University of Southern California Dissertations and Theses

An Investigation Into The Communication Style Of Suicidal Individuals

(USC Thesis Other)

An Investigation Into The Communication Style Of Suicidal Individuals

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been
microfilmed exactly as received 6 7 -3 9 9
DARBONNE, Allen Roy, 1936-
AN INVESTIGATION INTO THE COMMUNICATION
STYLE OF SUICIDAL INDIVIDUALS.
University of Southern California, Ph.D., 1966
Psychology, general
University Microfilms, Inc., Ann Arbor, Michigan
AN INVESTIGATION INTO THE COMMUNICATION STYLE
OF SUICIDAL INDIVIDUALS
by
Allen Roy Darbonne
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Psychology)
June 1966
UNIVERSITY O F S O U T H E R N CALIFORNIA
THE GRADUATE SCHOO L
UNIVERSITY PARK
LO S ANGELES. CALIFO RNIA 9 0 0 0 7
This dissertation, written by
Allen _ .Roy.. D arbonne.........
under the direction of hxs....Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by the Graduate
School, in partial fulfillment of requirements
for the degree of
D O C T O R OF P H IL O S O P H Y
Dean
Date.
June 1966
DISSERTATION COMMITT]
Chairman
ACKNOWLEDGMENTS
Grateful acknowledgment is made to the many people
who made this study possible. The author is indebted to
the chairman and other members of his dissertation commit
tee, Dr. Georgene Seward, Dr. Alfred Jacobs, and Dr. Murray
Wexler, for their helpful guidance and criticism. Sincere
appreciation is extended to Dr. Norman Farberow for his
assistance and to the Veterans Administration, where the
author is a staff member, for allowing time and providing
a setting which encourages free inquiry and research. The
cooperation of the staff members of the following organiza
tions was of invaluable assistance in providing data for
this study: the Los Angeles County Coroner's Office, the
Suicide Prevention Center, the Los Angeles State Mental Hy
giene Clinic, and the Benjamin Rush Center Division of Los
Angeles Psychiatric Service. A special note of thanks is
owed Mr. David Reynolds for his assistance in data collec
tion and to Mr. Ted Mordey and Mr. Peter Proctor for their
help with the reliability aspects of the investigation.
ii __ _ ____________________
iii
This study is dedicated to my wife, Ginny, whose encourage
ment and understanding was generously given from beginning
to final typing.
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ....................................... ii
LIST OF T A B L E S ..........................................vi
Chapter
I. THE PROBLEM............. 1
Theoretical Background
Previous Studies
Hypotheses
II. M E T H O D ..........................................22
Subjects
Instruments
Procedure
III. RESULTS ......... . . . . . . . . . . . . . . 36
Comparison of Commit-Suicide Group with
Non-Suicidal-Normal Group
Comparison of Suicide-Threat Group with
the Commit-Suicide Group and with the
Non-Suicidal-Normal Group
IV. DISCUSSION..................................... 45
Suggestions for Future Research
V. SUMMARY................. 59
iv
REFERENCES . „
_
Page
63
APPENDICES.............................................. 68
Appendix A: General Information
Questionnaire
Appendix B: Happy Note Form
Appendix C: Note Form for Non-Suicidal
Subjects
Appendix D: Note Form for Psychotherapy
Patients
Appendix E: Rank of Degree of Activity
in Suicide Method Assigned by
Individual Judges
Appendix F: Communication Style Scoring
Form Individual Data Collection Sheet
Appendix G: Percentage of Agreement of
Scores (Between Two Judges Scoring 20
Notes Independently)
I
LIST OF TABLES
Table Page
1. Mean Ranks of 18 Methods of Suicide..............31
2. Results of Comparison of C/S and N/S Groups . . 37
3. Results of Comparison of C/S and S/T Groups
and N/S and S/T G r o u p s........................ 41
4. Comparison of High and Low Suicide-Potential
S/T G r oups......................................54
vi
CHAPTER I
THE PROBLEM
Theoretical Background
Every 30 minutes, a person in the United States
kills himself. At least 20,000 suicidal deaths are known
to occur in this country annually. This figure is believed
to be a gross underestimation, with the true toll closer to
40,000 annually. Ranking tenth among the leading causes of
adult death in the United States, suicide is also a report
ed phenomenon in practically all civilizations. Suicide
has been known since earliest history. The Old Testament
refers to several cases of suicide which it neither con
demns nor condones. But the importance of suicide as a
human problem cannot be weighed by known statistics alone.
For each death by suicide there are unreported numbers of
persons adversely affected in its wake. No other form of
death leaves behind such long-lasting feelings of guilt,
shame, and distress in those closely and distantly related
2
to the deceased.
The bewildering question of suicide has attracted
the attention of many disciplines, including anthropology,
theology, philosophy, psychiatry, psychology, and socio
logy. The two approaches to suicide which have received
the greatest attention are the sociological and the psycho
analytic. The first organized formulation of the socio
logical point of view of suicide was by Durkheim (1897).
His interest was not in the individual, but in the effect
of society on the individual. Durkheim described three
types of suicide as follows: "egoistic" suicide, in which
the individual is not sufficiently integrated into his
society; "anomic" suicide, in which the individual's ad
justment to society is suddenly disrupted, as through an
economic depression; and "altruistic" suicide, wherein the
individual is overintegrated with society and he sacrifices
himself as in the case of a soldier on the battlefield or
the Japanese traditional suicide of honor.
The psychoanalytic position is based on the writ
ings of Freud (1925) whose interest was in the psyche of
the individual. Freud felt that the tendency toward sui
cide in a person is due to his unconscious sadistic im
pulses arising from strong love and hate feelings which he
has for someone else but which he directs against himself.
Psychoanalytic theorists who followed Freud have directed
increasing attention toward the influence of environment on
the individual. I
Contemporary students of suicide argue cogently
that suicide is neither an event of the psyche nor of the
society, but is an event which can best be understood as
r
an admixture of both (Shneidman & Farberow, 1960).
Shneidman and Farberow have encouraged researchers to ask
not only how a suicidal individual thinks and feels but
also how he integrates with his fellow man.
In recent years several investigators have pro
posed that an increased understanding of emotional disturb
ance and/or unusual behavior may be gained through a study
of the communicative style of individuals displaying this
behavior (Ruesch, 1961; Ruesch & Bateson, 1951; Bateson,
Jackson, Haley & Weakland, 1956; Sullivan, 1953). In gen
eral these theorists have in common a basic assumption that
style of communication is relatively stable for an individ
ual and is reflective of his style of living. Ruesch
(1957, 1961) has redefined psychotherapy, normal behavior,
and psychopathology in communication terminology. In his
opinion such terminology facilitates understanding the
4|
relationships between these three areas by considering themj
i
i
all on as nearly an operational basis as possible. Ap
proaching behavior from a communication viewpoint reduces
the distance usually found between theory and observation, j
Ruesch contends that, since therapists must observe com
munication and respond through communication, theory used
by therapists should be related to these operations.
Theorists from both the problem area of communica
tions and the problem area of suicide have begun to direct
attention toward the communicative behavior of suicidal
individuals. Hayakawa (1957) has suggested viewing the
suicide act itself as a communicative act. He states that
decisions toward suicide are preceded by some kind of logi
cal process, often conducted on erroneous premises. This
j
"sense" behind the suicide's decision, he feels can best be
understood from a communications standpoint. Jackson
(1957) in a comprehensive review of theories of suicide,
stresses the importance of communication factors in the
treatment and prevention of suicide when he states, "It is
doubtful if suicide, apart from its rational form, can ever
occur where adequate communication exists between the
therapist and his patient" (p. 19). Ruesch (1961) views
suicide as a symptom of disturbed communication. He
*.... 5]
recommends studying both the verbal and nonverbal messages j
of suicidal behavior in order to adequately assess its i
meaning. Ruesch treats suicidal patients as not necessari
ly synonymous with depressed patients. He has developed a
theory of the communicative style of depressed patients but
he has not as yet done so for suicidal patients.
Acute awareness of the significance of the communi
cative aspects of suicidal behavior by Farberow and Shneid
man (1961) is reflected in the title of their classic in
this field, The Cry for Help. Litman and Farberow (1961)
see the status of communication with the patient as one of
the prime considerations in an emergency evaluation of
self-destructive potentiality. According to them, when a
suicidal individual is able to communicate his feelings and
cry for help, the self-destructive danger may be high, "but
it is never so extreme as when the patient has given up and
-J
withdrawn and is no longer communicating." They also em
phasize that the hind of affect expressed along with pre-
suicidal communications must be studied. The suicidal act,
itself, is viewed by these investigators as an action com
munication. An itemized schedule for the assessment of
self-destructive potentiality which is derived to a great
extent from communication theory has been proposed by
6|
I
Tabachnick and Farberow (1961). They state, "The self- |
destructive behavior thus becomes, in part, a communication!
!
I
with a particular purpose and content directed toward a
!
specific audience" (p. 63). This approach emphasizes such
factors as the content, audience, degree of directness, and
purpose of the communication.
Adler (1958) formulated a theory of suicide which
calls attention to the suicidal individual's style of liv
ing and relating. Adler felt that suicidal individuals
have a life style which reflects heightened dependency
needs, low self-esteem and self-centered goals, a high de
gree of activity, and an expressive style of veiled aggres
sion which includes hurting the self in order to hurt and
manipulate others. The comprehensive quality of this
theory permits its use as a frame of reference for the
analysis of communication style. It is possible to re
phrase this explicit and comprehensive theory into opera
tional communications terms which can be tested by analyz
ing the communication behavior of suicidal and non-suicidal
individuals. If differences between communication style of
suicidal and non-suicidal individuals do exist, it seems
reasonable to assume that these differences will be demon
strated in unstructured communication situations, such as
— 7 l
i
in the content of a suicide note.
!
I
i
Previous Studies j
While much has been written of a theoretical nature!
i
considering the communicative aspects of suicidal behavior,
little systematic research has been done in this area. One
of the reasons for this may be researchers' reaction to the
complex methodological issues which Neuringer (1962) points
out as plaguing the entire field of suicide research. Fore
most among these is the fact that unlike most categories of
subjects studied, once the subject is known to qualify as a
"true" experimental subject, he is no longer around for ex
perimental manipulation. One method of investigation when
suicide is an accomplished fact has been the method of re
siduals, in which such residual effects as personal docu
ments, notes, etc. of the deceased are utilized as clues
pointing to the state of the individual when alive. While
Neuringer stresses that the method of residuals is not to
be demeaned, he also points to the inherent weaknesses of
poor control, observational distortion and unknown validity
found with this method. These are the problems which
Underwood (1957) has commented upon in relation to this
- * *
type of design.
The other most common method of investigation o_f__
suicide discussed by Neuringer (1962) is termed the method
of substitute subjects. In this method live subjects are
considered as being representative of suicidal persons.
This method assumes that suicide attempters, suicide
threateners, and other depressed patients are pale carbon
copies of individuals who commit suicide. Evidence contra
dicts this assumption (Farberow, 1950, 1955; Rosen, Hailes,
& Simon, 1954). Neuringer warns that it is logically and
empirically unsound and should not be utilized. The other
possible approach to experimental research in suicide is a
long-term follow-up method. This method would make certain
predictions about suicide, test a large number of individ
uals, and then wait until after their deaths to test out
the experimental hypotheses. The great pitfalls of this
I method are immediately apparent. Not only would problems
of practicality be severe because of the many years' dura
tion of such a study, but these years would also permit the
extensive confounding of experimental results by variables
related to time.
Considering these factors, the method of residuals
appears to be the most' effective method available for re
search. Consequently that method, specifically, the com
parison of suicide notes, is the method used in the present
J
9
study. More research has been conducted and more psycho^
logical literature has been directed toward suicide at-
tempters and "depressive-suicidal" patients than to actual
suicides. Although few, the studies of actual suicide by
the method of residuals have provided sufficient interest
ing findings to stimulate further investigation along these
lines. Farberow and Shneidman (1955) used information from
hospital case records and found certain behavioral differ
ences between attempted, threatened and completed suicides.
These results were consistent with Farberow's (1950) study
which did not include completed suicides. These studies
also reported their findings of the high incidence of
previous suicide attempts among suicide committers.
Shneidman and Farberow (1957) introduced a technique of
suicide note analysis and reported on a comparison between
genuine and simulated suicide notes in terms of Mowrer's
concepts of discomfort and relief. They found the DRQ
interesting, but of limited value in differentiating sui
cidal and non-suicidal individuals. This was primarily
attributed to the DRQ's insensitivity to the quality of
discomfort and relief. In this article they also reported
that data analysis has shown no significant difference in
economic, social-cultural, and personal factors between
ro
suicide committers who leave notes and those who do not.
In a chapter on the logic of suicide (1957b) these
researchers again compared the content of suicide notes and
|
pseudosuicide notes. They concluded that suicidal individ-j
uals use a type of logic which involves a fallacious iden
tification. They term it a "psychosemantic fallacy." This
fallacious identification between the self as experienced
by the self (I ) and the self as it feels itself experi-
s
enced by others (1Q) results in the suicide's acceptance
of erroneous premises and invalid conclusions. One of
these is the assumption that the suicide will be able to
experience after death the effect his expiration will have
upon others.
In a study concerning suicide and age Farberow and
Shneidman (1957) rated actual suicide notes on the basis of
Menninger's (1938) concept of the (1) wish to kill, (2)
wish to. be killed, and (3) the wish to die. They found
more intense interpersonal motives expressed in the notes
of younger suicides and more chronic depressive feelings
expressed by older suicides. In an analysis of 948 genuine
suicide notes in Los Angeles County, Shneidman and Farberow
(1960) found reasons for suicides and affect expressed in
notes differed significantly among nine sociological area
11
types. They stressed the unusual opportunity' offered by
suicide notes for investigators to obtain important infor
mation into the thoughts, feelings and behavior of suicidal
persons, written as they are in the very context of the
suicidal behavior. These researchers (1961) also made a
comprehensive statistical comparison between attempted and
committed suicides which occurred in 1957 in Los Angeles
County. Their data were derived from the coroner's files,
suicide notes, physician's records, and the records of Los
Angeles County and Municipal Emergency hospitals. Again
they found clear differences between suicide committers and•
suicide attempters, and they concluded that it was unwise
to combine these two groups as "suicides" in suicide re
search.
Veterans Administration case records were used by
Farberow, Shneidman and Leonard to investigate suicide
among schizophrenic mental hospital patients (1962) and
patients with malignant neoplasms (1963). This technique
was also used by Farberow, McKelligott, Cohen, and Darbonne
(1966) in a study of suicide among the chronically ill.
Darbonne (1963) studied the age variable in relation to
content of actual suicide notes and found significant dif
ferences between age groups. Osgood and Walker (1959)
studied the effect of motivation level upon language be
havior, assuming that actual suicidal individuals are under
heightened motivation. They compared actual suicide notes,
simulated suicide notes, and ordinary letters to friends
and relatives. Tuckman, Kleiner, and Lavell (1959) inves
tigated the emotional content of actual suicide notes and
found’that half of the notes contained positive affect and
25 percent expressed negative affect, and 25 percent were
neutral in affect. All investigators using suicide notes
have reported high reliability with this technique. In a
validational study, Tuckman, Kleiner, and Lavell (i960)
analyzed 63 suicide notes and found reasons for suicide
stated in the notes in agreement with information of an
informant in over 90 percent of the cases.
Hypotheses
The present study is an investigation into the
communication style of suicidal individuals. Based upon
the above background considerations it is assumed that an
individual's style of communication reflects his life
style. Two comparisons concerning communication style are
made. The first comparison is a test of the general hypo
thesis that the style of communication (and thus the life
style) of individuals who commit suicide differs from that
of non-suicidal individuals. The second comparison is
aimed at determining the similarities and dissimilarities
of the communication style of suicide-threatners with
actual suicides and with non-suicidal normals. As dis
cussed in a previous section of this paper, it is possible
to rephrase Adler’s theoretical statements concerning the
life style of suicidal individuals into communications
terms which can be tested by analyzing and comparing the
communication behavior of suicidal and non-suicidal indi
viduals .
The following is a verbatim summary of Adler's
(1958) theory regarding suicide, as it is expressed by
Ansbacher (1961) in his review of the Adlerian point of
view of suicide. After each statement of Adlerian theory
will be listed the translation of this life style theory
into testable minor hypotheses regarding communication
style.
I. "Pampered Life Style. Suicidal tendencies,
like depression, which is closely related, develop ’in in
dividuals whose method of living, from early childhood on,
has been dependent upon the achievements and the support of
others. They will always try to lean on others' (Ansbacher,
14
1956, p. 319). They expect and demand that others fulfill
their wishes. They have a pseudoconsciousness of their
value to others. They expect always favorable outcomes,
are poor losers, and are unable to play the game as equal
partners" (Ansbacher, 1961, p. 207).
Hypothesis la: Messages of suicidal individuals
will reflect their need to have things done for them. Con
tent related to others doing something for the communicator
or indicating the communicator expects or demands that
others fulfill his wishes will be found more frequently in
the messages (suicide notes) of suicidal individuals than
those of non-suicidal individuals.
Hypothesis lb: Messages of suicidal individuals
will reflect their concern regarding the emotional support
of others. Content related to whether or not the communi
cator feels others have supported him emotionally or have
let him down will be found more frequently in the suicide
notes of suicidal individuals than those of non-suicidal
individuals.
Hypothesis Ic: The suicidal person's difficulty
adapting to loss will be evident in his messages. Mention
of having lost something and being unable to do without it
will be found more frequently in the suicide notes of sui-
15
cidal individuals than non-suicidal individuals.
Hypothesis Id: The heightened importance of parent
figures will be notable in communications of suicidal per
sons. References to parents and/or authority figures will j
be found more frequently in the suicide notes of suicidal
individuals than in those of non-suicidal individuals.
II. "Inferiority Feelings and Self-centered Goal.
’Their self-esteem from childhood on is clearly low, as can
be concluded from their unceasing attempts to achieve great
est importance' (Ansbacher & Ansbacher, 1956, p. 319). The
suicidal person is ambitious and vain. The contemplation
of suicide gives him the feeling of mastery over life and
death. This feeling is 'the supreme expression of the goal
of superiority on the useless side of life' (Adler, 1929,
p. 55). His thinking is self-centered rather than problem-
or other-centered" (Ansbacher, 1961, p. 207).
Hypothesis Ila: Content directly indicating low
self-esteem (self-criticism), a reaction to it (self-
praise) , or indirectly demonstrated by an indication of an
attempt to achieve importance, will be found more frequent
ly in the messages of suicidal individuals than those of
non-suicidal individuals.
Hypothesis lib: The suicidal*s basic feelings of
low self-esteem and dependency are expressed as feelings of
being at the mercy of persons, life, and forces outside
himself, rather than at being in control of his own desti
ny. Comments to this effect will be found more frequently
in actual suicide notes than in the notes of non-suicidal
individuals.
Hypothesis lie: It is expected that the suicidal
individual's self-centered thinking and his dependence on
others for need satisfaction will result in his referring
to other persons, that is, using other-referents, in mes
sages more frequently (relative to the total message
length) than will be found in the messages of non-suicidal
individuals.
Hypothesis lid: It is expected that the suicidal
individual's self-centered thinking will result in his use
of self-referents more frequently (relative to the total
message length) in his messages than will be found in the
messages of non-suicidal individuals.
Hypothesis lie: The suicidal's self-centered goal
interferes with the formation of lasting social ties. Mes
sages of suicidal individuals will more frequently indicate
a loss or absence of social ties or a break in interperson
al relations than will the messages of non-suicidal indi
viduals.
III. "Degree of Activity. Among the maladjusted
the greatest amount of activity is found in criminals,
I
!
where again the murderer shows more activity than the pick
pocket. Among neurotics and psychotics, activity is gen
erally low; but again differences exist within the groups.
Activity is generally lowest in anxiety neurosis and
schizophrenia, greater in compulsion neurosis and depres
sion, and greatest in suicide" (Ansbacher, 1961, p. 207).
Hypothesis Ilia: Suicidal individuals will be ex
pected to communicate more in an unstructured communication
situation than non-suicidal individuals (except for socio-
pathic individuals). Actual suicide notes of suicidal
individuals will be longer than those of non-suicidal in
dividuals.
Hypothesis Illb: Suicidals are highly motor-action
oriented. Verbs of action, which involve actual physical
motor behavior rather than reflecting thought or feeling,
will be found more often in the messages of suicidal indi
viduals than those of non-suicidal individuals.
Hypothesis IIIc: Method of suicide attempt will be
more active for suicidal individuals than for non-suicidal
individuals. (Degree of activity of method determined by a
18
rating scale which is based on ratings of experienced sui
cide prevention clinicians.)
Hypothesis Illd: The individual notes of suicidal
persons will more frequently than the notes of non-suicidal
persons contain a greater number of verbs than adjectives.
This hypothesis is based on Boder's (1940) finding that the
adjective-verb quotient (Avq) is indicative of whether the
individual is more concerned with contemplation (adjec
tives) or with action (verbs other than copulas).
IV. "Veiled Aggression. 'The life style of the
potential suicide is characterized by the fact that he
hurts others by dreaming himself into injuries, or by ad
ministering them to himself' (Ansbacher & Ansbacher, 1956,
p. 324). Occasionally suicide is combined with open ag
gression as in cases where it is preceded by murder. Sui
cide can be taken as an act of reproach or revenge. In
this respect it is also similar to depression and to al
coholism and drug addiction, all of which are forms of
veiled attacks on others who will have the sorrow or care.
In suicide, as in depression, the life style employs com
plaints, mourning, and suffering to influence others by
creating sympathy. Thus we also find self-accusation and
begging for forgiveness, which are often included in the
notes left behind" (Ansbacher, 1961, p. 207).
Hypothesis IVa: Suicidal persons communicate some
awareness that their behavior has been harmful to others. j
|
Messages containing self-accusations and related to thoughts
that others have been sorrowed or hurt by their past be
havior will be found more frequently in the suicide notes
of suicidal individuals than in those of non-suicidal in
dividuals .
Hypothesis IVb: The suicide's concern regarding
the blame of others for his suicide will be apparent in
his communications. Messages related to blame for the
suicidal behavior will be found more often in the suicide
notes of suicidal individuals than in those of non--
I suicidal individuals.
|
Hypothesis IVc: The use of "suffering messages" to
influence others will be prominent in suicidal communica
tions. Content referring to the communicator's suffering
| will be found more frequently in the suicide notes of sui-
j cidal individuals than in those of non-suicidal individ-
| uals. I
| |
j Hypothesis IVd: The suicidal individual attempts
to make sure his veiled aggression hits his targets. Mes
sages related to the notification of a specific person
and/or addressed to a specific person will be found more
often in the suicide notes of suicidal individuals than
those of non-suicidal individuals.
Hypothesis IVe: Content expressing "direct anger"
will be found less frequently in the messages of suicidal
individuals than in those of non-suicidal individuals.
Hypothesis ivf: Content reflecting a general tenor
of veiled hostility, aggression, and resentment in some as
pect of the message will be found more frequently in the
messages of suicidal individuals than in those of non-
suicidal individuals. i
i
| In terms of those communication variables which
differentiate suicide committers from non-suicidal normals,
it is expected that a comparison of suicide threatners with
suicide committers and with non-suicidal normals will dem
onstrate the suicide-threat group's similarities and dis
similarities to suicide committers and to non-suicidal
normals. It is predicted that for those variables which
I
differentiate the suicide-threat group from the commit-
suicide group, the difference will be in the direction of
the non-suicidal-normal group. It is predicted that for
those variables which differentiate the suicide-threat
group from the non-suicidal-normal group, the difference
will be in the direction of the commit-suicide group.
i
i
i
i
i
CHAPTER II
METHOD
|
The general method of this investigation involved
three separate comparisons of suicide notes in terms of
the hypotheses concerning communication style which were
specified previously. The first comparison was between a
group of individuals who actually committed suicide (C/S
group) and a group of individuals who stated they had never
been suicidal (N/S.group). The next two comparisons were
in terms of those communication variables which had dif
ferentiated the commit-suicide (C/S) group from the non-
suicidal (N/S) group to a statistically significant degree.
| The first of these two comparisons was between a group of
individuals receiving outpatient psychotherapy who were
i
| threatening suicide (S/T group) and the cpmmit-suicide (C/S
group). The next comparison was between this suicide-
threat (S/T) group and the non-suicidal (N/S) group.
i
22
23
Subjects
As indicated in previous sections, the subjects of
this investigation consist of three groups: a commit-sui-
cide group (C/S); a non-suicidal-normal group (N/S); and a ;
suicide-threat group (S/T).
Commit-Suicide Group (C/S)
The C/S group was composed of 40 American-born
white males between the ages of 20 and 59 who had committed
suicide in Los Angeles County during the period from
July 1, 1963 to June 30, 1964. Of the 1177 persons who
were classified by the Los Angeles County Coroner as having
died by suicide during this period, 328 were known to have
left suicide notes, 192 of which were left by males.
Either the original or a facsimile copy of these suicide
notes were contained in the County Coroner's files. From
the total number of males leaving suicide notes, a random
I
sample of 40 was selected for this investigation. Random
! selection was based on the Kendall and Smith Table of
| Random Numbers (Edwards, 1960).
|
| Non-Suicidal-Normal Group (N/S)
I The N/S group is composed of 40 American-born
white males who stated that they had never felt suicidal
24
and had never seriously thought of taking their own lives.
In order to keep the sample as homogeneous as possible, in
addition to the variables of nationality, race, and sex,
the subjects were matched with the C/S group on the vari
ables of occupational level and age, plus or minus 5 years.
It was felt that these variables might affect the content
and style of written messages and thus confound the main
effect. Occupational level was used as an index of educa
tional level and status position as educational level is
not stated in the coroner's records. Support for this is
provided by Hollingshead and Redlich's (1958) finding that
the correlation between occupational level and educational
level is .72 and the correlation between occupational level
and class position is .88. Warner (1960) reported a corre
lation between occupational level and education of .77.
The present study used the Hollingshead and Redlich (1958)
modification of Alba Edward's system of classifying occupa
tional levels.
Subjects for the non-suicidal group had all volun
teered to write a make-believe suicide note. Subjects were
obtained from three sources: members of local unions, ac
quaintances of students in a class required of essentially
all freshmen at a large state university; and acquaintances
' 25]
of graduate students of two large local universities.
Strict precautions were exercised in eliciting the
make-believe suicide note, recognizing the moral, ethical,
and taboo overtones associated with suicide. As part of a
screening procedure, before the subject was told that the
study involved suicide, he was asked to complete a ques
tionnaire (see Appendix A) which was designed to detect any
suicidal ideation, behavior, or feelings, which would sug
gest that the task of writing a make-believe suicide note
might be contraindicated. If the subject answered "yes" to
either item 3, 4, 6, 7, 8, or 9 of the questionnaire he was
diverted by being asked to complete a form requesting he
write about his happiest experience (see Appendix B). If
the subject's responses indicated that he would not be up
set by writing a make-believe suicide note he was given a
|
form (Appendix C) which requested he write a make-believe
suicide note as if he were going to take his own life.
Suicide-Threat Group (S/T)
| The S/T group is composed of 31 American-born white
male psychotherapy patients who were currently threatening |
suicide. These were patients of three outpatient clinics
in Los Angeles which offer psychotherapy to suicidal indi
viduals. Due to the difficulty obtaining subjects of this
type, the S/T subjects were not matched for age and occupa
tional level with the C/S and N/S groups. Their age range
was 16 to 69, with a median age of 27.3. Their character
istics in terms of occupational level were as follows:
Range, level 1 to level 7; Median level, 4.5.
The suicide notes were obtained from the individual
patients by asking them to complete the suicide note form
(Appendix D) which was given as part of a battery of
psychological tests that had no apparent relationship to
suicide. This was done to reduce the possibility of ad
verse post-task reaction to writing a suicide note. As a
rule, this battery included a questionnaire (Appendix A),
the Hildreth Feeling and Attitude Scale (Hildreth., 1946) ,
i
and the Saxe Sentence Completion Test (Saxe, 1950). Fol
lowing the testing each patient was asked to comment on
i
j the tests and permitted an opportunity to ventilate any
J feeling generated by the suicide note task. The suicide
i
i
j
I notes were made available to therapists of the individual
|
i
patients so that they could be used as part of the diag
nostic and treatment procedure. This is consistent with
the treatment approach which encourages discussing directly
and in detail the patient's thoughts, feelings, and plans
regarding suicide. Follow-up was done where possible, and
it indicated no adverse reactions from having individuals
who are threatening suicide, write a make-believe suicide
note.
Instruments
Suicide Mote Form for Non-
Suicidal Normals
The suicide note form which was administered to the
N/S group is a modification of the form used by Shneidman
and Farberow (1957a). The modification involves the addi
tion of a statement to the subject to write as many notes
f
as he thinks he would write. It also involves a deletion
of a question asking the subject to specify the addressee
of the note.
The form consisted of a blank sheet of paper with
the following instructions typed across the top: "A study
is being done on the prevention of suicide. For this, it
is necessary to obtain many suicide notes written by normal
people I For this reason, you are asked to write below, in
your own words, the suicide note that you would write if
you were going to take your own life. Make your note sound
as real as you possibly can. Write what you think you
would write, and as many notes as you think you would j
write, if you were planning to commit suicide. 1. Before
you write the note, answer the following question: What
method would you use to take your life? _________________ .
2. Start your note here: (Use the back of this page and
additional paper if you wish.)"
Suicide Note Form for
Suicide Threateners
As indicated in previous sections, there have been
investigations which studied the simulated suicide notes of
non-suicidal normals, however, there are no studies in the
literature which use simulated suicide notes written by in
dividuals who are threatening suicide. The suicide note
| form which was administered to the S/T group in the present
study was generally identical with the form administered to
the N/S group. The differences include the deletion of the
first two sentences which refer to "normal people" and the
addition of the following first sentence: "Make believe
that you are thinking of taking your own life."
i
I
The Degree of Activity of
Suicide Method Scale
Adler did not explicitly define his concept regard
ing degree of activity. In order to test Hypothesis IIIc
of the present study, it was necessary to develop a scale
which could reliably differentiate suicide methods involv
ing a high degree of activity from those not involving a !
high degree of activity. Cattell's (1903) Ranking Method j
was used to order the 18 different methods of suicide used
or contemplated by the subjects of the C/S, N/S, and S/T
groups. In accordance with this method five psychologists,
who were experienced suicide prevention clinicians, were
asked to rank order the 18 suicide methods in terms of
Adler's concept of degree of activity.
Each psychologist was given 18, three by five cards
on each of which was stated one of the 18 suicide methods,
along with the following written instructions:
Instructions for Rating Degree of j
Activity of Suicide Method !
Adler considered "degree of activity" to be one of
the three primary factors of human behavior. He did
not, however, explicitly define this concept. Since he
viewed it in relative terms he did not feel the concept
should be quantified. In order to indicate Adler's
view of activity, several quotations from The Individ
ual Psychology of Alfred Adler (Ansbacher & Ansbacher,
1956) follow. After reading the quotations, (consider
ing the concept in terms consistent with the tenor of j
the quotations) sort the accompanying cards in order ofj
degree of activity involved in the suicide method
stated on each card.
"Although it is probably not possible to express the
degree of activity in quantitative terms, it is obvious
that a child who runs away from his parents, or a boy
who starts a fight in the street, must be credited with
3 0 j
j
!
a higher degree of activity than a child who likes to
sit at home and read a book."
"In problem children, the difference in activity is
already clear and should determine the correspondingly
different measures of education. Wild, unbridled,
stubborn, stealing, quarrelsome children obviously have
a greater degree of activity than shy, reticient,
frightened children, and those who are dependent on
others."
"Cases of sexual perversion may from the point of
view of activity, be found all the way from the lowest
to the highest degree. At the low extreme we find mas
turbation or fetishism, at the high extreme possibly
the lust murder. ..."
"Activity must not be confused with courage, al
though there is no courage without activity."
"Among the maladjusted the greatest amount of ac
tivity is found in criminals, where again the murderer
shows more activity than the pickpocket. Among neurot
ics and psychotics, activity is generally low; but agaih
differences exist within the groups. Activity is gen
erally lowest in anxiety neurosis and schizophrenia,
greater in compulsion neurosis. ..."
The five judges worked separately and independent
ly. The mean of the five ranks assigned to each method
were then computed. Table 1 lists the 18 suicide methods
and their mean ranks. Next, the methods were dichotomized
into high degree of activity versus not-high degree of ac
tivity. Methods with a mean rank of 6.4 or less were
classified as high degree of activity and those greater
than 6.4 were classified as not-high degree of activity.
The mean rank value of 6.4 was selected on the basis of
inspection, showing a difference of 2.2 between it and the
next higher rank. It also represented the rank value above
TABLE 1
MEAN RANKS OF 18 METHODS OF SUICIDE
(ORDERED FROM HIGH TO LOW DEGREE OF ACTIVITY.)
I
Mean
Rank
Method
2.0 Shot self with gun
3.6 Jumped from plane without parachute
3.8 Jumped from plane with parachute and didn't pull
cord
4.4 Hanged self
4.6 Crashed auto
5.8 Jumped from building
6.4 Cut wrists
8.6 Injected drug with hypodermic needle
9.0 Drowned" self
9.4 Asphyxiated self (in air-tight container or bag)
11.6 Ingested overdose of pills and carbon monoxide
11.6 Ingested poison
14.0 Inhaled carbon monoxide
14.2 Ingested overdose of pills (sleeping, tranquilizers
anti-depressants)
14.2 Ingested overdose of pills and alcohol
15.4 Gassed self in home
16.0 Ingested overdose of aspirin
16.4 Purposely exposed self to contract fatal disease
which were included all methods that were agreed on unani
mously by the five judges as being above the median posi
tion of 9.5 (see Appendix E).
Communication Style Scoring Form
Scoring categories based on the 19 hypotheses con
cerning communication style of suicidal individuals were
developed. Appendix F presents an example of the individ- j
ual scoring form which includes scoring instructions and
brief examples concerning each hypothesis. A suicide note
was scored "yes" or "no" for each category (with the ex
ception of six categories to be discussed shortly). That
is, each note was given one "yes" score if there was at
least one reference in the note which satisfied the scor
ing category. Otherwise, one "no" score was assigned. A
"yes" or "no" scoring method was used for most categories
to avoid the effect of note length and word repetition.
Certain of the hypotheses, specifically lie, lid,
and Illb, required comparisons of relative frequencies of
the references in the notes. For these categories a fre
quency count was made of the references in the individual
note and this frequency was divided by the total number of
words in the individual note, giving a ratio score for each
33
category in the note. This was done in order to control
for the effect of note length. In the case of category
Ilia, the comparison to be made was actually note length,
j
so that a simple frequency count of words was made. Cate- !
gory IIIc was scored as "yes," if the suicide note was
rated as high degree of activity, according to Table 1,
and "no" if it was not rated as high degree of activity.
Category Hid was scored "yes" if the frequency of verbs
exceeded the frequency of adjectives in the individual
note, and "no" if it did not.
Twenty notes randomly selected were scored inde
pendently by two advanced graduate students in clinical
psychology. The percentage of agreement between the two
judges for all scoring categories ranged from 60.0 percent
to 100 percent,' with a mean of 87.9 percent. Appendix G
presents the percentages of agreement between judges for
the individual categories.
Procedure
The individual actual suicide notes and make-
believe suicide notes were collected from the commit-
suicide, non-suicidal, and suicide-threat groups. In order
to eliminate rater bias during note scoring, a procedure
was followed throughout data collection which kept the_____
...... 34
scorer from seeing the note prior to scoring and from know
ing the suicide status of the note writer. After collec
tion of the notes, all were typed on the same typewriter
and on the same type of paper. Each note was then assigned
a code number, the numbers having been selected from a
table of random numbers (Edwards, 1960) . The original
suicide notes and the key list regarding the identity of
the coded and typed notes was retained by the secretary of
the Veterans Administration Central Research Unit for the
Study of Unpredicted Death until all scoring was completed.
After typing and coding, the suicide notes were shuffled
thorougly and presented to the scorer. The author scored
all of the notes. Reliability of the scoring method has
been previously discussed under the section entitled
"Ins truments."
| After all 111 notes were scored, they were sepa-
!
i rated into the three comparison groups, commit-suicide,
non-suicidal, and suicide-threat. The comparisons called
i
for by the hypotheses of this investigation were then made.
The first comparison dealt with the question whether cer
tain communication variables found in suicide notes could
differentiate individuals who commit suicide from non-
suicidal individuals. To answer this question, the C/S
group was compared with the N/S group relative to each j
hypothesis (except lie, lid, Ila, and Illb) by means of a
2 x 2 chi square item differentiation analysis. The C/S
j
group was compared with the N/S group relative to hypo-
*
theses lie, lid, Ilia, and Illb using Mood's Median Test
according to the procedure outlined in Siegel (1956).
The next procedure was to compare the suicide-
threat group with the suicide committers and with the non-
suicidal normal group in terms of those communication style
variables which were found to differentiate the suicide-
commit group from non-suicidal-normal group. First, the
C/S group was compared with the S/T group relative to those
variables differentiating the C/S and N/S groups. Follow
ing this, the N/S group was compared with the S/T group
relative to those variables differentiating the C/S and
N/S groups. As in the first comparison, a 2 x 2 chi square
item differentiation analysis was utilized with those hypo
theses for which it was applicable, and Mood's Median Test
was applied to the few hypotheses for which it was rele
vant.
CHAPTER III
RESULTS
Comparison of Commit-Suicide Group
with Non-Suicidal-Normal Group
The overall results of the comparison between the
C/S and E/S groups are given in Table 2. Ten of the 19
hypotheses were supported by the data (P< .05), a number
greater than would be expected to occur by chance. The
data also showed a trend in the direction of support for
hypothesis Ic (P<.10). Of all 19 hypotheses, only two
cases were found in which the observed frequencies were in
I the opposite direction of that which was expected. These
| were hypotheses lid and IVc.
i
!
| The results of the comparison between the C/S and
!
E/S groups will be presented in the general subgroups of
life style from which the communication style hypotheses
were derived:
I. Pampered Life Style (Dependency): Three of the
i
I
I
j four hypotheses related to dependent style of living and
36 ______________________
TABLE 2
| RESULTS OF COMPARISON OF C/S AND N/S GROUPS
j ____
Hypotheses
Freq.
C/S
Freq.
N/S
x2 P
la. Need others do for him 29 9 18.10 .0005
Ib. Needs emotional support 18 8 4.62 .025
Ic. Loss of something 12 6 1.79 N.S. (.10)
Id. Parent figures important 18 10 2.69 .05
Ila1 . Low self-esteem 12 11 0.00 N.S.
lib. At mercy of external forces 15 14 0.00 N.S.
lie. Concern about others 25 11
*
.01**
j lid. Self-referents 18 22
*
N.S.**
lie. Break in social ties > 11 4 2.95 .05
pEIIa. Communicate more 23 17
*
N.S.**
jlllb. Motor-action oriented 24 12
*
.05**
■II Ic.
1
High activity S/method 15 14 0.00 N.S.
itlld. Activity concern > contemplation 19 3 14.11 .0005
; IVa. Others hurt by past behavior 6 3 0.50 N.S.
| IVb. Concern regarding blame for suicide 12 2 7.01 .005
j IVc. Suffering mentioned 16 19 0.20 N.S.
| IVd. Targets named 35 18 14.31 .0005
IVe. Express direct anger 2 3 0.00 N.S.
1 IVf.
i
Veiled hostility 12 4 3.83 .05
i *Compared by Mood's Median Test.
! **Level of significance evaluated by reference to Mainland and Murray Table (1952).
T o i
38]
i
I
communicating were supported. The suicide notes of commit-!
i
suicides to a significantly greater degree than the non- j
i
i
I
suicidal normals reflected the need of suicides to have i
I
j
things done for them, the heightened concern of suicides
with whether or not they have received the emotional sup
port of others, and the importance of parent figures to
them. Worth noting also, was the trend (P<.10) of support
for the notion that suicides have more difficulty adapting
to loss than non-suicidal persons.
II. Inferiority Feelings and Self-Centered Goal:
Support was found for the hypothesis that the suicidal per
son's self-centered goal and dependency on others will re-
I
suit in his referring to other persons in the suicide notes
more frequently than do non-suicidal normals. Also sup
ported was the hypothesis that the self-centered goal of
suicidal persons leads to their communicating their experi
ence of a break or absence of social ties to a significant-
|
!
j ly greater degree than do non-suicidal normals. No statis-
! tically significant differences were found between the C/S
> and N/S groups in terms of content in the suicide note
!
directly expressing low self-esteem or self-praise. Neither
was there any difference found between the C/S and N/S
groups in terms of the number of suicide notes communicat-
\
ing feelings of being at the mercy of external forces, or
the number of self-referents occurring in the notes. j
III. Degree of Activity: Two measures of degree j
j
of activity were able to differentiate the commit-suicides i
from the non-suicidal normals to a statistically signifi
cant degree. It was found that true suicides used signifi
cantly more verbs which refer to actual physical motor be
havior in their messages than did non-suicidal normals.
In addition, the suicide notes of the commit-suicide group
more frequently contained a greater number of verbs than
adjectives than did the notes of the non-suicidal group.
Not supported by the data was the prediction that the sui
cide' s high degree of activity would result in greater
verbal activity, expressed in writing significantly longer
notes. No significant differences were found between the
C/S and N/S groups in terms of the rated degree of activity
of their suicide method.
IV. Veiled Aggression: The suicide notes of true
suicides to a significantly greater extent contained more
expression of veiled hostility than was found in the non-
suicidal-normal group's notes. The commit-suicides also
more frequently communicated concern regarding the blame of
others for their suicide. Adler's notion that suicidal
.... ~ "40
persons take care to see that the real target of their hos
tility is hit is supported by the finding that the C/S
group more frequently than the N/S group requested that
|
specific persons be notified of their suicide and/or ad
dressed their note to specific persons. Although not
statistically significant, a surprising finding was the
trend for the N/S group to mention their suffering more
than the C/S group. This observation was in the opposite
direction from that predicted for this variable. The fre
quency of messages concerning the note writer's feelings
that others had been hurt by his past behavior and the
amount of direct anger expressed in the notes did not dif
fer significantly for the C/S and N/S groups. Expression
of direct anger was almost totally absent from all suicide
notes.
Comparison of Suicide-Threat Group with
the Commit-Suicide Group and' with
the Non-Suicidal-Normal Group
The results of the C/S and S/T group comparisons
and the N/S and S/T groups comparisons in terms of those
|
variables which differentiated the C/S and N/S groups are
given in Table 3. The overall results of this comparison
support the hypothesis that suicide threateners, as a
TABLE 3
RESULTS OF COMPARISON OF C/S AND S/T GROUPS AND N/S AND S/T GROUPS
Hypotheses
Comparing C/S with S/T
x2 P
Comparing N/S with S/T
x2
la. Need others do for him 16.14 .0005 0.12 N.S.
Ib. Needs emotional support 0.00 N.S. 3.15 .05
Ic. Loss of something 1.16 N.S. (.15) 0.00 N.S.
Id. Parent figures important 0.00 N.S. 1.58 N.S. (.15)
lie. Concern about others 1.13* N.S. 3.28* .05
1 lie.
i
Break in social ties 0.21 N.S. 3.45 .05
llllb.
i
Motdr-action oriented 4.34* .025 0.42* N.S.
jllld. Activity concern > contemplation 3.67 .05 2.15 N.S. (.10)
| IVb. Concern regarding blame for suicide 0.19 N.S. 3.42 .05
I IVd.
I
Targets named 8.00 .005 0.34 N.S.
| IVf.
\
Veiled hostility 0.00 N.S. 4.15 .025
*Compared by Mood's Median Test.
group, are in some ways like suicide committers, but in j
i
other ways unlike them, and that these similarities and j
i
differences are detectable in their communication style. i
i
In all cases where the suicide-threat group differed to a
statistically significant degree from either the commit-
suicide group or the non-suicidal-normal group, the differ
ence was in the predicted direction. In other words, if an
individual variable (communication style item) differenti
ated the suicide-threat group from the commit-suicide
group, the difference was always in the direction of the
non-suicidal-normal group. If an individual variable dif
ferentiated the suicide-threat group from the non-suicidal-
normal group, the difference was always in the direction of
the commit-suicide group. Thus, the observed frequencies
of the S/T group in all of these cases fell between the ob
served frequencies of the C/S and N/S group. As was done
for the C/S and N/S comparison, the results of the compari
son between the S/T group with the C/S and the N/S group
will be presented in the four general life-style subgroups.
I. Pampered Life Style (Dependency): The notes of
suicide threatners expressed a need for others to do things
for the writer significantly fewer times than did the sui-
cide-commit notes. There was no significant difference
between suicide threateners and non-suicidal normals rela
tive to this variable. Communication of a concern about
emotional support from others was found more frequently in
the suicide-threat notes than in the non-suicidal-normal
notes, and the S/T notes were not unlike the C/S notes in
this respect. The suicide-threat notes did not differ from
the non-suicidal-normal and commit-suicide notes to a
statistically significant degree relative to the frequency
of references to having lost something or to the frequency
of references to parent figures, although both variables
showed strong trends which warrant reporting in view of the
eixploratory nature of this investigation. The S/T notes
were similar to the N/S notes and differed from the C/S
notes (P<.15) in terms of comments about having lost some-
I
thing. The S/T notes were similar to the C/S notes and
differed from the N/S notes (P<.15) in terms of the fre
quency parent figures were mentioned.
IX. Inferiority Feelings and Self-centered Goal:
The suicide-threat group more frequently than the non-
suicidal-normal group made reference to other persons.
They also more frequently communicated about a break or
absence of social ties. There was no difference between
the S/T group and the C/S group relative to these items.
44
III. Degree of Activity: Both measures of degree
of activity, frequency of motor-action verbs and the verb/
adjective ratio, indicated that the suicide-threat group
was less active than the commit-suicide group. Although
degree of activity did not differ to a statistically sig
nificant degree between the S/T group and the N/S group,
there was a trend (P< .10) for the S/T group, more fre
quently than the N/S group, to use relatively more verbs
than adjectives.
IV. Veiled Aggression: The suicide-threat group's
notes showed a significant difference from the non-
suicidal-normal notes by making more frequent comments
concerning who was to blame for their suicide and by more
i
frequently expressing veiled hostility. They were not sig
nificantly different from the commit-suicide notes regard
ing these features. In contrast, the S/T notes less fre
quently than the C/S notes referred specifically to the
i
targets of their messages.
CHAPTER IV
DISCUSSION
The focus of this study has been on two separate,
but related questions: first, does the communication stylej
of individuals who actually commit suicide differ from in
dividuals who say they have never seriously considered
talcing their own lives? Second, can similarities and dif
ferences between persons who threaten to commit suicide and
persons who actually commit suicide be detected in their
communication styles? The general results of this investi
gation support an affirmative answer to both of these ques
tions. A discussion of the findings relative to the indi
vidual minor hypotheses of this study may render more
definitive answers to these questions and to the next in
line of the hierarchy of questions: In what ways do the
communication styles of these three groups differ?
The findings indicate that persons who commit sui
cide, to a much greater degree than non-suicidal-normals,
................................................ ......... ......................................................................... — ----------------------------------------46
communicate in a style which focuses on content strongly
reflecting thoughts and feelings of dependency, even at the
chronologically mature age of these subjects. The depen
dent style is demonstrated in expressions of the importance
of others actually doing things for them, as well as sup
porting them emotionally. They have a relatively high pre
occupation with the loss of things. Parent figures seem to
be invested with a great deal of importance. While out
patient psychotherapy patients as a group, like the suicide
committers, are highly concerned about emotional support
from others and see parent figures as important, they are
unlike the suicides in that they appear less in need of
i
others doing things for them and less concerned about the
loss of valued objects. These characteristics may describe
something of what helps a person who comes for psychothera
py to work against his suicidal impulses. While the thera
pist may become the important parent figure to whom the
patient looks for emotional support, he is able at some
point to accept the responsibility for his therapy, having
some awareness of the importance of doing things for him
self. This dynamic along with the ego strength to adapt to
losses may help him acquire a more constructive communica
tion style than one of self-harm.
Adler's presentation of the life style of suicidal
individuals indicates that he considered low self-esteem
and self-centered goal to be functionally related concepts.
|
The findings of the present study suggest that the concept
of self-centered goal may, instead, be related to the pam
pered life style/dependency concept. This possibility
arises since the hypotheses related to low self-esteem were
not supported, while those related to dependency and self-
centered goal were supported by the data of this investiga
tion. The suicide notes of the commit-suicides more fre
quently than the non-suicidal-normals referred to a loss or
absence of social ties or a break in interpersonal rela
tions. This was in support of the hypothesis that the sui
cidal person's self-centered goals interfere with the for
mation of lasting social ties. In addition, the suicidal
person's reliance on others for need satisfaction is evi
dent in his referencd to other persons more frequently in
suicide notes than was found in those of non-suicidal
normals. The suicide-threatening patients wrote notes that
were similar to the commit-suicides and unlike the non-
suicidal normals in terms of these two variables. The fact
that suicide note content directly indicating low self
esteem (self-criticism) or a reaction to it (self-praise)
48
did not differ in frequency for suicide committers and non-
suicidal-normals may have at least three explanations.
First, low self-esteem may not be an important variable in
suicide. Second, low self-esteem may be an important vari
able for suicide, but it may not be differentiating since
it is ijnportant in so many emotional problems. Third, the
scoring category for this hypothesis may have reflected ex
perimenter naivete about the self-esteem concept and have
looked for too direct an.expression of this feeling or its
opposite. A different test may have been more revealing.
At present the second explanation seems most likely to be
accurate. Weight for this conclusion is gained from two
sources, the present data and a previous study. Inspection
of the data indicated that references to low self-esteem
were definitely present in the commit-suicide notes,
approximately one-third of the C/S group having made them.
Farberow (1950), in an earlier study, concluded that there
were no significant differences between his groups of sui-
1
| cide attempters, suicide threatners, and controls in terms
i
I
| of the variable of self-esteem, as measured by Maslow's
(1940) Self-Esteem Scale.
f
| Possibly the most important finding of this inves
tigation is that certain communication measures of degree
of activity were able to differentiate between groups of
suicide-committers and non-suicidal-normals, as well as
between groups of suicide-threateners and suicide-commit-
ters. The importance of this finding is that refinement of
these or similar measures may lead to the clinician's being
able to make early differential diagnosis between persons
who think about suicide and then go on to act on the im
pulse and persons who think about suicide but are unlikely
to act on the impulse. The notion of a difference between
these two types of "suicidal persons" has been in the
common sense repertoire of suicide prevention clinicians
for some time. However, all attempts to date to develop an
objective technique for assessing the likelihood of "acting
out on a suicidal impulse" have met an impressive lack of
success. To suggest at this point that the measures of
frequency of motor-action verbs and the verb-adjective
ratio are differentially diagnostic would, of course, be
premature. Much work would need to be done in terms of
validating these measures, and possibly eventually adopting
more refined or different measures. Nonetheless, even this
early type of success in measuring such a difficult vari
able is of interest.
A notion about suicide which has been generally
accepted by the psychodynamic schools of thought is that j
anger plays a part, and that the anger is typically turned
|
inward and veiled. This theory found support in the j
j
present investigation. The commit-suicide group, more j
frequently than the non-suicidal normals, included in
their notes messages of veiled hostility. They also more
frequently reminded the reader of the note that blame
should be assigned regarding the suicidal act. Further,
the suicide-committer took steps to see that the targets
of his suicidal message were specifically named. The
suicide-threat group was very similar to the commit-suicide
group relative to each of these variables except naming the
target of the message. For the latter variable they were
more like the non-suicidal-normals. The statistically non
significant differences between the commit-suicides and
non-suicidal-normals in terms of the number of references
to others having been hurt-by the note writer's past be
havior and the direct expression of anger are interesting
!
because so few members (only 11 percent) of both groups j
i
made such references. It is possible that suicidal personsj
are unaware or unconcerned that others have been hurt by
their past behavior. The difficulty in consciously per
ceiving a relationship between anger and suicide in Western
culture no doubt accounts for some of the reason anger, I
i
directly expressed, was almost totally absent from notes of!
all the groups studied. While not reaching statistical
significance, the observed tendency for non-suicidal-
normals to communicate suffering and misery more frequently
than the suicide commiters was an initially surprising
observation. On further thought, however, it is not diffi
cult to offer one explanation for this finding. It seems
highly plausible that the general public, as represented
by our non-suicidal-normal group, would assume that most
suicides are a result of what the press usually states as
their cause, suffering and despondency. If this assumption
is correct we could expect to find this idea frequently ex
pressed in the make-believe suicide notes.
We have noted that the suicide-threat group is both
similar and dissimilar to the commit-suicide group and to
the non-suicidal group. A question arises whether the
suicide-threat group is from a single population, relative
to ,the variable of suicide, or if it is possibly from two
or more populations, varying in degrees of suicidality. If
suicidality is considered as the degree of likelihood that j
an individual or a group of individuals will successfully
act-out the impulse to commit suicide, it seems reasonable i
to assume that the more suicidal a person is, the more he i
I
will have a communication style similar to the commit sui
cides. If it were possible to split the suicide-threat
into high-potential suicide threateners and low-potential
' ^
suicide threateners, would the former communicate more like
the C/S group and the latter communicate more like the N/S
group? While not a part of the original research question,
the author became too intrigued with this possibility to
let it wait for "future research."
To answer this question, a technique was needed to
dichotomize the suicide threateners into a high-potential
suicide-threat group (H S/T) and a low-potential suicide-
threat group (L S/T). This dichotomization needed to be
done using the available information on each S/T subject.
A lethality assessment technique developed at the Suicide
Prevention Center in Los Angeles and utilized by Hattem
(1963) and Scheitzach (1964) met the above criterion and
was selected to dichotomize the S/T group. The lethality
rating with this technique is derived from a consideration
of five variables: Suicidal Plan, Symptoms, Environmental
Stress, Resources (family, friends, etc.) and Character
(this includes the level of personality development and
ego-strength). Hattem reported that interjudge reliability
53
with this method was relatively high (yielding rank corre
lation coefficients ranging from .76 to .79). Using this
technique; the 31 subjects in the S/T group were split into
11 H S/T and 20 L S/T. From the 20 L S/T, 11 subjects were!
selected at random (using random tables) to be compared
with the 11 H S/T subjects. Having equal numbered groups
permitted use of the Mainland and Murray (1952) tables to
test the significance of difference of frequencies of com
munication variables observed in the groups. Table 4 pre
sents the results of the comparison between the high-
potential suicide threateners and the low-potential suicide
threateners. None of the communication variables were able
to differentiate the two groups to a statistically signifi
cant degree. All trends, however, except three which tied
frequencies, were in the expected direction. The fact that
not a single trend was counter to that which was predicted
suggests that larger sample comparisons might yield statis
tically significant differences.
While such a step is purely conjecture at this
point, it is tempting to posit some relationships between
the major variables which were found to be significantly
related to suicide in this study. Such a consideration may
offer insight into possible dynamics involved in the
TABLE 4
COMPARISON OF HIGH AND LOW SUICIDE-POTENTIAL S/T GROUPS3
Hypotheses
Frequency
H S/T
Frequency
L S/T
Pb
la. Need others do for him 6 2 N.S.
Ib. Needs emotional support 5 2 N.S.
Ic. Loss of something 1 1 N.S.
Id. Parent figures important 6 1 N.S.
lie. Concern about others 6 4 N.S.C
lie. Break in social ties 4 4 N.S.
Illb. Motor-action oriented 3 2 N.S.C
j
lllld. Activity concern > contemplation 3 2 N.S.
I IVb.
i
Concern regarding blame for suicide 1 1 N.S.
| IVd. Targets named 7 3 N.S.
j IVf.
i
Veiled hostility 5 1 N.S.
j r
I q
i N = 11 for each group.
lb
I Evaluated by reference to Mainland and Murray Tables (1952).
!
i C 2
j Compared by Mood's Median Test. (All others compared by X .) m
55
development of a personality prone to suicide. The data
suggest that individuals who commit suicide may have an
early history of being treated in such a way that they de
velop a feeling that they should be pampered by others. j
The predisposing experience of persons who display this
strong need for pampering may be related to poor handling
of dependency needs by the parents, where either excessive
or deficient "feeding" (or an inconsistent combination of
both) was the pattern. As a result of this uncertainty
about whether his survival needs will be met, the individ
ual retains his infantile self-centeredness and continuing
question concerning his survival. The feeling grows that
others do not really want to care for him, so if care is to
be gotten, others must be manipulated into giving it. Suc
cess at manipulation requires the development of a keen
awareness of others in a self-centered way. Any seeds of
a concept of social interest die out quickly' as they go un
fed. He learns to sense and use the unconscious guilt of
others. A surface feeling of his importance to others re
sults from his noting that others give him attention when
he hurts himself. Since attention is as close to caring as
the suicidal person thinks he can get, these experiences
become very important ones for him. The dependency of the
56
suicidal person differs from the dependency found in the
constellation of dynamics of numerous other emotional dis
orders. It is an active-dependency, rather than a passive-
dependency. A passive-dependent personality sits and waits
until others make food available to him. The suicidal per-!
sonality is active, but the activity is directed at manipu
lating others to meet his needs, rather than accomplishing
things to satisfy his own needs. This pattern results in
his being unable to develop any feelings of real independ
ence or self-worth, because he never accomplishes anything
for himself. Since he sees others as the only source for
getting his needs met, he is painfully aware that others
are also the potential frustrators of his needs. Hating
others for this set of events, he is nonetheless afraid to
express his anger directly for fear of losing what he does
have. Thus, a repetitious pattern of hurting himself to
use the guilt of others to hurt them and to manipulate them
becomes subjectively the most reasonable style of coping.
Certain implications for diagnosis and treatment of
suicidal persons may be drawn from the conclusions. It may
be possible to detect individuals with highly self-destruc
tive life patterns from an analysis of other message forms
more readily available in clinical settings. Response to
57
open-ended written applications for psychotherapy or tape
recordings of initial phone contacts are two obvious forms
which could be objectively analyzed. Information from
anamneses and projective tests could also be used to gain
insight into the patient's life style and communication
style. Treatment might be directed at helping the patient
become aware of his intrapunitive active-dependent coping
style, by making objective reference to his style of com
munication during the process of psychotherapy.
Suggestions for Future Research
The data reported here suggest several other inter
esting areas of research. A few of them are mentioned
below:
1. The present study utilized the suicide note as
the message form to be analyzed. It would be interesting
to apply this same technique to other message forms, such
as open-ended written applications for psychotherapy or
tape recordings of initial phone contact requests for
therapy, to see if suicidal and non-suicidal groups could
be differentiated on the basis of those communication vari
ables found differentiating in the present study. This
approach might even be extended to attempt to differentiate
58
other diagnostic groups on the basis of differences in com
munication style.
2. The results of the comparison between high and
low potent ial suicide threateners seems to warrant further ]
study with larger samples.
3. The present study excluded persons at both age
extremes as well as women. It would be interesting to see
if these groups had a style of communication which differed
from the samples studied.
4. The influence of other cultural variables on
the communication style of suicidal persons might be a
fruitful question to investigate.
5. Effort directed toward refinement and further
validating of the degree of activity measures which were
found to differentiate the groups in the present investiga
tion could be well-rewarded effort.
CHAPTER V
SUMMARY
The primary purpose of this investigation was to
test the general hypothesis that the style of communication
of individuals who commit suicide differs from that of non-
suicidal-normal individuals. A second goal has been to
compare outpatient psychotherapy patients, who threaten
■suicide, with suicide committers and with non-suicidal-
normals, in terms of those communication style variables
which were found to differentiate the group of suicide com
mitters from non-suicidal-normals.
Adler's theory of the life style of suicidal indi
viduals was used as a framework from which 19 hypotheses,
concerning the communication style of suicidal individuals,
were derived. These hypotheses concerned predictions that
the communication style of suicide committers would reflect
heightened dependency needs, low self-esteem and self-
centered goals, a high degree of activity, and an expres-
59
60
sive style of veiled aggression which includes hurting the
self in order to hurt and manipulate others. It was ex
pected that evidence of these communication styles would be
found more frequently in the messages of suicide committers
than in those of non-suicidal normals. Secondly, it was
predicted that persons who threaten suicide would communi
cate in some ways similar to those who actually commit sui
cide, while in other ways they would communicate in a man
ner more similar to non-suicidal-normal individuals. It
was predicted that, for those variables which differentiat
ed a group of suicide-threateners from a group of suicide-
committers, the difference would be in the direction of the
non-suicidal-normal groups. For those variables which dif
ferentiated a group of suicide threateners from a group of
non-suicidal normals, it was predicted that the difference
would be in the direction of the suicide-committer group.
The message form which was selected to be analyzed
and compared was the suicide note. This was done because
of the suicide note's unique quality of being a message
which is primarily structured by the writer and is available
from a large, number of persons who have taken their lives.
The subjects for the study were (a) 40 individuals
who had actually committed suicide, (b) 40 individuals who
61
indicated they were not then and had never been suicidal,
and (c) 31 outpatient psychotherapy patients who were
threatening to commit suicide. All subjects were American-
born white males. In addition, the commit-suicide group an
the non-suicidal-normal group were matched for age (within
5 years) and for occupational level.
The actual suicide notes were obtained from the
files of the Los Angeles County Coroner. Suicide notes
were obtained from the non-suicidal-normals and the suicide
threateners by asking them to write the suicide note they
would write if they were going to take their own lives. A
standard form was used to analyze the suicide notes in
terms of the 19 hypotheses. All analyses were done blind.
Inter-rater reliability tested by the percentage of agree
ment method yielded agreements on the 19 categories ranging
from 60 to 100 percent, with a mean of 87.9 percent.
| The overall results of the comparisons made in this
!
| study were in support of the major hypotheses. Only those
I
items related to self-esteem were not successful in differ
entiating between the commit-suicide group and the non-
suicidal-normal group. Ten of the 19 minor hypotheses were
supported in the comparison between these two groups. The
results of the comparisons of the suicide threateners with
the suicide-committers and with the non-suicidal-normals
demonstrated the predicted similarities and differences
between these groups. The variable which most strongly
distinguished the commit-suicide group from the suicide-
threat group was the greater degree of activity reflected
in the communication style of the suicide-committers. The
study also gave suggestions for future research.
i
R E F E R E N C E S
63
REFERENCES
Adler, A. Problems of neurosis. London: Routledge & Kegan
Paul, Ltd., 1929.
Adler, A. Suicide. J. indiv. Psychol.. 1958, 14. 57-61;
translated from Selbstmord. Int. zeitschrift indiv.
Psychol.. 1937, 15, 49-52.
Ansbacher, H. L. Suicide: the Adlerian point of view. In
N. L. Farberow & E. S. Shneidman (Eds.), The cry for
help. New York: McGraw-Hill, 1961.
Ansbacher, H. L., & Ansbacher, Rowena R. (Eds.). The indi
vidual psychology of Alfred Adler. New York: Basic
Books, 1956.
Bateson, G., Jackson, D. D., Haley, J., & Weakland, J.
Toward a theory of schizophrenia. Behav. Sci., 1956,
JL, 251-264.
Cattell, J. McK. Statistics of American psychologists.
Amer. J. Psychol.. 1903, 14, 310-328.
Darbonne, A. R. Suicide and age: a content analysis of
suicide notes. Paper read at Western Psychol. Ass.,
Los Angeles, April, 1963.
Durkheim, E. Le suicide. Paris: Librairie Felix Alcan,
1897; translated by J. A. Spalding & G. Simpson and
published as Suicide. Glencoe, 111.: Free Press, 1951.
Edwards, A. L. Experimental design in psychological re
search. New York: Holt, Rinehart & Winston, 1960.
Farberow, N. L. Personality patterns of suicidal mental
hospital patients. Genet. Psychol. Monogr., 1950, 42,
65
Farberow, N. L., & Shneidman, E. S. Attempted, threatened
and completed suicide. J. abnorm. soc. Psychol., 1955,
50, 230-231.
Farberow, N. L., & Shneidman, E. S. Suicide and age. In
E. S. Shneidman & N. L. Farberow (Eds.), Clues to sui
cide. New York: McGraw-Hill, 1957.
Farberow, N. L., & Shneidman, E. S. (Eds.). The cry for
help. New York: McGraw-Hill, 1961.
Farberow, N. L., Shneidman, E. S., & Leonard, Calista V.
Suicide among schizophrenic mental hospital patients.
Washington, D.C., V. A. med. Bull., 8_, March, 1962.
Farberow, N. L., Shneidman, E. S., & Leonard, Calista V.
Suicide among general medical and surgical hospital
patients with malignant neoplasms. Washington, D.C.:
V. A. med. Bull., 9_, February, 1963.
Farberow, N. L., McKelligott, J. S., Cohen, S., & Darbonne,
A. R. Suicide among patients with cardiorespiratory
illnesses. J. Amer. Med. Ass., 1966, 195, 422-428.
Freud, S. Mourning and melancholia. In S. Freud, Collect
ed papers. Vol. 4. London: Hogarth Press, 1925.
Hattem, J. The precipitating role of discordant interper
sonal relations in suicidal behavior. Unpublished
doctoral dissertation, Univer. Houston, 1963.
Hayakawa, S. I. Suicide as a communicative act. Etc.:
Rev, gen. Semant., 1957, 15, 46-51.
Hildreth, H. M. A battery of feeling and attitude scales
for clinical use. J. clin. Psychol.. 1946, 23, 214-221.
Hollingshead, A. B., & Redlich, F. C. Social class and
mental illness. New York: Wiley, 1958.
Jackson, D. D. Theories of suicide. In E. S. Shneidman &
N. L. Farberow (Eds.), Clues to suicide. New York:
McGraw-Hill, 1957.
66
Litman, R. E., & Farberow, N. L. Emergency evaluation of
self-destructive potentiality. In N. L. Farberow &
E. S. Shneidman (Eds.), The cry for help. New York:
McGraw-Hill, 1961.
I
Mainland, D., & Murray, I. M. Tables for use in fourfold i
contingency tables. Science, 1952, 116, 591^594.
Maslow, A. H. A test of dominance feelings (self-esteem)
in college women. J. soc. Psychol.. 1940, 12, 255-270.
Menninger, K. Man against himself. New York: Harcourt,
Brace, 1938.
Neuringer, C. Methodological problems in suicide research.
J. consult. Psychol., 1962, 26, 273-278.
Osgood, C., & Walker, Evelyn G. Motivation and language
behavior: a content analysis of suicide notes. J.
abnorm. soc. Psychol., 1959, 59, 58-67.
Rosen, R., Hailes, W. M., Sc Simon, W. Classification of
"suicidal" patients. J. consult. Psychol., 1954, 18,
359-362.
Ruesch, J. Disturbed communication. New York: Norton,
1957.
Ruesch, J. Therapeutic communication. New York: Norton,
1961.
Ruesch, J., Sc Bateson, G. Communication: the social matrix
of psychiatry. New York: Norton, 1951.
Saxe, C. H. The Saxe Sentence Completion Test. In H. M.
Grayson (Ed.), A psychological admissions— testing pro
gram and manual. Los Angeles: V. A. Center, 1950.
Scheitzach, J. An evaluative comparison of the dyadic re
lationship wherein one member engages in self-destruc
tive behavior. Unpublished doctoral dissertation,
Univer. of California at Los Angeles, 1964.
67
Shneidman, E. S., & Farberow, N. L. Comparison between
genuine and simulated suicide notes by means of the
Mowrer DRQ. J. gen. Psychol.. 1957, j66, 251-256. (a)
Shneidman, E. S., & Farberow, N. L. The logic of suicide.
In E. S. Shneidman & N. L. Farberow (Eds.), Clues to |
suicide. New York: McGraw-Hill, 1957. (b)
Shneidman, E. S., & Farberow, N. L. A socio-pathological
investigation of suicide. In H. P. David & J. C.
Brengelman (Eds.), Perspectives in personality re
search. New York: Springer, 1960.
Shneidman, E. S., & Farberow, N. L. Statistical compari
sons between attempted and committed suicides. In N.
L. Farberow & E. s. Shneidman (Eds), The cry for help.
New York: McGraw-Hill, 1961.
Siegel, S. Nonparametric statistics for the behavioral
sciences. New York: McGraw-Hill, 1956.
Sullivan, H. S. The interpersonal theory of psychiatry.
New York: Norton, 1953.
Tabachnick, N. D„, & Farberow, N. L. The assessment of
self-destructive potentiality. In N. L. Farberow &
E. S. Shneidman (Eds.), The cry for help. New York:
McGraw-Hill, 1961.
Tuckman, J., Kleiner, R. J., & Lavell, Martha. Emotional
contents of suicide notes. Amer. J. Psychol., 1959,
116. 59-63.
Tuckman, J., Kleiner, R. J., & Lavell, Martha. Credibility
of suicide notes. Amer. J. Psychol.. 1960, 116. 1104-
1106.
Underwood, B. J. Psychological research. New York:
Appleton-Century-Crofts, 1957.
Warner, W. L. Social class in America. New York: Harper
& Bros., 1960.
n
A PP E NDI CES
68
69
APPENDIX A
GENERAL INFORMATION QUESTIONNAIRE
Number
FILL OUT ALL THE QUESTIONS BELOW.
(Please note that it is not necessary that you sign
your name to this form.)
AGE (how old are you?)__________ .
SEX (Male or Female) ___________.
MARITAL STATUS (Are you single, married, divorced, sepa
rated or widowed?) __________________________.
BIRTHPLACE (In what country were you born?) _______________
(If U.S. in what state were you born?) ________
RELIGION (What religion are you?) ___________________________
RACE (White, Negro, Oriental) _______________________________
EDUCATION (What is the highest grade you completed?) ______
OCCUPATION (What is your present job or occupation?) ______
(How long have you had this occupation?) _______
(What were the last two occupations that you
worked ,at?)
1. 2. _____________________
ANSWER "YES" OR "NO" TO EACH OF THE QUESTIONS BELOW.
YES or
NO
1. Have you ever had any emotional or nervous
difficulties? ______
2. Have you had any treatment for these emotional
troubles? ______
3. Have you ever written a suicide note?_____________ ______
4. Have you ever thought of writing a suicide note? _______
5. Have you ever been severely blue or depressed
for periods of time? ______
6. Have you ever seriously threatened to take your
own life? ______
7. Have you ever made an actual attempt to end or
take your own life? (When? _________________ )
(What method? _) ______
69
8. Have you had thoughts of suicide occur to you
recently?
9. Would it bother you to read or think about
suicide now?
71
APPENDIX B
HAPPY NOTE FORM
Number
A STUDY IS BEING DONE COMPARING HAPPY AND UNHAPPY THOUGHTS.
FOR THIS, IT IS NECESSARY TO OBTAIN NOTES ABOUT YOUR HAP
PIEST EXPERIENCE. PLEASE WRITE BELOW, IN YOUR OWN WORDS*
ABOUT THE MOST ENJOYABLE EXPERIENCE THAT YOU HAVE EVER HAD.
TELL ABOUT THE MOST PLEASANT EXPERIENCE OR THE HAPPIEST
TIME YOU HAVE HAD IN YOUR LIFE. YOU CAN USE BOTH SIDES OF
THIS PAPER IF YOU WANT TO. BEFORE YOU WRITE, ANSWER THESE
TWO QUESTIONS FIRST:
1. When did this experience happen?
2. Where did this take place? _____
START YOUR NOTE HERE:
APPENDIX C
72
NOTE FORM FOR NON-SUICIDAL SUBJECTS
Number
A STUDY IS BEING DONE ON THE PREVENTION OF SUICIDE. FOR
THIS, IT IS NECESSARY TO OBTAIN MANY SUICIDE NOTES WRITTEN
BY NORMAL PEOPLE I FOR THIS REASON, YOU ARE ASKED TO WRITE
BELOW, IN YOUR OWN WORDS, THE SUICIDE NOTE THAT YOU WOULD
WRITE IF YOU WERE GOING TO TAKE YOUR OWN LIFE. MAKE YOUR
NOTE SOUND AS REAL AS YOU POSSIBLY CAN. WRITE WHAT YOU
THINK YOU WOULD WRITE, AND AS MANY NOTES AS YOU THINK YOU
WOULD WRITE, IF YOU WERE PLANNING TO COMMIT SUICIDE.
BEFORE YOU WRITE THE NOTE, ANSWER THE FOLLOWING QUESTION:
1. What method would you use to take your life? __________
2. START YOUR NOTE HERE: (Use the back of this page and
additional paper if you wish.)
73
APPENDIX D
NOTE FORM FOR PSYCHOTHERAPY PATIENTS
Number
Make believe that you are thinking of taking your own life.
Write below, in your own words, the suicide note that you
would write if you were going to take your own life. Make
your note sound as real as you possibly can. Write what
you think you would write, and as many notes as you think
you would write, if you were planning to commit suicide.
1. Before you write the note, answer the following ques-
tion: What method would you use to take your life?
__________________________________________________________________
2. Start your note here: (Use the back of this page and
additional paper if you wish.)
)
74
APPENDIX E
RANK OF DEGREE OF ACTIVITY IN SUICIDE METHOD
ASSIGNED BY INDIVIDUAL JUDGES
Suicide Method
A B
Judge
C
I
D E
Shot self with gun. 1 1 1 3 4
Jumped from plane without parachute. 7 4 4 1 2
Jumped from plane with parachute and
didn't pull cord. 8 3 5 2 1
Hanged self. 5 6 2 6 3
Crashed auto. 2 2 9 4 6
Jumped from building. 6 5 6 5 7
Cut wrist(s). 3 8 8 8 5
Injected drug with hypo. 4 11 10 9 9
Drowned self. 9 9 7 . 7 13
Asphyxiated self (in air-tight con
tainer or bag). 15 7 3 14 8
Ingested overdose of pills, CO. 11 13 13 11 10
Ingested poison. 10 12 14 10 12'
Inhaled CO. 16 17 11 15 11
Ingested overdose of pills (sleeping,
tranquilizers, antidepressants) 13 14 15 13 16
Ingested overdose of pills and
alcohol. 12 15 17 12 15
Gassed self in home. 17 18 12 16 14
Ingested overdose of aspirin. 14 16 16 17 17
Purposely exposed self to contract
fatal disease. 18 10 18 18 18
i
75
APPENDIX F
COMMUNICATION STYLE SCORING FORM
INDIVIDUAL DATA COLLECTION SHEET
Number
Freq.
Yes No Note content related to hypotheses
_____ _____ la. Content related to others doing some
thing for the communicator or indicating the
communicator expects others to fulfill his
wishes. Instruction. Memories of something
being done or not being done for communicat
or. (Ex. “You never did a thing for me. You
did so much. Bury me in. . . . Take care of.
. . . Take my. . . . Notify. Help me. Other
did/did not help. Cure me. My happiest
memory is when you . . . for me.")
_____ _____ lb. Content related to whether or not com
municator feels others have or have not en
couraged or supported him emotionally. (Ex.
"You let me down. I could never satisfy you.
Without you [her], life is meaningless. You
never understood. No one cares about me.
You did as much as you could. I could always
depend on you.")
_____ _____ Ic. Mention of having lost something and
being unable to do without it. (Ex. "Health,
money, job; loved-one" etc.)
_____ _____ Id. Contains reference to parents and/or
authority figures. (Ex. Parent, police, God;
president, therapist. Dr.)
_____ _____ Ila. Content indicating low self-esteem
(self-criticism) or a reaction to it (self-
praise) or shown in an indication of an at
tempt to achieve importance. (Ex. Fears
loss face or status, sees self as a burden or
76
APPENDIX F (continued)
Freq.
Yes No Note content related to hypotheses (Ila.
continued)
worthless. Self-praise. Self-criticism.
"Failed. You can find someone better than
me. Can never accomplish goals. God knows
I've tried, but I just can't make it.")
__________ lib. Comments reflecting feelings of being
at the mercy of persons, life and forces out
side himself, rather than being in control of
his own destiny. (Ex. "You [-or life-] made
me [-or wanted me to-] do this," rather than,
"I decided to do this"; I can't stand [-or
help-] it, rather than "I will not [or don't
want to-] stand it." "Life is driving me [or
doesn't accept me]," rather than "I am in
control [or I don't accept and am tired of
life]." "This is the only way," rather than,
"This is the best [or easy] way.")
______ lie. Use of other-referents in message.
(Ex. You, your, he, her, John, they, we, our,
God, Mother, wife, us, etc.)
__________ lid. Use of self-referents in message.
(Ex. I, me, mine, my, myself, own name, etc.)
_____ _____ lie. Content and tenor indicates subject
feels an absence of social ties or a break in
interpersonal relations. (Ex. "Lonely, re
jected, no one cares, can't do without her
[you]."
__________ Ilia. Message length. (Total number of
words.)
_____ _____ Illb. Number of verbs of action. Involve
physical motor behavior rather than reflect
ing thoughts, feelings, etc. Not active
verbs used only grammatically, like "he did
worry.") (Ex. Run, drove, give, write,
77
Freq.
Yes
APPENDIX F (continued)
No Note content related to hypotheses. (Illb
continued)
notify, work, tell, call, come, etc.)
IIIc. Rated degree of activity of method of
suicide attempt. (Method: _________________ .)
Hid. Number of adjectives ________ . Number
of verbs other than copulas __________.
IVa. Messages contained self-accusations
and/or comments that others have been sor
rowed or hurt by his past behavior. (Ex.
"I am sorry X hurt you or caused you trouble.
I know I've upset you. Maybe this will make
it up to you. Now X won11 bother you any
more. I've hurt so many people.")
IVb. Messages relating to blame (that some
one is or is not to blame) for the suicidal
behavior. (Ex. "You are/are not to blame.
You could have stopped me by . . . No one is
at fault but me.")
IVc. Messages referring to communicator's
suffering. (Ex. mourning, pain, loneliness,
hopelessness, hurt, confusion, bad luck,
sickness, fear; "can't find my place, tired,
just can't take this any longer.")
IVd. Messages related to the notification of
a specific person or addressed to a specific
person.
IVe. Messages expressing "direct" anger.
(Ex. "I hate you. You S.O.B., doing that
made me furious. I'd like to kill you. I
hope you are hurt as much some day. You are
cheap and dirty to treat me like this. I
hope you are tortured by the memory of this.")
Freq.
Yes
APPENDIX F (continued)
No Note content related to hypotheses. [
;
IVf. Content reflecting a general tenor of 1
veiled hostility, aggression, and resentment |
in some aspect of the message. (Ex. "Don't
let Jane have the children. I hope this is
what you wanted. She never cared for me.
You pushed me to this. Why couldn't you be
satisfied with me, I gave you everything. I
don't deserve your treatment. He lied about
me. I ashed you to stop drinking. I can't
stand this rejection. Enjoy your freedom.
If only you would have let me do what I
wanted. Now I won't be in your way. You are
right I am a burden, but I'll no longer bur
den you and all the others.")
APPENDIX G
|
PERCENTAGE OF AGREEMENT OF SCORING !
i
(BETWEEN TWO JUDGES SCORING 20 NOTES INDEPENDENTLY) i
Scoring Category Percentage of Agreement
la. 85
Ib. 70
Ic. 95
Id. 85
Ila. 60
lib. 80
lie. 90
IIcl. 85
lie. 85
Ilia. 100
Illb. 90
IIIc. 100
m a . 90
IVa. 95
IVb. 90
IVc. 95
IVd. 100
IVe. 95
IVf. 80
l

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Sex-Role Preferences Of Early Adolescents In Relation To Adjustment

PDF

An Experimental Study Of Aggression Habit Patterns And Sex Role

PDF

First Steps In An Attempt To Construct An Objective Test Of Character Structure

PDF

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

PDF

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

PDF

Self-Worth, Future Goals, Mood And Time Perception

PDF

Prediction Of Therapeutic And Intellectual Potential In Mentally Retardedchildren

PDF

Delinquency As A Function Of Intrafamily Relationships

PDF

Behavioral Seriousness And Impulse-Control Balance In Delinquency

PDF

Attitudes Of Ministers And Lay Leaders Of The American Baptist Conventionof The State Of Washington On Selected Social Issues

PDF

Acculturation And Value Change

PDF

A Validation Study For The Feshbach And Roe (1968) Test For Empathic Ability

PDF

The Effects Of Sex, Assigned Therapist Or Peer Role, Topic Intimacy, And Expectations Of Partner Compatibility On Dyadic Communication Patterns

PDF

Experimenter Expectancy Effect Examined As A Function Of Task Ambiguity And Internal Versus External Control Of Reinforcement

PDF

The Effect Of Conditions Of Risk, Internal Versus External Control Of Reinforcement, And Sex On Binary Choice Probability Learning

PDF

A Temporal Approach-Avoidance Conflict In An Academic Test Situation

PDF

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

PDF

The Use Of Multiple Objective Measures In Survey Research As A Means For Predicting Respondent Future Behavior

PDF

Verbal Reports Of Emotional States And Onsets And Offsets Of Conditioned Stimuli

PDF

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

Asset Metadata

Creator Darbonne, Allen Roy (author)

Core Title An Investigation Into The Communication Style Of Suicidal Individuals

Degree Doctor of Philosophy

Degree Program Psychology

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

Tag OAI-PMH Harvest,psychology, general

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Seward, Georgene H. (committee chair), Jacobs, Alfred (committee member), Wexler, Murray (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-225991

Unique identifier UC11360125

Identifier 6700399.pdf (filename),usctheses-c18-225991 (legacy record id)

Legacy Identifier 6700399.pdf

Dmrecord 225991

Document Type Dissertation

Rights Darbonne, Allen Roy

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