinquent in the specific situation in which it
develops is again expressed and should be considered when
examining causes of violence in youth.
Chang (1972) looked into criminal behavior in
Korea and found that the majority of crimes occurred in
the afternoon. Weather was also listed as a contributing
factor since about fifty percent of all larceny-theft cases
occurred in clear skies. Seasonal factors were also listed
since property crimes increased in winter,, while acts of
violence increased sharply in the summer. To combat the
problem, Chang (1972) suggested a re-engineering of the
environment in such a way as to make the cost-factor in the
commission of a crime to be greater than the potential
gain.
Several studies have looked into crime and punish
ment. A project by Levin (1972) revealed data suggesting
that offenders receiving probation have lower rates of
recidivism than those who were incarcerated. Also, incar
cerated offenders receiving shorter sentences had lower
rates of recidivism than those with longer sentences. The
author suggested a program whereby those juveniles with a
history of assault could be screened out. Probation would
be granted, then, to all of the remaining juvenile first
offenders. Assigned programs would include intensive
counseling and efforts designed to develop positive
36
self-fulfilling prophecies. If this were done, Levin (19725
foresees a drop in the recidivism rate of thirty-three
percent.
Another program concentrating upon reform in
delinquency institutions was in effect in the 1920's in
Germany. Stress was placed on individualization, group
bonding, and the development of a sense of responsibility
for each other. Konopka (1971) reported that no distinc
tion was made between education or treatment and there was
a strong combination of what has become, known as casework
and group therapy. Punishment was used as a last resort
and was in the form of communicating to the offender the
disappointment of staff in the youth's behavior. The
emphasis of the program was on the restorative power that
can only be found in the relationship of one man to another.
This program was not able to continue, however, because of
the rise of the Nazi Party and the destruction of all re
form movements.
In this study, juvenile delinquency was looked at
in terms of the characteristics of involved youths, possible
causes of the problem, and suggested solutions. In the
area of etiology, minimal attention was given to genetic or
biological influences. This will be covered in the next
section, dealing with the special phase of juvenile delin
quency known as the juvenile fire setter.
37
The Juvenile Fire setter
In the review of general articles in the area of
juvenile delinquency, it was ascertained that a variety of
causes have been suggested as to why the problem exists.
In the area of the juvenile fire setter, the same
quandry exists. One has a variety of areas to choose from,
depending upon his particular orientation. Much of the
earlier research was psychoanalytically oriented and is
still in use today as a major force in the etiology of the
juvenile fire setter. More recently, the trend has turned
toward a social learning point of view with the backbone of
the theory enhanced by a modification technology. Others
have continued to look into the areas of biological and
physical phenomena to explain questions of causation. In
this section of the study these particular philosophies
will be presented, reviewed, and their relationship to this
investigation will be made evident.
Psychoanalytic Theories
Psychoanalytic explanations of fire setting have
predominated the literature, perhaps due to their major
influence in the entire field of psychiatry and psychology.
Using this approach, heavy emphasis is placed upon the
sexual and symbolic nature of fire setting.
38
Axberger (1973) viewed fire setting within a model
of sexuality, assuming that it was symbolic of sexual
activity. Fire setting, in this sense, is viewed as a
result of repressed sexual impulses, and often occurring
during periods of self-enforced sexual abstinence. This
model also views fire as a symbol of consuming emotion and
as a supreme form of god-like destruction. The fire setter
may not only be experiencing sexual conflict but may also
lack the facility to represent this symbolic nature in the
form of art. Instead, these unconscious symbolizations are
translated into the criminal action of fire setting.
Historically, fire has often been associated with
sexual passion. Freud (1932) wrote that it was necessary
for man to renounce his homosexually-tinged desire to
extinguish fire by a stream of urine. This was seen as
the only way man could possess himself of fire. Urinating
to extinguish fires was seen as erotic. The creation of
fires was viewed by Freud (1932) as symbolic of sexual
activity.
Researchers, using this model, would view arson as
the result of thwarted sexual impulses. Stekel (1924) used
this model when he wrote that the relatively young age of
fire setters was directly the result of the awakening and
lack of gratification of sexuality. Fire setting was
39
seen as a symbolic solution of the conflict between
instinct and reality.
This model was used by Simmel (1949) in an analysis
of a twenty-one year old fire setter. The fire setting
activity was viewed as a result of repressed masturbatory
impulses.
Gold (1962) also used this same model in his
research. He wrote that fire setting was either the result
of repressed sexual impulses or a primary sexual excitant.
In children, the psychoanalytic model has been
translated into a study of the relationship between
enuresis and fire setting, the former being seen as a
juvenile form of sexual behavior. Kaufman, -Heims and
Reiser (1961) found slightly less than half of their fire
setting subjects to be enuretic.
Michaels and Steinberg (1952) found a higher rate
of arson among enuretic delinquents than among nonenuretic
delinquents. They also found that arson was more likely to
be a reoccurring symptom in those delinquents who were
enuretic.
Siegelman (1969) also researched this area but
found different results. In her data, only sixteen percent
of recidivist fire setters had been enuretic beyond the
age of six, while the figure was thirty percent for
nonrecidivists.
40
Michaels and Goodman (1934) looked into the rela
tionship of enuresis in the population at large to deter
mine if there were greater numbers among fire setting
youths. They found that twenty-four percent of children
attending a recreational camp had a history of this same
problem.
Bakwin and Bakwin (1960) found that twenty-six
percent of unselected hospital outpatient children between
the ages of four to twelve, wet themselves during the day,
at night or both.
Similar findings were also found in a study by
Oppel, Harper and Rider (1968). They conducted a twelve-
year longitudinal study of 992 children born in Baltimore
in 1952. No evidence of a special relationship between
enuresis and fire setting or sexual behavior in children
and fire setting could be substantiated.
Looking into sexuality and fire setting, Medvecky
(1970) presented a case of a patient diagnosed as a
psychopathic personality with a weak intellect and psycho-
sexually infantile. The fire setting behavior in this case
was determined to be sexually motivated.
This study reviewed the literature in fire setting,
beginning with psychoanalytic studies and ending with the
current emphasis upon family disorganization. Heath,
41
Gayton and Hardesty (1976) began work by suggesting that if
Freud's hypothesis about the relationship of fire setting
to sexuality and enuresis were correct, then fire setters
should often be enuretic. It was found, that only fourteen
percent of fire setters were actually bedwetters. This
percentage was not found to be any higher for any other
group of emotionally disturbed children. Heath, Gayton
and Hardesty (1976) also reported that psychoanalytic
writers have switched their emphasis to aggressive drives
and defects in impulse control, object relations and
reality relationships. The investigation was concluded
with a request for more study into the area of family re
lationships and suggestions for treatment strategies.
Fire Setting and Emotional Disturbance
The emphasis in recent years has moved from a
sexual orientation into explanations concerning social
learning determinants, genetic and biological causes, and
fire setting as a symptom of emotional disturbance.
Lester (1975) examined various aspects of fire
setting, including clinical studies on the subject, general
characteristics of fire setters, theories of fire setting,
and the relationship of fire setting to sexual behavior.
He found that fire setters did not differ in diagnosis from
other types of criminals, and did not appear to have any
more problems in their heterosexual behavior. There was
42
some evidence found of a greater history of enuresis and
urethral disorder in the fire setters, however. Arsonists
who were criminally insane were more likely to have experi
enced institutional care as children and to have had more
difficulty with interpersonal relationships than fellow
patients. Fire setters tended to have committed offenses
against property rather than against persons. Fire setters
who were sexually excited by fires formed a very small pro
portion of the total fire setter population.
In a study of the relationship between mental
illness and fire setting, Mavromatis and Lion (1977) found
a high incidence of alcoholism in the fire setting group.
There was also a relationship between fire setting and
mental retardation. Those who were retarded or whose fire
setting behavior assumed a particularly ritualistic pattern
as part of a fetish, demonstrated a poor prognosis for
recovery.
Mecir (1974) also looked into the relationship of
mental illness and fire setting behavior. After reviewing
the case histories of seven patients with different family,
psychological and medical backgrounds, it was found that
arson was a serious symptom of a more complex psycho
pathology or behavior disorder.
Within the past twenty years,, the theme of arson as
a symptom has become quite prominent in the literature.
43
Robins and Robins (1967) viewed fire setting as the result
of four possible motives including: profit, to cover up
another crime, revenge, and pyromania. It was this last
category that was felt as the second most predominant
motive: seventy percent of the fires, in the pyromaniac
category, were committed by persons who were either
psychotic or mentally defective. Many of the psychotics
were diagnosed as schizophrenic. Alcoholism was also re
lated to fire setting behavior in the pyromaniac group.
Laplante (1969) also viewed the fire setter as a
pyromaniac. He felt that theories regarding etiology could
be divided into four major categories: psychoanalytic,
general, theories of neuroses and psychoneuroses, and
special theories.
It was believed that fire setting behavior was not
a specific disease, i.e., pyromania, but a symptom of a
different problem. Assuming this logic were correct, it
becomes more evident as to why there were so many diverse
opinions of how to deal with the fire setter. The behavior
was simply a symptom of another underlying problem, which
must first be addressed. The treatment must be related to
the problem, rather than to the act of fire setting.
Siomopoulos (1978) also looked into the distribu
tion of various offenses among various psychiatric dia-
noses in a sample of male indicted felons. Schizophrenia
44
was the leading diagnosis among the total population as
well as among the accused for every offense except arson.
Organic brain syndrome was associated exclusively with
crimes of high violence, whereas mental retardation was
found primarily among those accused of lesser crimes. No
specific illness was related to fire setting in this parti
cular study.
Gunderson (1974) reviewed the case histories of six
manic patients involved in fire setting activities. It was
his hypothesis that the fire setting was due to improper
management by the staff: a need of tighter control and
more restriction. When these changes were instituted in
the hospital, the fire setting behavior was eliminated.
In Poland, Psarska (1972) looked into mental ill
ness as it related to fire setting in that particular
country. In this study fifty mildly mentally retarded male
adults who were charged with various offenses, including
arson, homicide, sexual acts, and who were under psychiatric
observation, were studied. Some of the personality charac
teristics identified as common to all subjects included
feelings of .inferiority and a related need to compensate
for these feelings. Inappropriate environmental attitudes
were also found to be characteristic of the subjects, and
it was suggested that this factor intensified the patient’s
deviant tendencies.
45
Another study in Poland by Fleszar-Szumigajowa
(1969) looked at arson and personality characteristics.
Out of a total.of three hundred eleven cases, only seven
were found to be mentally normal. Schizophrenia was the
most frequently.occurring diagnosis, followed by mental
deficiency, personality disorder and alcoholism.
In addition to the relationship of mental illness
to fire setting, Vandersall and Wiener (1970) also looked
at the home setting, enuresis, I.Q., organic factors, type
of fires, and ego structures of twenty juvenile arsonists.
No consistent precipitating stresses were identified.
There was no one personality type represented and no
specific sexual conflicts. The one consistent factor was
the often temporary breakdown of controls in the child
before the act of fire setting. Vandersall and Wiener
(1970) recommended that therapy concentrate upon the under
standing of the inner conflict by the child. Also there
was the recommendation for the necessity of reinstituting
appropriate controls and a need for support.
In one of the landmark studies in fire setting,
Pasternack (1971), working for the National Naval Medical
Center in Bethesda, Maryland, found that of thirty-four
cases referred for a period of one year for dangerous
behavior, there were a combination of factors signaling
possible future tendencies toward violence. Background
46
behavior in these subjects included parental brutality,
school and learning difficulties, personal experience with
violent death, compulsive fire setting, and the torturing
of pets. When these behaviors were evident in children,
they were important indications of possible violent
behavior in adult life. The individuals who were studied
tended to be somewhat schizoid, have poor ego-defense
mechanisms, sadomasochistic relationships with spouses or
relatives, a history of fugue states, and a readiness to
attack others. Crimes of violence often occurred when the
individual had been threatened by others, especially when
combined with a history of alcoholism and drug abuse.
Pasternack (1971) concluded that a combination of any of
the factors previously mentioned should be used by mental
health professionals to detect potential criminal behavior.
Another study of psychological factors and fire
setting was completed by Kammerer, Singer and Michel (1967),
In a study of seventy-two cases, such psychological dis
turbances as dementia, organic illness, alcoholism, and
psychoses often were part of, the motivation behind the
act itself. A further analysis of the data suggested four
possible motives for fire setting in addition to the
%
psychological disturbances already mentioned: pleasure,
power, aggression, and vicarious experience.
47
An article by Justice, Justice and Kraft (1974)
reviewed a triad of childhood symptoms, including enuresis,
fire setting and cruelty to animals, which have been
generally accepted in the literature as predictive of vio
lence in adulthood. In the research, Justice, Justice and
Kraft (1974) suggested that other symptoms as well may
serve as even more useful predictors. Using as a back
ground upon which to draw data, a review of the literature,
interviews with people working with troubled youths, and
in-depth interviews with eight convicts, the following
symptoms were identified: fighting, temper tantrums,
school problems, truancy, and interpersonal difficulties.
Children with these behaviors are giving early warning
signs that a serious emotional disturbance may be involved
and may possibly erupt into violent behavior.
Environmental Theories
In addition to sexual and - emotional disturbance
theories, disturbed family background or environmental in
fluences have also taken an important place in the litera
ture as possible causative factors.
Levin (1977) looked at the psychological charac
teristics of fire setters. In his research he noted that
the conclusions of experts in the field have often been
contradictory but that there were certain aspects of the
problem about which there is some agreement: fire for
48
---------------
profit is a motive for behavior. Elimination of the
possibility of profit would either wipe out or reduce this
aspect of the fire setting problem. Levin (1977) noted
that many fire setters come from broken homes and have had
harsh and frustrating lives as young children. He indicated
that there was some tendency for fire setters to be young,
to have low intelligence, to have been bedwetters as
children and to have physical deformities. The desire for
revenge against a person, an institution or society may be
a major motive or a contributing factor. Since fire
setting is a secret crime and does not require confronta
tion with the victims, it is often committed by the weak
or cowardly.
Wagner (1974) looked into motives behind incendi
arism and also challenged the assumption that sexual per
version was the motive behind arson. He concluded that the
basic motivation could be found in chronic social frustra
tion with or without a personal need for attention.
Another researcher viewing fire setting in terms of
social interactions or with a sociological perspective was
Richard Eisler (1972)„ He described a crisis-oriented
family therapy with a fourteen year old boy who had con
fessed to setting a series of grass fires. Eisler (1972)
viewed the fire setter’s behavior as representative of
disturbed social or family relationships and suggested that
49
treatment focus on the social relationships rather than on
the disturbed behavior of the fire setter. After six
sessions, treatment was terminated; after the revealing of
a series of unresolved crises in the family which had led
to the eruption of symptoms in the boy. Eisler (1972)
viewed the fire setting behavior as a desperate nonverbal
communication to his family; concluding that intensive
family-oriented intervention may produce significant and
durable results within a short period of time.
Tapia (1971) reported the case histories of
eighteen children who had been referred to the University
of Missouri Medical School for treatment of cruelty to
animals. Fire setting was one of the behaviors that seemed
to accompany the act. Other behaviors included destruc
tiveness, bullying, fighting and stealing. All of the sub
jects were normally intelligent young boys with a mean age
of nine years and six months. In his review, Tapia (1971)
found that etiological factors ranged from strictly bio
logical causes as organic brain syndrome to strictly en
vironmental causes. It was this latter category that was
t-
the most common.. The majority of subjects came from chaotic
home situations with aggressive parental models.
Fire setting as a function of the environment was
addressed by Kalus (1978) in her analysis of Hawaii's
secondary schools and related discipline problems. In a
50
study of twenty-one high schools, problems such as truancy,
burglary, vandalism, smoking, drug use and fire setting
were reported. Fighting and disorderly conduct was also
occurring with moderate frequency and eight teachers had
been assaulted by students in the previous year. Kalus
(1978) found that the school's enrollment size was posi
tively correlated with the crime rate. As the school
became larger and more crowded, the crime rate increased.
No association was found between school-community relations
and the crime rate. A correlation was established,
however, between high parental interest in school and low
crime rate. It was recommended- that school crime could be
reduced through higher expectations by principals for
academic and behavior standards, lower enrollments, and
renewed efforts to get parents more involved and interested
in school activities.
Trasier (1978) also looked at. the role of the
school and its relationship to delinquency specifically in
Great Britain. Acts of arson were occurring regularly but
there was very little other serious crime in the British
schools. Trasier (1978) found many of the'same conditions
in Great Britain as Kalus (1978) found in Hawaii. British
schools utilize a system that generates stress and
frustration. These are the conditions upon which a
delinquent subculture can develop, persist, and may
51
ultimately be one of the major causes of the fire setting
often reported.
Coleman (1978) viewed the same problem and its
relationship to the American system of education: that, the
role of the school was highly influential in its develop
ment. Juvenile crime was described as a subcultural
phenomenon that is usually committed by groups with their
own distinctive attitudes, values, and perspectives.
Addressing problems such as drug use, vandalism, and theft,
Coleman (1978) suggested a restructuring of educational
institutions to reduce the student1s sense of alienation
and status deprivation. He also recommended the creation
of programs to encourage involvement in other activities,
rather than in activities common to deviant subcultures.
Juvenile crime and the Chinese were studied by
Joe and Robinson (1978). In the investigation of Chinese
youth gangs operating in the Vancouver area, it was found
that most of the youths had recently arrived from Hong Kong.
The learning problems that these youth experienced was one
of the major causes of the behavioral problems that subse
quently occurred in the schools. Joe and Robinson (1978)
made recommendations for the improvement of educational
provisions for Chinese youth as one of the ways to decrease
the rise of juvenile crime.
52
In 1978, the role of the school in the development
of juvenile crime was studied by the Vocational Foundation,
Inc., in a White Paper on Unemployment, Education and Crime.
As a result of interviews with clients and experts, it was
determined that inner-city youth between the ages of six
teen to nineteen are heavily prone to robbery, violence,
and other types of crime. Some of the reasons cited for
this problem included the changing composition of the
United States, the utter collapse of the system of juvenile
justice, and the increasing inaccessibility of entry-level
employment to dropout youth. Recommendations included the
improving of the predictability of the justice system, and
the providing of youths with the opportunity to go straight
in a productive way, especially in the area of jobs.
Approaches to solving the problem included an attack on the
exclusive use and heavy reliance upon education and testing
requirements before a job can be obtained, and the sugges
tion that a realistic assessment of the opportunities af
forded by conventional schooling be made.
Glicken (1978) also reviewed the area of juvenile
crime and its relationship to the school system. He found
that children who are identified as pre-delinquent often
show early warning signs of aggression, theft, learning
problems, and acting out behavior in the classroom. He
suggested that remediation focus on self-concept, student-
53
to-student relationships, and the locus of control model.
With this restructuring, Glicken (1978) foresees a de
creasing number of predelinquent children who will later
develop antisocial symptoms.
Behavioral Programs
Another research area that is related to the study
of environmental causes is the role of parental modeling
in the development of juvenile fire setting. In his review
of the literature, Fineman (1980) discussed the learning
to burn model. Citing Kafry (1977), Fineman developed a
model composed of four steps leading to the development of
a fire setting child.
1. The first step involves predisposing factors, in
cluding both environmental and constitutional sources.
Some of these factors might include family, school and be
havioral problems for the former category and chromosomal
abnormalities, physical problems, enuresis, and possibly
sexual dysfunction for the latter. Given this set of fac
tors, the child must next learn to burn. Fineman (1980)
has noted that there is a relatively high incidence of fire
play and fire interest in the child population at large.
By age’three, this interest is exceptionally high among
boys.
2. The next step in the model requires the child to
begin to play with matches. In order for this to occur at
54
such a young age, poor parental supervision must be in
volved. It is also possible that- the child is imitating
parental behavior when he first begins to strike the match.
The manner in which the child's first interest in fire is
directed, may be one of. the major determinants in the de
velopment of future fire setting behavior. At this age,
the child may be taught that playing with fire is accep
table .
3. The next step in the model involves the antecedent
or crisis event which actually precipitates the act of fire
setting. These are age-related and may include such novel
situations as lighting candles or cigarettes or the more
serious events of a death in the family, a divorce, or the
addition of a new step-parent, all of which may lead to the
development of anger and frustration in the child. Teen
agers are frequently motivated by peer pressure and
antagonism for authority.
4. The final step in the model involves the reinforce
ment for the fire setting act. For the young child, this
might involve the establishment and the general observation
of environmental change. For others, the attention that the
fire draws from parents and peers may be enough. For the
more disturbed, the sensual reinforcement of the flames,
the power of the fire, and the magic of the incendiary
transformations are reasons enough to repeat the act.
55
Given these sets of circumstances, the learning to burn
model predicts that the fire setting response will remain.
Daly (1975) used a parent-training model as a
comprehensive early intervention program for families with
problem children. Youth who had been referred to a clinic
for such problems as aggressiveness, disobedience, theft
and truancy, were placed in a program with their parents.
The major goal of the program was to teach the adults new
skills to modify their child's behavior in an effort to
prevent future problems. Group as well as individual
training techniques were used: parent training, consulta
tions, youth skill training and school contacts. As a re
sult of this program, Daly (1975) reported that there was
a significant impact on changing the youth's behavior as a
direct result of parental intervention.
Finally, similar results were reported by Knappe
and Jurth (1969). Fire setting was viewed as a normal
phenomenon in the development of male children. In the
study which included children and juveniles who had set
fires, a high incidence of mental retardation was found.
These children had imitated careless acts of adults in
using matches around incendiary materials. In the recom
mendations, Knappe and Jurth (1969) suggested that
instruction of both parents and children in fire preven
tion was strongly indicated.
________ 5£
Patterson (1976) developed a behavioral treatment
program that can be useful in the area of fire setting as
well as in other types of juvenile crime. In the research,
Patterson (1976) found that in aggressive families,
aggressive behavior is reinforced by parents. Because of
this, intervention strategies are stressed in the area of
parenting skills, including contingent positive and nega
tive reinforcement. The teaching of prerequisite social
skills for children are also indicated. With this orien
tation, the therapist would view fire setting as a behavior
motivated by the dynamic interaction between the child's
developmental history as well as immediate environmental
events. Using this approach, treatment would be concen
trated in the area of changing the family milieu. Other
treatment strategies would include stimulus and reinforce
ment control, and changing tendencies toward fire play.
Personnel ascribing to this orientation would stress the
development of programs in fire education to be used during
the early formative years. The intention of such programs
would be the prevention of maladaptive behavior and the
elimination of fire setting as a response, to various ante
cedent conditions.
Organic Causes
Organic causes for ciminal and psychopathic beha
vior continue to be postulated and much research in the
57
current literature is devoted to. this important area.
Hreidarsson (1973) reviewed the case of a mentally retarded
male with Klinefelter's Syndrome and a presumptive Y/15
chromosomal translocation. His personality development was
different from others in his family who were mentally re
tarded but did not have the translocation. He was more
immature, impulsive, and more difficult to get along with
than were his siblings. He had committed arson on several
occasions and had a record of crime against property. He
also had a strong sexual drive, being bisexual, exposing
himself and masturbating openly. Chromosome abnormality
continues to be one of the areas studied in the etiology of
fire setting and criminal behavior.
Chromosomes of male patients in a security prison
were studied by Bartlett (1968) in Oxford, England. A
chromosomal survey of 204 male inmates in a security prison
revealed five individuals of abnormal karyotype. All of
the inmates were being treated for psychiatric disturbances
and all were adjudicated for some sort of crime. Both of
the XYY subjects were homosexual with schizoid personali
ties. Arson was part of the offense or previous convictions
of both XYY and XXY subjects.
Brain-wave activity was studied by Hurley and
Monahan (1969). They found an excess of slow waves in the
resting E.E.G. record in a high percentage of arsonists.
58
No particular conclusion was drawn from this finding and
brain-wave activity continues to be researched today.
Kido (1970) studied a group of six murderers and
one arsonist and found a low threshold for epileptic
seizure discharges when seizure-inducing drugs were
administered. These low thresholds for drug-induced
seizures were related to recidivism and murder in a
larger adolescent delinquent sample. Kido (1970) believed
that these low thresholds may represent a biological pre
disposition to criminal activity and, in particular, a
manifestation of the fighting instinct and aggressive
drive. These findings are in need of additional data be
fore they can become more than a source of speculation.
Woolf (1977) looked at the effects of physical
abnormalities on social behavior and fire setting. He
reviewed the history of an arsonist suffering from Moebius
Syndrome or a congenital facial diplegia. The patient was
suffering from a complete bilateral paralysis of the facial
nerves, so that he was. unable to show facial expressions
and his eye movements were severely limited. This male
patient had done fairly well during the first 20 years of
his life. At that time, however, he encountered social
rejection by his fellow employees. As a result, he set
several small fires in the laundry room in the nurses' home
where he lived. Woolf (1977) speculated that abnormalities
59
may form the backbone of a problem but it might be environ
mental causes that finally push the.fire setter to act.
Indeed, if this is the case, one cannot continue to study
the problem in isolation, but must view it as part of a
total picture combining various theories of possible
causation.
Studies by Siegelman (1969) and Siegelman and
Folkman (1971) suggested that there was a high incidence of
allergies and respiratory problems in individuals who
engage in fire setting and related acts of delinquency as
a response to particular environmental conditions. Again,
researchers have come to view the problem in combination
with other factors, rather than to lay it at the doorstep
of one particular pet theory.
Summary
Following the general discussion about the nature-
nurture controversy and the role that the locus of control
model plays in the fire setting model, studies were revieved
about the characteristics and various causes of fire
setting. The discussion began with sexual explanations as
first postulated by Freud (1932). Studies were then pre
sented in the area of emotional disturbance as the under
lying motive, environmental explanations, parent modeling
theories, and genetic theories. Treatment models build
from the particular theorist's orientation.
60
Researchers espousing a genetic-physical point of
view would concentrate upon locating the exact offending
cause and applying appropriate remedial procedures, such as
a new drug or a new procedure designed to cure, or at
least to alleviate, the symptoms of the problem.
The environmental specialist would want to look at
the behavior under which it would occur. The child would
not be treated as an isolate but family members, school
conditions, environmental stresses would all be studied
and considered. After an evaluation of all impinging fac
tors, remedial techniques would be applied. Researchers
using this model might want to be involved in family
therapy techniques.
Those who hold to a sexual explanation may want to
find a more appropriate outlet for this natural drive. If
the case is one of sexual repression, the goal would be to
alleviate the problem.
Perhaps the researcher would suggest insight
therapy, wherein the patient could learn to understand the.
motives behind his fire setting acts, and to direct his
sexual energies toward a more satisfying and natural con
clusion.
Those who believe that parental modeling and poor
parental supervision is involved, would most likely assume
a course along the behavior modification paradigm. The
61
theorist would want to involve parents in a program designed
to teach more effective and consistent skills.
Whatever theory one accepts, it must be understood
that the problem exists and is very real. On a daily
basis, firemen across the country are losing their lives
because of manmade disasters. Perhaps future researchers
would do better to concentrate their energies toward fire
education and programs in fire prevention, along with the
designing of scales that would screen out potential fire
setters. This is the type of research that is needed now
and this is the goal to which this investigation was direc
ted .
Hypothesis Generation
In the literature review, research was presented
suggesting possible personality and motivational charac
teristics of the juvenile fire setter and the sociopath.
One of the theories presented suggested that sociopaths
view their world as outside of their immediate control.
Using this model, questions for this investigation were
generated in the area of locus of control and it was as
sumed that the fire setters and sociopaths in this study
would be externally motivated. Juvenile delinquents have
been described as having problems in such sociopathic
behaviors as truancy, running away, use of alcohol and
drugs. Questions were developed from this source of data
62
to formulate a scale testing sociopathic behavior. Other
studies suggested that sociopaths and fire setters would
have, difficulties in peer relationships, family and inter
personal communications, psychological well-being, either
of the neurotic or psychotic type, accident proneness,
bedwetting problems, tendencies toward'violent behavior,
low intellect, school-related problems, and many other
effected life areas.
In the following chapters these areas of research
interest have been formulated into testable hypotheses. It
was the goal of this particular study to translate some'of
the literature into operationally defined hypotheses so
that they could be tested, and ultimately be assembled
in such a manner as to be useful to the policeman on the
beat, the fireman fighting a fire, and the school counselor
in his/her counseling. In Chapter III these will be pre
sented in testable form.
63
CHAPTER III
METHODOLOGY
Overview
This chapter deals with important methodological
background and procedures involved in. the study. The first
section presents a description of the methodology used in
the research. This is followed by a discussion of the
research design. Other sections of this chapter consider
the selection of subjects? instrumentation and procedures .
used? a description of the data collection, recording,
processing and analysis; methodological assumptions? and
possible limitations of the study. '
Research Methodology
The method of research used in this study was
correlational. Isaac and Michael (1979, p. 14) define
this approach as "an investigation to the extent in which
variations in one factor correspond to variations in one
or more other factors."
In this study correlation coefficients were used
to determine whether fire setters responded differentially
64
to various personality characteristics as compared to
groups of normals and sociopaths. The relationship between
a fire setter's and his parent's responses to a question
naire was studied to determine if they were different from
responses given by the other two groups.
Research Design
For this study, it was determined that the indepen
dent variables would include the three groups of subjects,
namely: the fire setters, sociopaths, and the normals,
using both the subjects themselves and their parents. The
three groups were presented with a questionnaire which
would also be considered in the independent variable
category. The dependent variables included the subject's
responses to the parental and children's form.
On the parental questionnaire, there was a total of
seven dependent variables, since parents had to indicate
whether they and their children exhibited certain charac
teristics "never," "sometimes" or "frequently."
On the children's form, there were nineteen
possible dependent variables, since one of the tasks was to
determine where the subjects fell on different personality
scales. Categorical variables used in this study included
sex, ethnic status, and socioeconomic levels, while the
continuous variables were intelligence and grade levels.
65
The study was designed using the mailed question
naire and research interview format with the former pro
cedure most often used with the parental form, and the
latter used with the children's. This design was selected
since it was the most appropriate method in which to gather
data, making comparisons among the groups possible in this
study. This design also allowed the gathering of data
amenable to factor analysis, analysis of variance, and
analysis of covariance procedures. This design was chosen
as the best procedure to use in the testing of the follow
ing null hypotheses:
1. Juvenile fire setters would not score highly on
a scale designed to measure soeiopathic tendencies.
2. Juvenile fire setters would not score highly on a
scale designed to measure fire setting interests
and activities.
3. Fire setters and sociopaths should score well on
a scale designed to measure good positive adjust
ment .
4. Juvenile fire setters should score poorly on a
scale designed to measure the speed with which
activities are done.
5. Juvenile fire setters as well as sociopaths should
not score highly on a scale designed to measure
a high degree of psychotic tendencies.
66
7.
8.
9.
10.
11.
12.
13.
Fire setters and sociopaths should not score
highly on a scale designed to measure degree of
interest in childlike activities.
Fire setters and sociopaths should not score
highly on a scale designed to measure a high
degree of accident proneness.
Juvenile fire setters, as well as sociopaths,
should score highly on a scale designed to measure
truth-telling„
Juvenile fire setters and sociopaths should
score highly on a scale designed to measure a
high degree of fears and frightened behavior
patterns.
Juvenile fire setters and sociopaths should
score highly on a scale designed to measure a
high degree of responsibility-taking.
Juvenile fire setters and sociopaths should
score highly on a scale designed to measure
a high degree of positive social interactions.
Sociopaths and fire setters should score poorly
on a scale designed to measure a high degree of
male interests or machismo.
Fire setters and sociopaths should score highly
on a scale designed to measure a high degree of
positive self concept.
67
14. Fire setters should score highly on a scale
designed to measure good positive socialization
skills.
15. Sociopaths and fire setters should score differently
from normals on questions designed to measure degree
of depression.
16. Fire setters and sociopaths should score highly
on questions concerning positive parental inter
actions, similarly to normals.
17. Sociopaths will not score higher on questions of
violence than fire setters or normals. Fire
setters should not score higher on questions about
violence than do normals.
18. Sociopaths and fire setters will not score lower
than normals on questions designed to measure a
high degree of anxiety.
19. Older fire setters do not demonstrate more signs
of emotional disturbance than do younger fire
setters.
20. Parents of fire setters and sociopaths will not
respond with a higher incidence of maladaptive
behaviors than parents of normal subjects.
21. Parents of fire setters and sociopaths will not rate
their own lives as more unhappy and disturbed than
parents of normal juveniles.
68
Selection of Subjects
A total of 132 boys participated in this study.
Only boys were chosen for participation based upon data in
the literature which suggested that only seven percent of
arsonists were girls. The subjects were divided into
three groups
1. Normal Group - 69,
2. Fire setters Group - 32,
3. Sociopathic Group - 31„
The normal group was selected from volunteers of
students who attended Nevin Elementary School and South
Gate Junior High School in the Los Angeles Unified School
District. The experimenter randomly chose classrooms from
the total possible number of classes in grades two through
nine. The experimenter then spoke to each classroom and
gave out parent permission-to-participate forms. Children
in the elementary school were given a treat for participa
tion. At the junior high, a small monetary reward was used.
The elementary school used in the study was predominantly
Hispanic in population, with thirty percent of the students
as Black. There were no Caucasians registered. The school
received Title I Federal Funds as an Educationally Impacted
School. Teachers were paid extra compensation for working
at Nevin Elementary School, since it was also designated
under the Los Angeles Desegregation Program as an Urban
69
Classroom Teacher School. This meant that the school was
considered hard-to-staff, and extra funds were designated
for that purpose. The junior high was made up of a popu
lation consisting of sixty percent Hispanics, thirty
percent White, and the remaining numbers distributed among
Blacks and Asians. The socioeconomic status of the school
at the time of the study was predominately middle class.
As the study progressed, it was determined that
more older normal subjects would be needed. These subjects
were taken randomly from volunteers among the faculty mem
bers of the two schools and from the psychologists working
with the Los Angeles Unified School District, Area 2.
Parents would volunteer if they had a son between the ages
of sixteen and eighteen.
The sociopathic group was represented by adjudica
ted juvenile delinquents incarcerated at Camp Kilpatrick,
a maximum security facility operated by the Los Angeles
County Probation Department. These subjects were not
randomly selected. The experimenter reviewed all the entry
data cards of the youths to determine the ages, ethnic
status and background histories. Only subjects below the
age of eighteen were selected. Since this was a security
camp situation, there were no juveniles less than thirteen
years of age. The experimenter, by examining data cards
eliminated those youths who had arson or fire setting
20.
tendencies. There was an effort made to have an equal
representation of Hispanic, Black and Caucasian subjects.
However, according to camp officials, there were more
Blacks and Hispanics than Caucasians in the program.
Because of the ethnic situation, the experimenter was un
successful in obtaining an equal representation of ethnic
groups in the sociopathic population.
The fire setter subjects were also not randomly
selected by the examiner. Because of extreme difficulties
in obtaining community cooperation, the experimenter was
able to gather but a relatively small number of subjects.
Most of these were taken from the incarcerated populations
at Camps Kilpatrick, Miller, and Turner, all part of the
Los Angeles County Probation System. These youths were all
serving time for arson-related activities or had arson in
their backgrounds.
A few subjects were available within the local
school system and were being educated in the Educationally
Handicapped Classes. Two subjects were obtained when the
experimenter participated in interview situations with the
Santa Ana Fire Department. All of the other subjects were
interviewed by persons other than the experimenter,
suggesting, the possibility of a loss of experimenter con
trol, interfering variance, and experimenter error.
Subjects were interviewed by probation officers, Los Angeles
71
County Fire Inspectors, and a Fountain Valley, California,
Fire Marshall. Each experimenter was given a set of in
structions in order to standardize the format. When the
interview was over, the auxiliary experimenters sent the
results to this writer for collection and analysis.
Because of this situation, no control of ethnic, economic
or other variables was possible.
The parental questionnaire was sent by mail with
instructions and a stamped self-addressed envelope to all
the parents of the normal and sociopathic groups. All of
the parents who had children in the camp situation for
arson also received their questionnaire by mail. The re
maining parents of fire setters were instructed by the
probation officers as to the nature of the study and asked
to participate. Parents of children involved with the fire
departments were interviewed by the fire officials and
were told that the questionnaire was part of the investiga
tive process. In no case was any pressure applied to
parents to complete the form.
Instrumentation
There were two instruments used in this study.
The parent questionnaire was developed as part of a funded
project of the Federal Emergency Management Agency,
specifically, the United States Fire Administration, Office
of Planning and Education. This questionnaire was developed
72
as part of an interview format to be used by fire department
personnel in an effort to distinguish between those fire
setting children who were merely curious and those who were
potentially more disturbed and in need of psychological
help. This questionnaire was first published as part of a
training manual entitled Interviewing and Counseling ,
Juvenile Firesetters (Fineman, et al. 1979). The question
naire lists 88 possible behaviors to which the parent re
sponds with a circle under a "never" behavior; a "some
times" occurring behavior or a "frequently" occurring
behavior. After the parent completes the checklist, the
fire department official scores the responses by adding
them up and dividing them into one of three categories.
According to the manual’s instructions, if there
are many behaviors listed under the "frequently" occurring
category, this is an indication of a referral needed to an
outside source. If the responses are mainly in the "never"
category, the problem may li& in the area of educational
interventions that are needed and may be taught by the
interviewing officer. It was the purpose of this study to
determine if the questionnaire did indeed distinguish
between those who are normal and those who are the more
disturbed fire setters. In an effort to control for the
possibility of false positives, a third group was added,
the sociopaths. This was to control for the possibility
that the tests were measuring sociopathic behaviors alone.
_______________ ___________ 73
The children's questionnaire was developed by the
examiner and covered various personality traits commonly
associated in the literature with sociopaths and fire
setters. The questionnaire had a two-fold purpose, to
identify specific questions that discriminated fire setters
from sociopaths and normals and to develop scales of
behaviors in which fire setters tended to belong. By
identifying specific questions to which fire setters tended
to respond, it was hoped that a future instrument could be
developed for use by fire department officials in their
investigation of suspected juvenile arsonists. The valida
tion of the personality scales was an attempt to determine
if previously studied identifying characteristics of fire
setters were correct.
Data Collection, 'Processing and Analysis
Each subject was interviewed by the examiner, the
probation officer, or the fire department official only
once. The questions were read to the subject and a yes or
no was circled as the response. When subjects said
sometimes they were told to choose the answer that was the
most correct. In some cases, no response was recorded
when the subject declined to answer or did not think that
the question applied to him. For the parental question
naire, the purpose of the study was outlined in a cover
letter enclosed in the envelope containing the questionnaire
74
with a stamped self-addressed envelope. Parents of normal
subjects were told that their help was needed in an effort
to determine those behaviors that are most characteristic
of their normal sons. Parents of sociopaths and arsonists
in camp situations were told that their sons had been
interviewed and their help was being solicited in an effort
to determine ways in which help could be provided to
juveniles before they ended up in a camp situation. Proba
tion officers told parents that a research project was
being done and asked for their cooperation. Parents in
volved in the fire department interview program were told
that the form was a part of the entire process.
At the top of the parental questionnaire, standard
ized instructions were given as to how to score the desired
responses. Although parents were encouraged to answer the
entire questionnaire, many were returned with blank spaces,
or with n/a written by a response, or, in some cases, with
every response marked. In this latter case, the question
naire could not be scored and was not used in the data
analysis. A Spanish and English copy of the questionnaire
and cover letter were available.
Data processing was done at the University Computer
Center, located at the University of Southern California.
The Statistical Package for the Social Sciences (SPSS) and
Statistical Analysis System (SAS) programs were used to
____________ 75
generate the following descriptive information:
1. ages of respondents,
2. grades of respondents,
3. respondent's ethnic group,
4. socio-economic levels,
5. intelligence quotients,
6. group categories,
7. frequency tables of all questions in the
children's questionnaire with information obtained
as to the mean, median, mode, standard deviation,
standard error, and range for all data,
8. frequency tables on all questions in the
parent's questionnaire,
9. frequency tables on the nineteen scales on the
children's questionnaire generated by factor
analysi s.
Data processing was also used to obtain the
following inferential information:
1. factor analysis of the children's questionnaire
with two methods:
(a) factor matrix using principal factor with
iterations ,
(b) varimax rotated factor matrix,
2. one-way analysis of variance with nineteen
children's factors.
___________________________________________ 76
3. one-way analysis of variance with all 100 items
on the children's questionnaire,
:4. analysis of variance with the modified LSD proce
dure to determine equivalence of groups for:
(a) age of respondents,
(b) respondent's grade in school,
5. chi square used to determine equivalence of
groups in:
(a) socio-economic levels,
(b) respondent's intellectual quotients,
6. one-way analysis of variance on seven parent
factors with the Duncan multiple range test to
test means of significant difference among groups,
7. one-way analysis of variance on the total 88 item
parent questionnaire with the Duncan multiple
range test,
8. one-way analysis of covariance with the 19 children
factors on the 13 to 17 year old scale using age as
a covariant, and the Least Squares Means model for
testing significant means,
9. one-way analysis of covariance with the 100 items
on the children's questionnaire for 13 to 17 years
old using age as a covariant and the Least Squares
Means model,
______________________________________________________________________ 17.
10. one-way analysis of covariance on the seven parent
factors using age as the covariant and the Least
Squares Means procedure,
11. one-way analysis of covariance on. the 88 parent
items using age as the covariant.
Methodological Assumptions
The following assumptions were made in this study:
1. The research design, control methods, subject
selection procedures, and data processing techni
ques used in this investigation were appropriate to
the intent of the study.
2. Subjects used in this study were a representative
sample of the population from which they were
selected.
3. Data were accurately gathered, recorded, stored
and analyzed.
4. The data yielded by the study were appropriate for
analysis by inferential parametric and nonparame-
tric statistical procedures.
5. Results of the data analysis were interpreted
fairly and accurately.
6. The conclusions drawn from data analysis followed
logically and accurately from the objective
analysis and did not reflect experimenter bias.
78
7. It was assumed that the test instruments designed
and selected for use in this study yield informa
tion considered valid and reliable for the purpose
of the investigation.
8. It was assumed that the procedures used in this
study were effective in discriminating among fire
setters, sociopaths and normals.
9. It was assumed that the results of the study could
be applied in the future prediction and fifferen-
tiation of fire setters.
10. It was assumed that the subjects in this study
understood, the questions that were read to them.
11. It was assumed that the conceptual models studied
in this research were valid and constituted a
viable framework suitable for testing.
12. It was assumed that both parents and children
who served as subjects answered all questions
fairly and accurately.
13* It was assumed that all persons who served as
examiners in this study followed the guidelines
and used the standardized format.
Limitations of the Study
The following limitations existed and limited the
generaiizibility of the study's conclusions:
1. The number of subjects available for study was
small, thereby limiting the strength of the
79
of the conclusions that were drawn from the
statistical procedures.
2. Very limited cooperation was available from local
fire, police, and probation departments, causing
not only a limited number of subjects to be
studied but introducing a possible biasing factor.
3. This study was done in a year when fire setting
activity by juveniles was down appreciably over
the preceding two years.
4. The normal subjects were drawn only from paid
volunteers and the children of faculty members,
possibly causing a biasing factor in the
selection of subjects.
5. Not all parents responded to the parental
questionnaire. Of those responding, not all
parents answered all the questions on the
questionnaire. Both of these factors limit the
generalizibility of the questionnaire.
6. The questions that were finally determined to be
valid in terms of differentiating among the
three independent groups were not standardized
on an appropriate sample with national norming.
7. The limited size of the sample did not allow for
the control of all relevant variables.
80
8. The examiner was unable to administer all the
tests, possibly inducing an experimenter bias
factor.
9. The examiner limited the range of the age groups
that were represented in the study. Therefore,
study results cannot be assumed to be valid
for children under seven or over eighteen.
10. Conclusions drawn from the data analysis were
limited to the extent to which assumptions under
lying those procedures had been satisfied.
Summary
Using a correlational design, three groups of
subjects, fire setters, normals and sociopaths were given
children’s and parent’s questionnaires to complete. The
first group was interviewed on a person-to-person basis
while the last responded by mail.
Using this format, answers were sought to determine
whether or not the three independent groups responded
differentially on the children's and parent's question
naires .
In addition, the children's questionnaire was
composed of 19 personality scales, based upon a factor
analytic procedure. A total of 132 boys participated, all
of whom ranged in age from seven to seventeen.
81
Using descriptive and inferential statistics, data
processing was completed. It was assumed, that the methodo
logical assumptions, underlying the appropriate use of
various statistical procedures were met.
Limitations of the study were outlined and mainly
involved the lack of subjects willing to participate, lack
of cooperation from community agencies, and a lack of con
trol over experimental bias. It was suggested that study
results be interpreted in view of these limitations.
82
CHAPTER IV
ANALYSIS AND INTERPRETATION OF FINDINGS
Descriptive Analysis
In order to run a descriptive analysis on the data
gathered for categorical variables, a coding system had to
be devised for computer operation. For the ethnic category
White subjects were listed as ones, Black subjects were
listed as twos, and Hispanic subjects were listed as threes
and the one Asian subject was listed as ' four.
For intelligence, the following system was devised:
1. highest group for intelligence 1
2. middle group fors intelligence 2
3. lowest group for intelligence 3
The first group included those individuals with an intellec
tual score above 110; the middle group included those from
80 to 110; and the lowest included all those before 80.
Socio-economic status were listed using the same
coding system: individuals reporting income above $40,000
were listed in the highest group; those with incomes between
$12,000 and $40,000 were listed in the middle group, and
83
all the rest were listed in the lowest group.
Results of the descriptive analysis were summarized
in Table 1. Incomplete data was programmed as missing
values, and usually resulted from a lack of information re
garding that particular trait.
Table 1
Means, Medians, Modes, and Standard Deviations
of Categorical Variables
Standard
Category Label • Mean Median Mode______Deviation
Age 12.863 13.333 14.000 2. 502
Grade 7.108 7.591 9.000 2. 580
Ethnic 2.008 2.000 1.000 0.861
Socio-economic 2.584 ’ 2.728 3 . 000 0.612
Intelligence 2.036 2.028 2.000 0.602
Group 1.712 1.457 1.000 0. 825
The analysis of results indicated that the mean age
of the subjects was 12.9 years old with most of the respon
dents fourteen years of age. The bulk of scores fell
between the eleven and fifteen year range. The distribu
tion was slightly negatively skewed, indicating that there
were more older subjects .in the study.
The average grade of the subjects was seventh with
a majority of youths falling within the seventh to ninth
grade range. The overall spread was from second to first
84
year college with nine subjects in the former grade and one
in the latter. Again, the distribution was slightly nega
tively skewed, suggesting a tendency toward more subjects
in the higher grade levels.
The ethnic distribution was approximately equal for
Whites and Hispanics with 46 subjects falling in the former
category and 45 in. the latter. Blacks were somewhat
smaller in representation since a total of 37 participated.
Asians were the most under-represented with only one sub
ject participating. Data were missing on three subjects
for this variable.
For socio-economic status, it was determined that a
majority of subjects fell within the poverty range with the
resulting distribution negatively skewed. Data were not
available for all subjects on this particular variable.
Intelligence information was the most incomplete of
all the categorical variables. A specific request had to
be made by the probation officer before a subject was
tested. In some cases, this may have been the result of
the officer’s feeling that a more serious deficit existed.
As such, not all subjects were tested. Overall, this
should have led to some skewing of data toward the lower
end of intellectual ability. A review of the table,
however, indicates that this was not the case, since the
85
average score among the 56 cases for whom data were
available was within the 80 to 110 or average I.Q. range.
Information regarding group status can be pulled
from Table 1. The preponderance of subjects fell within
the normal range with a total of 69 youths participating.
Fire setters were second in number with a total of 32
subjects, and socio-paths were the smallest with 31 parti
cipants. The distribution was positively skewed.
Previously cited data indicated that the majority
of subjects in the study were normals who were white in
terms of ethnic status, or normal intelligence, and within
the poverty level of socio-economic status. Overall, they
were older in age, with the average being 13 and in the
seventh to ninth grades.
Evaluation of Initial. 'Equivalence'
of Groups
Using one-way analysis of variance, a check for
initial equivalence of groups was run for the variables of
ages of respondents and their grades. The modified LSD
procedure was then used to determine significant mean dif
ferences. The results of the analysis are in the following
table:
86
T a b le 2
Analysis of Variance for
Age of Respondents
Source of
Variation
Sum of
Squares df
Mean
Square F ratio
Between
Groups 106.9246 2 53.4623 9.685*
Within
Groups 706.5982 128 5.5203
Total 813.5227 130
Note:
*Significant at the .0001 level
87
T a b le 3
Analysis of Variance for
Respondent's Grade in School
Source of
Var iation
Sum of
Squares df
Mean
Square F ration
Between
Groups 11.3268 2 56.1634 9.559*
Within
Groups 746.1603 127 5.8753
Total 858.4871 129
Note:
*Significant at the .0001 level
An analysis of the previous two tables indicate
that the three groups were not equivalent. The normals and
fire setters were significantly younger than the sociopaths.
These two groups were not different from each other, how
ever. Differences also existed among the groups in terms
of grade levels. Again, normals and fire setters were not
significantly different from each other but were different
from the sociopaths since they tended to be older.
Additional analyses of variance were run to deter
mine if significant differences existed between normals and
88
fire setters with respect to age and grade level, using a
seven to twelve year old grouping and a thirteen to seven
teen year grouping. The results of those analyses are
listed in the following tables:
Table 4
Analysis of Variance for
Age of Respondent
(7 to 12)
Source of Sum of Mean
Variation Squares elf Square F ratio
Between
Groups 7.0360 1 7.0360 2.386
Within
Groups 135.6305 46 2.9485
Total 142.6665 47
89
T a b le 5
Analysis of Variance for
Grade of Respondent
{7 to 12)
Source of
Var iation
Sum of
Squares df
Mean
Square F ratio
Between
Groups 1.5000 1 1.5000 0.465
Within
Groups 148.4789 46 3.2278
Total 149.9789 47
90
Table 6
Analysis of Variance for
Age of Respondent
(13 to 17)
Source of Sum of Mean
Variation Squares df Square F ratio
Between
Groups 17.7285 2 8.8643 6.918*
Within
Groups 102.5125 80 1.2814
Total 120.2410 82
Note:
*Signif icant at the .01 level
91
T a b le 7
Analysis of Variance for
Grade of Respondent
(13 to 17)
Source of Sum of Mean
Variation Squares dT Square F ratio
Between
Groups 5.8495 2 2.9248 1.531
Within
Groups 149.0392 78 1.9108
Total 154.8887 80
An analysis of the preceding tables indicated that
normals and fire setters did not significantly differ with
respect to age and grade level when the groups were
analyzed restricting the age range from seven to twelve.
The older groups did differ significantly with respect to
age but not grade. In this analysis,, the fire setters were
older than either the normals or the sociopaths. The data
92
indicated that a control for age was necessary in any sub
sequent analysis.
To continue the analysis for initial equivalence
among groups, chi square was used to analyze data with re
spect to ethnic, socio-economic, and intelligence factors
for all the subjects. The guiding hypothesis that the
three groups would not differ in terms of ethnic status was
not maintained. Among the normal subjects, there was a
higher percentage of Hispanics. In the fire setting group,
there were more Blacks. These results were interpreted as
an artifact of the sampling situation, rather than an in
dication that more fire setters are White than Black and
more normals are Hispanic than White.
In terms of income, the three groups did not differ.
For the chi square analysis, it was determined that the
majority of subjects fell within the poverty range. In
intelligence, the three groups did not differ. The majority
were in the average range.
The analysis was then run for the same three
variables using the younger, or seven, to twelve year old
grouping, alone. This analysis indicated that most normals
were Hispanics while the majority of fire setters were
t
White. There were no sociopaths in this grouping. In
socio-economic status, there was a significant difference
between the two groups, with the normals falling into the
93
lowest category while the fire setters fell into the middle
income range. Again, this was considered to be a result of
the sampling situation. With respect to intelligence level,
there were no differences. The majority of subjects were
in the average range.
With the older subjects, there were significant
differences among the three groups in ethnicity:
the majority of normals were White,
the fire setters were White,
the sociopaths were Black.
With respect to income, a statistically significant differ
ence was established since the sociopaths were poorer than
the other two groups. There were no significant differences;
in intelligence, however.
In summary, an analysis of initial equivalence of
groups revealed that the three groups were not equivalent
and a need for a controlling factor for age among the older
subjects was indicated. Other factors, such as ethnic
status were also not equivalent but were considered to
be an artifact of the sampling situation from which they
were drawn, rather than a statistically significant
finding„
94
Hypothesis Testing of. the
Parent 'Questionnaire' (7-12)
The developers of the parent questionnaire had
weighted individual responses to each particular stimulus
or question, so that a circle around a frequently occurring
behavior in one question may be highly significant of emo
tional difficulties* A circle around another frequently
occurring behavior on another question may not have been
significant* In order to include this information in the
statistical data, a general linear model procedure was
used. There were a total of 88 questions on the question
naire. The questions were divided into two groups, those
dealing with parent behaviors and those concerning children
and their actions. Eighty of the questions concerned the
latter's actions while the other eight were concerned with
the former category.
Using a coding procedure, the responses to each -
question were counted, and totals were drawn for the number
little concern, definite concern and extreme Concern
items. This was done separately for the children's as
well as the parent's responses. Then, totals were found
for both groups in the little and definite concern cate
gories. There were no responses to be marked for extreme
concern in the parent's section of the form. With this
procedure, it could be determined whether or not the
95
scoring system, as developed, was indeed measuring what it
purported to measure. An analysis was then run for each
individual question to determine those that were answered
significantly different among the three groups using the
younger and older age differentiation already described.
A one-way analysis of variance was run for each dependent
variable. Table 8 lists the findings of the seven analyses.
Table 8
Analysis of Variance for the Number of
Little, Definite, and Extreme Concern Items for
Children and Parents (7-12)
Dependent
Variable F PR >F
Little Concern (€) 13.94 0.0017*
Definite Concern (C) 18.44 0.0005*
Extreme Concern (C) 9. 66 0 .0064*
Little Concern (P) 6.11 0.0243*
Definite Concern (P) 4.25 0.0548*
Total of Little Concern (CP) 16.53 0.0008*
Total of Definite Concern (CP) 17.46 0.0006*
Note:
*Signifleant at the .05 level
96
A review of Table 8 indicated that the scoring
system, as devised, does indeed differentiate among the
two groups. There was a statistically significant differ
ence in the number of little concern items that normals
answered as compared to fire setters. These same findings
occurred for all categories of responses.
In order to determine how each group answered,
the Duncan’s Multiple Range Test was run using an alpha
level of .05 to determine significance.
In Table 9, a differential analysis by group is
presented.
Table 9
Analysis of Means to Determine
Differential Group Responses (7-12)
Dependent Significant Significant
Variable Mean N Group Status
Little Concern (C) 6-7. 300* 10 Normal
Definite Concern (C) 21.000* 9 Fire setter
Extreme Concern (C) ■2. 444* 9 Fire setter
Little Concern (P) 6.100*
10
Normal
Definite Concern (P) 3.333 9 Fire setter
Little Concern (CPI) 73.400* 10 Normal
Definite Concern (CP2) 24.333* 9 Fire setter
Note:
*Significant at .05
97
The table indicated that the groups differed as
anticipated. It would be expected that normals would have
more little concern items than fire setters. This dif
ference was in fact sustained. It would also be expected
that fire setters would have more definite concern items
than normals. This difference was also sustained. Al
though the grouping for numbers of definite concern items
for parents was significant, there were no significant
differences between the parents of the fire setters and
those in the normal category for this classification. This
was the only unexpected finding in this analysis.
An analysis of variance was then run for each
individual question or stimulus on the 88 item parent
questionnaire to determine those questions that signifi
cantly discriminated between the two groups.
Table 10 shows the results of that analysis.
98
T a b le 10
Analysis of Variance for
Individual Questions (7-12)
Dependent
Variable F . . . PR ^ F Mean Group
Hyperactivity 8.82 0.0095 1.5555 Fire
Behavior Problems 32.21 0.0001 2.0000 Fire
Impatience 7.16 0.0160 1.4444 Fire
Lying 4.47 0.0495 1.3333 Fi re
Excessive Anger 5.35 0.0335 1.5555 Fire
Sleep/Waking Problems 4.47 0.0495 1.3333 Fire
Fantasizing 11.18 0.0038 1.5555 Fire
Poor or no Eye
Contact 7.16 0.0160 1.4444 Fire
Child has seen
Therapist 17.89 0.0006 1.6666 Fire
s.?Learning Problems 6.40 0.0223 2.0000 Fire
Impulsiveness 10.00 0.0060 1.5555 Fire
Panic When Fire
Out of Control 6.40 0.0223 2.0000 Fire
Out of Touch with
Reality 10.00 0.0060 1.5555 Fire
Stealing 5.33 0.0346 1.375
Cruelty to Children 5.33 0.0346 1.375 Fire
Di sobedience 5.33 0.03 46 1. 3 75 Fire
Fighting with Peers 4.76 0.0454 1.375 Fire
Long History of
Behavior Problems 22.06 0.0003 1.7142 Fire
Poor Loser 6.07 0.0263 1.625 Fire
Good in Sports 4.76 O’. 0454 1/375 Fire
Child Shows Off 4. 20 0.0596 1.37 5 Fire
From the information presented in Table 10, it was
learned that a total of 21 questions significantly dis
criminated between fire setting and normal children. The
differences were as hypothesized. Fire setters were more
hyperactive, had more behavior and learning problems in
school, and were often impatient and impulsive than normal
children. Fire-setting children tended to show off, were
often poor losers, may steal, and often were involved in
incidents involving cruelty to other children.
One of the interesting findings^evident from the
data was that everyone of the fire-setting children was
described by his parents as demonstrating learning and
behavioral problems in school, and as panicking when the
fire got out of control, an unexpected finding.
All of the normal subjects reported that they were
good in sports. A statistically significant difference was
found in this area also, suggesting that many fire setters
are not good athletes.
The results of the parent's questionnaire have
broad implications for people working in the school system.
When children begin to exhibit a high degree of anger, show
learning and behavior problems, are impulsive, and demon
strate such psychological problems as excessive; fantasizing,
and sleep difficulties, they are providing danger signs.
100
If those of us who are employed in the school
system could begin our interventions at. this time, we may
be able to head off a potential arsonist or high-risk child
in general. The cost would be little when compared to the
damage in property and loss of lives that inevitably occur
when the fire setter strikes.
Hypothesis 'Testing, of the
Parent Questionnaire (13-17)
Essentially the same procedure was followed to
determine significance for the parent questionnaire with
the older subjects. This grouping was composed of three
groups, however, including the sociopaths. Since age was
previously found to be a source of confounding variance,
an analysis of covariance procedure was used as the test
statistic.
In Table 11 the results of the analysis on the
number of little, definite, and extreme concern items are
presented«
101
T a b le 11
Analysis of Covariance for the Number of
Little, Definite and Extreme Concern Items (13-17)
Dependent
Variable F PR >F
Little Concern (C) 8.33 0.0008*
Definite Concern (C) 3.64 0.0001*
Extreme Concern (C) 3. 86 0.0277*
Little Concern (P) 5.41 0.0075*
Definite Concern (P) 3.34 0.0437*
Little Concern (CPI) 8.90 0.0005*
Definite Concern (CP2) 12.98 0.0001*
Note:
^Significant at o05 or better
A review of Table 11 indicated that the scoring
system did indeed differentiate among the three groups in
terms of total parental responses to the number of little,
definite and extreme concern items that their children
demonstrated. Using the Least Squares Means procedure, a
differential analysis was made to determine exactly where
the differences lay. For the little concern items, normals
responded differentially to either of the fire setters or
102
the sociopaths. These latter two groups, however, did not
differ from each other. On the number of definite concern
items, all three groups differed significantly from each
other. On the number of extreme concern items, fire setters
were distinctly different from the other two groups in that
they had more of these items than either the normals or the
sociopaths. Interestingly, the normals and sociopaths did
not differ significantly from each other.
On the number of little concern items for the
parents, the normals were different from either the fire
setters or the sociopaths in that they had more of these
items, as would be expected. Fire setters and sociopaths
were not significantly different from each other. For the
category of definite concern items related to parental be
havior, fire setter’s parents significantly differed from
normals but not from sociopaths. Fire setters, as hypothe
sized, had more of these items than either of the other
groups. On the totals, fire setters and sociopaths were
different from normals in the number of little concern
items that were answered. As expected, normals had more
of these items than either of the other two, but neither
the fire setters nor the sociopaths differed from each
other.
For the total number of definite concern items,
parental and children's questions, all three groups differed
_____ 103
significantly. Fire setters had the most number of
definite concerns, sociopaths the second highest number
normals had the fewest number of definite concerns.
An analysis was then run for each of the 88 items
on the parent’s questionnaire to determine, those that were
the most indicative of fire/setting and other sociopathic
disturbances.
104
T a b le 12
Analysis of Covariance for
Individual Questions on the
Parent's Questionnaire (13-17)
Dependent
Variable F PR y F
2 Lack of Concentration 3. 43 0.0406
4 Behavior Problems in School 9.68 0.0003
5 Impulsiveness 8.33 0.0008
6 Impatience 5. 04 0.0103
18 Lying 5.73 0.0058
19 Excessive-uncontrolled Anger 8. 66 0.0006
20 Child Shows Violence 4.90 0.0117
21 Stealing 21. 55 0.0001
22 Truancy 11.14 0.0001
24 Child Shows Cruelty
toward Children 3.03 0.0580
29 Runs Away from
Home and School 5.99 0.0048
30 Disobeys 5.12 0.0096
31 Long History of
Behavior Difficulties 10. 59 0.0002
33 Expresses Anger
by Hurting Self 6.16 0.0045
35 Easily led by Peers 46.82 0.0001
36 Jealousy 5.33 0.0082
40 Shows Off 6.34 0.0036
49 Anxiety 4.17 0.0214
48 Sleep or Waking Problems 3.69 0.0324
73 Child has seen a Therapist 9.04 0.0005
13 Depression 4.23 0.0204
41 Severe Depressions
and Withdrawals 4.53 0.0158
83 . .S t range,, Qual i ty
about Child 3.32 0.0447
105
An analysis of the results indicated that 23
questions significantly differed in terms of responses
from the three independent groups*
On several of the questions, fire setters responded
differently from the other two groups. In other cases,
sociopaths responded differently. There were no clear-cut
trends in the data. Fire setters and normals differed
significantly on lack of concentration, jealousy,
demonstrating behavior problems in school, impulsiveness,
running away from home and school, disobeying, truancy, and
in being easily led by others.
Fire setters and sociopaths did not differ from
each other on these questions, many of which cover socio-
pathic behaviors generally accepted in the literature as
part, of the definition of the antisocial personality. Fire
setters and sociopaths did differ on the expression of ex
cessive and uncontrolled anger, with the former group some
what higher in this behavior than the latter. They also
differed on a tendency to show violence with fire setters
again higher on this trait than the sociopaths. Fire
setters are more likely to have severe depressions or
withdrawals than are the sociopaths. The same finding was”
evident in the area of impatience. Tendencies toward
showing off were again more characteristic of the fire
106
setter than the sociopath. They also have more feelings
of anxiety, and are more likely to express their anger by
hurting themselves. Interestingly, fire setters are more
likely to experience sleep or waking problems. In all of
these cases, sociopaths and normals did not differ fron
each other. Only the fire setters deviated among the
three groups.
Certain questions were answered differently by all
three groups. One such question concerned a long history
of behavior difficulties. Fire setters were the most
likely to have a response in the affirmative, while normals
were the least likely to have a long history of behavior
problems. Sociopaths were in the middle.
A surprising finding occurred in response to the
frequency of stealing behavior. Fire setters were the
most likely to exhibit this behavior while normals, again,
were the least likely. Again, sociopaths were in the
middle.
Another question which distinguished among the
three groups concerned the child having seen a therapist.
Fire setters were the most likely to respond that they
had visited a therapist; normals were the least likely,
and sociopaths were in the middle.
In summary, it would appear that the major scoring
system of the parent's questionnairre, that of counting the
107
number of responses to each category of little, definite,
and extreme concern items, does indeed do as it says it
does. The system differentiates among those who are in
need of further referral and those who are exhibiting
behaviors of little concern. Overall, as hypothesized,
fire setters and sociopaths tended to score similarly,
while normals were usually the group to score somewhat
differently. As expected, normals were most likely to
demonstrate behaviors of little concern while fire setters
and sociopaths demonstrated behaviors of definite or
extreme concern. On the number of extreme concern items
for children and the total number of definite concern items
for children and parents, all three groups differed signi
ficantly, with fire setters in the,, highest range.
On a question-by-question analysis of the parent's
questionnaire, using the thirteen to seventeen year olds,
fire setters and sociopaths again tended to score similarly.
This was in keeping with the original hypothesis that fire
setters and sociopaths are very similar. On several items,
however, fire setters distinguished themselves. They are
more likely to seek professional treatment, more likely to
have depressions, and more than likely to have a long
history of behavior problems.
, 1 - Q R .
Effects of Factor Analysis Upon the
Children1s Questionnaire
In order to set up scales on the children's ques
tionnaire to determine if significant behaviors could be
lumped together, a factor analysis was run. The analysis
was run twice, with the first consisting of a free factor
analysis and the second one based upon the experimenter's
determination as to which items went together. The free
factor analysis allowed the computer to choose those items
that fit together. Using this format, a total of 32
factors were listed, representing a total of 78.7 percent
of the common variance.
The second analysis was chosen by the examiner
sinceit required, the computer to fit the 100 items on
the children's questionnaire into 11 factors. These factors
were chosen on the basis of the hpotheses presented in
Chapter One and as a result of the Review of the Litera
ture. Those factors included:
1. hyperkinesis,
2. fire interests,
3 • psychotic behaviors
4. sematic complaints,
5. peer relationships,
6. school problems,
7. family problems,
10 9 /
8. locus of control,
9. sociopathic tendencies,
10. self concept,
11. general personality factors.
Using a predetermined correlation coefficient of
.45, items were clustered together with their appropriate
factors. Using this method, a total of 46 questions were
identified and used. The criteria for determination of
items with factors was then changed and a lower correla
tion coefficient was selected in order to account for more
of the questions. Face validity was also used in this de-
c
termination. Using this procedure, a total of 19 factors
was finally selecteddand the questions testing those factors
were determined and lumped together. In Table 13 these
factors are identified and those questions that were grouped
with that factor are indicated. A negative sign in front
of a number indicates that a no response is expected in
keeping with that factor.
110
Table 13
Factors and
Belonging to
Questions
that Factor
Factor Question Numbers
1. Sociopath 3, -24, 15, 34, 32, -46
50, -59, 72, -83, -84, 63
2. Fire 16, 81, -48, 86, 79, 28, 96
3. Positive Adjustment 1, 8, -14, 33, 38, 57,
85, 95, 100
4. Psychotic 55, 68, 87, 93
5. Getting Things Done 20, 27, 91
6. Child-like 19, 26, 43
7. Accident-prone 53, 71
8. Negative Behavior
and Fears 60, 62, 80
9. Truth 11, 25
10. External Locus
of Control 9, 22, 74
11. Responsibility 29, -31
12. Peers 5, 78, 97
13. Machismo 21, 36, 45
14. Socialization 17, 35, 40, 52, 58, 76
15. Self-concept 12, 51, 64, 100
16. Depression 18, 23, 37, 44, 56, 75,
92, 94
17. Parents 66, 73
18. Violence 54, 77
19. Anxiety 82, 90, -99
111.
A review of Table 13 indicates that a total of 81
questions were identified and placed into factor categories.
Using this information, a one-way anova was run to deter
mine if the factors as identified were significant.
In Table 14, these results are presented for the
seven to twelve year olds. Table 15 is presented for the
older boys of thirteen to seventeen years, using the anova
as the test statistic.
112
T a b le 14
One-way Analysis of Variance
With 19 Children's Factors (7-12)
Factor F PR >F
Sociopathie 28.34 0.0001*
Fire setter 145.46 0.0001*
Positive Adjustment 4.65 0.0364*
Getting Things Done 0.60 0.4441
Psychotic 0.11 0.7442
Child-like 3.11 0.0847
Accident-prone 1.80 0.1863
Truth Teller 1.93 0.1719
Negative Behavior/Fears 0.02 0.8810
Locus of Control 2. 51 0.1201
Responsi b i1i ty 12. 53 0.0009*
Getting Along with Others 1. 53 0.2227
Machismo 0. 55 0.4609
Self-concept 0. 61 0.4381
Socialization 2.88 0.0967
Depression 0.04 0.8504
Parent Relationships 0. 65 0.4234
Violence-prone 2.90 0.0952
Anxiety Response 2.03 0.1612
Note:
* denotes significance
113
T a b le 15
One-way Analysis of Covariance
With 19 Children's Factors (13-17)
Factor F PR >F
Sociopathic 12.04 0.0001*
Fire setter 7.63 0 . 0009*
Positive Adjustment 2. 46 0.0919
Getting Things Done 0.03 0.9685
Psychotic 0.56 0.5756
Child-like 1.09 0.3425
Accident-prone 8.13 0.0006*
Truth Teller 5.06 0.0086*
Negative Behaviors 0.40 0.6743
Locus of Control 1. 91 0.1547
Responsibility 2. 25 0.1122
Getting Along with Others 0.85 0.4316
Machismo 9. 60 0.0002*
Self-concept 2.17 0.1209
Socialization 0.41 0.6646
Depression 1.00 0.3716
Parent Relationships 1.00 0.3736
Violence-prone. 1.31 0.2745
Anxiety 1.46 0.2388
Note s
*Significant
114
A review of Table 14 indicates that four scales were
significant in distinguishing among the groups in the
younger category, and five scales were significant for the
older subjects. Interestingly, only two of the scales
overlapped between the younger and older subjects. For the
younger children, the fire setters, as expected, were more
likely to score highly on the sociopathic, and fire setting
scale while the normal subjects did better in the areas of
the acceptance of responsibility and on the scale designed
to measure degree of positive adjustment. None of the
other scales were significant, confirming the null
hypotheses for each of the remaining fifteen identified
factors.
For the older subjects, five scales were signifi
cant, with results as expected. On the sociopathic scale,
fire setters were significantly different from normals, but
not from the sociopaths. Normals were significantly dif
ferent from both groups and were the most likely to score
poorly on this scale. Interestingly, fire setters and
sociopaths had least squares means scores that were very
close to each other with sociopaths only slightly ahead
(0.5210-0.5468).
On the fire setter scale, fire setters were again
significantly different from normals but not from
sociopaths. Normals were again significantly different
___________________________________________________________________________________________________________________________115__
from the sociopathic group. As expected, normals were the
least likely to score highly on this test. Sociopaths who
did not differ from fire setters, were somewhat less
likely than the fire setters to obtain a higher score
(sociopaths 0.2038; fire setters 0„2538).
A scale that was significant for the older groups
but not the younger subjects was accident-proneness.
Fire setters and sociopaths did not differ significantly,
but both were different from the normals. As expected,
normals were the least likely to score highly on the scale.
Fire setters tended to score in the middle while socio
paths were the most likely to be accident-prone. Again,
as explained, the fire setters and sociopaths continued to
score similarly.
The truth teller scale was also significant in the
same direction as the three previous scales. Normals were
the most likely to tell the truth. Although fire setters
and sociopaths were not significantly different, the trend
of responses suggested that the latter groups would be the
least likely to tell the truth.
The final scale to achieve significance was the
machismo or male interests scale. Fire setters and socio
paths did not differ significantly. Fire setters also did
not significantly differ from the normals. Only the normals;
differed significantly from the sociopaths who were the
116
most likely to score highly on a scale designed to measure
degree of interest in traditionally male oriented activi
ties. This was also the only scale where fire setters and
normals did not significantly differ.
Summary
In conclusion, a factor analysis was run for the
children's questionnaires, using the analysis of variance
technique for the younger subjects and analysis of covari
ance for the older groups. A total of 19 factors was
determined using a predetermined correlation coefficient
as a cut-off point as well as face validity. Four scales
were determined to significantly discriminate between the
fire setters and normals in the younger group and five
scales for the older groups. In the younger category,
results were as hypothesized for the four scales which
included sociopathic, fire setter, positive adjustment,
and responsibility-taking. The null hypothesis was re
tained for the remaining 14 sclaes. For the older group,
five scales including sociopathic, fire setter, accident-
proneness, truth teller, and machismo interests were signi
ficant with results as expected. In the majority of cases,
fire setters and sociopaths tended to score similarly, as
hypothesized in Chapter One.
117
Hypothesis Testing of the
Children1s Questionnaire (7-12)
In Table 16 the results of the analysis of variance
for the children's questionnaire, ages seven to twelve,
are reported. Only those questions which discriminated
between the normal and fire setting subjects were listed.
Table 16
Analysis of Variance of the Questions
Which Discriminated Between Normal and
Fire Setting- Subjects
(7-12)
Question Dependent
Number Variable F PR >F
7 I Love to Read 7.76 0.0077
14 I Sleep the Whole Night
Through 4.45 0.0403
16 I Have Set Some Fires (No error)
17 I Have Lots of Friends 14.91 0.0004
22 ' I Have No Control Over
the Things I do 9.95 0.0029
24 I Have Run Away From Home 14.91 0.0004
28 I Like to Play with Matches 26.98 0.0001
29 I Love to Daydream 8.73 0.0049
31 People Can Count on me
to do Things 7.05 0.0108
34 I Earn my Good Grades 14.91 0.0004
36 I Have been in Many
Fist Fights 8.60 0.0052
37 I Wish I Were More
Like my Friends 4.50 0.0395
118 ‘
Table 16
(continued)
41 Sometimes I Walk in
my Sleep 3.90 0.0544
43 I Get Excited When I
See Fire Burning ■ . 5.55 0.0228
46 I Have Taken Drugs 6.50 0.0143
48 I Really Enjoy Doing
Good Work in School 5.44 0.0242 -
50 .I Always Try to do
What my Parents Say 14.91 0.0004
52 I am Happy Most of
the Time 5.44 0.0242
53 I Have Had a Lot of
Accidents 4. 47 0.0399
58 I Feel Good When I am
with Other People 6.70 0.0129
59 I Like to Miss School 22. 50 0.0001
63 I'd Never Take Anything
Without Paying 39.13 0.0001
64 Everything I do is Great 5.73 0.0208
72 I Love Math 7.76 0.0077
79 When Angry I Feel Like
Burning something 9.42 0.0036
81 I Enjoy Fooling Around
With Matches 13.65 0.0006
82 When Angry I Feel Like
Throwing Things 9.16 0.0040
83 I Would Like to
Try Alcohol 10.82 0.0020
86 I Have Set Fires that
I Couldn't Put Out 39.13 0.0001
95 I Love to Swim 6.10 0.0173
119
A total of 30 questions was found to significantly
discriminate between the normal and fire setting subjects
in the younger groups. Using the Duncan's Multiple Range
Test, it was determined that normal subjects are more
likely to enjoy reading, have more friends, and be more
likely to enjoy math than are fire setters. Normals are
also more likely-to do as their parents request. They also
seem to enjoy doing good work in school. Normals are more
likely to assume credit and to accept the responsibility of
earning good grades in school. In other words, they
demonstrate signs of internal locus of control perceptions.
Normals are more inclined to describe themselves as happy
most of the time, and feel that they can be counted upon to
do things when needed. Normals are more likely to indicate
a fondness for swimming than are fire setters. This latter
group, even at this young age, is more likely to express a
desire to consume alcohol or may have already done so.
They are also more likely to indicate a desire to stay home
from school. Normal subjects indicated that they enjoy
being around other people, more so than the fire setting
subjects, who expressed a wish to be more like their
friends. Normals are more likely to feel that they control
their own actions and their own destinies than do fire
setters, who spend more of their time daydreaming. Fire
120
setters are more likely to admit to having taken drugs than
are normals. They are also more likely to indicate that
they have taken things from stores without paying. Fire
setters are more likely to have accidents and definitely
more inclined to be involved in fist fights.
On the questions dealing specifically with fire,
the results were as expected. Fire setters were more
likely to respond in the affirmative that they have set
some fires. They are more likely to play with matches,
and are more inclined to feel like burning something when
they are angry. Interestingly, an unexpected result was
found when normals reported that they were more likely to
get excited when they saw a fire burning than were the fire
setters. The opposite would have been expected. As
hypothesized, fire setters are more likely to indicate that
they have set fires that get out of control than are
normals.
On other questions of a general nature, normals are
more likely to sleep the whole night through and have
fewer problems with sleep walking than are the fire settera
They are more likely to hold positive feelings about them
selves, and when angry, are less likely to feel like
throwing things than are their fire setting counterparts.
121
Summary
Fire setters, overall, did tend to respond as ex
pected on most of the significant questions. They were
more likely to answer affirmatively on the fire setting
and sociopathic questions than were the normals who were
more likely to answer appropriately on questions con
cerning positive life adjustment, good family relation
ships and a good adjustment to school life.
Hypothesis Testing of. the
Children's Questionnaire (13-17)
Results of the analysis of covariance for the
thirteen to seventeen year olds reporting only those
questions that discriminated among the three groups:
fire setters, sociopaths and normals. The results are
reported in Table 17.
122
T a b le 17
Analysis of Covariance of the
Questions Which Discriminated Between
Fire Setter, Sociopath and Normal Subjects
(13-17)
Question Dependent
Number Variable PR >F
8
11
16
24
21
28
31
35
36
46
53
I always pay attention
in School 4.57 0.0133
I love to play with
other kids 4.66 0.0123
My parents and I do a
lot of things together 3.14 0.0489
It's usually not my fault
when I get into trouble 6.90 0.0017
I always tell the truth 7.15 0.0014
I have set some fires 22.24 0.0001
I have run away from home 4.95 0.0094
I live with one parent 3.88 0.0247
I like to play with matches 3.31 0.0417
People can count on me
to do things 3.80 0.0267
I trust a lot of people 4.67 0.0121
I have been in many fist
fights 27.01 0.0001
I have taken drugs 22.79 0.0001
I have had a lot of
accidents 8.49 0.0005
123
Table 17
(continued)
63 I'd never take anything
without paying for it 18.96 0.0001
71 I have been in the
hospital a lot 3.33 0.0409
72 I love math 3.43 0.0373
75 I wish I had as much
stuff as my friends have 5.85 0.0043
83 I would like to try
alcohol 8.48 0.0005
84 Sometimes I stay home
when I am well 4.88 0.0101
86 I have set fires I
couldn't put out 7.46 0.0011
99 I feel bad when I do
something wrong 6.79 0.0019
A total of 24 questions were found to significantly
discriminate among the three groups. Not all of the
questions were the same as the younger group, although,
interestingly, the general categories or underlying themes
were very similar. Using the Least Squares Means analysis,
it was determined that normals differed significantly from
124J
sociopaths in aswering the question, "My parents and I do a
lot of things together." Normals were the most likely to
respond in the affirmative while fire setters, who did not
differ from either of the other two, were in the middle,
and sociopaths, the most likely to respond in the negative.
Normal subjects were more likely to trust other
people than either fire setters or sociopaths, with these
last two groups not differing significantly. On another
question in the general area of positive self-adjustment or
self-concept, such as "I feel good when I am with other
people," fire setters and normals scored similarly while
sociopaths differed significantly from both of the other
groups. This was a rather unexpected finding. It had
been assumed that if fire setters were very much like
sociopathic subjects, they would score similarly as did
the sociopaths. On this question, this was not the case.
In a question related to positive school adjust
ment (I love math), normals differed only from sociopaths
in that the former group was much more likely to respond
in the affirmative. A trend toward a significant differ
ence was found between the fire setters and normals but the
level of significance was just beyond the accepted standard
for this investigation (.05).
125
In response to the question, 111 love to play with
other kids," a result similar to the last question was ob
tained. Normals were significantly different from socio
paths with the former group again, much more inclined to
answer in the affirmative. There were no significant dif
ferences with the fire setters with either of the other
two categories.
Fire setters were significantly different from
normals on the question, "I always tell the truth," with
normals being significantly different from sociopaths. On
this question, fire setters and sociopaths tended to
respond similarly. Normals were the most likely to answer
yes to this question, while the other two groups would re
spond with the opposite answer.
Normals again differed significantly from sociopaths
on the question, "People can count on me to do things."
Although it has been hypothesized that fire setters and
sociopaths would respond similarly, here again that re
sponse was not found.
On seven questions that revolved around a theme of
positive adjustment, in most instances, the normals would
differ significantly from the sociopaths but there would be
no differences between the fire setters and either of the
other two groups.
126
Only three questions dealing with fire setting
turned out to be significant. In response to, "I have set
some fires," all three groups differed significantly.
However, this must be construed as an artifact of the draw
ing of the sample, since this was one variable carefully
screened in the choosing of the subjects. In response to,
"I have set fires that I couldn't put out," fire setters
and normals differed significantly. Normals answered no
almost all the time. Fire setters and sociopaths demon
strated a slight trend in this direction of responding, but
apparently, there were enough yes answers to this question
to cause the demonstrated significant differences.
This same finding was also evident in response to
"I like to play with matches." Fire setters and sociopaths
scored similarly while normals differed significantly from
each of the other two groups. Fire setters would most
likely respond in the affirmative while normals would give
the opposite answer.
Several questions fell into a large grouping known
as sociopathic. On the question, "I have run away from
home," fire setters and sociopaths were again similar in
their rate of responding. Fire setters differed: from nor
mals who again differed from the sociopaths.
This same finding was evident again on the question,
"I have been in many fist fights." This was also the case
__________ ____________________________________________________________ 127
on the question, 111 have taken drugs." Normals were the
least likely to respond in the affirmative while sociopaths
were the most likely to respond with a yes answer.
This statistical finding was also evident on the
questions, "I'd never take anything without paying for it,"
and "I would like to try alcohol." On the former question,
as expected, the normals were the most likely to respond
yes while the sociopaths were the most likely to engage in
stealing activities. On the latter question, an interes
ting finding was evident. Although not statistically sig
nificant, it was the fire setters who were more inclined to
state that they had tried or would like to try alcohol,
rather than the sociopaths. It would have been expected
that the latter group would have responded in the affirma
tive with greater frequency, perhaps making the result
statistically signif ican't.
The response trend of only normals differing signi
ficantly from either fire setters or sociopaths was again
evident on the question, "I feel- bad when I do something
wrong." Normals, again, were the most likely to say yes,
with sociopaths the least likely to respond in such a man^
ner. As before, fire setters and sociopaths were not
statistically different.
128
Another question that produced some interesting re
sults was "I like to miss school." Only normals and socio
paths differed, with fire setters not differing from either
group. This was one of the few questions where fire set
ters did not clearly distinguish themselves in either
direction.
Several questions revolved* around illnesses and a
tendency toward accident-proneness. In answer to the
question, "I have had a lot of accidents," the same trend
in responding was again evident. Fire setters and socio
paths were significantly different from normals in their
responses, but were not different from each other. Normals
were more likely to say no to this question than were the
other two groups. In response to, "I have been in the
hospital a lot," normals differed only with respect to
sociopaths. As cited in an earlier case, fire setters did
not clearly establish themselves as different from either of
the other two groups. As before, normals were more likely
to say no to this question than were the sociopaths, an
expected finding.
Fire setters again did not clearly differentiate
themselves from either sociopaths or normals on question 84,
"I sometimes stay home even when I am well." The only
differences of significance were between normals and socio
paths with the latter group responding affirmatively.
123
On a question designed to measure tendency toward
hyperactivity, as well as truth telling, fire setters
differed significantly from the normals but not from the
sociopaths. As before, normals were statistically differ
ent from sociopaths in response to, "I always pay attention
in school." Normals were the most likely to say yes, to
this question, with fire setters the least likely to
respond affirmatively, an interesting finding but not of
significance statistically.
On a question designed to measure locus of control
and the assumption of responsibility, an interesting find
ing was apparent. In response to, "It is usually not my
fault when I get into trouble," sociopaths differed from
both the normals and fire setters who did not differ from
each other. Sociopaths, by responding no to this negatively
stated question were assuming responsibility for their
actions. Fire setters were the most likely to say yes,
thereby negating their responsibility for their behavior.
Normals were in the middle. Several reasons may be enter
tained about this unexpected finding. Although the experi
menter tried to explain to subjects the meaning of a yes
or a no on this question, it is possible that many did not
understand it, even with the explanation. Another reason
for this unexpected finding could have been the Probation
Department's emphasis on the assumption of responsibility
130
for one's own behavior. Since all of the sociopaths were
inmates at camp facilities and regularly attended sessions
where this was discussed, this may have influenced the
response given for this question.
On a question about family relationships, specifi
cally, "I live with one parent," sociopaths differed
significantly from fire setters and normals who did not
differ from each other. Sociopaths, as expected, were the
most likely to say yes to this question. Fire setters and
normals did not establish a clear differentiation in their
responses.
Finally, a question designed to measure degree of
self-satisfaction, "I wish I had as much stuff as my
friends," revealed a significant difference only between
normals and sociopaths, with the latter group responding
no more often than the former. It had been anticipated
that the sociopaths would indicate a greater desire to have
more things than either of the other groups, either due to
poor economic conditions or as an expression of envy and
dissatisfaction. In actuality, this did not turn out to be
the case. When interviewed by the examiner, the only ex
planation given by the subjects was usually that they had
enough stuff already.
131
Summary
To summarize, 24 questions were found to signifi
cantly discriminate among the three groups. Although they
were not all the same as the younger group, many of the
major themes underlying the grouping of questions were
similar. It had been hypothesized that the older group
would be significantly.more disturbed than the younger fire
setters. This apparently was not the case. Overall, fire
setters did respond similarly to sociopaths, as hypothe
sized. However, there were some instances where this was
not the case, and only normals and sociopaths differed. If
the fire setters, as anticipated, in the older group were
more disturbed than the younger group, it would seem that
more fire setters would have distinguished themselves on
questions dealing with fire setting tendencies and psycho
tic ideation. This, in fact, did not turn out to be true,
suggesting that no inference can be made at this time about
qualitative differences between younger and older fire set
ters.
Summary
Of all the subjects participating in the study,
the majority fell within the normal classification, mean
ing no fire setting behavior, nor were they adjudicated
offenders known to the Probation Department. The majority
132
were White in terms of ethnic status, of normal intelli
gence, and within the poverty level of socio-economic
status. The mean age of subjects was 12.9 years. An
analysis of initial equivalence of groups revealed that the
three older groups were not the same and a need for a con
trolling age-factor was indicated. In addition, subjects
needed to be divided between younger and older groups,
since there were no sociopaths under the age of thirteen.
There were other variables upon which the three groups
differed. They were seen as artifacts of the drawing of
the sample and not as valuable statistical findings. The
hypothesis testing of the parent's questionnaire for the
younger subjects revealed that the scoring format per
formed as expected. In addition, a total of 21 behaviors
were found to significantly discriminate between the two
groups of subjects. Most of the findings were as hypothe
sized .
On the parent's questionnaire for the older sub
jects, the scoring system again was validated as hypothe
sized. For this age group, a total of 23 questions were
found to significantly discriminate among the groups. On
this form, many of the questions found the fire setters
and sociopaths responding similarly. In some cases, the
behaviors significantly discriminated among all three
groups.
133-
Analyses were run for the children's questionnaire.
A total of four scales proved to be significant for the
younger subjects. They included sociopathic, fire setter,
positive adjustments and responsibility. For the older
subjects, five scales were significant, including socio
pathic, fire setter, accident-proneness, truth teller,
and tendencies toward machismo interests. On the older
scale, as hypothesized, fire setters and sociopaths tended
to score more similarly than normals and fire setters.
When individual questions were analyzed, 30 were
found to discriminate between the fire setting and normal
groups, suggesting that these may be the most valuable for
fire personnel and others to use in the screening of poten
tial children with fire-setting tendencies. For the older
subjects, 24 questions were found to discriminate. In most
cases., fire setters and sociopaths scored similarly and
both groups were usually significantly different from nor
mals. On a few questions, this finding was not maintained
and fire setters and normals scored similarly. Overall,
the hypothesis that qualitative differences would emerge
between the younger and older fire setters with the latter
group being much more disturbed, did not hold out and the
null hypothesis was maintained.
134
CHAPTER V
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS
Summary
Problem
This investigation was directed toward a current
nationwide problem: arson. It has been described as the
nation's fastest growing crime and one of the nation's
most serious human-made disasters.
This study looked at one aspect of this rather
large problem, the juvenile fire setter. According to
local experts, this single group has been responsible for
at least 35 of every 100 fires set. Various reasons have
been entertained as to why juveniles engage in this form
of antisocial behavior. Despite a wealth of research in
the area, little of it has actually been used by those pro
fessionals in the field who deal with the juvenile fire
setter on a daily basis. The research abounds with
theories but devotes little time to practicality.
A group of researchers from various fire depart
ments and local mental health facilities addressed this
problem by viewing the fire setter as being motivated
13 5
primarily by two different factors. Younger children,
especially boys, are often curious about fire and may en
gage in various forms of fire play. Older boys, as well as
younger, may also be motivated by an entirely different
reason, that of serious emotional disturbance. The group
of researchers known as the Fire Services and Arson Preven
tion Committee used this paradigm as the background for the
development of an interviewing system designed for use by
fire department personnel. A parental questionnaire de
scribing potential fire setting characteristics is included
in that interviewing system. Through the use of this
format, the researchers suggested that fire setters could
be divided into two major categories:
1. those who would benefit from educational
prescriptions,
2. those whose fire setting behavior is so serious
that outside interventions are necessary.
This study was designed to validate the question
naire. In addition, a children's questionnaire was de
veloped to determine if certain characteristics were more
indicative of juvenile fire setting behavior than others.
By developing a series of questions that effectively dis
criminated among fire setters, normals, and sociopaths, it
was anticipated that a screening instrument could be
developed to be used as part of the interviewing system
1 26
already in use. These, then, were the goals to which this
research study was directed.
The following questions were asked at the beginning
of this study:
1. Would juvenile fire setters respond any differently
to a series of questions about personality and motivational
factors than either juveniles convicted of other crimes or
normals?
2. Would there be qualitative differences between
younger and older fire setters?
3. Would parents of juvenile fire setters respond any
differently to a series of questions about personality and
motivational factors than either parents of juveniles
convicted of other crimes or parents of normals?
4. Are juvenile fire setters a distinct group or do
they belong to the larger category defined as sociopathic?
5. Would juvenile fire setters respond any differently
to a variety of scales of behavior than either juveniles
convicted of other crimes or normals?
Review of the Literature
In the review of the literature, the age-old con
troversy concerning the nature-nurture problem was presen
ted as it related to this particular study. Theories have
been generated that were totally biological in orientation
137
while others view the problem strictly in terms of environ-
mental or sociological causes. Rotter's (1977) interpre
tation of the locus of control model was presented and it
was suggested that juveniles engaged in fire setting and
other illegal activities are externally oriented. Socio
paths, as viewed in this investigation, are people unwilling
to assume responsibility for their own actions. The blame
is laid elsewhere. Other characteristics of the sociopath
were also presented, and they were defined as individuals
always in trouble, profiting neither from experience nor
punishment, and maintaining no real, loyalties to any per
son, group or code. The subject in this study described
as antisocial was assumed to be incorrigible, committing
thefts, truant from school, often running away from home,
associating with 'bad companions, and often displaying
physical aggression. He was described as exhibiting an
early onset for these behaviors. Frequent and repetitive
antisocial acts were assumed to be occurring, and the lack
of deep emotional attachments was also described as part of
the personality makeup. Finally, the absence of any other
psychiatric illness that could be used to explain these
behaviors was assumed to be evident.
Research into the etiology of juvenile delinquency
t
was presented, and it was demonstrated that many of the
same reasons suggested for the rise in antisocial behavior
138
could also be used in the explanation of fire setting ac
tivity. Although theories have ranged from sexual to bio
logical causes, current research has described the patho
logical fire setter as coming from a confused, disturbed,
and unstable family background, causing predisposition
toward a variety of antisocial acts (Fineman, 1980).
The history of the field has been rooted in psychoanalytic
explanations commencing with Freud’s (1932) hypothesis
that fire setting was sexually motivated. Out-of-control
aggression was another analytical concept proposed as a
possible cause of fire setting. Again, it was assumed the
act was motivated by sexual conflict. Studies were then
presented that viewed fire setting as rooted deeply in
emotional disturbance.
Environmental theories were reviewed and the role
of the family began to assume major importance. School
and delinquency was also studied and suggestions were pre
sented by various authors as to why education is no longer
relevant to several segments in our society.
Behavioral programs citing the learning to burn
model were reviewed and the role of the parent using appro
priate reinforcement strategies was emphasized. The review
of the literature concluded with a discussion about
possible organic causes and it was noted that promising
research is currently being conducted and may one day lead
139
to a genetic or totally physiological explanation for many
types of criminal behavior.
Methodology
Using a correlational design, a total of 132 boys
were separated into groups, based upon their past histories.
The normal boys numbered 69, with no fire-setting history.
The sociopathic group of 31 boys were incarcerated in
camps run by the Los Angeles County Probation Department.
The fire setting group of 32 came from the camp system,
Los Angeles Unified School District's special education
classes, and referrals from Los Angeles, Los Angeles County
and Orange County fire departments. The children were
individually .administered a 100-item questionnaire of the
yes and no variety.
Parents of the subjects received in the mail a
copy of the parent's questionnaire with a cover letter de
scribing the purpose of the study. The form was available
in English and Spanish. When the results were returned,
they were analyzed by computer using the SAS and SPSS
packages. Every attempt was made to standardize the admini
stration procedures. However, in certain instances, sub
jects were administered the questionnaire by someone other
than this examiner which could possibly introduce some
error variance.
___________________________________________________140
Descriptive and inferential information was obtained
such as the ages, grades, ethnic, socio-economic status,
and intellectual levels of the subjects. Other test sta
tistics used included factor analysis, analysis of variance,
and analysis of covariance.
The study was limited by a lack of participation
among local agencies who were most concerned about privacy
laws.
Due to the small size of the sample, results
should not be generalized beyond the scope and breadth
covered in this particular investigation.
Conclusions
At the beginning of this study, various questions
were raised that were to guide the development of the in
vestigation. Those questions will be reviewed and the
findings presented. All results must be interpreted with
caution, however, since many of the hypotheses were tested
with only a few questions, easily influenced by possible
lie factors, experiment-wide error, and various other error
factors.
Originally, the question was asked, "Would juvenile
fire setters respond any differently to a series of ques
tions about personality and motivational factors than
either juveniles convicted of other crimes or normals?"
14V
Although the findings were not clearcut and some overlapping
existed, overall, it can be said that fire setters and
sociopaths responded similarly on most questions. In some
cases, fire setters and normals responded similarly but
this was not in the majority of situations. A few questions
did discriminate fire setters from the other two groups.
The personality and motivational factors that were found
to be significant in this investigation are presented later
in this chapter.
T^G nuH hypothesis was maintained on the question,
"Would there be qualitative differences between younger and
older fire setters?" Both younger and older subjects
demonstrated similarities in the types of problems and
situations in which they found themselves. Although there
were differences in terms of questions that were found to
significantly discriminate the fire setters from the nor
mals, overall, the underlying themes were the same. In
no cases were any differences discovered that would have
suggested that older fire setters were profoundly more
disturbed than their younger counterparts.
The null hypothesis was maintained on every question
on the children's form that concerned seriously disturbed
behavior. On the parent's questionnaire there were indica
tions of emotional disturbance for both the younger and
older subjects, with the latter group presenting a few more
142
symptoms than the former group. Overall, however, there
was not a pronounced difference between, the younger and
older fire setters in terms of severity of emotional ill
ness. Therefore, the null hypothesis was maintained on
this question.
On the question, "Would parents of juvenile fire
setters respond any differently to a series of questions
about personality and motivational factors than either
parents of juveniles convicted of other crimes or parents
of normals," the answer was found to be yes, in at least
several instances. For the younger group, parents of
normals had more little concern items than parents of fire
setters. Parents of this latter group had more definite
and extreme concern items than the normals. In the older
group, parents of fire setters significantly discriminated
in their answers on the total number of definite concern
behaviors listed on the questionnaire. They were also
significantly different from the normals and sociopaths on
the number of extreme concern items. Thus, on this
question, the null hypothesis was rejected and the research
hypothesis maintained.
The answer to the question, "Are juvenile fire
setters a distinct group or do they belong to a larger cate
gory defined as sociopathic," is not clearcut nor precise.
For the younger subjects, there was no way to measure this
143,
variable and hence, no answer can be given. For the older
subjects where three groups, including one sociopathic did
exist, the answer must be that in general fire setters
scored similarly to sociopaths. Although there were in
stances where this was not the case, overall, the socio
pathic tendency was established.
Finally, on the question, "Would juvenile fire
setters respond any differently to a variety of scales of
behavior than either juveniles convicted of other crimes
or normals," the answer was a qualified yes. For the
younger fire setters, there were no ambiguities. On four
scales, sociopathic, fire setter, positive adjustment, and
responsibility, the fire setting group did respond in a
manner that was significantly different from the normals.
For the older subjects, five scales were found to discri
minate, including the sociopathic, fire setter, accident-
proneness, truth teller, and machismo interests. In terms
of significant differences among groups, however, the re
sults were not as clear. Fire setters did respond differ
ently than normals on all five scales of behavior. They
did not, however, respond differently than juveniles con
victed of other crimes. For this question, then, the null
hypothesis was partially maintained.
14 4
In addition to the preceding series of questions
that were asked, this investigation was guided by a number
of research hypotheses. These are reviewed and their find
ings presented.
1. The answer to. the first hypothesis, that juvenile
fire setters would score highly on a scale designed, to
measure sociopathic tendencies, was maintained. For both
the younger and older fire setters, the sociopathic scale
was identified as significant through factor analysis,
analysis of variance, and analysis of covariance procedures.
2. The second hypothesis concerned tendencies toward
pyromania. It was suggested that juvenile fire setters
would score highly on a scale designed, to measure this ten
dency. This hypothesis was maintained, for both younger
and older subjects. However, for the older group, fire
setters did not distinguish themselves from the sociopaths.
Both groups have interests in the area of pyromania.
3. It was hypothesized that both fire setters and
sociopaths would score poorly on a scale designed to measure
positive adjustment, since they were basically unhappy.
The research hypothesis was maintained for. the younger
group but not for the older subjects. As expected, fire
setters did not score as well as normals in. the area of
positive adjustment. No significant differences were found
for the older group, and the null hypothesis was accepted.
145
4. The literature suggested that one area often
characteristic of juvenile fire setters was that of hyper
activity, and a research hypothesis had been drafted such
that the fire setting group should have scored highly on
a scale designed to measure the speed with which activities
are completed. This hypothesis could not be accepted on "
the basis of the data generated in this ,study. No signifi
cant differences were found among the groups on this vari
able .
5. Since one area of research specifically suggests
that emotional disturbance is a basic etiological reason
for fire setting behavior, fire setters and sociopaths
should have scored significantly on a scale designed to
measure a high degree of psychotic tendencies. Normals,
using this theoretical model, therefore, should have scored
poorly. On the basis of the data gathered in this investi
gation, the research hypothesis had to be rejected. No
differences were found between fire setters and normals on
a scale designed to measure degree of psychotic tendencies.
6. It was assumed that fire setters and sociopaths
would be immature in their development. Therefore, they
should have scored highly on a scale designed to measure
childlike tendencies. In fact, this was not the case, and
the null hypothesis was maintained. Fire setters were no
more interested in childlike activities than normals were.
_____________________________________________________________________________________1A£_
7. A high degree of accident-proneness was hypothe
sized for both sociopaths and fire setters. In the litera
ture, this was one of the personality variables identified
among children with antisocial tendencies. For the
younger group, this did not turn out to be the case and the
null hypothesis was accepted. For the older group, how
ever, this result did occur. Fire setters and sociopaths
scored higher on a scale designed to measure a high degree
of accident-proneness than normals.
8. It had been suggested that fire setters and socio
paths would score poorly on a scale designed to measure
truth telling since sociopaths are defined as having little
or no conscience. The results were varied. For the
younger group,, the hull hypothesis was accepted. There
were no significant differences between normal and fire
setters. For the older subjects, a significant difference
did occur. Fire setters and sociopathic subjects, as
expected, did not do as well on a scale designed to measure
truth telling than did normals.
9. It had been hypothesized that fire setters and
sociopaths would not score as highly on a scale designed
to measure degree of fears and negative behaviors. This
had been suggested by the literature which stated that the
antisocial personality was essentially free of other
psychiatric disturbances that could possibly explain the
147
maladaptive behaviors. From this study, however, that
could not be ascertained. Neither the younger nor the
older subjects demonstrated any significant differences on
a scale designed to measure a high degree of fear.
10. It had been suggested that fire setters and socio
paths are externally motivated. Therefore, they should
have scored poorly on' a scale designed to measure degree of
responsibility-taking. For the younger subjects, the re
search hypothesis was accepted. Normals were more likely
to have an internal locus of control orientation and be
more willing, to assume responsibility for their actions
than were.fire setters. This was not the case for the
older group, however. No significant differences were
found among the three groups in this personality variable.
11. Since fire setters and sociopaths were assumed to
be loners with few friends and few positive interactions,
it had been hypothesized that they would score poorly on a
scale designed to measure degree of positive social inter
actions. In fact, this did not turn out to be the case.
Neither the younger nor the older groups demonstrated any
significant differences on this personality variable. The
null hypothesis was accepted.
12. It had been suggested that fire setters and socio
paths would indicate a high degree of interest in male-
oriented activities, since they might be concerned about
148
projecting a bravado or machismo image. For the younger
fire setters, this did not turn out to be the case.
There were no differences between normals and fire setters
on this variable. For the older subjects, the null hypothe
sis could be rejected, but with limitations. While this
scale did turn out to be significant, it was not in the
manner that was hypothesized. Only the normals differed on
this scale. Sociopaths and normals were significantly
different from each other. Fire setters were not different
from either of the other two groups.
13. Since sociopaths and fire setters are involved in
so many negative interactions with parents, police, and
others, it was hypothesized that they would score poorly on
a scale designed to measure positive self-concept, since
this variable is often defined as our internalization of
how we feel that others perceive us. In fact, this did
not turn out to be the case. There were no significant
differences in self-concept among the three groups for
either the younger or the older subjects.
14. Using the same reasoning as presented in Hypothesis
13, it had been expected that fire setters and sociopaths
would score poorly on a scale designed to measure positive
socialization skills. In fact, this did not turn out to be
the case. Fire setters and sociopaths did not score signi
ficantly different from the normal subjects, younger or
older.
149
15. Fire setters and sociopaths are often defined as
free of other types of personality disorders that could be
used as possible explanations for the deviant behavior.
Therefore, it had been hypothesized that normals, fire
setters, and sociopaths would score similarly on a scale
designed to measure depth of depression. In some cases,
this hypothesis was maintained. There were no differences
among the three groups that were significant. Since the
hypothesis held that there would be no differences among
the groups, it was accepted.
16. Normals would have better relationships with the
parents. Sociopaths and fire setters, on the other hand,
should have the opposite. Therefore, the latter groups
should have scored poorly on a scale designed to measure
degree of positive parental interaction. For all groups,
however, the null hypothesis had to be accepted. There were
no significant differences in degree of positive parental
interactions.
17. It had been suggested that sociopaths would score
the highest on a scale designed to measure tendencies
toward violent behavior. Fire setters were hypothesized to
fall in the middle while normals should have scored the
lowest. In fact, this did not turn out to be the case.
There were no significant differences among all the groups
on this variable.
' 150
18. Normals should score higher on a scale designed to
measure degree of anxiety, than either of the sociopaths or
the fire setters since it was assumed that anxiety often
accompanies neurotic behavior patterns or a highly de
veloped superego or conscience. In fact, this did not
turn out to be the case, and the null hypothesis had to be
accepted.
19. It had been hypothesized that younger fire setters
often act from curiosity and older fire setters from emo
tionality. Therefore, it was assumed that the older fire
setters would demonstrate more signs of emotional distur
bance in their responses to questions than the younger
group. 30 questions on the children’s form were found to
be significant for the younger group, and 24 were found
for the older subjects. There was some overlapping between
the two age levels, primarily on questions concerned with
fire setting, sociopathic behaviors, school-related ques
tions, and in the acceptance of responsibility for one’s
own actions. The other questions did not deal with areas
concerning serious personality disturbances. In general,
the null hypothesis had to be accepted. There were no
real definable differences between the younger and older
fire setting subjects in terms of their responses on the
questionnaire in the area of emotional disturbance.
151
20. Parents of fire setters and sociopaths were hypothe
sized to have a higher incidence of maladaptive behaviors
on a questionnaire than were parents of normal subjects.
For both the older and younger groups, the null hypothesis
was rejected. Parents of fire setters did indeed have a
higher incidence of definite and extreme concern items than
did parents of normal subjects. For the older subjects,
parents of normals had more little concern items. For the
definite concern category, fire setters had the greatest
number of questions answered affirmatively. Sociopaths
were in the middle while normals had the fewest. In the
extreme concern category, parents of fire setters again had
the greatest number of responses. Interestingly, socio
paths and normals did not differ significantly from each
other.
21. It had been hypothesized that parents of fire
setters and sociopaths would rate their own lives as more
unhappy and disturbed than parents of normal juveniles.
Using questions concerning parental relationships asked on
the parent's questionnaire, it was determined that for the
older subjects, the experimental hypothesis could be
accepted. Parents of normals listed the greatest number of
little concern items. In the category of definite concerns
in parental behavior, fire setters differed significantly
from normals but not from sociopaths, who did not differ
152
significantly from normals. For the younger groups, the
experimental hypothesis could be partially accepted.
Parents of normals had the greatest number of little
concern items about their own behaviors. Although the
definite concern category was significant, there were no
significant differences between the fire setting and normal
groups of parents.
There were other conclusions that could be drawn
from the analysis of. the data,: using, the same precautions
in analysis as first stated in the beginning of this sec
tion. The interview format, as formulated by the Fire
Services and Arson Prevention Committee of the California
State Psychological Association, does indeed help fire
personnel and others to distinguish between those children
who may simply need some education interventions and those
who are more seriously disturbed. The parent’s question
naire provides this information as per the scoring system
which divides behaviors into categories of little, definite,
and extreme concerns.
The children's questionnaire can help provide fire
personnel with valuable data, especially when the most
significant questions are drawn out and reassembled into a
new format. For the younger subjects, 21 questions were
found to discriminate between normals and fire setters.
153:
Many behavioral characteristics can be identified
in the literature. In this investigation, it was determined
that younger fire setters tend to be more hyperactive, im
patient, have more behavior problems in school, and lie
more often than normals. They show excessive anger, have
sleep and waking problems, fantasize more often, and have
little or no eye contact as compared to their normal
counterparts. Fire setters are more likely to have seen
a therapist, be impulsive, steal, be out of touch with
reality, and have learning problems. They are often de
scribed by their parents as showing off, being poor in
sports, a poor loser, and having a long history of behavior
problems. They often fight with their peers, disobey, show
cruelty to other children, and tend to panic when fire
begins to get out of control.
For the older fire setter, he is more likely to
demonstrate a lack of concentration, show cruelty to
children, have a strange quality about him, be jealous, and
have more behavior problems in school than his normal
counterparts. Fire setters are more likely to run away
from home, or school, disobey, be depressed, lie, be truant
from school, and be more easily led by peers than normals.
On these characteristics, the fire setter will not be
unlike juveniles who commit other types of crimes. This
profile suggests that the fire setter, at least among the
15:4
older children, has many of the same characteristics of
sociopaths.
Older fire setters are significantly different
from normals and sociopaths in the amount of uncontrolled
anger they show, in the number of severe depressions or
withdrawals that they experience, in their tendency toward
violence, impulsiveness, impatience, feelings of anxiety,
and in the number of sleep or waking problems that they
often have. On these characteristics, fire setters
especially distinguished themselves from sociopaths. Since
many of these behaviors are common to subjects often called
neurotic, it may be suggested that older fire setters, at
least in this study, had more symptoms of emotional dis
turbance than did the sociopaths or the normals. Thus, it
can be seen that the hypothesis often present in the litera
ture indicating that emotional disturbance is behind much
of the fire setting behavior, was again evident in this
investigation .
In general, results gathered during this investi
gation suggest that normals, fire setters, and sociopaths
differed significantly in the areas of a Tong history of
behavior difficulties, in stealing behavior, and as to
whether the child had been referred to a therapist.
Interestingly, on the history of behavior difficulties,
155
fire setters would be more likely to exhibit this problem
than either sociopaths, who fell in the middle, or normals
who exhibited the least amount of problems.
Another unusual finding in this data was that fire
setters were the highest in the area of stealing behavior,
with sociopaths again in the middle and normals least. As
in the previous case, it would have been expected that
sociopaths would have scored the highest in this
particular behavior.
In terms of the child being referred for profes
sional help, a similar result was obtained. Fire setters
were the most likely in this group of subjects to have
seen a therapist. Again, sociopaths were second and
normals last. These results possibly suggest that the fire
setter was more disturbed than the other groups or could
have been the result of sample selection bias, in that
most.of the fire setters were in higher socioeconomic
levels than the other groups, and more Caucasians were
represented. These factors alone could have been responsi
ble for the larger number of fire setters seeking profes
sional help. This finding may also have been the result
of parent awareness of disturbed behavior, and parent
adeptness in seeking professional help.
Most interestingly, an in-depth analysis of the
data indicated that fire setters, as expected, scored
156
highly on a scale designed to measure interest in fire
activities. In this, they were distinctly different from
the normals. They did not differ significantly from the
sociopaths, however. This finding is the more unusual
since the sociopaths were chosen as subjects only after
their history revealed an absence of fire setting activity.
These results, suggest, however, that sociopathic subjects
were also interested in activities related to fires, as
were the fire setters, and gave impetus to the hypothesis
that fire setters were a branch of antisocial or socio
pathic behavior.
Additional analysis of the data indicated that the
younger, as well as the older, fire setters did not score
highly on a scale designed to measure psychotic tendencies.
Sociopaths also did not score highly on this particular
scale. Assuming the hypothesis that fire setters are very
disturbed, as in reported in much of the literature, this
finding is somewhat surprising. There were two possible
conclusions that could have been drawn from this result.
It could have been assumed that all of the subjects in this
study were free of psychotic tendencies, or it could have
been assumed that the subjects did not. understand the
questions that were being asked of them. It is the second
explanation that made the most sense. Many of the younger
subjects clearly did not understand what it meant to read
15 7'
people's minds, or did not understand when asked if they
ever saw things that others did not. In response to the
last question, many of the answers were yes. When asked
to explain what they saw, the subjects would say car acci
dents, snakes in the hills, or other very literal events.
There were other areas that produced unexpected
results. It has been widely reported in the literature
that sociopaths have difficulties with peer relationships,
and interpersonal familial relationships. Thus, it would
have been expected that sociopathic and possibly fire
setting subjects in this study would have demonstrated these
attributes. In fact, they did not. Again, a wide variety
of reasons could be entertained to explain the finding.
Most likely, the results were produced either by a
sample that was unique and not representative of the general
population, or the result of a lie factor. It may have
been this latter factor that produced several of the unex
pected findings that occurred in this investigation. It is
often said that people try to place themselves in a position
guaranteeing the best possible light. Most likely, that
was occurring here. More investigation would be needed to
determine if the finding were genuine or the result of a
Hawthorne effect.
158
Recommendations
This investigation produced several unexpected
results: findings that went against current thought as
expressed in the professional literature.
It was determined that the parental questionnaire
did indeed discriminate among children in need of additional
referral and those in need of educational intervention. It
did determine that fire setters could be effectively dis
criminated from normals below the age of thirteen, and in
many cases, above the age of thirteen. The findings were
not as clear when it came to the sociopathic group. In
many cases, the fire setting subjects scored similarly to
those juveniles convicted of different crimes. This was
not true in all situations, however. On several questions,
fire setters either distinguished themselves totally from
the other two groups, or scored similarly to the normals.
The children's questionnaire was based upon a
review of the literature and covered many of the attributes
often applied to those involved in antisocial activities.
Many of those personality characteristics did not, in fact,
turn out to be correct. Future research would need to
replicate the procedures and methodology used in. this
investigation applied to a new sample of subjects. Only
then could it be determined if the findings in this study
159
were accurate or based upon error variance. The authors
of the interview format being used nationwide by fire de
partments indicated that there were primarily two types
of fire setters,
1. curiosity seekers,
2. those in need of psychiatric help.
Although this investigation indirectly assessed
this concept, it did not, in fact, attempt to determine if
it were true. Additional research will need to operation
ally define the term curiosity fire setter as distinguished
from the emotionally disturbed subject. Once the defini
tions are determined, fire setters could be interviewed
with a prearranged series of questions.
Using this format, it could then be determined
whether a fire setter had in fact engaged in the activity
solely on the basis of curiosity or due to emotional dis
turbance. This would add another piece of knowledge to the
small but growing pile of "useful" information about the
juvenile fire setter.
This investigation determined that, there were
certain questions answered by the children who served as
subjects that discriminated among fire setters, normals,
and in many cases, sociopaths. These questions are now
ready to be reassembled and combined with the parental
questionnaire to form a new package. Before this can be
16 CT
done, however, the questions need to be revalidated using
the same procedures utilized in this study with new sub
jects. Before this new investigation begins, it is
recommended that community support be sought around the
country so that the questionnaire can be standardized on a
much larger, more representative population. This process
should not be too difficult since the parental questionnaire
is already in use in many different states. Although this
study was hampered by a lack of involvement among fire de
partment personnel, the new procedure should meet with much
greater success. Community resistance in this study
centered around the large number of questions that were
asked on the children's form. The new interview sheet
would be one-third the length of the previous questionnaire
which would greatly shorten the time in giving the test.
This study was also hampered by community officials
who felt that right to privacy laws were being violated.
Future researchers would need to reassure their contacts
as to the privacy of any findings, including the releasing
of any names, pictures, or in any way identifying the
children.would.would serve as subjects. If the proper
groundwork is laid, there should be little difficulty in
enlisting community help and support for an investigation
as worthy as this one.
161
It is recommended that the results of this study
detailing the picture of the juvenile fire setter be used
in school situations by investigators to screen out poten
tial fire setters. Using a longitudinal format, investiga
tors should then follow the subjects for several years to
determine if the subjects do exhibit fire setting tenden
cies in later life. It is only through this method that
the predictive validity of the questionnaire could be
established.
This investigation included a sociopathic group
only among the older children. Fire setting subjects
scored similarly to the sociopathic group in many cases.
It might be said that fire setters are simply another type
of antisocial juveniles. This statement cannot and should
not be made, however, until a sociopathic group is tested
along with the younger subjects also. If similar findings
are produced, a more definitive hypothesis could be
developed. From this investigation, it could be stated
that fire setting subjects distinguished themselves from
normals in the younger group in that they had more signs
of emotional maladjustment. In the older group, although
there were definite indications of emotional difficulties,
overall, fire setters did not distinguish themselves from
sociopaths. It might be possible to assume that younger
162
fire setters are more disturbed. As they become older,
they develop sociopathic tendencies. This statement could
only be made, however, after additional research involving
this concept.
As previously noted,, this study revealed results
contradictory to currently-held precepts about the juvenile
fire setter. Future research should concentrate in the
areas of positive adjustment levels, hyperactivity,
psychotic tendencies, immature behavior patterns, fears,
locus of control concepts, peer relationships, self-concept
studies, depression, parental and social relationships of
the fire setter.
The results of this investigation may have been
due to other factors already discussed. It is also
possible that the current profile of the fire setter as an
angry youth seeking to get revenge, or seeking sexual
gratification may not be correcto
There is a question that the current literature,
depicting the fire setter as a loner, angry with everyone,
and with low self-esteem may be misleading.
There is a question whether or not the fire setter
is a sociopath, with no conscience and no concern for
society.
There is a question whether or not the correct pro
file of the fire setter as emotionally disturbed, beginning
163
at an early age, with problems at home, school, and in peer
relations is valid.
Is the fire setter someone who can be identified
at an early age, treated, and rehabilitated before he
continues with his deadly pursuit?
These questions have been raised in this study,
with no definitive conclusions. Perhaps a new investigation
could provide that final piece of information.
. 1 6 4
BIBLIOGRAPHY
16 5
BIBLIOGRAPHY
Addah, M. , & Benezeeh, M. Schizophrenia and delinquency:
a medicosocial investigation. Annales Medico-
Psychologiques, January 1977, 1 (11) , 1-33 .
American Psychiatric Association, 1980. Diagnostic and
Statistical Manual of Mental Disorders. Third Edition.
Washington, D.C.: American Psychiatric Association,
1980.
Anthony, H. The association of violence and depression in
a sample of young offenders. London: British Journal
of Criminology, 1969, J3(4) , 346-365.
Axberger, G. Arson and fiction: a cross-disciplinary
study. Psychiatry, August 1973, 36, 244-264.
Axberger, G. The arsonist, real and imagined. Annales
Medico-Ps'ychoiogiques, January 1973, 9 , 1-25.
Bakwin, H & Bakwin, : R. Clinical management of behavior
disorders in children. Philadelphia: W. B „ Saunder
Company, 1960.
Barracato, J. Fire...is it arson? The Aetna Casualty and
Surety Co., 1979.
Bar-tal, J., & Darom, B. Pupil *s attributions of success
and failure. Child Development, March 1979, 264.
Bartlett, D. Chromosomes of male patients in a security
prison. Nature, 1968, 219, 351-354.
Bender, L. Children and adolescents who have killed.
American Journal of Psychiatry, 1953, 116, 510-513.
Bender, L. Firesetting in children. Aggression,
Hostility and Anxiety in Children, 1959, 116-137.
Block, J., & Folkman, W. Fire arid children: learning
survival skills. (U.S.D.A. Forest Service paper
PSW-119), Berkeley, California, 1976.
166
Boling, J., & Brotman, C. A fire-setting epidemic in a
state mental health center. American Journal of
Psychiatry, September 1975, 132(1), 946-950.
Bolton, W.. Rural juvenile delinquency: problems and needs
in East Tennessee. Tennessee University, Knoxville
Agricultural Experiment Station. June 1978. (ERIC
Document Reproduction Service No. ED-165-916)
Bradley, G. Domain specific aspects of locus of control:
implications for modifying locus of control orienta
tion. Journal of School Psychology, 1977, 1_5(1), 18.
Brown, T. Juveni1e fire-setters. Paper presented at the
Arson and Fire Investigators Seminar. Berkeley,
California, June 1962.
Bugental, W., & Senker, B. Causal attributions of hyper
active children and motivational assumptions of
behavior: change approaches and. evidence for an
interactionist position. Child Development, 1977,
j48, 874.
Cemef, I. Case studies of socioeconomic and personality
characteristics of juvenile delinquent, users of
volatile substances. Dissertation Abstracts
International, 1973, 34(3-A), 1382-1383.
Chang, D. Environmental influences on criminal activity
in Korea. Criminology, November 1972, J, 338-352.
Chapman, B., and Boersma, G. Learning disabilities, locus
of control and mother's attitudes. Journal of
Educational Psychology, 1979, 71(2), 250.
Cisler, R. Crisis intervention in the family of a fire
setter. Psychotherapy: Theory, Research & Practice,
Spring 1972, 9(1), 76-79.
Coleman, J. Deviant subcultures and the schools.
Theoretical Perspectives on School Crime, I (5).
Washington: Department of Health, Education and
Welfare, February 1978.
Daly, P. A comprehensive early intervention program for
families with problem children, 197 5. (ERIC Document
Reproduction Service No. ED-150-519).
16 7
Davidson, H. Pathological firesetters: irresistible
impulse and criminal responsibility. Paper presented
to American Academy of Forensic Sciences, Chicago,
Illinois, February 24, 1954.
Denholtz, M. At home aversion treatment of compulsive
firesetting behavior: a case report. In R. Rubin,
Advances in Behavior Therapy; Proceedings of the
Fourth Conference of the Association for Advancement
of Behavior Therapy. New York: Academic Press, 1972.
Deukmejian, G. Arson Burns Everyone. Crime Prevention
Center, California Department of Justice, August 1980.
Eisler, R. Crisis intervention in the family of a fire
setter. Psychotherapy: Theory, Research & Practice,
1972, 9(1), 76-79.
Eysenck, H. Personality Structure Measurement. San Diego:
R.R. Knapp, 1964, 365.
Fenichel, O. Pyromania, the psychoanalytic theory of
neurosis. New York: W.W. Norton & Co., 1945.
Fetrow, R., & Fetrow, A. How a pre-trial facility can
destroy the self-esteem of the juvenile. International
Journal of Offender Therapy and Comparative
Criminology, 1974, 18(3), 227-232.
Finchan, B., & Barling, M. Locus of control and generosity
in learning disabled, normal achieving, and gifted
children. Child Development, 1978, j49, 530.
Fine, S., & Louie, D. Juvenile firesetters: do the
agencies help? American Journal of Psychiatry,
April 1979, 136(4-A), 433-435.
Fineman, K. Firesetting in childhood and adolescence.
Psychiatric Clinics of North America, 3(3),
December 1980, 482-500.
Fineman, K., Brudo, C., Brudo, E., Morris, C.,
Michaelis, L., & Day, J. Interviewing and counseling
juvenile firesetters the child under seven years of
age. Federal Emergency Management Agency,
United States Fire Administration, Office of
Planning and Education, June 1979, 11-59.
1 6 8
Fineman, K. Some psychological aspects of child caused
fires. California Fireman Magazine, 1976, 1(1),
596-606.
Finkelstein, B. Offenses with no apparent motive.
Diseases of the Nervous System, 1968, „29(5), 310-314.
Fire Education Planning. U.S. Department of Commerce,
National Fire Prevention & Control Administration,
Public Education Office, May 1977.
Fleszar-Szumigajowa. The perpetrators of arson in forensic-
psychiatric material. Polish Medical Journal, 1969,
8^(1), 212-219.
Folkman, W. Children with matches: a Study on fires in
the Los Angeles National Forest Area. U.S. Forest
Service, Pacific Southwest Forest & Range Experiment
Station, 1966, (PSQ-109).
Fray, P. Crimes and offenses by primitive reactivity.
Annales Medico-Psychologiques, May 1970, 3.(5), 701-718.
Freud, S. The acquisition of power over fire. Inter
national Journal of Psychoanalysis, 1932, 13^, 405-410.
Gerson, W. , & Schwidder, W. A juvenile arsonist. Praxis
de Kinder Psychologie und Kiriderpsychiatrie, 1953,
2, 202-212.
Glicken, M. Understanding and treating pre-delinquent
behavior in the school setting. Journal of the
International Association of Pupil Personnel Workers,
March 1978, 22(2), 72-75.
Gold, L. Psychiatric profile of the firesetter. Journal
of Forensic Sciences, October 1962, 7^, 404-417.
Goldwyn, J. Impulses to incendiarism & theft: a case
report. American Journal of Psychiatry, 1973, 86,
1093-1099.
Greenberg, H. Pyromania in a woman. Psychoanalytic
Quarterly, April 19 66, 3_5, 2 5 6-262.
Gregory, B. Effects of locus of control and type of
reinforcement on programme instruction performance of
adolescent boys. Journal of Educational Research,
Summer 1976, 70, 45-49.
1 6 9
Grinstein, A. States in development of control over fire.
International Journal of Psychoanalysis, 1952, 33
1-5.
Gunderson, J. Management of manic states: the problem of
fire setting. Psychiatry, May 1974, 3J7, 137-146.
Hallahan, D. Selective attention and locus of control in
learning disabled and normals. Journal of Learning
Disabilities, April 1978, 1jL(4), 47.
Hammond, N. Socialization and domestic violence.
Michigan Women's Commission, NCSA 1978 (0391).
Haney, J. Prime time target: juvenile fire setters.
Upper Arlington, Ohio: Officer-in-charge,
Fire Prevention Bureau, 1976.
Heacock, D. The black slum child and the problem of
aggression. American Journal of Psychoanalysis,
Fall 1977, 36(3), 219-226.
Heath, G., Gayton, W., & Hardesty, V. Childhood Fireset
ting. Canadian Psychitric Association Journal, 1976,
21, 229-237.
Heilman, D. Enuresis, firesetting & cruelty to animals.
American Journal of Psychiatry, June 1966, 122,
1431-1435.
Hogan, P. The adolescent mind and arson: part I and II.
Fire and Arson Investigator, 1975, 26.
Holland, C. Elimination by the parents of firesetting
behavior in a seven year old boy. Behavior Research
and Therapy, 1969, 7, 135-137.
Hreidarsson, A. Presumptive Y/15 translocation and mental
retardation in a family with a case of Klinefelter's
Syndrome. Journal of Mental Deficiency Research,
September 1973, 17(3-4), 163-170.
Hurley, W., & Monahan, T. Arson: the criminal and the
crime. British Journal of Criminology, 1969, 9(1)
4-21.
Issac, S., & Michael, W. Handbook in research and
evaluation. San Diego: Edit Publications, 1979.
17 0.
Joe, D., & Robinson, N. Chinese youth gangs: an
investigation of their origins arid activities in
Vancouver schools. Paper presented to the American
Educational Research Association Annuel Meeting,
Toronto, March 1978.
Justice, B., Justice, R„ , & Kraft, I. Early-warning signs
violence: is a triad enough? American Journal of
Psychiatry, April 1974, 131(4), 457-459.
Kafry, D. Children with matches: preliminary analysis
of findings. Department of Psychology, University
of California at Berkeley, 1977.
Kalus, J. Analysis of Hawaii secondary school discipline
variables. Unpublished dissertation, Walden
University, 1978.
Kammerer, T., Singer, L., & Michel, D. Incendiaries:
criminological, clinical, and psychological study
of 72 cases. Annales Medico-Psychologiques, 1967,
1(5), 687-716.
Kanner, L. Fire setting. Child Psychiatry, Fourth Edition.
Chicago: C.C. Thomas Co., January 1972, 669-672.
Katz, P. Patterns in the development of juvenile
delinquency. Corrective Psychiatry & Journal of
Social Therapy, 1972, T8(2), 10-18.
Kauffman, D., & Walsh, G. Youth in force in the labor
force. Washington, D.C.: Employment and Training
Administration, 1978.
Kaufman, I. Crimes of violence and delinquency in
schizophrenic children. American Academy & Child
Psychiatry Journal, 1961, 31, 124-136.
Kaufman,-I., Heims, L., & Reiser, D. A re-evaluation of
the psychodynamics of firesetting. American Journal
of Orthopsychiatry, 1961, 20, 42-54.
Kido, M. An EEG study of delinquent adolescents with
reference to recidivism.and.murder. Folia
Psychiatricaet Neurologlca Janponica, 27(2), 1970,
77-84.
Kido, R., & Iverson, J. Los Angeles County Juvenile
Probation Officers’ opinions of the California
Juvenile Court Law. Unpublished Master of Social
Work Thesis, University of Southern California, 1964.
______________________________________ .__________________ 1 7 1
Knappe, V., & Jurth, E. Children and juveniles as
pyromaniacs. Psychiatric, Neurolgie urid Medizinische
Psychologie, 1969, 21.(12), 463-471.
Knight, B., & West, D. Temporary and continuing delinquency
in Briton. British Journal of Criminology,
January 1975, 15(1), 43-50.
Kohn, M., & Sugarman, N. Characteristics of families
coming to the family court on PINS petitions.
Psychiatric Quarterly, Spring 1978, 50 (1), 37-43.
Konopka, G. Reform in delinquency institutions in
revolutionary times; the 1920’s in Germany. Social
Science Review, September 1971, '4J5(3), 245-258.
Koznar, J. Family rearing in relationship to dissocial
behavior and its prevention. Psychologia a
Pato-psychologia Dietata, 1978, JL1(1) , 45-52.
Kulich, B. A look at a juvenile prison. Praxis der
Kinderpsychblogie und Kinderpsychiatrie, April 1975,
24.(3) , 107-110.
Laberge, P. A case of fire-fetishism. Information
Psychiatrique, 1974, 50(10), 1021-1024.
Laplante, J. The pyromaniac. Laval Medical, December
1969, 40(10), 1062-1072.
Lawrence, D., & Winschel, P. Locus of control: implica
tions for special education. Exceptional Children,
April 1975, 483.
Lester, D. Firesetting. Corrective & Social Psychiatry,
1975, 21 (2) , 22-26.
Levin, B. Psychological characteristics of firesetters.
Center for Fire Research, National Bureau of
Standards, 1977.
Levin, B., & Martin, D. Crime and punishment and social
science. Public Interest, Spring 1972, 21_, 96-103.
Lewis, N., & Beck, K. Locus of control: task expectan
cies and children's, performance following failure.
Journal of Educational Research, March 197 8, 71,
207-210.
1 7 2
Lewis, N., & Yarnell, H. Pathological firesetting.
Nervous & Mental Disease Monographs. New York:
New York Coolidge Foundation, 1951.
Liska, A. Causal structures underlying the relationship
between delinquent involvement and delinquent peers.
Sociology & Social Research, October 1973, 58(1),
23-26.
Los Angeles Times. Lois Timnick: arsonists act out of
rage and frustration, December 10, 1980, pp. 1,
3, 20.
Lowenstein, L. A study of child arsonists: their
diagnosis and treatment, (ERIC Document Reproduction
Service No. ED-175-199)
Macht, L., & Mack, J. The firesetter syndrome.
Psychiatry, 1968, 31, 277-288.
Magee, J. Pathological arson. Scientific Monthly,
1933, 37, 3 58-361 *
Mavromatis, M., & Lion, J. A primer on pyromania.
Journal of Clinical Psychiatry, November 1977,
38 (11) , 954-955.
McKerracher, D., & Dacre, J. A study of arsonists in a
special security hospital. British Journal of
Psychiatry, July-December 1966, 112, 1151-1155.
Mecir, J. Mentally disturbed juvenile arsonists.
Ceskosiovenska Psychiatric, 1974, 7£(3), 158-166.
Medvecky, B., & Marossyova, E. Causistic contribution
to pyromania. Ceskosiovenska Psychiatric, December
1970, 66 (6) , 360-363.
Michaels, J. Enuresis in murderous aggressive children
and adolescents. Archives of General Psychiatry,
1961, 5, 490-493.
Michaels, J., & Goodman, S. Incidents & intercorrelations
of enuresis and other neuropathic traits in so-called
normal children. American Journal of Orthopsychiatry,
1934, 4, 79.
1 7 3
Michaels, J., & Steinberg, A. Persistent enuresis &
juvenile delinquency. British Journal of
Delinquency, 1952, 3^ 114-123.
Oppel, W. , Harper, P., & Rider, R. Social, psychological
and neurological factors associated with noctural
anuresis. Pediatrics, 1968, 4 2 , 627.
Pasternack, S. Evaluation of. dangerous, behavior of active
duty servicement. Military medicine, February 1971,
136(2), 110-113.
Patterson, G. The aggressive child: victim and
architect of a coercive system. In L . Hamerlynck,
L. Handy, & E. Mash (Eds.), Behdvior modification and
families. New York: Bruner/Mazel, 1976.
Pospiszyl, K. Application of H.J. Eysenck's Maudley
Personality Inventory in the study of delinquent
youths. Przealad Psychblogiczny, 1973, 16(2), 193.
Psarska, A. Some aspects of delinquency in the mentally
retarded. Psychiatria Polska, March 1972, 6(2),
151-157.
Rawlings, M. Self-control and interpersonal violence: a
study of Scottish adolescent male severe offenders.
Criminology, May 1973, 11(1), 23-48.
Robbins, E., & Robbins, L. Arson with special reference
to pyromania. New York State Journal of Medicine,
1967> § 1 ' 795-798.
Robins, L. Deviant children grown up: a sociological and
psychiatric study of the sociopathic personality.
Baltimore: Williams & Wilkins, 1966.
Rotter, L. Domain specific aspects of locus of control:
implications for modifying locus of control orienta-
tion. Journal of School Psychology, 1977, 1 5 (1) , 18.
Russell, D. Juvenile murderers. International Journal
of Offender Therapy & Comparative Criminology, 1973,
17 (3) , 235-239.
Sakamoto, M. A dynamic psychopathology of firesetting
in schizophrenic children. Osaka City Medical
Journal, 1960, 6(1), 59-69.
174
Schafer, W. Approaches to juvenile delinquency prevention
and treatment in rural settings. Rural and Smalltown
Delinquency: a new understanding and approaches.
(ERIC Document Reproduction Service No. ED-15068)
Schindler, S. Family constellation and aggressive conduct.
Zeitschrift fur Klinische Psychologie und Psycho-
therapie, January 1975, 2_2 (2), 180-182.
Schmideberg, M. Pathological firesetters. Journal of
Criminal Law , & Criminology, 1953, j§4_(l) r 30-39.
Schmideberg, M. Juvenile murderers. International Journal
of Offender Therapy & Comp a rat iv e C r i in inology, 1973 ,
17 (3) , 240-245.
Schuberth, R. Aggression and its motivational background.
Schweizerische Zeitschrift fur Psychologie und ihre
Anwendungen, 1974, 32(2), 139-149.
Scott, D. Children and firesetters. The Psychology of
Fire. New York: Scribner & Sons, 1974.
Seaman, S., & Seaman, J. Deviance avowal and disavowal
in the situation of deviant and non-deviant company.
New Orleans: Loyola University, 1978.
Shaw, C. Firesetting. The psychiatric disorders of
childhood. New York: Meredith Publishing Company,
1962.
Siegel, L. Case study of a thirteen year old firesetter:
a catalyst in the growing pains of a residential
treatment unit. American Journal of Orthopsychiatry,
1952, 2_7, 396-410.
Siegelman, E. Children who set fires: an exploratory
study. Resources Agency of California Department of
Conservation, Division of Forestry, 1969.
Siegelman, E., & Folkman, W. Youthful firesetters:
an exploratory study in personality & background.
USDA Forest Service Research, Pacific Southwest,
Forest & Range Experiment Station, Berkeley,
California, 1971, PSW 230.
Simmel, E. Incendiarism. Psychoanalysis Yearbook, 1949,
261-274.
1 7 5
Siomopoulos, V. Psychiatric diagnosis & criminality.
Psychological Reports, April 1978/ 42: (2) , 5 59-562.
Sorrells, J. Kills who kill. Crime and Delinquency,
July 1977, 23(3), 312-320.
Stekel, W. Peculiarities of behavior. Psychopathic
Sexualis, 1924, 11.
Stott, D., & Wilson, D. The adult criminal as juvenile.
British Journal of Criminology, January 1977, 17(1),
47-57.
Sturup, G. The diagnosis & treatment of a pyromaniac.
International Journal of Social Psychiatry, 1955
1(1), 54-59.
Sutton, B. Child firesetting behavior. Children's
Psychology Hospital Studies, Austin, Texas,
March 1974.
Tapia, F. Children who are cruel to animals. Child
Psychiatry & Human Development, Winter 1971, 2(2),
70-77.
Tennent, T. Female arsonists. British Journal of
Psychiatry, 1971, 119, 497-502.
Timnick, L. Arsonists act out of rage & frustration.
' Los ArtgCles Times, December 10, 1980, pp. 1, 3, 20.
Trasler, G. School, delinquency, and the youth culture
in Britain and North America. Washington, D.C.:
Department of Health, Education and Welfare,
February 1978.
Vandersall, T., & Wiener, J. Children who set fires.
Archives of General Psychiatry, January 1970, 22
63-71.
Virkkunen, M. On arson committed by schizophrenics.
A.C.T.A. Psychiatry Scandanavia, 1974, _50, 105-160.
Vocational Foundation, Inc. Our turn to listen: a white
paper on unemployment, education and crime, based on
interviews with New York City teenage dropouts.
New York: Vocational Foundation, Inc-, 1978.
176
Wagner, G. On the motivation of incendiarism.
Psychiatrie, Neurologie, un Medizinesche Psychologie,
March 1974, 26(3), 155-164.
Walshe-Brenna, K. A socio-psychological investigation of
young murderers. British Journal of Criminology,
January 1977, 17(1), 58-63.
Wardrop, F. Pathological firesetting. Archives of
Criminal Psychodynamics, 1955.
Warner, G. A few representative cases of pyromania.
Psychiatric Quarterly, 1932, '6>, 675-690.
Wax, D., & Haddox, V. Sexual aberrance in male
adolescents manifesting a behavioral triad considered
predictive of extreme violence: some clinical
observations. Journal of Forensic Sciences, 1974,
19, 102-108.
Weissman, M., & Siegel, R. The depressed woman and her
rebellious adolescent. Social Casework, 1972, 53(9),
563-570.
Welsh, R. The use of stimulus satiation in the elimina
tion of juvenile firesetting behavior. In A. Graziano
(ed.) Behavior Therapy with Children. Chicago:
Aldine-Atherton, 197i.
Winget, C., & Whitman, R. Coping with problems:
attitudes toward children who set fires. American
Journal of Psychiatry, April 1973, 130 (4) , 442-445.
Wolford, M. Some attitudinal, psychological and socio
logical characteristics of incarcerated arsonists.
Fire and Arson Investigator, 1972, _22(4), 1-30.
Woolf, P. Arson & Moebius syndrome: stigmatization.
Medicine, Science and the Law, 1977, ]L7, 68-70.
Yarneil, H. Firesetting in children. American Journal of
Orthopsychiatry, April 1940, 1C), 2 72-286.
Zabcynska, K. A longitudinal study of the development of
juvenile delinquency. Polish Psychological Bulletin,
1979, 8(4), 239-249.
1 7 7
APPENDIXES
178
APPENDIX A
PARENT QUESTIONNAIRE IN ENGLISH
179
COPY
PARENT QUESTIONNAIRE
PARENT(s): Please fill out this form As Soon As lbs si hie. Make a circle around the
answer— "never", "sometimes", or "frequently" that best describes your child for every
question. Ask any questions you have. We want to know if the child shows any of the
following behaviors. When marking the form, consider all parts of the child's life
(at home, at school, etc), where these behaviors might be present.
Behavior Never Sometimes Preauentlv
Hyperactivity 01 01 C2
lack of Concentration 01 C1 C2
Learning Problems (home or school) C1 02 C2
Behavior problems in school C1 G2 02
Impulsive (acts before he thinks) C1 C1 C2
Impatient Cl C1 02
Accidents C1 01 02
Convulsions or spells C1 02 02
Wets during day C1 02 02
Extreme mood swings C1 01 02-3
Need for security 02 01 02
Need for affection 02 01 02
Depression 01 02 02-3
Unusual movements-tics 01 02 02
Stuttering 01 02 02
Bed wetting (after age 3) Cl 02 . 02
Soiling (after age 3) 01 02 02
lying 01 01 02
Excessive & uncontrolled anger 01 02 02
Violence 01 02 C2
Stealing 01 02 02
Truancy
01 02 02
Cruelty to animals
01 C2 C2-3
1 8 0 '
C O P Y
Behavior_______________________ Never_______ S one times____Frequently
Cruelty to children Cl C2
C2-3
Fighting with peers C1 C1 C2
Fighting with siblings C1 C1 C2
Destroys toys of others C1 C2 C2
Destroys own toys Cl Cl C2
Buns away from home—school C1 C2 C2
Disobeys di Cl C2
.Long history of severe behavioral
difficulties C1 C2 C2
Child is a poor loser C1 C2 C2
Child expresses anger by
hurting other*s things C1 C2
C2-3
Child expresses anger by hurting
self or something he likos C1 C2-3 C3
Child has been in trouble with
the police Cl C2 C2
Easily led by peers Cl Cl C2
Jealousy C1 C1 C2
Temper tantrums C1 C1 C2
Doesn*t play with other children C1 C1 C2
Shows off C1 Cl C2
Severe depressions or withdrawal Cl C2 C3
Child is good in sports C2 C1 Cl
Shyness Cl Cl C2
Extreme goodness C1 Cl C2
Sexual .activity with others C1 C2 C2
1 8 1
•COPY
3.
Behavior___________________Never Sometimes____ Frequently
Stomachaches- C1 C2 C2
Nightmares C1 C2 C2
Other sleep or waiting problems Cl C2 C2
Anxiety C1 C1 C2
Fantasizing C1 Cl C2
Poor or no eye contact C1 C2
C3
Child has twitches (eyes, face, etc) Cl C2 C2
Crying C2 C1 02
Nail biting C1 C1 C2
Vomiting C1 C1 C2
Thumb sucking C1 C1 C2
Aches & pains C1 C1 C2
Chewing odd things C1 Cl C2
Constipation C1 C1 C2
Diarrhea
01
C1 C2
Masturbation Cl C1 C2
Curiosity about fire C1 C1 C2-3
Plays with fire Cl Cl C2-3
Panicked whon fire out out of
control C2 C2 C1
Fires set some distance from
child's home C2 C2 01
Child proud or boastful
regarding his fire setting C1 C2 C2
Stares at fires for long
periods of time 01 C2
C3
Daydreams or talks about fire C1 C2 C2-3
Unusual look on child's face
as he frequently stares at fires C1 C2-3 C3
182.
C O P Y
Behavior Never Sometimes Freouentlv
Family discord PI P2 P2
Father or mother absent P1 P2 P2
Family has moved with child F1 P2 P2
Child has seen a therapist C1 02 C2
Other family member has seen
a therapist P1 P2 F2
Parent has serious health problem P1 P2 P2
Marriage is unhappy PI P2 P2
Mother's discipline is effective P2 P1 PI
Father's discipline is effective P2 PI P1
Unusual fantasies C1 C2
G3
Strange thought patterns C1 G2 C3
Speech biszare, illogical or
irrational C1
03 C3
Out of touch with reality C1
03 C3
Strange quality about child 01 C2 G3
Self-imposed diets C1 C1 C2
Sleep walking C1 C2 C2
Phobias C1 C2 C2
Fears C1 C1 C2
Child plays alone C1 C1 G2
1 8 3
APPENDIX A
PARENT QUESTIONNAIRE IN SPANISH
1 8 4
COPY
CUFSTJONARK* CF Ln$ °ADRFS
F/DRFSt Favor de 1 "Lenar osta forma tan nmnto como nosible.
Circula la resouesta - "nunca", "a. veces", o "frecuentamente
lo cual mejor describe su hi jo/a oara cada oregunta. Haga
cualquier oregunta que tenga. Nosotros queremos saber si el n.ino
muestra el siguiente coir.oortamiento. Guando contests, consid^ra
tndas oartes de la v.ida del ntfio (er casa* en la escuela, etcetera)...
*onde esta conducts, oudlsra nanifestarse.
Conducta Nunca A Veces Frc cue ntamente
oobre nergieo Cl Cl C2
escaz de concentracion Cl C1 C2
Droblemas con adquirir
cenocimientos (en casa e «n
la escuela) Cl C2 02
nroblemas con comnortamiento
en la escuela Cl CP 02
imoulsivo (hacer sin nensar) Cl Cl 02
imoaciente Cl Cl 02
accidentes Cl Cl 02
convulsiones o ataques Cl C2 02
se moja durante el dfa Cl C2 02
excesivo cambio de hunor Cl Cl 02---3
necesidad oara seguridad c? Cl C2
necesid^d oara afeccion CP Cl C2
denresi^n Cl C2 C2-3
. . . /. .
novimientos insolitos-tics Cl C2 C2
tartamudez Cl C2 C2
moja la cama(desoues de la edad
.de 3) Cl C2 C2
ensueia(desoues de la edad de 3) Cl
C2 C2
mentira Cl Cl 02
c^lera excesiva sin freno Cl 02 02
violencia Cl 02 02
roba
Cl 02
02
185
C O P Y
a..
Conducta Nunca A Veces Frecuentamente
anda a la tuna Cl C2 C2
crueldad a los animales Cl C? C2-3
irhumanidad a los ninos Cl CP C2-3
se oelea con los companeros Cl Cl C2
ss oelea con los Hermanns y
las Hermanas Cl Cl C2
destruye juguetes de otros Cl C2 C2
destruye sus mismos juguet.es Cl Cl C2
huye del hogar~de la escuela Cl C2 C2
desobedece Cl Cl C2
historia lanera de rroblemas
severos con el comoortamiento Cl C2 C2
el nino cuando pierde, se hace
desgraciado Cl C2 C2
el^nino muestra la colera por
danarse mismo o por danar algo
queridn Cl C2-3 C3
el nino s,e hallo en aouro con
la Policia Cl C2 C2
rae conforne facilrr.ente a los ;
jcompaneros
jse deja llevar de la nano por
JLos comparieros Cl C1 C2
celos Cl Cl C2
rabieta (del temoeramento) Cl Cl C2
no juega con otros ninos Cl Cl C2
se da farol/hace teatro Cl Cl C2
derresinn severe o retirada Cl C2.
C3
el nino tiene exito en los
deportes C2 Cl Cl
timidez Cl Cl C2
bondad extrema Cl Cl C2
coito/acto sexual
Cl C2 C2
186
C O P Y
3-
Conducta Nunca A Veces Frecuentamente
dolores del estomago Cl
C2 C2
pesad illas Cl C2 C2
otros r>roblemas de dormir o
de despertar Cl C2 C2
ansiedad Cl Cl C2
tiene ideas fantsfsticas (irreales)Cl Cl C2
hay ©obre o falta contact© de
los ojos Cl C2
C3
el nino tiene contracciones
nervjlosas(de los ojos, de la cara,
etcetera) Cl C2 C2
11 ora C2 Cl C2
se come las urns Cl Cl C2
vomita Cl Cl C2
chupa el dedo gordo Cl Cl C2
dolores Cl Cl C2
mastica cosas extranas Cl Cl C2
estrenimiento Cl C! C2
d iarrea Cl Cl C2
mas turbac ion Cl Cl C2
curiosidad acerca del incendio Cl Cl C2-3
juega con lumbre (fosforos.)'/
materia combustible Cl Cl C2-3
aterrorizado cuando el fuego
sc puso libre (no controlado) C2 C2 Cl
prende fuegos a largo distancia
de la cas del nirlo C2 C2 Cl
el nino es orgulloso o jactancioso
acerca de su habilidad de prender
fuegos Cl C2 C2
fija la vista en los fuegos
para much© tiempo Cl C2
C3
ensuena o habla del fuego Cl C2 C2-3
187
C O P Y
4.
Conducta Nunca A Veces Frecuentamente
mirada extrana en la cara del
nino cuando frecuentamente
e] fi.ia la vista en los fueees Cl C2— 3
_ _ _C3
discordia entre la familia PI P2 P2
ausencia del padre o de la madre PI P2 P?
la familia se ha movido con el _
nino PI P2 P2
el nino ha visto a un teraoista Cl C2 C2
otras oersonas en la familia
han visto a un teraoista PI P2 P2
el oadre tiene oroblemas serios
con la salud PI P2 P2
el matromonio es inf ell?. PI P2 P2
la discrclina de la mama es
efectivo P2 PI PI
la discrolina del papa es
efectivo P2 PI PI
fantasia extraHas Cl C2
C3
molde extrano de pensar Cl C2
C3
habla fantastica, ilogica,
irracional (absurda) Cl
C3 C3
perdio contacto con la
realidad Cl C3 C3
el nino tiene calidad extrana
de su orcpio nersonalidad Cl C2 C3
none si mismo a dieta Cl Cl C2
somnambulismo Cl C2 C2
fobias Cl C2 C2
aprensiones Cl Cl C2
el nino juega solito Cl Cl C2
18 8
APPENDIX B
CHILDREN’S QUESTIONNAIRE
1 8 9
C O P Y
Child's Name ________________________ Birthdate_______________ Age_____
Grad e Ethn i c *_______SES **_____ Inte 1 le c tua 1 _______Group S ta tus»*»
(if known!
1. I like sports. yes no
2. My favorite holiday is the 4th of July. yes no
3.
I always pay attention in school. yes no
4. People who put fires out are heroes. yes no
5.
T love to play with other kids. yes no
6. T eat very little. yes no
7. I love to read. yes no
8. My parents and I do lots of things together. yes no
9. It's usually not my fault when I get into
trouble, (usually needs explanation) yes no
10. I like to take chances. yes no
11 . I always tell the truth. yes no
12. I think I'm great. yes no
13.
I hurt animals sometimes. yes no
14. I sleep the whole night through. yes no
15.
T have problems paying attention in
school. yes no
16. I have set some fires. yes no
17. T have lots of friends. yes no
18. I get tired easily. yes no
19. I like most boys. yes no
20. I am a good speller. yes no
21. I live with one parent. yes no
22. I have no control over the things I do
(needs to be explained)
yes
no
23. I am usually sad. yes no
24. T have run away from home. yes no
25. I never lie. yes no
-* it— ----------------; -----; -----------------------------------------
1white; 2 black; 3 hispanic **write parent's occupation and gross
l^ec*rlL| \{\cxycn& } V < m o o ?q•
\ N o rm q lj a P vre. Se.++e.r, 3 . S o cA o p ^ h h
190
C O P Y
26. I love to play with toys. yes no
27.
I always get things done in time. yes no
28. I like to play with matches. yes no
29. I love to daydream. yes no
30. My stomach gets upset sometimes. yes no
31.
■People can always count on me to do things. yes no
32. I let my teachers push me around. yes no
33.
My parents love me. yes no
3^.
I earn (work hard) my good grades. yes no
35.
I trust a lot of people. yes no
36. I have been in many fist fights. yes no
37.
I wish I were more like my friends. yes no
38. I believe in God. yes no
39. I always need to be doing something. yes no
bo. I love to watch fireworks. yes no
Sometimes I walk in my sleep. yes no
^2. I like most girls. yes no
*3.
I get excited when I see a fire burning. yes no
Teachers are often saying things that make
me feel sad. yes no
^5. Football is my favorite sport. yes no
k6. I have taken drugs. yes no
^7. . My parents hit me. yes no
^8. I really enjoy doing good work in school. yes no
^9. I feel like I have to win when I am
playing a game. yes no
50. I always try to do what my parents say. yes no
51.
I really like the way I look. yes no
52. I am happy most of the time. yes no
53.
I have had a lot of accidents. yes no
191
C O P Y
5k. I could watch a fire burn all day long. yes
55.
I sometimes see things that others don't see,. yes
56. I bite my nails. yes
57. I enjoy fishing. yes
58. I feel good when I am with other people. yes
59. I like to miss school. yes
60. Some people in my family drink too much
liquor. yes
61. When I do something really well, it"s
usually just a matter of luck. yes
62. Sometimes the dark scares me. yes
63.
I would never take anything from a store
unless T paid for it. yes
6k. Everything I do is great. yes
65.
I sometimes think bad things. yes
66. My mother is always telling me to be
quiet. yes
67. Fires really cause a lot of problems. yes
68. I sometimes hear things that others don't
hear. yes
69. Sometimes I like to suck my thumb. yes
70. I enjoy watching movies. yes
71.
I have been in the hospital a lot. yes
72. I love math. yes
73.
My parents argue a lot. yes
7k. I would do better in school if my teachers
liked me more. yes
75. I wish I had as much stuff as my friends
have. yes
76. My parents like my friends. yes
77. I love t.v. shows that have lots of
murders. yes
78. I get along great with my brothers and
sisters.
yes
3.
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
no
192
COPY
79. When I'm angry, I sometimes feel like burning
something. yes no
80. Somebody keeps telling me to do bad
things. yes no
81 . I enjoy fooling around with matches. yes no
82. Sometimes, I get so angry I feel like
throwing things. yes no
83. I would like (or have tried) to try
alcohol. yes no
84 > I sometimes stay home from school even
when I am well. yes no
85.
I live in a happy family. yes no
86. I have set fires that I couldn't out out. yes no
87.
Sometimes I think people are watching me. yes no
88. My parents like the way I dress. yes no
89.
I am easily embarrassed. yes no
90. I wish I could leave everything and
just travel all the time. yes no
91. I finish things very fast. yes no
92. I sometimes dream about fires. yes no
93.
Sometimes I think I can read minds. yes no
94. Sometimes I wet the bed (at night). yes no
95.
I love to swim. yes no
96. I feel better when I see fire burning. yes no
97. I like most of my teachers. yes no
98. Sometimes when I'm bad, IMm not sure
what my parents are going to do
about it (i.e., will they be mad
or not). yes no
99. I feel bad when I do something I'm not
supposed to. yes no
100. I plan to live until I am very, very old. yes no
193

Asset Metadata

Creator Dudek, Suzanne Robins (author)

Core Title A correlation study leading to the development of a scale useful in the prediction of potential juvenile fire setters

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Education

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

Tag OAI-PMH Harvest,Psychology, clinical,psychology, psychometrics

Language English

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c26-509556

Unique identifier UC11247102

Identifier usctheses-c26-509556 (legacy record id)

Legacy Identifier DP24883.pdf

Dmrecord 509556

Document Type Dissertation

Rights Dudek, Suzanne Robins

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