Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Verbal Conditioning Of Dependence-Independence In Domiciled Veterans
(USC Thesis Other)
Verbal Conditioning Of Dependence-Independence In Domiciled Veterans
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
This dissertation has been
microfilmed exactly as received g 9 -6 3 7
ORLANDO, Nita Ethel, 1929-
VERBAL CONDITIONING OF DEPENDENCE-INDEPEN*
DENCE IN DOMICILED VETERANS.
University of Southern California, Ph.D„ 1968
Psychology, clinical
University Microfilms, Inc., Ann Arbor, Michigan
VERBAL CONDITIONING OF DEPENDENCE-INDEPENDENCE
IN DOMICILED VETERANS
by
Nita Ethel Orlando
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 1968
UNIVERSITY O F S O U T H E R N CA LIFO RN IA
T H E G R A D U A TE S C H O O L
U N IV E R S IT Y PA R K
L O S A N G E L E S . C A L IF O R N IA 9 0 0 0 7
This dissertation, written by
Nita Ethel Orlando
under the direction of h ..3 .V .. 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 I L O S O P H Y
Dean
_ June, 1968
Date...... ......
DISSERTATION COMMITTEE
V
,Chairman ' / .C hair
| ACKNOWLEDGMENTS
I
; A special debt of gratitude is due to Dr. Philip.
i Carman, Chief of the Domiciliary Psychology Service, for
I his inexhaustible administrative ingenuity, and to my
| Chairman, Dr. Leslie Havran, for his inexhaustible patience.
I To the members of my committee, Drs. Marjorie Void and
! Henry Slucki, a note of thanks for their integrity at the
I moment of truth.
I thank all those serving with the Veterans Admin-
i
istration who gave so generously of their time to assist in
| the production of this research: the Domiciliary veterans
I who participated in the study, Drs. Frank Risch, Donald
;Spiegel, and Carrie Chapman. Among many others, two
veterans contributed significantly: Jack Elder, Engineer,
and Miles Flickinger, Man Friday. Finally, for his
constant psychological guidance and support, thanks to
Dr. B. Pupulis.
ii
TABLE OF CONTENTS
Page
1 ACKNOWLEDGMENTS.................................. ii
i
i
ILIST OF TABLES...................................... vi
IlIST OF FIGURES.........................................viii
I Chapter
; I. THE PROBLEM.................................. 1
Introduction
The dependent variable: dependency
Preliminary question
The independent variable: verbal
conditioning
Preliminary question
Statement of the Problem
Background: change in domiciliary
administrative function
Behavioral effects of institu
tionalization
Professional manpower shortage
Summary
II. SURVEY OF THE LITERATURE..................... 11
Dependent-Independent Dimension
Definitions of dependency
Research on dependency-
independency dimension
Summary
Verbal Conditioning and Positive
Reinforcement
Research on positive reinforcement
contingencies
Summary
III. METHOD OF DESIGN.............................. 33
Purpose
Definitions
Chapter
Operational definition:
! dependent variable
j Operational definition:
I independent variable
j Factorial Design
Hypothesis I
Hypothesis II
Hypothesis III
Hypothesis IV
Method
j Subjects
| Pre-test measures
Procedure
i Post-test measures
| IV. RESULTS.....................................
j Characteristics of the Experimental
! and Control Groups
; Hypothesis I
Hypothesis II
Hypothesis III
Hypothesis IV
V. DISCUSSION OF RESULTS ......................
Hypothesis I
Hypothesis II
Hypothesis III
Hypothesis IV
Butterfly Results
Administrative Implications of
Butterfly Data
VI. SUMMARY AND CONCLUSIONS . . . .............
Summary
Conclusions
Treatment Implications of the Study
Research Implications of the Study
I BIBLIOGRAPHY.......................................
APPENDICES....................................... .
APPENDIX A. Minnesota Multiphasic Personality
Inventory Dependency Scale Items . .
Page
44
67
79
92
100
102
iv
Page
APPENDIX B. Items from Spiegel's Dependency and
Self-Assertion Scale................ 106
APPENDIX C. Instructions to R a t e r ................ 108
|
| APPENDIX D. Master Table of Individual Subject
i Scores: De Scale, SPE, and
Butterfly D a t a...................... 117
APPENDIX E. Supervisors Performance Evaluation . . 120
LIST OF TABLES
Table Page
1. Factorial Design ............... ....... 34
2. Rotating Schedule of Group Meetings........... 42
3. Summary of Means for Age, Length of
Institutionalization and t-Values........... 45
4. Source Table for Analysis of Variance,
Pre-Post Dependency Scores .................. 46
5. Summary of Dependency Scale Score Changes,
Pre- and Post-Testing........................ 47
6. Summary Table: Frequency of Statements of
Independence by Subjects Over T i m e ......... 49
7. Chi Square 2 X 2 Contingency Table:
Frequency of Response Over Experimental
Time, Weeks 1 and 2 .......................... 52
8. Chi Square 2 X 2 Contingency Table:
Frequency of Response Over Experimental
Time, Weeks 3 and 4 .......................... 52
9. Chi Square 2 X 2 Contingency Table:
Frequency of Response Over Experimental
Time, Experimental and Control Groups .... 52
10. Summary of Rating Changes: Pre-Post SPE . . . 54
11. Pre-Post Difference Scores: Dependency Scale
and Supervisors1 Performance Evaluation . . . 55
12. Dependency Difference Scores and Median Age . . 57
13. Dependency Difference Scores and Mean Stay . . 59
14. Ranked Dependency Scores of Experimental and
Control Groups ................... 61
■V. w
Table Page
15. Chi Square 2 X 3 Contingency Table: The
Relationship of Pre-Test De Scores to
the Number of Butterflies Issued:
Groups I, II, I I I ......... — . .............. 64
16. Summary of Medians, Mean Changes and Mean
Numbers of Butterflies to Pre-Test, Oral
Interview, and Post-Test...................... 66
vii
LIST OF FIGURES
Figure Page
1. Frequency of Response: Statements of
Independence .............................. 50
2. Optimum Range for Verbal Conditioning as a
Function of Initial Dependency Level . . . 62
viii
CHAPTER I
THE PROBLEM
Introduction
The Dependent Variable: Dependency
Clinical diagnoses and treatment of the psycho
neuroses, functional psychoses, and behavior disorders rest
heavily upon the theoretical construct of a dependence-
independence dimension (Fenichel, 1945; Levy, 1943; Saul,
1944). Coleman (1964) regards the stressful developmental
change from dependence to self-direction as the core
conflict in contemporary life. He considers that the
dependence-independence conflict is more important than
physical privations or psychological frustrations as a
source of potential maladjustment:
. . . All of us have to make the transition from
the dependent, protected status of childhood to
adult status and responsibility, with all that
this implies in terms of independence and self
directed activity. . . . Many of those who cannot
achieve independence become life-long inmates of
mental hospitals, where their needs are taken care
of and where they do not have to face the respon
sibilities and problems of normal adult life. . .
(p. 153)
Alexander (1946) maintains that the external pres
sures in our western culture for independent, self-assertive
I 2
| achievement and the individual's internal needs for other-
i
;directed dependence and security are polar opposites. This
!
!polarity "has been found to be the nuclear conflict in such
j
i widely different conditions as neurotic criminality and
some types of psychosomatic diseases" (p. 69).
Maslow and Mittleman (1941) equate dependency in
!hospitalized adults with a pathological helplessness that
is "always potentially catastrophic in intensity." It
i
:forces patients to automatically overestimate the strength
;of the professional staff and to unrealistically "seek
restoration by dependency on a stronger person, by blind
faith in him, by a desire for his constant presence and
: his undivided affection and esteem" (p. 91).
Arieti (1959) places conflicts over dependence-
' independence in a central position in the psychodynamics of
tubercular patients, peptic ulcer patients, in both
neurotic and involutional depressives, and extremely so in
alcoholics. The deliberate inducement of urgent dependency
needs has become one of the Three-D effects (Dependency,
Debility, and Dread) required for successful brainwashing
of American prisoners abroad (Arieti, 1966).
;Preliminary Question
The dimension of dependence-independence may thus
be viewed as a major theoretical variable in the clinical
concept of personality malfunction. The preliminary
3
i
l
■ statement of the problem therefore becomes: Can the con-
? struct of dependence-independence be defined, isolated,
I and experimentally manipulated to produce specifiable
I changes in human behavior?
The Independent Variable:
Verbal Conditioning
Speculations of the relationship of pain and
!pleasure as determinants of man's behavior is as old as
jrecorded thought (Brinton, 1950; Russell, 1945). The
Greco-Western tradition of ethical hedonism, the affirma
tion that the attainment of happiness is the highest good,
! was developed into psychological hedonism, the postulate
that man seeks pleasure and avoids pain, by Jeremy Bentham
|(Chaplin and Krawiec, 1962).
Freud (1919) incorporated this psychological
;hedonism of the utilitarians in his pleasure principle, a
principle by which the Id, as a sub-system of the person
ality, seeks to gratify instinctual drives. Gratification,
:or pleasure, is derived when tension aroused by the
accumulation of physiological energy is discharged. The
pleasure principle is essentially a tension-reduction model
(Ford and Urban, 1965).
Boring (1950) views Freud's pleasure principle as a
"hedonism of the future," with purpose conceptualized as a
"pull" to action. The contemporary experimental psycholo
gists, on the other hand,opposed Freud by stressing
purpose as a "push" to behavior; their "hedonism of the
past" looked to prior events for the determinants of
actions (p. 706).
It remained for Thorndike (1911) to focus on the
behavioral consequences of pleasure and pain and to bring
the "hedonism of the past" to a central position in the
modern psychology of learning with the formulation of his
theoretical Law of Effect. This law was based on careful
observation of animal problem solving behavior. Animal
learning of the correct response is "stamped in"; the
incorrect response is "stamped out" as a result of the
effects that followed the response:
. . . Of several responses made to the same situa
tion, those which are accompanied or closely
followed by satisfaction to the animal will, other
things being equal, be more firmly connected with
the situation so that, when it recurs, they will
be more likely to recur.
. . . Those which are accompanied or closely
followed by discomfort will, other things being
equal, have their connection with that situation
weakened, so that when it recurs, they will be
less likely to occur. The greater the satis faction
or annoyance, the greater the strength or weakening
oIT the (stimulus-response) bond. . . (Thorndike,
1911; p. 244). [Italics mine.]
"Satisfaction" and "annoyance" as determinants of
learning are further carefully defined by Thorndike:
. . . By a satisfying state of affairs is meant one
which the animal does nothing to avoid, often doing
things which maintain or renew it. By an annoying
state of affairs is meant one which the animal does
nothing to preserve, often doing things which put
an end to it. . . (1913; p. 2).
i The more recent empirical law of effect is a state-
j ment of the fact that when an operant response (behavior
ithat is identified by its consequences acting upon the
environment) is followed by an appropriate reinforcement,
; the strength of the operant response is increased. The
I appropriate reinforcing stimulus may be positive (added to
the conditioning situation) or negative (removed from the
conditioning situation). Its only requirement is that it
increases the probability of a response, whether approach
;or avoidance.
The operation of reinforcement is, according to
Skinner (1950), . . no theory. . . . It simply speci
fies a procedure for altering the probability of a chosen
iresponse" (p. 200). Reinforcement, then, is an empirical
phenomenon, operationally demonstrable and repeatedly
verified by an imposing body of research. For an overview,
the following sources are readily available: Ferster and
Skinner, 1957; Hilgard and Marquis, 1961; Keller, 1954; andj
j
Verplanck, 1955a. j
)
j
Preliminary Question j
The problem therefore becomes: Can the operation
of reinforcement be experimentally manipulated to produce
specifiable changes in the dimension of dependent-
independent behavior?
Statement of the Problem
j
i
|Background: Change in Domiciliary
iAdministrative Function
! The West Los Angeles Domiciliary consists of a
|group of residence halls within the Veterans Administration
| complex of medical and neuropsychiatric hospitals. The
j Domiciliary was originally a retirement home for men who
jhad honorably served 20 years in any branch of the mili-
|tary. However, since World War II, a drastic change has
; taken place in the character and composition of the
| residential population. According to Carman (1968),
i retired career men now account for only 15 per cent of the
j
I 2,700 veterans in the Domiciliary. The balance of the
I
: members are the non-service-connected medically disabled,
the socioeconomic failures, alcoholics, and the psycholog
ically disturbed; over half of the population is presently i
I
i i
: composed of neuropsychiatric transfers in remission. j
i
This unofficial change in function from retirement j
: care to custodial and rehabilitative care is reflected in
the admission rate and age at admittance. Prior to World
i
j
; War II, a member was . admitted once, or perhaps twice, j
* !
; during his entire lifetime. Today, the number of |
■re-admissions has increased to three to five re-admissions,
per member, during any given three-year span. The age
range has dropped from an average of 45 years to a mean of
30. Thus a much younger and less stable population i
has evolved.
Neither the civilian culture nor the military
consider a 30-year-old male as retirement material. It is
; primarily because of this decrease in mean age that
|administrative and psychological emphasis has shifted to
i
I rehabilitation and return to the community.
I
'Behavioral Effects of
;Institutionalization
■ Underlying the return to the community rationale is
|
I a backlog of unofficial observation: after five years of
residency, Domiciliary members who are under retirement
i age tend to become "institutionalized.” However, data on
! the effect of prolonged stay in an institution upon the
I individual is not limited to unofficial observation. Many
of the effects of institutionalization, upon varying age
I groups, have been empirically determined.
Among others, Bakwin (1949), Bowlby (1951), and
Spitz and Wolf (1946) have studied emotional disturbance
!and physical retardation of infants in institutional
settings. Beres and Obers (1950) and Ainsworth (1962) have
j found a high rate of character disorders and psychoses in
institutionalized adolescents. Adult hospital patients,
regardless of their original nosological diagnoses, develop
an additional "institutional neurosis" (Barton, 1959);
Williams (1960) also emphasizes this auxiliary symptomatol
ogy as a function of prolonged institutionalization.
8
According to these empirical data, the effects of
i institutionalization are characterized by dependency,
i
!apathy, passivity, submissiveness, and a weak sense of
i
jpersonal identity. The effects appear to have a common
I factor, an underlying psychological transaction between the
I institution and the veteran. In this transaction the
veteran relinquishes effective control and self-direction
of behavior in exchange for the support of the institution.
;The dimension of dependence-independence is thus one aspect
: of the institutionalization effect.
Pragmatically, then, the reduction of dependency
has implications for the probability of success in return
ing veterans to the wider community, hopefully before
institutionalization has had its effect. The extent of
ithe problem involved in successfully implementing the
return-to-the-community rationale, however, becomes appar
ent when one examines the ratio of staff-to-veteran
population.
Professional Manpower Shortage — ;
I
The Domiciliary Psychology Staff consists of one
Chief, two research psychologists, two vocational psycholo
gists and three clinical psychologists. There are two
placement officers and a psychological technician who
handle job referrals and psychometric testing, respectively.!
; i
; Clerical help, either civil service or veterans on the
!
■Rehabilitation Incentive Therapy payroll, varies from
|3 to 5.
The ratio of professional psychological treatment
i
iavailable to the Domiciliary population thus becomes
1/245, or one staff member to every 245 veterans. However,
each staff member has only 20 hours per week available for
jtreatment, since one half of the week is devoted to
research, administration, conferences, etc. Assuming that
'the full 20 hours per week for each staff member is dis
tributed equally, this means that each psychologist must
:see one patient every five minutes in order to cover the
field.
In view of this professional manpower shortage, it
‘appears desirable to investigate the possibility of
maximizing the use of available treatment time. One method
is to utilize the group treatment mode. This mode lends
itself to the lowering of the staff-to-patient ratio by I
i
affording several patients the opportunity of interaction
with one therapist. In addition, there is the possibility
|
that such economy by the use of groups may be further j
maximized by the specific utilization of positive reinforce-;
ment to increase the probability of independent behaviors.
| S u m m a r y
| It is theorized that the operant conditioning
t
|model, as described by Skinner, may have implications for
|verbal conditioning of the dependence-independence dimen-
i
jsion. Further, it is hypothesized that the operation of
■ positive reinforcement may be a relevant variable in the
:reduction of dependent behavior.
Thus the design seeks to explore simultaneously
i
i experimental and applied aspects of the research problem.
;The experimental aspect explores the role of positive
reinforcement as a factor in reducing dependency and in
!increasing responses of independent behavior. The applied
I aspect explores increases of independence, or reduction of
dependence, in terms of the Domiciliary1s return to the
community rationale.
CHAPTER II
SURVEY OF THE LITERATURE
Dependent-Independent Dimension
Modern dynamic formulations no longer emphasize
dependency as a product of fixation at an instinctual stage j
of development. For Freud, fixation was a state of psycho
logical arrest at an infantile level; the dependent person
ality was conceptualized as one who had been fixated at the
oral-anal stages (Freud, 1949).
These stages, their fixations, and the resulting
behavior, according to psychoanalytic thought, are deter
mined by the unconscious and therefore inaccessible to
conscious self-inspection. Unless the techniques of
psychoanalysis are applied, behavior governed by an uncon
scious oral-anal fixation (i.e., dependent behavior) is I
i
inflexible and rigid. Of this behavior, Kubie (1950) j
t
asserts: j
|
. . . It never learns from experience. It cannot
be altered by argument of reason or persuasion or
exhortation or rewards or punishment, and not even
by its own successes and failures. . . . This
behavior can learn nothing. It cannot change or
develop or grow. It is enslaved . . . (p. 31).
! Psychology has moved away from such fixed determin-
;ism; the concept of dependency is based instead on the
i
! irrefutable fact that the human infant is born helpless and
| remains so for a period of time. This "biological depend
ency" (Maslow and Mittleman, 1941, p. 26) upon adults for
; physical survival is augmented by a psychological depend
ency upon significant others in his environment (Sullivan,
;1953). Unlike biological dependency, psychological depend-
; ency is largely influenced by learning and socio-cultural
expectations of behavior (White, 1959).
The combination of maturational and social
:pressures impose on the individual specific tasks to be
mastered at different life periods. For example, learning
to walk and talk are specific physical tasks of early
childhood; the child learns concurrently which behaviors
are socially approved or disapproved through a tacit system
of rewards and punishments (Mussen et ad., 1963). The
pre-adult period is thus a long transition in mastery of
physical, social, and emotional skills, from incompetence I
:to competence, and from the dependence of childhood to the I
i
independence of adulthood (Kagen and Moss, 1962). !
:Definitions of Dependency
The sources cited appear to agree that psycholog- j
ical adjustment or maladjustment depends on the individ- j
ual's degree of success or failure in making this dependence
to independence transition. Agreement on the definition of
dependency itself, however, is sparse; operational defini
tions are rare.
Sullivan (1947) equates dependency with a "sought
end-state" of security or the maintenance of security:
. . . The pursuit of security . . . [and] . . . the
full development of personality along the lines of
security is chiefly founded on the infant's discov
ery of his powerlessness to achieve certain desired
end states with the tools, the instrumentalities,
which are at his disposal.
. . . From the disappointments in the very early
stages of life outside the womb . . . comes the
beginning of this vast development of actions,
thoughts, foresights, and so on, which are calcu
lated to protect one from a feeling of insecurity
and helplessness in the situation which confronts !
one . . . (p. 10).
Maslow and Mittleman (1941) assert that a primary
characteristic of the dependent individual is perceptual
distortion; he consistently sees others as stronger, more
competent and decisive than he. Dependent tendencies are
therefore defined as a "need for emotional or authoritative
support . . . the dread of loneliness" (p. 392).
Bernadin and Jessor's (1957) definition of depend
ency centers on reliance on significant others for
approval, assistance, and "conformity to the opinions and
demands of others" (p. 64). Similarly, Snyder (1963)
defines dependent behavior in therapy sessions "as behavior
involving the need for more help than the average person
requires in meeting problem situations" (p. 2).
Warn (1958) maintains that dependency can be
i defined by the amount of help that an individual seeks:
j . . . Emotional dependency is that complex of
! processes, which operate to arouse subjective
! cravings to so relate oneself to others, that
effective control and direction of one’s behavior
is relinquished in exchange for restrictive
monopoly of the flow of love and affection from
others . . . (p. 15).
; Learning theorists Dollard and Miller (1950) postu-
i
i late dependency as a learned drive. First, defining
■ learned drive:
' . . . As the result of learning, previously neutral
cues gain the capacity to play the same functional
role in the learning and performance of new
responses as do primary drives, such as hunger and
thirst . . . (p. 78).
Dollard and Miller then continue:
. . . In the light of the fact that the required
social conditions for learning exist in the family,
it seems reasonable to advance the hypothesis that
the related human motives of sociability, depend
ence, need to receive and show affection, and desire
for approval from others are learned . . . (p. 92).
[Italics mine.]
Dependency as a learned drive is acquired, accord- j
i
ing to Sears et al. (1953), through the dyadic interaction i
between child and parent, and through: j
i
. . . repeated rewarding and at least occasional
punishing of an instrumental act . . . [to] . . .
produce a drive to perform that act and secure
the environmental results that normally result
from it . . . (p. 180).
; Navran (1951) distinguishes two classes of depend-
j
ency, primary and secondary. He defines primary dependence I
as the child's initial way of relating to parents or
parent-surrogates. Secondary dependence is defined as
"a dependence which is overly intense when the individual
becomes an adult" (p. 8). Secondary dependence is, by
definition, a maladaptive response mode.
Research on Dependency-
Independency Dimension
Kagan and Moss's (1962) 30-year longitudinal study
of 89 children, from their birth to early adulthood, is the
most comprehensive treatment of dependency to date. From a
comprehensive data bank of observations, interviews, and
test batteries they were able to analyze the stability of
selected personality variables through childhood and
adolescence. More important, they were able to predict the
relationship between a childhood behavioral dimension, such
as passivity-dependence, and the probability of its occur
rence in adulthood.
Dependency, measured primarily in terms of the
tendency to seek aid, affection, and reassurance from
adults, is significantly stable for girls from 0-3 to
adolescence (r = .64) but not so for boys (r = .04), who
shift sharply from dependence at 3-6 to the socially
defined male assertive behavior. Predictably, then,
dependent behavior in adulthood, measured by withdrawal
from stressful situations, concern with financial security,
and marked reliance on love object or parents, was signif
icantly higher for women (r = .55) but not for men j
16
I
(r = .05).
! Such data make it possible to postulate a learned
!secondary drive of dependence, as did Beller (1955) and
i Livson and Mussen (1957), although the two studies repre
sented systematically opposed theoretical viewpoints:
In a psychoanalytically oriented investigation of
I"ego control" and its relationship to overt dependency and
|aggression, Livson and Mussen postulated that the final
value of overt expression is a combined function of
;strength of drive and inhibitive strength of that drive.
They assumed equal inhibitive valences for aggressive,
dependent, and sexual behavior in the American middle-class
child.
Ego control, as measured by cosatiation and delay
of gratification indices, was equated with socialized
behavior; secondary drives would not therefore be expressed
overtly. Their hypothesis was not supported: no signifi- j
cant relationship was reported between overt dependency and
amount of ego control; a significant negative correlation, J
however, existed between overt aggression and ego control. |
Beller hypothesized the existence of two separate
secondary drives, dependence and independence. He tested
;the assumption that specific components of early parent-
child interactions contributed to secondary drive develop
ment . For a secondary dependence drive, the child's |
primary drives are first reduced by need satisfaction; |
17
since primary drive reduction is also tension reducing, the
parental agent becomes a generalized stimulus for dependent
behavior on the part of the child.
For a secondary independence drive to be developed,
the child's primary drives are not always reduced by the
adult. The child must overcome environmental obstacles and
thus develops goal-oriented behavior. It is implied that j
the absence of the parental agent becomes the stimulus for
independent behavior on the part of the child. Beller's
data analysis of ratings obtained on a nursery school
sample supported his assumption of specific components and
two separate drives; there was a moderately negative corre
lation between dependence and independence, not completely
inverse.
The role of dependency in physical illness and
disablement has been studied by several authors. By means
of pre-operative ratings, Bard (1955) was able to differen
tiate dependent from independent female patients scheduled j
for radical mastectomy. More significantly, he was able to
predict the post-operative emergence of "psychogenic j
invalidism”; dependent women thus had a slower rate of
recovery and a longer hospitalization.
Rothberg (1965) found that the relationship between
dependency and delayed recovery did not hold for male
patients following surgery. She hypothesized that depend- j
ence and anxiety would increase under the stress of
I " “ 18 “ ”
i
j
:hospitalization and surgery, causing disruptive post-
!operative bodily symptoms. She found no significant
relationship between dependency, anxiety, or recovery
I rates.
i Warn (1958) confirmed the hypothesis that tubercu-
i
jlars and epileptics would manifest more dependent behavior
jthan normals. Contrary to expectation, paraplegics were
!not significantly different in dependency from normals,
although the former constituted the most seriously handi-
i
'capped group in the study. He suggested that the "visibil
ity1 1 of the paraplegics was a major factor in their
dependency adjustment.
Green (1966), distinguishing between "instrumental
dependency" and "emotional dependency" in totally blind
males, hypothesized that blind adults were higher than
normals in both ID and ED as a function of length of blind-
!
ness. He found the relationship to hold for ID and length j
(r = .47) but that ED decreased as the length increased j
i
(r = -.35), indicating a form of adaptation learning. j
j
Gordon’s (1953) findings of a relationship between j
j
t
behavior in a social situation and behavior in a laboratory j
setting led him to conclude that the dimension of
dependence-independence was a unified one, related to the
continuum of maturity. Those subjects whose behavior was
ranked by their physicians and by themselves as being
socially dependent were also dependent in the laboratory
19
Rod and Frame test, as measured by the degree of dependence
|on external cues. The Ss designated independent socially
|
|exhibited comparable independence in the laboratory.
|Caruth (1959) failed to condition subjects defined as
dependent by Gordon's scale. Her hypothesis that dependent
subjects would rely more heavily on positive verbal
reinforcements as cues for response patterns in a Taffel-
type situation was not supported.
Spiegel and Litrownik (1968) explored perceptual
susceptibility of schizophrenics to group pressure under
;Asch-type conditions. They found conforming error to be
significantly higher for dependent, non-assertive schizo
phrenics than for independent, self-assertive schizo
phrenics. Hardison and Purcell (1959) induced stress
conditions for a digital task by negative verbal reinforce
ment. Their hypothesis that performance by independent
subjects would be less affected by negative reinforcement
was not supported. They did add, however, a dimension of
flexibility-constriction to the data on dependency. The
subjects in their study who were independent, as measured
by the deference score on the EPPS, were also more
;flexible. Flexible independent subjects tended to improve
performance under stress while constricted dependent
subjects did not.
Bernadin and Jessor (1957) also used negative
verbal reinforcement in the form of critical comment, for
i 20
la finger maze task. Dependent subjects showed a signifi
cantly higher performance deficit under these conditions
;than did independent subjects. Contrary to Spiegel's
|findings, they did not exhibit conformity in an Asch-type
condition.
Bernfield and Guthrie (1959) induced dependency
ihypnotically to explore behavior in a pseudotherapeutic
I situation. Their hypothesis that dependent patients would
;ask more questions, discuss fewer topics, and display less
self-confidence was supported. j
Zuckerman, Levitt and Lubin (1961) attempted to j
clarify the construct of dependency as an inferred process
which gives rise to measurable phenomena. Concurrent
validity of objective-projective tests purported to measure
1 dependency was assessed by comparing these tests against an
external criterion rating; construct validity was assessed
by factor analyzing correlations between all the tests in j
i
the battery and the external criterion. j
A broad factor of "General Dependency" emerged, as
i
well as specific subclasses, i.e., deference, succorance,
etc. The maj or dependency factor showed moderate to high
loadings on direct (objective) tests but negligible load
ings on the indirect (projective) tests. The authors
conclude from their extensive statistical analysis that the
I construct of dependency can be inferred and measured by j
existing psychometric tests but validity correlations tend j
i~'"' ......................... 21 ""
I
| to decrease as a function of their indirectness. The more
! obj ective, direct measures yield greater concurrent and
!construct validity.
Research Using the De Scale
Navran (1954) attempted to isolate and define
secondary dependence by means of a rationally derived
: Dependency Scale (De Scale), constructed by a factor
| analysis of existing diagnostic items in the Minnesota
j Multiphasic Personality Inventory (MMPI) . Research and
validation studies of the MMPI as an objective personality
test may be found in several readily available references
land will not be discussed here (Welsh and Dahlstrom, 1956;
; Dahlstrom and Welsh, 1960).
The De Scale consists of 57 MMPI items relevant to
dependency, exclusive of other nosological categories, and
with a specified scorable direction of response for each j
item (Appendix A). The scale has been found to success- |
!
fully discriminate between neuropsychiatric patients and j
normals, the former group scoring significantly higher in |
dependence (p = c.001). It also differentiated between two|
different socio-occupational groups, both groups under the
same medical diagnosis of ulcers: the first, a civilian
ulcer group which verbalized dependence; the second, a
naval enlisted ulcer group which verbally denied dependence!
i
(p = <.01; Navran, 1951). j
In combination with other indices, Warn (1958) used
the scale in his study of emotional dependency in the
physically disabled. Zuckerman, Levitt and Lubin (1961)
included it in their psychometric dependency assessment
battery, which was subjected to analysis for concurrent and
construct validity. In a later study of the relationship
of an acquiescence response set to dependency, Zuckerman
and Eisen (1962) utilized the measures from the previous
methodological investigation; they found that the De Scale
was not influenced by acquiescence set even though items
were answered "true" or "false."
Button (1956) reported a significant difference
between neuropsychiatric patients and alcoholics
(p = >.001). In his study neuropsychiatric patients scored
higher in De than either normals or alcoholics. The
finding that alcoholics scored high but not significantly
higher than normals lent support to the hypothesis that
dependency is a definite conflict area for alcoholics, "but
they are not particularly concerned about it" (p. 274).
Fordyce (1953) hypothesized that the higher the
pre-test De, the longer the patient would remain in psycho
therapy. He found higher De scores for "persistent"
patients than for those who voluntarily terminated
(P = >.05). Conversely, Nelson (1959) was able to predict
patient drop-outs from psychotherapy by using the De Scale
as part of a larger test battery.
23
I
! Jacobs et al. (1966) investigated the psychological
effects of prolonged bed rest on personality variables,
under simulated outerspace conditions. A significant rise
! in De scores (p = >.10) supported the hypothesis that
: enforced curtailment of activity and independence would
iaugment a class of regressive responses, i.e., dependency.
Orlando and Steele (1967) investigated differences
jbetween two systematically opposed treatment modes, psycho-
I analytic group therapy and reality group therapy. The
i
hypothesis that De scores would increase for the psychoana
lytic group, due to the transference phenomena, and would
decrease for the reality group was not statistically
supported.
McGuigan (1959) evaluated the psychological and
sociological effects upon American college students of a
.
year spent in Europe. Among other findings, De scores
increased for the experimental group and decreased for the
"at home" controls. The results suggested that the devel
opment of independence was relatively unhampered "at home" j
and that a tendency towards submissive social adjustment j
was a product of intercultural experience.
Summary
Dependency is generally defined as overt or covert
behavior directed towards obtaining the guidance, nurtur- j
ance and support of significant others. The writer !
i 24
I
I subscribes to Navran1s definition of primary and secondary
I dependence: primary dependence, as the original state of
| the human infant in its relationship to adults; secondary
I dependence, as a maladaptive response mode in adults.
i
It is submitted that secondary dependence can be
'defined in terms of the probability of its greater occur
rence as a response class in adulthood. It is to be
remembered that dependency is a hypothetical construct
;which is inferred to exist, although it is not fully
observable at the present time. There is some agreement
that the construct gives rise to measurable phenomena with
predictable consequences.
Verbal Conditioning and
Positive Reinforcement
The recognition of the key role played by reinforce
ment in the operant conditioning paradigm has raised a
related and equally important question. There is a wide
range of behavior which is not, apparently, directly
j
reinforced. If reinforcement is a necessary and sufficient!
j
condition for learning, how is the learning of these other
behaviors accounted for?
According to Hill (1963), this type of learning is
j
a function of an important aspect of reinforcement, or j
reinforcers: they can themselves be conditioned. "If a j
|
stimulus occurs repeatedly with a positive reinforcer, it !
I 25
i
I tends itself to acquire the capacity to reinforce behavior"
|
;(p. 62). That is, a stimulus that is unrelated to the
i
;original reinforcing stimulus or the operant response is
jconditioned, by contiguity; it acquires reinforcing value
as an independent and non-specific conditioner of new and
unrelated responses (Keller and Schoenfeld, 1950). j
These secondary reinforcers allow for the emergence .
;of generalized reinforcers. Generalized reinforcers are of
;systematic importance in operant behavior, for they account
:for the conditioning of responses several levels away from
the original reinforcing stimulus. Thus, even in what
appear to be independent events, learning occurs as a
function of generalized reinforcers. According to Skinner
I(1953), reinforcement is still a necessary and sufficient
condition for learning.
Generalized reinforcement is a major variable in j
verbal behavior. Skinner (1957) has pointed out that j
i
;language involves a reciprocal relationship between speaker
and listener. They mutually reinforce each other, often
directly on a verbal level; often indirectly, through
generalized reinforcers indicating comprehension and
approval of the other's words. Skinner's acceptance of
verbal behavior as a complex system of operant response,
capable of being emitted and reinforced, was the starting
point of an impressive body of literature devoted to the ;
; i
modification of verbal behavior within an operant j
26
conditioning paradigm.
Krasner (1965a) defines verbal conditioning as:
"The systematic application of social reinforcements / to
influence the probability of another person emitting a
specifiable verbal behavior" (p. 213). As an experimental
phenomenon, verbal conditioning has a short history of
little more than a decade; yet a review of the literature
yielded over 300 published studies. An example of the
unprecedented expansion in this area is Krasner1s brief
introduction to Verbal Conditioning and Psychotherapy
(1965b). Singling out only his own and his colleagues'
work, he cites the following investigations of relevant
variables:
Subject personality variables: Ullmann, Krasner,
and Weiss, 1963.
Examiner differences: Krasner, Ullmann, Weiss,
and Collins, 1961.
|
The relationship between awareness and performance: j
Krasner, Weiss, and Ullmann, 1961.
The effects of instructional sets: Ekman, Krasner,!
and Ullmann, 1963. j
Generalization effects: Ullmann, Krasner, and
Collins, 1961.
Atmosphere effects: Weiss, Krasner, and Ullmann,
1960, 1963.
Pleasantness of emotional words: Ullmann, Krasner,
and Gelfand, 1963.
!
Physiological concomitants of changes in emissions j
of emotional words: Ullmann, Krasner, and !
Sherman, 1963. I
27
j
Changes in response class reinforced: Krasner
and Ullmann, 1958.
! Placebo effects: Gelfand, Ullmann, and Krasner,
I 1963.
!
' Relationship to other behavior influence situa
tions such as hypnosis (Weiss, Ullmann, and
Krasner, 1960; Bentler, O'Hara, and Krasner,
1963.
Discussions of the broader field of verbal condi-
itioning itself are contained in these and other readily
iavailable sources (Greenspoon, 1962; Williams, 1964;
!Krasner, 1958). For the purpose of this dissertation it is
proposed to survey only the literature bearing upon the
contingencies of positive reinforcing stimuli in the verbal
i conditioning paradigm. I
;Research on Positive Reinforce
ment Contingencies
Verplanck's (1957) working definition of reinforce-
. j
ment will be the model for the survey: ". . . Stimuli can j
only be classified as reinforcing after it has been demon- j
i
strated that their use produces modifications in behavior |
! j
under the stated conditions ..." (p. 25).
i
Greenspoon's (1951) initial research of verbal
!conditioning within a Skinnerian paradigm used four
different contingencies: the now famous positive reinforc-
i
ing verbal stimulus "mmm-hmm"; a negative verbal reinforce-|
ment, "huh-uh"; a visual stimulus of a red light flash and
an auditory stimulus of a 190-cycle tone. The response j
I class to be reinforced was that of plural noun emission.
; This response class significantly increased in frequency
ias a function of the positive reinforcing stimuli of verbal
japproval, the light, and the tone; it decreased as a func-
; tion of the negative reinforcing stimulus, "huh-uh."
The positive reinforcing stimulus of "mmm-hmm"
i has been used singly or in combination with other verbal
approval phrases to increase the probability of occurrence
; of a given response class: Salzinger and Pisoni (1958)
: effectively increased the frequency of self-referred affect
statements by schizophrenic patients in a structured clin
ical interview setting. Following a 10-minute operant
period, emission of affect was reinforced by 'Wm-humm"
and "I see"; the number of responses during extinction was j
i
a straight-line function of the number of reinforcements. j
Plural nouns and adverbs were reinforced by
|
"mmm-hmm" or "good" by Wilson and Verplanck (1956) ; an j
i
additional experimental group received reinforcement by
the E's pencil-and-paper recording of the emitted response
classes. The rate of response in both reinforced classes
increased, but not significantly so in comparison to the
; controls. However, in Experiment II, reported in the same
study, semantic categories of "travel-words" or "living-
■ J
I
!thing-words" were increased significantly by the reinforce
ment of "mmm-hmm" or "good" (p = C.005).
j 29
!
In the elicitation of opinion statements by tele
phone, Hildum and Brown (1956) found that "good" was
;effective in influencing "pro-responses"; "umm-hmm" was
not. Mandler and Kaplan (1956) obtained mixed results as
: a function of whether or not Ss perceived "mmm-hmm" as a
positive or negative reinforcement. If "mmm-hmm" was
perceived as positive, the response class of plural nouns
increased significantly; if perceived as negative, rate of
iresponse decreased. Caruth (1959) hypothesized that
;"umm-hum" would be used as a cue for a conditioned response
pattern more often by dependent individuals in a Taffel
:situation than by independent individuals. Her hypothesis
was not sustained.
The reinforcing effect of "good" in the acquisition
of pronouns of self-reference (I, We) was also demonstrated
when the verbal reinforcer was shifted to another class of
responses (He, They) by Cohen et al. (1954). The authors
point out that the reinforcer "good" was not paired experi
mentally with a primary reinforcer and therefore cannot be
considered operationally as a true secondary reinforcer.
Following McGeoch and Irion (1952) they agree, however,
that "good" may be considered as part of a class of general
social approval symbols and hence, a secondary reinforcing
stimulus.
The use of "good" as a social reinforcer capable of j
i
i
satiating and/or depriving the "social drives" of children j
i 30
j
! was investigated by Gewirtz and Baer (1958). Their results
| showed a higher rate of learning of a non-verbal response
I
! under deprivation (social contact) than under satiation.
| Klein (1954) reported significant results, conditioning and
extinguishing a personal pronoun response class in neuro
psychiatric patients, with "good" as the reinforcing
stimulus.
Nuthman (1959)7 ' a'Ctlinjited to establish the differen-
| tial reinforcement value of "good" and a visual (flashing
;light) stimulus. The verbal stimulus caused an increase in
; acceptance-of-self references on a true-false personality
: test; the light blink was not conducive to conditioning,
iWickes (1956) used a combination of verbal and non-verbal
reinforcing stimuli to test the hypothesis that human
movement responses can be elicited from a Rorschach-type
protocol, and significantly modified by reinforcement in
the "standardized" testing situation was confirmed. The
verbal stimuli used in sequence were "good," "fine," and
"all right"; the non-verbal, head-nodding, smiling, and
leaning forward. All other verbal and non-verbal behavior
;was restrained. Both classes of stimuli increased the
I frequency of M responses: the verbal condition group was
significantly higher than the controls, beyond the .025
level; the non-verbal condition, beyond the .005 level.
Verplanck (1955) chose as a general class of
reinforcing stimuli statements of agreement or paraphrase
i to condition the emission of opinions. The agreement
I reinforcer was defined by "I agree,1 1 "You're right," and
["That^s so." Paraphrase reinforcement was a repetition of
the Ss1 statement in other relevant words. Agreement with
the Ss increased the rate of opinion; paraphrasing was
effective but not equivalent reinforcement. Disagreement
ior silence during extinction reduced the rate of response.
j (
:Differences between groups were significant (p = .01).
Verplanck concluded: "... The verbal behavior of
a speaker, apparently without regard to its content or
setting, is under the control, not only of the speaker him-
j
self, but also of the person with whom he is conversing"
! (p. 675) .
In a later study, Verplanck (1956) maintained that
"any event of short duration whose incidence in time is
■under the control of E may be used as a reinforcing stimu
lus if Ss is instructed properly" (p. 73). Consequently he
reported on a variety of positive reinforcement procedures j
! ( i
for conditioning human behavior. The reinforcing stimulus
may be conventional ("good," buzzer, or lights); the j
reinforced response, unconventional (nose touching, chin j
tapping, neck-tie straightening).
Studies of positive reinforcers other than verbal
stimuli, however, are sparse. There are isolated reports
of reinforcement by cartoons (Hovorka, 1950), lights
(Sidowski, 1954), bell tones (McNair, 1957), music j
; 32
I
!(Jeffrey, 1955), and toys (Estes, 1945).
I Summary
i
' Verbal conditioning is the systematic application
j of a reinforcing stimulus to specifiable operant response
classes. Conditioning is said to occur when the verbal
response probabilities are changed by experimental stimuli,
which are contingent upon the selected response.
It is submitted that by the operant conditioning
iprinciples of generalization and generalized reinforcement,
there is a probability that changes in verbal behavior have
consequences for changes in other classes of behavior.
Experimental studies of the dependence-independence
dimension have primarily focused on problems of measure
ment. Subjects have been separated by various indices into
dependent or independent behavior response classes (Gordon, j
|
1953; Navran, 1954). The bulk of the review of the litera- j
ture concerns predictions which have been made for the j
i
differential behavior of dependent or independent subjects j
under various experimental conditions. In two cases j
(Jacobs et al., 1966; Bernfield and Guthrie, 1959) depend
ency has been actually induced or augmented.
A survey of the literature has not revealed a
single study which explores the reduction of dependency,
by the operation of positive verbal reinforcement. j
CHAPTER III
I METHOD AND DESIGN
Purpose
It was the purpose of this dissertation to explore
experimentally the increase or decrease in the dependent
Ivariable, dependency in domiciled veterans, by the manipu
lation of the independent variable of verbal conditioning,
using positive reinforcement.
I Definitions
■ Operational Definition:
Dependent Variable
Dependency is defined operationally as an increase j
in the total number of directional responses (true-false)
measured by the Dependency Scale (De) of the Minnesota
Multiphasic Personality Inventory (MMPI). Conversely,
independence is defined as a decrease in these response
scores.
34
Operational Definition:
Independent Variable
Verbal conditioning: Positive reinforcement is
defined operationally as the presentation of the verbal
reinforcing stimuli "good1 1 or "umm-hmm" to the occurrence
of an operant; e.g., response class of independent state
ments emitted by the subject. A statement of independence
is defined as follows: any man's statement of self
reference indicating an action on his own behalf, in any
area, will be defined as a statement of independence. This
statement may be in any tense--past, present or future.
Factorial Design
Following the Analysis of Variance Fixed Effects
Model I (Hayes, 1963), this study is planned as a 1 X 2
factorial design. The factor is reinforcement, with two
levels, zero reinforcement and partial reinforcement.
TABLE 1
FACTORIAL DESIGN
Control
Group
Experimental
Group
Zero
reinforcement
Partial
reinforcement
(I, III) (II, IV, V, VI)
35
Hypotheses
! The following hypotheses were based on the above
!design:
I Hypothesis I
It was hypothesized that verbal conditioning, using
positive reinforcement, would decrease dependency in
domiciled veterans, where dependency is defined operation
ally by De scores. As the response class of independent
■ statements is reinforced verbally, so will De scores
decrease generally.
Hypothesis I was tested as the treatment effect.
If reinforcement is a relevant independent variable in the
reduction of De, there will be a difference between groups;
:e.g., the experimental groups receiving partial reinforce
ment, and control groups, zero reinforcement.
Theoretically, the effect between groups is
obtained as a function of the presence or absence of
reinforcement. Reinforcement is present for Groups II, IV,
V, and VI, so conditioning should occur; reinforcement is
absent in Groups I and III, so no conditioning should
occur.
Specifically, it was hypothesized that the partial-
reinforcement Groups II, IV, V, and VI would show a
significant decrease in dependency. It was hypothesized
that the zero-reinforcement Groups I and III would show no
i 36
I
I significant change in De.
|Hypothesis II
i
It was hypothesized that verbal conditioning would
!decrease De scores as a function of the subject's age. The
younger the subject the higher the probability that the De
scores would decrease generally; the older the subject, the
less decrease, or no change. This hypothesis was limited
to the experimental groups.
Hypothesis III
It was hypothesized that verbal conditioning would j
; i
decrease De scores as a function of length of institution
alization. The shorter the duration of Domiciliary stay,
the higher the probability that De scores would decrease
generally; the longer the duration, the less decrease or no
i
change. This hypothesis was limited to the experimental j
groups. i
|
Hypothesis IV
It was hypothesized that verbal conditioning would
have a differential effect upon subjects at extreme ends of
the initial De score distribution. The lower the pre-test
| |
score, the higher the probability that there would be no^ j
change in De scores. Conversely, the higher the pre-test
score, the higher the probability that there would be a j
i
decrease in De scores generally. This hypothesis was j
limited to the experimental groups. j
| 37
j
Method
|Subjects
There were 24 subjects in the study assigned to two
control and four experimental groups of four subjects each.
;They were selected from the male patient population of the
Domiciliary. The criteria for subject selection were
formulated so as not to interfere with group participation:
1. Members officially diagnosed as epileptic,
having a chronic brain syndrome, or being physically
deficient in speech or hearing were excluded.
2. Selection included, however, those whose
:medical diagnoses were not expected to interfere with
participation, such as gastrectomy, arthritis, chronic
alcoholism, or service-connected disablement.
3. Members officially diagnosed as psychotic were
excluded. Diagnoses of chronic schizophrenic reaction or
schizophrenia in remission were not accepted.
4. Selection included, however, all other neuro
psychiatric diagnoses which were not expected to interfere
with participation, such as personality disorder, anxiety
I reaction, and depression.
5. Ambulatory members only, with an age range of
40 to 50 years of age.
6. First admissions only, with length of institu
tionalization after initial admission between one to five
■ ' years.. ...... . . . . _ ..... .... ................... .
38
a. The first-admission requirement controlled
for an unofficial chronicity; e.g., there
is a subset of veterans admitted and
discharged several times within one year.
b. The minimum requirement of one year
controlled for other extraneous factors
involved in early discharge from the
Domiciliary, such as spontaneous remission.
c. The maximum requirement of five years
controlled for official chronicity.
7. All Ss were on minimal drug medication, since
i
heavy sedation might be expected to interfere with
participation.
Pre-test Measures
In preparation for final selection, subjects were
pre-tested on the De scale embedded in the 556 items of j
j
the MMPI (Appendix A). General screening consisted of j
i
rejection of all subjects with a standard t-score of 90 or j
more in those areas which might be expected to interfere
i
with group participation. The areas of rejection were I
i
represented by the following MMPI scales: j
i
I
; j
Nosological Diagnostic i
Scale No. Category
4 Psychopathic deviate
6 Paranoia
8 Schizophrenic
9 Manic-depressive
39
!
A pre-test rating on the Supervisor Performance
Evaluation (SPE) was used as an external criterion for
I concurrent validity. The SPE, currently in use at the
iDomiciliary (Appendix E), consists of a set of 10 factors
relevant to patients' performance on assigned job-details
within the Domiciliary. Such factors as initiative,respon
sibility, etc., are rated on a scale of 1 (lowest) to 5
(highest). The total possible score is 50; the higher the
total rating, the more satisfactory the man's performance
• as viewed by his supervisor. Supervisors rated both
experimental and control subjects without knowledge of the
groups to which subjects were assigned.
Procedure
Thirty-five subjects meeting the selection criteria
and pre-test screening were "butterflied" for an individual
interview with the E. A butterfly, in veteran idiomatic j
language, is a mandatory order to appear before a staff I
j
member. Butterflies carry the highest priority; failure to j
j
keep such appointments may lead to disciplinary charges j
varying in severity from restriction of the member's j
i
I
leisure activities to his dishonorable discharge from the !
I
Domiciliary. |
In order to control for bias and set, neither !
j
subjects nor supervisors were informed of the research j
|
goal. The manifest purpose presented to the members j
40
! through the personal interview was oriented, therefore,
towards the exploration of problems in the setting of the
Domiciliary. They were assured that any problem which they
brought to the group meetings would be given personal
attention by the Chief of the Psychology Service.
Prior to the experiment a detailed search of the
~ records yielded an N of 50 men who met all of the selection
!criteria. Of these 50, five were hospitalized and unavail
able. All remaining 45 men were butterflied to appear for
:pre-testing on the MMPI. At this point, irregularities of
individual responses to the butterflies were noted. Not
j
everyone responded to the first mandatory butterfly. Those
who did not were re-butterflied until they appeared or the
arbitrary maximum of 6 butterflies was reached.
Of the 45 available subjects, 10 never appeared.
The remaining 35 subjects evolved into the following major |
j
groups: !
Group I took the pre-test (N=6).
Group II took both the pre-test and the post-test ;
(N = 5) j
Group III took the pre-test, post-test, and eventu-j
ally became experimental and control groups i
(N = 24).
Each man in Groups I, II, and III who took the
pre-test was butterflied for an oral interview. Of these
35 subjects, 29 appeared and agreed to participate in the i
!
exploration. These subjects were assigned randomly to one J
41
of the control groups of the four experimental groups.
i
After the first meeting, however, loss of subjects reduced
the two control groups to a total of N=8 and the four
! experimental groups to a total of N=16. There were equal
N's in all groups. The groups were also randomly desig-
j nated as follows:
Control Groups I, III
Experimental Groups II, IV, V, VI j
Total treatment time for all groups was held
| constant. Each group met for a total of 16 hours, one hour
a day, four days a week, for four consecutive weeks.
Permutations controlling for time of day, day of week, and
'fatigue effects are shown in Table 2.
The experimental groups II, IV, V, and VI were
1 t
! j
verbally reinforced for statements of independence by j
!
the E. "Good" and "umm-hmm" were used as positive j
i
; reinforcement to the occurrence of the independence |
i
operant. The control groups I and III were not verbally
reinforced for statements of independence; however, to
control for level of reinforcement across groups, "good"
i and "umm-hmm" were used at random in the control condition
;at an approximate rate of five reinforcements per hour to
any response class other than statements of independence.
All groups were structured by presenting topics for j
discussion chosen from Spiegel's 5-point Guttman Scale
measuring dependency and self-assertion (Appendix B).
TABLE 2
ROTATING SCHEDULE OF GROUP MEETINGS
(Randomly derived; four meetings weekly, groups
designated MA" "B" "C" "D" "E" and "F".)
Week 1 Week 2
Mon Tue Wed Thu Mon Tue Wed Thu
1st session D E B A C F E D
2nd session A C F B E D F C
3rd session B A E F A C D B
4th session E F C D B A C F
5 th session F B D C F B A E
6 th session C D A E D E B A
Mon
Week 3
Tue Wed Thu Mon
Week 4
Tue Wed Thu
1st session B C A F A D E B
2nd session F E D B B A F D
3rd session E A B C F C D E
4th session D F C A D B C A
5 th session C B E D E F B C
6 th session A D F E C E A F
43
I
:Reinforcement of verbal behavior resulting from the topics
was present or absent according to the appropriate experi
mental or control condition. All sessions were tape
recorded, with the subjects' permission, to provide verbal
i conditioning data on the frequency rate of statements of
independence.
:Post-test Measures
To evaluate changes in dependency as a function of
the independent variable manipulated, the same psychometric
index used in pre-testing was used for post-testing; e.g.,
the MMPI De Scale. To achieve concurrent validity, scores
obtained on the psychometric instrument were compared with
an independent set of scores from the external criterion,
Supervisors’ Performance Evaluation (SPE).
CHAPTER IV
RESULTS
Characteristics of the Experimental
and Control Groups
The 24 subjects in the sample were originally
divided into six groups of four, two being control groups
and four experimental. For the purpose of data analysis
it was necessary statistically to treat the two control
groups as one group (N = 8) and the four experimental
groups as one (N = 16). From a design point of view, this
regrouping was justified since the E's behavior was held
constant for all groups according to the experimental or
control condition.
The selection criteria included an age range from
40 to 50 years and length of institutionalization from one
to five years. Length of institutionalization was compar
able for both groups, with a mean duration of two years.
Since the two groups differed from each other in duration
by only a tenth of a decimal, no t-test was computed.
Subjects in the control group were slightly younger
(mean age =45) than the experimentals (mean age = 47.4).
44
i 45
iThis difference, however, was not statistically significant
! at the .05 level (t = 1.27). The pre-test De scores again
I showed both groups to be comparable, the t-test between
!means being insignificant (t = 0.245).
Since the two groups were comparable in age, length
of institutionalization and initial level of dependency,
the null hypothesis that the two groups were drawn from the
same population was accepted.
TABLE 3
SUMMARY OF MEANS FOR AGE, LENGTH OF
INSTITUTIONALIZATION AND t-VALUES
Group*
Mean
Age Range t-obtained
t-required
at .05
E
N=16
47.1 8
1.27 4.30
C
N=8
45.8 10
Group
Mean
Duration Range t-obtained
t-required
at . 05
E
N=16
C
N=8
2.0
2.1
4
3
Not necessary to compute
Group
Mean Pre
De Scores Range t-obtained
t-required
at .05
E
N=16
25.75 33
0.245 2.074
C
N=8
24.5 34
*E = Experimental; C = Control.
46
Hypothesis I
The hypothesis that verbal conditioning using
positive reinforcement would decrease dependency in
domiciled veterans, where dependency is defined operation
ally by De scores, was accepted.
A comparison was made of pre- and post-test differ
ence scores on the De scale between the experimental and
control groups. As a result of verbal conditioning, the
experimental group would be expected to decrease in De
scores; the controls to increase or show no change. The
difference between groups was significant. A one-way
analysis of variance on the 1 X 2 factorial design
(Table 4) yielded an F-ratio of 6.93, one-tailed test
(p = .025) .
TABLE 4
SOURCE TABLE FOR ANALYSIS OF VARIANCE,
PRE-POST DEPENDENCY SCORES
Source SS df MS F
Treatments 165.02 1 165.02 6.93*
(between 1.....
groups)
Error 520.81 22 23.67
(within
groups)
TOTALS 685.83 23
*Significant at the .025 level.
47
A more detailed analysis of the data also supports
Hypothesis I (Table 5). The experimental group, which
received partial positive reinforcement for statements of
independence, decreased in dependency from pre- to post
test. The total reduction in the sum of De scores was
-55, or a mean change of -3.4. The control group, in the
absence of reinforcement, increased in dependency from
pre- to post-test. The total increase in the sum of De
scores was +17 points, or a mean change of +2.1. An
analysis of each subject’s score, pre-and post-De, is given
in Appendix D.
TABLE 5
SUMMARY OF DEPENDENCY SCALE SCORE CHANGES,
PRE- AND POST-TESTING
Sum of Sum of
Raw Scores Raw Scores Net Mean
Group N Pre-test Post-test Change Change
Experimental 16 422 367 -55 -3.4
Control 8 196 213 +17 +2.1
The hypothesis that the independent variable of
verbal conditioning would reduce dependency is further
:substantiated by a 25 per cent sampling of the tape
recorded data. For each group a tape recording of an
hourly meeting was chosen randomly from the four available
for each week and transcribed. A frequency count of
I statements of independence evoked or spontaneously emitted
Iwas tabulated for each subject and each group. The
|
|frequency analysis of verbal behavior within the group
|setting was used as an index of public verbal behavior.
It follows logically from the hypothesis that the
: rate of emission of independent statements should vary as
a function of the presence or absence of reinforcement.
In the presence of reinforcement, rate of emission would be
expected to increase, while in its absence, to decrease or
!show no change. The data obtained support this assumption
(Table 6). Subjects in the experimental group show a
marked increase from Week 1 to Week 4. The control group
reduces to zero. The difference in trend between the two
groups is shown graphically in Figure 1.
In order to determine whether the four-week experi
mental span was of optimum efficiency in the reduction of
|dependency, the public verbal behavior (frequency of
response) was dichotomized into two two-week periods. The
obtained frequencies for both groups were compared to the
mean hypothetical expected frequencies for Weeks 1 and 2.
:A separate obtained-to-expected comparison was made for
Weeks 3 and 4. Finally, Weeks 1 and 2 were compared to
Weeks 3 and 4.
A Chi Square 2 X 2 contingency table was computed
for each condition. For Weeks 1 and 2 combined, the
difference between experimental and control groups in
TABLE 6
I SUMMARY TABLE: FREQUENCY OF STATEMENTS OF INDEPENDENCE
BY SUBJECTS OVER TIME
Ss
F r e q u e n c y
Total Week 1 Week 2 Week 3 Week 4
5 0 1 0 0 1
E 6 0 2 3 0 5
Group 2 7 1 1 1 5 8
8 0 0 1 3 4
t=tb-
13 1 1 0 1 3
E 14 1 1 4 2 8
Group 4 15 2 3 2 9 16
16 0 0 4 8 12
t=3'9"
17 3 3 2 1 9
E 18 0 0 0 0 0
Group 5 19 0 0 2 3 5
20 0 0 2 2 4
T=1S
21 1 0 1 3 5
E 22 1 1 0 0 2
Group 6 23 0 1 0 2 3
24 0 0 0 1 1
N=16
10 1'4 . 22 40 T=ll
E=86
1 0 3 2 0 5
C 2 1 0 0 0 1
Group 1 3 1 2 0 0 3
4 0 0 0 0 0
T=9
9 1 0 0 0 1
C 10 2 1 2 0 5
Group 3 11 0 0 0 0 0
12 1 1 0 0 2
N=8 6 7 4 0 T=H
C=17
wm.
'frequency of response was not significant (p = .20).
;Verbal conditioning increased emission of independent
| statements at the end of two weeks, but was not rigorously
t
significant (Table 7). For Weeks 3 and 4 combined the
difference between groups was significant beyond the .001
'level. Verbal conditioning affected rate of emission more
strongly at the end of Weeks 3 and 4 than at the end of
Weeks 1 and 2 (Table 8). When comparing Weeks 1 and 2
!combined with Weeks 3 and 4 combined the difference in
emission rates between the two groups was well beyond the
!.001 level (Table 9).
Rate of reinforcement was comparable for both
groups. The experimental group received positive
:reinforcement for statements of independence at a mean rate
of five per hour. The controls received random positive
reinforcement to any response class other than statements
i
of independence at a mean rate of five per hour. A t-test j
i
between means was not significant (t = .062).
The E and an independent rater, separately,
analyzed the transcribed data to determine the frequency of
emission of statements of independence. A reliability
coefficient of .924 was obtained for the judgments of the
;E and the independent rater on statements of independence.
The index of error was .020. The transcribed reinforcements
were omitted from the rater's copy, to insure that the j
I
rater's evaluation of statements of independence would not j
CHI SQUARE 2 X 2 CONTINGENCY TABLES:
FREQUENCY OF RESPONSE OVER
EXPERIMENTAL TIME
TABLE 7
WEEKS 1 AND 2
Obtained Expected
E
(21.25)
24
(21.25)
18.5
42.5
C
(15.75)
13
(15.75)
18.5
31.5
37 37 N=7 4
Chi Square = 1.664*
df = 1
Alpha at .20 = 1.642
*Significant at the .20 level.
TABLE 8
WEEKS 3 AND 4
Obtained Expected
E
(47.5)
62
(47.5)
33
95
C
(18.5)
4
(18.5)
33
37
66 66 N=132
Chi Square = 31.572*
df = 1
Alpha at .001 = 10.83
^Significant at the .001 level.
TABLE 9
EXPERIMENTAL AND CONTROL GROUPS
E C
Weeks
1 & 2
24
(30.89)
13
(6.107)
37
Weeks
3 & 4
62
(55.1)
4
(10.89)
66
86 17
N=Tt)3
Chi Square = 14.53
df = 1
Yates Corrected = 12.51*
Alpha at .001 = 10.83
*Significant at the .001 level.
i 53
I
■be biased by noting the Efs periodic reinforcements. As a
I further precaution, other statements made by the E were
jomitted at random. Instructions to the rater may be found
i
:in Appendix C, as well as a sample of the scored data.
Evidence that changes in dependency are reflected
in other areas of behavior is gained from a statistical
analysis of the pre-post supervisors' ratings of subjects
j
ion the Supervisors' Performance Evaluation. When compared
i on pre-test SPE scores, both groups were comparable.
\A t-test between means was insignificant (t = 0.39). The
;null hypothesis was accepted that the experimental and
control groups were drawn from the same population.
It follows from Hypothesis I that changes in
I dependency through the presence or absence of positive
!reinforcement should be reflected in other areas of behav
ior. A comparison was made of pre- and post-test differ
ence scores on the Supervisors' Performance Evaluation
(SPE), the concurrent validity index chosen. In theory,
those subjects exposed to the independent variable should
show increases in work performance outside of the experi
mental sessions, while the control group should decrease or
’show no change.
The difference between groups was highly signifi
cant at more than the .001 level. A Mann-Whitney U-test
was computed (df = 8,16) with a U of 44, and a U' of 84.
The experimental group greatly increased in work
„ 54
performance from pre-_to post-test and the control group
did not.
A summary analysis of the data throws into clearer
focus the phenomenon of behavioral change reflected in work
performance. A summary of mean scores and mean changes for
both groups, from pre- to post-test, on the SPE is shown in
Table 10.
TABLE 10
SUMMARY OF RATING CHANGES: PRE-POST SPE
Su pe rvi so rs (SPE)
N Mean Pre Mean Post Difference Mean Change
E
N=16
C
N-8
30
31
36.8
33.6
+105
+21
+6.5
+2.6
There is a clear line of demarcation between the
two groups as reflected in these mean changes. The experi
mental group (Table 5) showed an extremely high increase in
work performance characteristics (Table 10). By contrast,
the controls (Table 5) increased only very slightly in work
performance (Table 10). The obtained difference scores for
the SPE support the behavioral change predicted by
Hypothesis I.
The SPE as a concurrent validity index is, in
55
;addition, a strong one. A Chi Square was computed on the
I pre-post difference scores for both the De scale and the
I
t
ISPE. Decreases in dependency were accompanied by increases
I in work performance. The Chi Square, corrected for con
tinuity according to McNemar (1949), was 7.585 in a 2 X 2
contingency table. It was significant between the .01 and
;the .001 levels of confidence. The SPE appears, therefore,
i to be a valid concurrent validity index of behavioral
jchanges in dependence, as it is highly correlated with the
i
;De scale. Such data further support Hypothesis I (Table
11) .
TABLE 11
PRE-POST DIFFERENCE SCORES: DEPENDENCY SCALE
AND SUPERVISORS1 PERFORMANCE EVALUATION
C D
r— I
C t f
O
VI
C D
p 0 +
SPE
0 +
14
7
1
2
N = 24
df = 1
Chi Square = 7.58*
*Significant at <.01,
i 56
i
i Hypothesis II
I It was hypothesized that verbal conditioning would
j decrease De scores as a function of the experimental
: subjects' age. Specifically, it was predicted that the
; younger the subject, the higher the probability that De
scores would decrease after the subjects' exposure to the
experimental condition. Also it was predicted that the
i older the subject, the less decrease would be expected.
The selection criteria included an age range of 40
i to 50 years. "Younger" was arbitrarily designated to refer
to those men below the median age of 45.0 years; "older"
subjects were those above the median of 45.0 years to 50
years of age.
The subjects actually obtained, however, skewed to
; the upper limit of the criteria range with a mean of 47.4.
Thus, more than 66-2/3 of the experimental subjects were
above the mean of 47.4 at the time of the experimental
condition.
A Chi Square 2 X 2 contingency table was computed
to determine the relationship between increases and/or
decreases in dependency and the subjects' age. Change in
dependency was represented by the De difference scores,
positive or negative, from pre- and post-test data. The
age variable was dichotomized into category 1, above median
age of 45.0, and category 2, below median age of 45.0.
57
The results were not significant. Corrected for continuity
the Chi Square was .250.
TABLE 12
DEPENDENCY DIFFERENCE SCORES
AND MEDIAN AGE
+0
-
Above
Median
Age
(45.0)
3 10
Below
Median 2 1
Age
N = 16
df = 1
Chi Square = .250
|
The specific prediction that the younger the sub- j
i
ject the higher the probability of decrease in dependency j
|
was not sustained. Hypothesis II was not supported. !
i
i
Hypothesis III j
It was hypothesized that verbal conditioning would
decrease De scores as a function of institutionalization.
It was predicted specifically that De scores would decrease I
i
for the experimental group in proportion to the length of I
the subject’s Domiciliary residence: the shorter the |
duration, the more decrease; the longer the duration, the |
58
less decrease.
The selection criteria included a minimum institu
tionalization of one year and a maximum of five. The
minimum requirement controlled for extraneous factors
involved in early discharge from the Domiciliary, such as
spontaneous remission, which would confound the independent
variable of verbal conditioning. The maximum requirement
of five years controlled for official chronicity or the
institutionalization effect.
"Short" duration of institutionalization subjects
were arbitrarily designated as those below the median of
3.0 years; "long" duration subjects were those above the
median of 3.0 to 5.0 years of institutionalization. The
subjects actually obtained, however, skewed to the lower
limit of the selection criteria range, with a mean of 33
months. Although one of the 16 experimental subjects
matched the mean exactly, there were one and one-half times
as many subjects below the mean as there were above it.
Thus the actual sample consisted predominantly of men with
short duration of institutionalization.
To determine the relationship between decreases in
dependency and length of institutionalization a 2 X 2
contingency table was computed. Change in dependency was
again represented by De difference scores, positive or
negative, from pre- and post-test data. The length of
institutionalization variable was dichotomized into
I category 1, above mean length of 33 months and category 2,
;below mean length (Table 13). Yates' correction for
I continuity was included in the computation.
i
i TABLE 13
DEPENDENCY DIFFERENCE SCORES AND MEAN STAY
+0 -
Above mean
stay of
33 months
1 3
Below mean
stay
3 9
N = 16
df = 1
1 Chi Square = 8.10*
*Significant between the .01 and .001 level.
The difference between groups of short duration and
long duration was highly significant. Men below the mean
iduration of 33 months decreased in dependency more than
those above the mean. Hypothesis III was accepted.
Hypothesis IV
It was hypothesized that verbal conditioning would
:have a differential effect upon subjects at extreme ends of
the original De score distribution. Specifically, it was
predicted that the lower the experimental subjects1
pre-test scores, the greater the probability that there
would be no change in De since subjects at the independent
pole were already close to the possible extreme score of
independence. Conversely, it was predicted that the higher
the pre-test De score the greater the probability that
there would be a decrease; i.e., the more dependent subject
would condition more readily.
The distribution of pre-t^6t De scores for the
experimental group varied from a minimum of 7 to a maximum
of 39, with a mean of 25.7 (Table 14). Upon post-test,
increases were found at both the lowest and the highest
ends of the continuum. Thus the hypothesis as stated was
not supported. Nevertheless, a detailed analysis of the
data for the experimental groups reveals a solid block of
decreases in dependence within the range cut off by the two
lowest and three highest scores (Table 14 and Figure 2).
Within the experimental group alone, all subjects within
the range defined by the cut-off scores (15 to 36)
decreased in dependency.
This finding suggested that the original hypothesis
might be valid only for the middle range of dependency,
reasoning that the very extremely dependent would be too
solidly entrenched in their dependence to be moved to
change by the verbal conditioning effort. For them, the
verbal conditioning could very well be a threat to their
style of life which might cause them to react with
61
TABLE 14
RANKED DEPENDENCY SCORES OF EXPERIMENTAL
AND CONTROL GROUPS
Group Pre Post Difference
EXPERIMENTAL
Two
lowest
7
8
8
5
+1
+3
15
16
17
26
27
28
31
31
31
34
36
12
14
17
17
22
22
27
23
29
24
28
-3
-2
0
-9
-5
-6
-4
-8
-2
-10
-8
75%
Three
highest
38
38
39
34
41
44
-4
+3
+5 N=16
CONTROL
Two
lowest
7
11
8
9
+1
-2
15
19
28
31
19
13
28
30
+4
-6
0
-1
50-50
Two
highest
40
45
52
54
+12
+19 N=8
; ................................. 63
! increased statements of dependence. At the low end of the
| spectrum, the subjects already espousing independence
i
I would not be expected to move significantly downward
;because they already were psychologically at the point
which the E desired. A Spearman Rho rank order correlation
for the 11 experimental subjects in the center of the De
; score continuum reached a value of -5.3, which was signif
icant at only the .10 level. This result is in the
; predicted direction but not at a level of significance
; which is sufficient to accept the new hypothesis under
consideration.
An unexpected but related line of evidence is
yielded by the analysis of the relationship between initial
dependency level and the number of butterflies needed to
obtain the involvement of subjects in the study (Table 15) .
It will be recalled that 35 subjects were butterflied as
prospective participants. The six men in Group I failed to
respond to a total of six butterflies after being pre
tested. There were an additional five subjects (Group II)
who took pre- and post-test in response to varying numbers
of butterflies. Group II did not participate in either the
]experimental or control condition. Group III contained the
;experimental and control subjects.
: |
The data in Table 15 indicate that there is a
significant relationship between the initial level of |
dependency and the amount of work required (in the form of j
64
!butterflies) to obtain their cooperation. Those men scor-
!ing below the median in pre-test De scores required a
i
! significantly greater number of butterflies than those
| scoring above the median. The results show that the lower
!
i the initial level of dependence, the higher the amount of
work; the higher the dependency, the less work. This
I relationship yielded a Chi Square of 8.36, significant at
!the .05 level (Table 15). Independent subjects, those
j
I below the median of pre-test De scores, required signifi-
i
•cantly more effort to obtain their presence than subjects
with a higher level of initial dependency.
TABLE 15
CHI SQUARE 2 X 3 CONTINGENCY TABLE: THE RELATIONSHIP
OF PRE-TEST DE SCORES TO THE NUMBER OF
BUTTERFLIES (Bf) ISSUED:
GROUPS I, II, III
Number of Bf to Pre-test De
Obtained Frequencies
Above Median
of Pre-test
De Score
1 - 2 Bf
15
(12.3)
3 - 5 Bf
3
(2.6)
6 Bf
0
(3.1)
Below Median
of Pre-test
De Score
9
(11.7)
2
(2.4)
6
(2.9)
N = 35
Median = 26.0
Chi Square = 8.36*
df = 2
*Significant at the .02 level.
Table 16 summarizes the data with respect to
butterflies, De scores, and SPE ratings. It can be seen
that Group I, who took the pre-test only, had the lowest
median De score of 14.5. They also required the greatest
number of butterflies to appear for pre-test, oral inter
view, and post-test. They were the least cooperative,
refusing to take part in the latter two aspects of the
design. Group II, with a median De score of 19.5, took
both the pre-test and post-test but declined to participate
in the research. This group required an average of 3.5
butterflies per subject, noticeably fewer than Group I, but
greater still than those subjects in Group III. Within
Group III the eight controls had a median De score of 23.5
and the experimental group, 29.5. These two subgroups
required the fewest number of butterflies. Thus the rela
tionship first delineated in Table 15 is elaborated in
Table 16.
TABLE 16
SUMMARY OF MEDIANS, MEAN CHANGES AND MEAN NUMBERS OF BUTTERFLIES TO
PRE-TEST, ORAL INTERVIEW, AND POST-TEST
De SPE M/Bf
Group
Md
Pre
Md
Post Diff
Mean
Change
Md
Pre
Md
Post Diff
Mean
Change
Pre
De Oral
Post
De
I
N=6
14.5
- - -
31.0 31.0 0 0 4.5 6 6
II
N=5
19.5 38.0 +31 +6.2 29 25 -15 -3.0 3.6 3.4 1.8
III
Experimental
N=16
29.5 22.5 -55 -3.4 29 35.5 +105 6.5 1.5 1.3 1.5
Control
N=8
23.5 23.5 +17 2.1 29 34.5 +21 2.6 1.5 1.3 1.5
ON
O N
CHAPTER V
j
j
! DISCUSSION OF RESULTS
I
Hypothesis I
The central focus of this research design was to
!isolate and manipulate experimentally the construct of
:dependence-independence in domiciled veterans. Following
the operant conditioning model, positive reinforcement was
experimentally manipulated to produce specifiable changes
in dependent-independent verbal behavior. The major
'hypothesis of the research design thus focused on the
reduction of dependency, not only in verbal behavior, but
;in other related areas of institutional behavior.
Hypothesis I, therefore, tested the effect of
verbal conditioning on dependency, where dependency was
defined operationally as increase or decrease in the total
number of directional responses measured by the Dependency
Scale (De) of the MMPI.
It was predicted that the experimental group, as a
result of verbal conditioning, would decrease in De scores
I
:from pre- to post-test. It was predicted that the control
group, in the absence of verbal conditioning, would
increase, or show no change. To compare pre- and post-test
difference scores on the De Scale, a one-way analysis of
variance was computed. The difference between groups was
significant at the .025 level of confidence. The null
hypothesis was rejected and Hypothesis I was accepted. In
the sample studied, verbal conditioning reduced dependency
in domiciled veterans as reflected by a decrease in De
scores.
A corollary assumption underlying Hypothesis I was
that changes in written responses to a psychometric instru
ment, such as the De scale, would be accompanied by changes
in public verbal behavior. As dependency increases or
decreases on the scale, so should the rate of emission of
independent statements vary in the group setting. The more
dependency reduces, the higher the rate of such statements;
the more dependency increases, the lower the rate. It was
predicted specifically that the experimental group would
increase their rate of emissions of statements of independ
ence, as a function of positive reinforcement and the
controls, in the absence of reinforcement, would be
expected to show a decrease, or no change. This expecta
tion was supported at the .001 level of confidence. Verbal
conditioning did, in fact, increase the emission of public
! verbal behavior within the group setting. However, the
data suggest that the four-week experimental span repre-
■ y
sents the minimal amount of time needed to produce this
r ~ ..................... ...... ' _ ' ~ 69
;significant effect. The present design leaves open the
| question of what length of time would produce a maximal
| effect.
S The design also leaves open the question of dura-
i
;tion of change, after achieving a significant effect. The
design did not include systematic efforts to capitalize on
I the reduction of dependent behavior achieved by verbal
conditioning. It would appear, however, that if the insti-
|tution were organized so as to support actively such
ichanges with swift administrative follow-up, the reduction
in dependency might result in more patients achieving the
desired return to the community. Conversely, if experimen-
|tally induced changes in behavior are not actively
:supported by the administration, there is a high probabil
ity that the subjects would revert back to their pre-
experimental operant levels of behavior, diluting the rate
of return.
Changes in verbal responses to the De Scale and
public verbal behavior in the group setting also were
accompanied by changes in other areas of behavior. The
assumption underlying the administration of the Supervisors
|Performance Evaluation was that changes in dependency, in
;the presence or absence of reinforcement, would be reflected
! • i
:in other behavioral areas, such as work performance. When
compared on pre- and post-SPE difference scores the differ
ence between groups was highly significant (p = c.001).
70
The experimental group not only decreased in dependency but
also showed a high increase in quality of work performance;
the control group, on the other hand, increased in depend
ency and increased only slightly in quality of work
performance. Such data strongly support the hypothesis
that the reduction in dependency effected by verbal condi
tioning produces behavioral changes in the related area of
work p er formanc e.
Hypothesis II
Hypothesis II was designed to test the nebulous but
popular concept in psychology that only the young are
susceptible to change. It was predicted specifically that
the younger the subject the higher the probability of a
reduction in dependency; the older the subject, the less
likelihood of change. Since the Domiciliary population now
consists of men considered too young to retire, the age
group chosen ranged from 40 to 50 years. Within this age
range population, arbitrarily designated as middle aged,
subjects were dichotomized as younger or older, according
to whether they fell below or above the median of 45.0
years.
The younger experimental subjects, presumably more
flexible, were expected to decrease in dependency as a
result of exposure to the operation of verbal conditioning;
the older subjects, presumably more rigid, were not
71
expected to change. No difference between groups obtained.
Although Hypothesis II was not supported by the
data, the treatment implications are positive. The appli
cation of verbal conditioning, using positive reinforce
ment, has been shown to reduce dependency in middle aged
veterans. The lack of support for Hypothesis II implies
that the method is useful, irrespective of the age of the
subject in the 40-50 age range.
Hypothesis III
Hypothesis III was designed to explore the rela
tionship of dependency and the institutionalization effect.
Subjects chosen for the study had been institutionalized at
the Domiciliary from one to five years. Specifically, it
was predicted that the institutionalization effect would be
minimal at the lower limit of one-year stay and maximum at
the upper limit of five-year stay. The subjects at the
lower limit, presumably less affected by institutional
life, were expected to decrease in dependency. The
subjects at the upper limit, presumably more entrenched in
the institutional life, were not expected to change.
Reduction in De scores was expected to be optimal below the
mean of the obtained sample, 33 months. For those in the
experimental group who decreased in dependency, two-thirds
were below the mean of 33 months. A Chi Square of 8.10
72
was significant between the .01 and the .001 levels of
confidence. Hypothesis III was supported.
Taken together with the lack of support for the age
factor, the positive support for Hypothesis III suggests
that men between the ages of 40 and 50 can be changed in
the direction of independence, but that the prospects for
such movement are brightest if the length of stay prior to
the application of the verbal conditioning regimen does
not exceed 33 months. This finding implies that maximum; „
’ . : ' £ *
results may be obtained by conditioning independent
behavior as soon as possible, between admittance and the
33-month cut-off point. There is a higher probability
that the shorter the duration of stay, the greater will be
the efficiency of verbal conditioning in the reduction of
dependency.
Hypothesis IV
Hypothesis IV predicted that verbal conditioning
would have a differential effect upon subjects at the
extreme ends of the original De score distribution sampled.
Since low scores represented subjects already close to the
desired pole of independence, they were not expected to
change. High scores, however, representing maximal depend
ence, would have a higher probability of decrease. This
hypothesis was not supported.
73
The data revealed that subjects initially scoring
high in dependency increased rather than decreased their
scores upon post-test. This suggests that these subjects
are so entrenched in their syndrome that a paradoxical
effect occurs. Attempts to decrease their dependency by
means of verbal conditioning act as if to threaten their
security. The result is a psychological resistance,
manifested by increased De scores. The minor increases
shown by subjects initially scoring at the very low end of
: the De scale continuum can be seen as chance changes which
; might be consistent with their maintenance of their initial
psychological position.
The emergence of a curvilinear relationship between
initial levels of dependency and subsequent changes in
dependency indicates that the lower slope of the regression
is a U-shaped function. This finding suggests an optimum
range of initial dependency level, within defined cut-off
points, where verbal conditioning has a higher probability
in effecting a decrease in De scores.
This hypothesis was subjected to statistical test
ing and there appeared to be the predicted relationship
within the middle ranges of the De score continuum.
Although the statistical test for this group of scores did
not reach a rigorous level of significance, the result is
promising enough to warrant further exploration in
subsequent research.
74
Butterfly Results
The analysis of the relationship between initial
level of dependency, as measured by the De Scale, and the
amenability of the members to treatment was an unforeseen
by-product of the design. Since all subjects butterflied
in Groups II and III took the pre- and post-test De it was
possible to compare initial level of dependency with mean
change in De whether or not subjects had participated in
the experimental condition. It was also possible to
compare mean changes in work performance for all Groups I,
II, and III, since SPE evaluations were available on each
subj ect.
As can be seen from Table 16, there is no De mean
change data available for the maximally independent Group I
(Md = 14.5). Mean change for pre- and post-SPE is zero.
Group II, whose median was close to that reported for
normal subjects (Navran, 1951) did not participate in the
experimental condition, although they took the pre- and
post-De test. They show an increase in dependency scores
and a decrease in work performance ratings. Within Group
III the experimental subjects, with an initially high
dependency level (Md = 29.5), show a decrease in De and a
substantial increase in work performance. The initial
level of dependency for the controls was also in the high
range (Md =23.5). In the absence of verbal conditioning,
however, they show an increase in dependency and a small
increase in work performance.
The significant statistical difference between the
experimental and control groups for pre-post De scores and
pre-post SPE scores has already been discussed.
Group II, taking the pre- and post-tests, had not
been subjected to any of the experimental conditions.
Table 16 shows that Group II increased in dependency and
decreased in work performance more than the control group
itself. A statistical comparison of mean changes in De and
SPE between the controls and Group II was not significant,
although the trend was unmistakable.
Supervisors of Group II subjects were questioned on
the reasons for the SPE lowered ratings over the four
weeks. One subject was in the process of transferring to
another VA Domiciliary and no longer appeared to be inter
ested in his job detail. Another subject had gone on an
alcoholic binge, missing time at work. The remaining three
subjects had been downgraded primarily for indifference to
their details. A typical supervisory comment was "doing
satisfactory work. What is hard to understand is his
indifference to almost everything and lack of enthusiasm."
The implications arising from this analysis of the
butterfly problem are twofold:
1. The higher the initial level of dependency, the
more readily subjects respond to participation in
Domiciliary research. Of the three major groups involved
with respect to the butterfly data, Group III was the high
est in initial dependency, with a pre-test median of 28.0.
This group not only required the least work to obtain their;
appearance but also stayed to become the experimental and
control subjects in the research. Group II was more inde
pendent, with an initial dependency median of 19.5, and
required more work to obtain their presence. This group
complied with pre- and post-testing but refused to partic
ipate in the research. Group I was the most independent
of the three groups with a pre-test median of 14.5. This
group required the most work in the form of butterflies
and never appeared again after the pre-test.
These data imply that level of dependency may be
an important variable in the decision-making of subjects
who may be considered for research participation. The
more independent subject is less likely to respond to
butterflies or take part in research. The resulting sample
of subjects who actually participate may be heavily
weighted in the direction of dependency.
2. There may be a paradoxical relationship between
initial level of dependency and degree of entrenchment in
the institutional life. The most independent subjects in
Group I, who responded only to the pre-test butterflies,
appear to be a subset in the larger Domiciliary population.
There is a high probability that this subset of high-
independence subjects are men who paradoxically fight to
77
maintain their dependency on the Domiciliary environment,
resisting any attempts to change the status quo of "three
hots and a flop." Refusal to answer mandatory butterflies
evidently reflects this resistance.
Administrative Implications
of Butterfly Data
There are important implications for administrative
planning in the use of such initial dependency levels, as
measured by the De Scale and data on response to butter
flies. Knowledge of the relationship that exists between
these levels and the involvement of the men themselves in
treatment programs allows a more precise selection of
subjects for future conditioning programs. Those men at
the very low end of the De score continuum respond inde
pendently; their independent verbalizations can be used as
an indicator that they will not cooperate readily with
efforts to change their Domiciliary status.
The more moderately dependent men in the middle
range will participate; they offer the greatest prospect
for change via verbal conditioning. The most openly
dependent men may cooperate but will be heavily threatened
by treatment procedures and will respond with a paradoxical
increase of dependent behaviors. In the interest of
efficiency, the De scale may thus be used as a screening
device to designate those subjects most likely to benefit
from programs designed to return the veteran to the
community.
CHAPTER VI
SUMMARY AND CONCLUSIONS
Summary
Method and Procedure
The primary research purpose of the design was to
explore the dimension of dependence-independence in
domiciled veterans. Dependency was operationally defined
as an increase over time in the total number of directional
responses (true-false) measured by the Dependency Scale of
the MMPI.
When pre-post test differences were negative in
sign, the subject was said to have decreased in dependency,
becoming more independent. When pre-post test differences
were positive in sign the subject was said to have
increased in dependency, becoming even more dependent.
The independent variable in the 1 X 2 factorial
design was partial positive reinforcement of the response
class of statements of independence emitted in group
settings. Positive reinforcement was operationally defined
as the presentation of the verbal reinforcing stimuli
"good" or "umm-hmm" to the occurrence of the operant.
79
Statements of independence were operationally defined as
any man's statement of self-reference indicating an action
on his own behalf, in any area.
There were 24 subjects in the study, chosen from
the Domiciliary veteran population according to these
specified criteria: ambulatory males between 40 and 50
years of age, in contact with reality, and an institution
alization length of one to five years. The 24 subjects
were randomly assigned to two control groups and four
experimental groups. These were subsequently compressed
into one control group (N=8) and one experimental group
(N=16) for the purpose of statistical analysis.
Both the experimental and control groups met for a
total of 16 hours, one hour a day, four days a week, for
four consecutive weeks. The experimental subjects received
positive reinforcement, "good" or "umm-hmm" for the emis
sion of statements of independence. To control for level
of reinforcement the control group received random positive
reinforcements of independence. The approximate rate of
this random reinforcement was five per hour.
To evaluate changes in dependency as a function of
verbal conditioning, all subjects were administered pre-
and post-tests on the De scale embedded in the MMPI. As a
concurrent validity index they were also pre- and post-
measured on the Supervisors Performance Evaluation (SPE),
to explore the relationship between increases or decreases
in dependency and behavioral changes in other areas, such
as job performance. All groups were structured in the
elicitation of independence statements by presenting
uniform topics from Siegel's Dependency and Self-Assertion
Scale. All sessions were tape recorded to provide a
frequency analysis of emission rate of independent state
ments .
Hypotheses
Four hypotheses were postulated:
Hypothesis I: It was hypothesized that verbal
conditioning, using positive reinforcement, would decrease
dependency in domiciled veterans, where dependency is
operationally defined by the Dependency Scale of the MMPI.
As the response class of independent statements is verbally
reinforced, so will De scores decrease generally.
Hypothesis II: It was hypothesized that verbal
conditioning would decrease De scores as a function of the
subject's age. The younger the subject, the higher the
probability that the De scores would decrease generally;
the older the subject, the less decrease, or no change.
This hypothesis was limited to the experimental groups.
Hypothesis III: It was hypothesized that verbal
conditioning would decrease De scores as a function of
length of institutionalization. The shorter the duration
of Domiciliary stay, the higher the probability that De
82
scores would decrease generally; the longer the duration,
the less decrease, or no change. This hypothesis was
limited to the experimental groups.
Hypothesis IV: It was hypothesized that verbal
conditioning would have a differential effect upon subjects
at extreme ends of the initial De score distribution. The
lower the pre-test score, the higher the probability that
there would be no change in De scores. Conversely, the
higher the pre-test score, the greater the probability that
there would be a decrease in De scores generally. This
hypothesis was limited to the experimental groups.
The major hypothesis postulated that verbal condi
tioning, using positive reinforcement, would decrease the
operationally defined variable of dependency. Specif
ically, conditioning should occur in the presence of
reinforcement, reducing dependency in the experimental
group. No conditioning should occur in the absence of
reinforcement for the control group; dependency should
increase or show no change. These results obtained.
Hypothesis I was accepted at the 2.5 per cent level of
confidence.
A corollary assumption underlying Hypothesis I was
that changes in written responses to the De scale would be
accompanied by changes in verbal behavior within the group
setting. Such changes would be reflected in rate of
emission of independent statements. Specifically, the
..................................... •... '............................ "..'...............'.............. " .... 83
experimental group was expected to increase in rate of
self-referred verbal behavior in the presence of positive
reinforcement. The control group, in the absence of
reinforcement, should decrease, or no change. The expecta
tion was supported at the .001 level of confidence.
Although verbal conditioning affected rate of
emission significantly at the end of two weeks, the maximum
effect occurred by the end of four weeks. The difference
between the first two weeks and the last two weeks was
significant beyond the .001 level.
The reduction of dependency in the experimental
condition was accompanied by significant change in the
related area of job performance. As measured by pre- and
post-Supervisors' rating (SPE), the experimental group
decreased in dependency and increased in work performance;
the controls increased only slightly. The SPE as a
concurrent validity index was highly correlated with the
De Scale, pre-post difference scores for both the De Scale
and the SPE being significant between the .01 and the .001
levels of confidence.
The reduction in dependency postulated by Hypoth
esis I was therefore supported by pre-post De changes in
the predicted direction, an increase in public verbal
behavior within the group setting, and increase in pre-post
evaluations of work performance. Verbal conditioning,
using positive reinforcement, reduced dependency in
.................................. ' ....84
domiciled veterans. This effect was reflected in each of
the three separate areas of written response behavior,
public verbal behavior, and job-performance behavior.
Hypothesis II explored the relationship between
changes in dependency and age of the subjects. Specif
ically it was predicted that the presence of positive
reinforcement would decrease dependency maximally at the
lower limit of the 40-50 age range. Subjects above the
median of this range were not expected to significantly
decrease. The results did not support the hypothesis.
There were no significant differences in reduction of
dependency between younger and older subjects. Verbal
conditioning, using positive reinforcement, reduced
dependency throughout the middle age span in the sample.
Hypothesis III explored the relationship between
changes in dependency and individual length of institution
alization. Specifically, it was predicted that the maximum
reduction in dependency would occur in those subjects at
the lower limit of the institutionalization span, which
ranged from one to five years. This hypothesis was
supported. The operation of positive reinforcement reduced
dependency to a significantly greater extent in veterans
whose length of stay was under 33 months than it did for
veterans who had been domiciled over 33 months.
Hypothesis IV explored the relationship between
changes in dependency and the verbal conditioning of
85
extremely dependent or extremely independent subjects.
Specifically, it was predicted that highly dependent
subjects would show more reduction in dependency in the
presence of positive reinforcement than would the maximally
independent subjects.
Although this hypothesis was not supported, the
data revealed a curvilinear function between initial
dependency level and reduction in dependency. This implies
that there is an optimum range of initial dependency level,
within defined maximum and minimum cut-off points, where
verbal conditioning effects a consistent reduction of
dependency. In the sample studied, the optimum range
varied from four raw score points below the normal mean of
19 to 17 above. The probability that this range of initial
dependency level, as measured by the De Scale, is a predic
tive index for the likelihood of response to conditioning
for independence reached only the .10 level of confidence
in this study. While not rigorously definitive, it is
considered a finding worthy of further research explora
tion.
Conclusions
86
Verbal conditioning, using the operation of posi
tive reinforcement, significantly affected the dimension
of dependence-independence in domiciled veterans within a
group setting. The reduction of dependency was clearly
evidenced in three areas of behavior:
1. Decreases in pre-post psychometric testing on
the De Scale.
2. Increases in public verbal behavior of self-
statements of independence.
3. Increases in work performance behavior.
Maximum reduction of dependency occurred at the
end of a four-week period with groups meeting for one hour
daily, four days a week. Verbal conditioning reduced
dependency most effectively in non-organic middle-aged
veterans up to 50 years, who had been institutionalized up
to 33 months. The De Scale of the MMPI may be used as a
predictive index for selection of subjects within an
optimum range for reduction of dependency.
Treatment Implications of the Study
There are several implications for administrative
efficiency which issue from the data. As we have seen in
Chapter I, the dimension of dependence-independence may be
viewed as a major theoretical variable in the clinical
87
concept of personality malfunction. Secondary dependence,
as defined by Navran (1951), is by definition a maladaptive
response mode in the adult.
The construct of dependency in domiciled veterans
has become a crucial one, in view of the steady transition
in the domiciled veteran population from that of the older
honorably retired to that of the younger psychologically
disturbed. A concomitant development of a professional
manpower shortage in the Domiciliary, where the staff-to-
patient ratio is 1/245, has accentuated the need for
systematic efforts to reduce dependent behavior and facili
tate veterans’ restoration to the wider community.
Administrators in neuropsychiatric hospitals are
continually faced with the problem of formulating econom
ical and feasible treatment programs for heterogeneous
populations. Using the verbal conditioning paradigm, it
may be possible to train non-professionals, such as work
supervisors and section heads, to work with veterans in
groups, which would be a step toward equalizing the treat
ment staff-to-patient ratio. In this way a substantial
segment of the domiciled population can be brought into a
total treatment milieu. The results of the research
strongly indicate that verbal conditioning, using positive
reinforcement with a daily group setting, may be an effi
cient and powerful method. It is capable of being
incorporated into such a total treatment program. If
88
staffing shortages do not make verbal conditioning efforts
feasible for the total Domiciliary population, the results
of this study offer both test and length of stay guidelines
which may be used to maximize the therapeutic return for
the treatment effort expended.
A subset of data evolved from subjects' varying
irregularities in response to "butterflies" (mandatory
orders to appear for the research), This subset revealed
a relationship between a subject's initial level of depend-;
ency and the amount of work required in the form of butter
flies issued to obtain his cooperation. The higher the
level of initial independence (Md = 14), the more work
required to obtain appearance. The higher the level of
initial dependency (Md = 28), the less work required.
Subjects high in independence refused to participate in the
research. Subjects high in dependency remained as experi-
mentals and controls.
The design left open the question of sustaining
conditioned changes in independent behavior. Administra
tive follow-up and support of the verbally conditioned
behavior may very likely be a prerequisite to maintaining
these changes. Otherwise, independent behavior may return
to the pre-operant level. However, at this point, no
definitive evidence is available to deal with this issue.
89
Research Implications of the Study
Replication of a similar research design raises
several questions:
1. Future replications of the design should incor
porate a larger N. The present study was limited to six
_groups of four each, essentially a very small sample of
the total 2,700 veterans presently residing at the
Domiciliary. Larger N's would assure greater statistical
generalization to this larger population.
2. The age range in the sample studied was 40-50
years. There still remain men under 40 and over 50 in the
Domiciliary. Would the operation of positive reinforcement
be effective for men not included in the age range studied?
3. The institutionalization span in the sample
studied was one to five years. There still remain the
freshly admitted men and those who have settled in for more
than five years. Would the operation of positive
reinforcement be effective for such veterans?
4. The Domiciliary population is preponderantly
male. The examiner in this study was female. Would a male
examiner obtain the same results? Or would women examiners
be more effective with this particular population?
5. The examiner was within the same age range as
the subjects. Would a younger examiner be as effective?
An older one?
90
Data bearing on these questions would be meaningful
in terms of administrative staffing. If male examiners are
as effective as female examiners, or young examiners as
effective as older ones, no change in the hiring policy is
needed. But if it obtains that women are more effective
with this domiciled population, more women should be hired.
If older women are more effective, then older women should
be hired. Such results would definitely serve as valuable
guides for Veterans Administration Domiciliary administra
tors .
In any event, the positive findings obtained in
this study certainly make worthwhile further research
exploration of the use of verbal conditioning as a treat
ment procedure.
B I BLI OG RA PHY
91
BIBLIOGRAPHY
Ainsworth, Mary D. The effects of maternal deprivation:
A review of findings and controversy in the context
of research strategy. In World Health Organiza
tion, Deprivation of maternal care: A reassessment
of its effects^ Geneva: World Health Organiza-
tion, 1962. Pp. 97-165.
Alexander, R., and T. French. Psychoanalytic therapy.
New York: Ronald Press, 1946.
Arieti, S. American handbook of psychiatry. Vol. II.
New York: Basic Books, Inc., 1959.
_________. American handbook of psychiatry. Vol. III.
New York: Basic Books, Inc., 1966.
Bakwin, H. Emotional deprivation in infants. J. of
Pediatrics, 1949, J5, 512-521.
Bard, M. The use of dependence for predicting psychogenic
invalidism following radical mastectomy. J. Nerv.
& Ment. Pis., 1955, 122, 152-160.
Barton, R. Institutional neurosis. Bristol: John Wright
& Sons, Ltd., 1959.
Beller, E. E. Dependency and independence in young
children. J. Genet. Psychol. , 1955, 87^, 25-35.
Beres, D., and S. J. Obers. The effects of extreme
deprivation in infancy on psychic structure in
adolescence: A study in ego development.
In R. S. Eissler et al. (eds.), The psychoanalytic
study of the chilcT- Vol. V. New York: Inter-
national University Press, 1950. Pp. 212-235.
Bernadin, A. C., and R. Jessor. A constructor validation
of the Edwards personal preference schedule with
respect to dependency. J. Consult. Psychol., 1957,
21, 63-67. ---------------
92
.... "..... '.............. ' ." . 93
Bernfield, Judith M. , and G. M. Guthrie. Experimental
control of dependency by post hypnotic suggestion.
J. of Clinical Psychol., 1959, 15, 114-115.
Boring, E. G. A history of experimental psychology.
New Yorlci Appleton-Century-Crofts , Inc . , 1950.
Bowlby, J. Maternal care and mental health. Monograph
Series No. Geneva: World Health Organization,
1952.
Brinton, C. Ideas and men: The story of western thought.
Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1950.
Button, A. D. A study of alcoholics with the MMPI.
Quarterly J. of Studies on Alcohol., 1956, 17,
263-281.
Carman, Philip C. Psychologists: A change in the total
institution. Proposed symposium to be presented
at Western Psychological Association, March, 1968,
San Diego, California.
Caruth, Elaine G. The relationship of dependency to verbal
learning without awareness. Diss. Abs., 20, 1959.
Pp. 748-749.
Chaplin, J. P., and T. S. Krawiec. Systems and theories of
psychology. New York: Holt, Rinehart & Winston,
Cohen, B. D., H. I. Kalish, J. R. Thurston, and E. Cohen.
Experimental manipulation of verbal behavior.
J. Exp. Psychol. , 1954, 47_, 106-110.
Coleman, J. C. Abnormal psychology and modern life.
New York: Scott, Foresman & Co., 1964. 634 pp.
Dahlstrom, W. G., and G. S. Welsh. An MMPI handbook.
Minneapolis: University of Minnesota Press, 1960.
Dollard, J., and N. E. Miller. Personality and psycho
therapy . New York: McGraw-Hill, 1950.
Estes, K. W. Some effects of reinforcement upon verbal
behavior of children. Unpublished doctoral
dissertation, University of Minnesota, 1945.
Fenichel, 0. The psychoanalytic theory of neurosis.
New York: Norton, 1945.
94
Ferster, C. B., and B. F. Skinner. Schedules of reinforce-
ment. New York: Appleton-Century-Crofts, Inc.,
T 9 3 7 .
Ford, D. H., and H. B. Urban. Systems of psychotherapy.
New York: Wiley & Sons, Inc. , l9f>6 .
Fordyce, W. Study in Progress at VA Regional Office, Fort
Snelling, St. Paul, Minnesota. Reported in a
written personal communication to L. Navran, dated
December 30, 1953.
Freud, S. Inhibitions, symptoms and anxiety. London:
Hogarth Press, 1 9 3 6 .
_. An outline of psychoanalysis. New York:
Norton, 1949.
_________. (1919) Beyond the pleasure principle. In J.
Strachey (ed), Standard edition of the complete
psychological works. Vol. 18. London: Hogarth
p^s'S7"T95t:---------
Gewirtz, J. L., and D. M. Bair. Deprivation and satiation
of social reinforcers as drive conditioners.
J. Abn. S c . Soc. Psychology, 1959, 57_, 165-172.
Gordon, B. An experimental study of dependence-
independence in a social and a laboratory setting.
Unpublished doctoral dissertation, University of
Southern California, 1953.
Green, E. B. Dependent behavior in the blind adult.
Piss. Abs., 27 (5-b), 1966. Pp. 1606-1607.
Greenspoon, J. Verbal conditioning and clinical psychol
ogy. In A. J. Bachrach (ed.), Experimental
foundations of clinical psychology. New York:
Basic Books, Inc., 1962. Pp. 5l0-553.
________ . The effect of verbal and nonverbal stimuli on
the frequency of members of two verbal response
classes. Unpublished doctoral dissertation,
Indiana University, 1951.
Hardison, J., and K. Purcell. The effects of psychological
stress as a functional of need and cognitive
control. J. of Person. , 1959, 27^, 250-258.
Hayes, W. L. Statistics for psychologists. New York:
Holt, Rinehart & Winston, 1963. 719 pp.
94
Ferster, C. B., and B. F. Skinner. Schedules of reinforce-
ment. New York: Appleton-Century-Crofts. Inc.,
1937.
Ford, D. H., and H. B. Urban. Systems of psychotherapy.
New York: Wiley & Sons, Inc., 1956.
Fordyce, W. Study in Progress at VA Regional Office, Fort
Snelling, St. Paul, Minnesota. Reported in a
written personal communication to L. Navran, dated
December 30, 1953.
Freud, S. Inhibitions, symptoms and anxiety. London:
Hogarth Press, 1936.
An outline of psychoanalysis. New York:
Norton, 1949.
(1919) Beyond the pleasure principle. In J.
Strachey (ed), Standard edition of the complete
psychological works. Vol. 18. London: Hogarth
Press, 1955.
Gewirtz, J. L., and D. M. Bair. Deprivation and satiation
of social reinforcers as drive conditioners.
J. Abn. & Soc. Psychology, 1959, 57, 165-172.
Gordon, B. An experimental study of dependence-
independence in a social and a laboratory setting.
Unpublished doctoral dissertation, University of
Southern California, 1953.
Green, E. B. Dependent behavior in the blind adult.
Piss. Abs., 27 (5-b), 1966. Pp. 1606-1607.
Greenspoon, J. Verbal conditioning and clinical psychol
ogy. In A. J. Bachrach (ed.), Experimental
foundations of clinical psychology. New York:
Basic Books, Inc., 1962. Pp. 510-553.
_________. The effect of verbal and nonverbal stimuli on
the frequency of members of two verbal response
classes. Unpublished doctoral dissertation,
Indiana University, 1951.
Hardison, J., and K. Purcell. The effects of psychological
stress as a functional of need and cognitive
control. J. of Person., 1959, 21_> 250-258.
Hayes, W. L. Statistics for psychologists. New York:
Holt, Rinehart & Winston, 1963. 719 pp.
95
Hildum, D. C. , and R. W. Brown. Verbal reinforcement and
interviewer bias. J. Abn. & Soc. Psychology, 1956,
53, 108-111.
Hilgard, E. R., and D. G. Marquis. Conditioning and
learning. New York: Appleton-Century-Crofts,
Inc. , 1961.
Hill, W. F. Learning: A survey of psychological inter
pretations. San Francisco: Chandler Publishing
Co., 19b3.
Hovorka, E. J. A study of stimulus-generalization in human
operant behavior. Unpublished master's thesis,
Indiana University, 1950.
Jacobs, A. J., Carol Bruner, Nita Orlando, and J. P.
Meehan. Changes in motivational and emotional
states as a function of prolonged bed rest.
Paper given at Western Psychological Association,
Long Beach, 1966.
Jeffrey, W. E. New technique for motivating and reinforc
ing children. Science, 1955, 121, 371.
Kagen, J., and H. A. Moss. Birth to maturity. New York:
Wiley, 1962.
Keller, F. S. Learning (reinforcement theory). New York:
Doubleday, 1954.
, and W. N. Schoenfeld. Principles of psychology.
New York: Appleton-Century-Crofts, Inc., 1950.
Klein, S. Conditioning and extinction of operant verbal
behavior in neuropsychiatric patients. Unpublished
doctoral dissertation, Indiana University, 1954.
Krasner, L. Studies of the conditioning of verbal
behavior. Psychol. Bull., 1958, 55_, 148-170.
________ . Verbal conditioning and psychotherapy. In
Research in behavior modification. New York:
Holt, Rinehart & Winston, 1965b.
________ , and L. P. Ullmann. Research in behavior modifi-
cation. New York: Holt, Rinehart & Winston,
1965a.
96
Kubie, L. X. The neurotic conflict and human adaptation.
In H. Hans and Gertrude M. Kurth (eds.), Elements
of psychoanalysis. Cleveland: World Publishing
Co., l95o.
Levy, D. M. Maternal overprotection. In N. D. C. Lewis
and B. L. Pacella (eds.), Modern trends in child
psychiatry. New York: International University
Press, 1945.
Livson, M. , and P. H. Mussen. The relation of ego control
to overt aggression and dependency. J. Consult.
Psychol. , 1957, 55_, 66-71.
McGeoch, J. A., and A. L. Irion. The psychology of human
learning. New York: Longmans, Green, 1953.
McGuigan, F. J. Further study of psychological changes
related to intercultural experiences. Psychol.
Rep., 1959, 5, 244-248.
McNair, D. M. Reinforcement of verbal behavior. J. Exp.
Psychol. , 1957, 5_3, 46-49.
McNemar, Q. Psychological statistics. New York: Wiley
and Sons, 1949.
Mandler, G., and W. K. Kaplan. Subjective evaluation and
reinforcing effect of a verbal stimulus. Science,
1946, 124, 582-583.
Maslow, A. H., and B. Mittleman. Principles of abnormal
psychology. New York: Harper & Sons, 1941.
Mussen, P. H., J. J. Conger, and J. Kagan. Child
development and personality. New York: Harper &
Row, 1963.
Navran, L. A rationally derived Minnesota Multiphasic
Personality Inventory scale to measure dependence.
Unpublished doctoral dissertation, Stanford
University, 1951.
_________. A rationally derived MMPI scale to measure
dependence. J. Consult. Psychol., 1954, 18, 3,
192.
Nelson, J. W. Dependency as a construct: An evaluation
and some data. Diss. Abs., 19, 1959. Pp. 2149-
2150.
97
Nuthmann, A. M. Conditioning of a response class on a
personality test. J. Abn. & Soc. Psy., 1957, 54,
19-23.
Orlando, H., and R. Steele. Dependency as a function of
group treatment mode. Unpublished pilot study,
University of Southern California, 1967.
Postman, L. The history and present status of the law of
effect. Psychol Bull., 1947, 44, 489-563.
Rothberg, June S. Dependence and anxiety in male patients
following surgery: An investigation of the rela
tionship between dependency, anxiety, and physical
manifestations of recovery following surgery in
male patients. Diss. Abs., 27 (2-B), 966.
Pp. 525-526.
Russell, B. A history of western philosophy. New York:
Simon & Schuster, 1945.
Salzinger, K. Experimental manipulation of verbal
behavior: A review. J. Gen. Psychol., 1959, 61,
65-94.
_________, and S. Pisoni. Reinforcement of affect responses
of schizophrenics during the clinical interview.
J. Abn. & Soc. Psychol., 1958, 57^, 84-90.
Saul, L. J. Physiological effects of emotional tension.
In J. McV. Hunt (ed.), Personality and the behavior
disorders. Vol. I. New York: Ronald Press, 1944.
Sears, R. R., J. W. M. Whiting, V. Nowlis, and P. S. Sears.
Some child rearing antecedents of aggression and
dependency in young children. Genet. Psychol.
Monogr., 1953, 47, 135-236.
Sidowski, J. B. Influence of awareness of reinforcement on
verbal conditioning. J. Exp. Psychol., 1954, 48,
355-360.
Siegel, S. Non-parametric statistics for the behavioral
sciences. New York: McGraw-Hill, 1956.
Skinner, B. F. The behavior of organisms: An experimental
analysis" New York: Appleton-Century, 1938.
Are theories of learning necessary? Psychol.
Rev., 1950, 57, 193-216.
98
Skinner, B. F. Science and human behavior. New York:
Macmillan^ 1953. 430 pp.
_. Verbal behavior. New York: Appleton-Century-
Crofts, Inc., 1957.
Snyder, W. U. Dependency in psychotherapy. New York:
Macmi1lan, 1963.
Spiegel, D. E., and A. J. Litrownik. The effects of
dependency and self assertiveness of schizophrenic
patients on susceptibility to group influence in
perceptual tasks. J. Clinical Psychol., 1968, 24,
12-16.
Spitz, R. A., and K. M. Wolf. Anaclitic depression: An
inquiry into the genesis of psychiatric conditions
in early childhood, II. In A. Freud et al. (eds.)
The psychoanalytic study of the child. Vol. II.
New York: International University Press, 1946.
Pp. 313-342.
Sullivan, H. S. Conceptions of modern psychiatry.
Washington, D.C.: The William Alanson Wnite
Psychiatric Foundation, 1947.
_________. The interpersonal theory of psychiatry.
New York: Norton & Co., 1953.
Thorndike, E. L. Animal intelligence: An experimental
study of the associative processes in animals.
Psychol. Rev. Monogr., 1898, Suppl. 2, No. 8, 1-16
Animal intelligence. New York: Macmillan,
----------------19 i t :------------- -------- ----------
________ . The psychology of learning. Educational
Psychology. Vol. II. New York: Teachers College
1913.
Verplanck, W. S. The operant, from rat to man: An intro
duction to some recent experiments on human
behavior. Trans. New York Acad. Sci., 1955a, 17,
594-601.
________ . The control of the content of conversation:
Reinforcement of statements of opinion. J. Abn. &
Soc. Psychol., 1955b, 51, 688-776.
The operant conditioning of human motor
behavior. Psychol. Bull., 1956, 5^3, 70-83.
99
Verplanck, W. S. A glossary of some terms used in the
objective science of behavior. Suppl. to Psychol.
Rev., 1957, 64, No. 62, Part 2, 42.
Warn, L. J. A comparative investigation of dependency in
epilepsy, paraplegia and tuberculosis. Unpublished
doctoral dissertation, University of California,
Los Angeles, 1958.
Welsh, G. S., and W. G. Dahlstrom (eds.). Basic readings
on the MMPI in psychology and medicine.
Minneapolis: University of Minnesota Press, 1956.
White, R. W. Motivation reconsidered: The concept of
competence. Psychol. Rev., 1959, 66^, 297-333.
Wickes, T. A. Examiner influence in a testing situation.
J. Consult. Psychol. , 1956, 20^, 23-25
Williams, J. H. Conditioning of verbalization: A review.
Psychol. Bull., 1964, 62^, No. 6, 383-393.
Williams, R. The prevention of disability in mental
disorders. Mental Health Monogr. 1, PublicHealth
Service Publication No. 924, U.S. Department of
Health, Education, and Welfare, 1960. Pp. 1-39.
Wilson, W. C., and W. S. Verplanck. Some observations on
the reinforcement of verbal operants. Amer. J.
Psychol., 1956, 69, 448-451.
Zuckerman, M. Levitt, and B. Lubin. Concurrent and
construct validity of direct and indirect measures
of dependency. J. Consult. Psychol., 1961, 25,
No. 4, 316-323.
Zuckerman, M. , and Barbara Eisen. Relationship of
acquiescence response set to authoritarianism and
dependency. Psychol. Rep., 1962, LO, 95-102.
A P P E N D I C E S
100
APPENDIX A
MINNESOTA MULTIPHASIC PERSONALITY INVENTORY
DEPENDENCY SCALE ITEMS
101
MINNESOTA MULTIPHASIC PERSONALITY INVENTORY
DEPENDENCY SCALE ITEMS
Item No.
9 I am about as able to work as I ever was.
19 When I take a new job, I like to be tipped off on
who should be gotten next to.
21 At times I have very much wanted to leave home.
24 No one seems to understand me.
41 I have had periods of days, weeks, or months when I
couldn't take care of things because I couldn't
"get going."
65 I loved my father.
67 I wish I could be as happy as others seem to be
70 I used to like drop-the-handkerchief. .
79 My feelings are not easily hurt.
82 I am easily downed in an argument.
86 I am certainly lacking in self-confidence.
98 I believe in the second coming of Christ.
100 I have met problems so full of possibilities that I
have been unable to make up my mind about them.
107 I am happy most of the time.
138 Criticism or scolding hurts me terribly.
141 My conduct is largely controlled by the customs of
those about me.
158 I cry easily.
102
103
Item No.
163 I do not tire quickly.
165 I like to know some important people because it
makes me feel important.
170 What others think of me does not bother me.
180 I find it hard to make talk when I meet new people.
189 I feel weak all over much of the time.
193 I do not have spells of hay fever or asthma.
201 I wish I were not so shy.
212 My people treat me more like a child than a
grown-up.
236 I brood a great deal.
239 I have been disappointed in love.
259 I have difficulty in starting to do things.
264 I am entirely self-confident.
267 When in a group of people I have trouble thinking
of the right things to talk about.
304 In school I found it very hard to talk before the
class.
305 Even when I am with people I feel lonely much of
the time.
321 I am easily embarrassed.
337 I feel anxiety about something or someone almost
all the time.
338 I have certainly had more than my share of things
to worry about.
343 I usually have to stop and think before I act even
in trifling matters.
357 I have several times given up doing a thing because
I thought too little of my ability.
104
Item No.
361 I am inclined to take things hard.
362 I am more sensitive than most other people.
369 I like to let people know where I stand on things
375 When I am feeling very happy and active, someone
who is blue or low will spoil it all.
382 I wish I could get over worrying about things I
have said that may have injured other people's
feelings.
383 People often disappoint me.
390 I have often felt badly over being misunderstood
when trying to keep someone from making a mistake
394 I frequently ask people for advice.
APPENDIX B
ITEMS FROM SPIEGEL'S DEPENDENCY
AND SELF-ASSERTION SCALE
105
ITEMS FROM SPIEGEL’S DEPENDENCY
AND SELF-ASSERTION SCALE
1. I would rather have a job in which the work is planned
for me by someone else than a job in which I have to
figure everything out for myself.
2. As a rule I would not hesitate to do something other
people feel is wrong as long as I feel that my goal
or purpose in doing it is a good one.
3. I prefer the sort of job where I am told to do rather
than give orders myself.
4. I almost always speak up and say what I think, even
when I know others may disagree with me.
5. I have a strong desire to be completely independent
and on my own.
6. It is difficult for me to voice my opinion in a group
when the opinion of the other group members is
different than mine.
7. I have a strong need to solve my problems without
help from anyone else.
8. I never back down as long as I am convinced that I’m
right.
9. I don't like to have other people give me advice or
suggestions.
10. I usually assert myself so strongly that I often make
people a little angry or annoyed with me.
11. One way or another, I usually get what I want.
12. I am willing to lose friends or make enemies if
necessary in order to achieve what I believe are
worthwhile purposes or goals in life.
106
107
13. As I rule I would not hesitate to do something other
people feel is wrong as long as I feel that my goal
or purpose in doing it is a good one.
14. I am very easily discouraged.
15. It is difficult for me to voice my opinion in a group
when the opinion of the other group members is
different from mine.
16. I feel that life is so much more pleasant when one
tries to agree with people rather than argue with
them.
17. I am so uncertain of almost everything in life that it
is very difficult for me to have a definite opinion or
take a firm stand on any issue.
18. I often accomplish a difficult task by sheer deter
mination and will power.
19. I never back down as long as I am convinced that I'm
right.
20. I am willing to work very hard day after day to
accomplish something which I feel is worthwhile.
21. I usually assert myself so strongly that I think I
often make people a little angry or annoyed with me.
22. I feel that there are no goals in life worth
struggling to reach.
23. There is really very little in life that I have a
strong desire to accomplish.
24. I am a very frustrated person because I almost never
get my own way about things.
25. Life is such a struggle that much of the time I feel
as if I just can't go on anymore.
26. I almost always speak up and say what I think even
when I know others may disagree with me.
APPENDIX C
INSTRUCTIONS TO RATER
108
INSTRUCTIONS TO RATER
The attached typewritten protocol represents
excerpts from group meetings in which different men engage
in dialogue. You will choose from these excerpts those
statements which you evaluate to be statements of independ
ence . A statement of independence is defined as follows:
Any man's statement of self-reference indicating an
action on his own behalf--in any area--will be defined as a
statement of independence. The statement may be in any
tense, past, present or future.
Independent statements are further categorized as
"spontaneous" or "evoked." If the subject makes the state
ment of independence without reference to a specific
stimulus, it is categorized as a "spontaneous" statement of
independence. For example, "I want to get a job soon," or
"I straightened out my pension problem" are spontaneous
statements of independence.
If the subject makes the statement of independence
as a response to a question on the part of another subject
it is categorized as an "evoked" statement of independence,
even though the response may be a simple "yes" or "that's
right." For example, one member may ask, "Did you tell him
off?" If the other member replies, "That's right," the
109
110
latter has emitted an evoked statement of independence.
In order to determine if a statement is to be categorized
as spontaneous statement of independence, you must have on
the transcript both the question and the answer.
At the present time we are interested only in a
gross measurement of the frequency with which both
spontaneous and evoked statements of independence are made.
It will not be necessary therefore for you to keep a record
of which member makes the statement, but only how many
times such statements are made within a given hour. On the
left-hand side of the protocol are two columns: "Spontane
ous" and "evoked." For each statement that you evaluate as
a statement of independence, place a tally mark beside the
statement in the appropriate column and underline the
statement with a pencil.
For experimental purposes, a statement is defined
and counted by the measure of "utterance unit." An utter
ance unit is usually defined as stretches of speech
bounded by a change in speaker. For example, Member A may
say several sentences consecutively. When Member B speaks
(or interrupts) that is the end of Member A1 s utterance
unit. Under normal circumstances you would evaluate
whether or not a statement of independence existed within
Member A*s utterance unit.
However, a variation of the utterance unit proce
dure will be followed with the present protocol, since you
will be working with visual, rather than auditory, data.
An utterance unit will be defined as stretches of speech
bounded by three elipses, thus: . . . You will evaluate
the utterance unit thus bounded to see if it contains a
statement of independence. If you judge that there is more
than one statement in the utterance unit, underline the two
or more statements and place one tally only in the appro
priate column.
SAMPLE DATA SCORED BY E 112
Group 5, Session 3, Week #3
N. . . . I want to know the conditions under which you best
work. Would you rather have a job where things are planned
for you or where you get the job idea and do most of the
decision-making yourself?
517.... Jl like to figure things out myself.
N. . .. You would?
517.... Except in an incident where I don't think I am
capable and I will give it to someone higher.
N. . . . How about you Mr. E?
S19. . . . _I like to be on my own.
N.... Good.
519.... Things I don't know I give to someone higher up,
but things I_ know, 1 might as well do them myself.
N.... If someone sort of hangs over you all the time?
519.... No, I don't go for that at all.
N.... Mr. F., would you rather have a job figured out for
you or do it yourself?
520.... I_ would rather go ahead and figure it out mvself.
You can't teach someone how to do anything. You can show
them how to do it and then they can figure it out them
selves.
N.'v . . What kind of a job did you have outside that let you
figure things out pretty much yourself?
520.... I_ drove a bus for 15 years. They made the schedule
but 1^ did it the way 1_ thought it should be done.
N.... Good. I get the impression that you men like to give
orders and not take them.
520.... I don't know where you get that. I can give and
take them.
N.... There are some people who are much more comfortable
if they don't have to make decisions.
113
518.... Oh, that is true. They don't want to take the
responsibility.
N. ... Right.
517.... Like my job I have right now. I give orders
alright, but I don't put it in the phrase of an order.
1 put it in the phrase of a request. X ask them to do
something.
114
SAMPLE DATA SCORED BY RATER
iGroup 5, Session 3, Week #3
|N.... I want to know the conditions under which you best
iwork. Would you rather have a job where things are planned
!for you or where you get the job idea and do most of the
;decision-making yourself?
| S17.... I . like to figure things out myself.
IN.... You would?
i
i
i S17.•.. Except in an incident where I don't think I am
|capable and I will give it to someone higher.
N. . .. How about you, Mr. E?
519.... I like to be on my own.
519.... Things I don't know I_ give to someone higher up,
i but things I know, I might as well Ho them myself.
iN. ... If someone sort of hangs over you all the time?
|S19.... No, I don't go for that at all.
I
N. . .. Mr. F., would you rather have a job figured out for
i you or do it yourself?
!S20.... I_ would rather go ahead and figure it out myself.
You can't teach someone now to do anything. You can show
them how to do it and then they can figure it out them
selves .
|N.... What kind of a job did you have outside that let you
figure things out pretty much yourself?
520.... I drove a bus for 15 years. They made the schedule
i but I dicT it the way I thought it should be done.
j
N.... I get the impression that you men like to give orders
and not take them.
:S20.... I don’t know where you get that. I can give and
take them.
N. . .. There are some people who are much more comfortable
if they don't have to make decisions.
S18.... Oh, that is true. They don't want to take the
responsibility.
S17. . . . Like my job I have right now. I give orders
alright, but I don*t put it in the phrase of an order.
I put it in the phrase ofTa request. I ask-"them to do
something.
APPENDIX D
MASTER TABLE OF INDIVIDUAL SUBJECT SCORES
DE SCALE, SPE, AND BUTTERFLY DATA
MASTER TABLE OF INDIVIDUAL SUBJECT SCORES: DE SCALE, SPE, AND BUTTERFLY DATA
Butterfly
Group Group
Ss
De SPE Bf
Pre Post
Change
Diff Pre Post
Change
Diff Pre
Pre
Int
Post
De
25* 16 38 38 0 5 6** 6**
26 12
■ - -
32 30 -2 6 6 6
I
_
27 16
- -
30 30 0 4 6 6
(N=6)
-
28 12
- -
43 43 0 5 6 6
_
29 13
- -
30 29 -1 4 6 6
-
30 22
- -
29 32 +3 3 6 6
Md=14.5 Md=31.0 Md=31.0 M=0 M=4.5 M=6 M=6
31 16 14 -2 30 25 -5 5 4 3
_
32 32 38 +6 30 25 -5 3 3 2
II
33 37 43 +6 29 24 -5 3 2 1
(N=5)
_
34 15 28 +13 28 34 +6 4 5 2
-
35 41 49 +8 28 22 -6 3 3 1
T=+31 T=-15
Md=19.5 MD=38 M=H-6.2 Md=29 Md=25 M=-3 M=3.6 m=3.4 M=1.8
ci
1 19 13 -6 29 35 +6 1 1 1
Ci
2 28 28 0 27 36 +9 2 2 2
Ci
3 31 30 -1 40 37 -3 2 1 1
III
ci
4 15 19 +4 36 34 -2 1 2 2
(N=8)
9 7 8 +1 29 30 +1 1 1 1
C
10 11 9 -2 29 34 +5 1 1 1
11 45 54 +9 30 17 -13 2 1 1
C3
12 40 52 +12 28 46 +18 2 1 1
J
T=+T7 T=+2l
Md=23.5 Md=23.5 M=2.1 Md=2’ 9 Md=34.5 M=+2.6 m-i.5 M-l. 3 M=1.5
*Ss numbers over 24 are arbitrarily assigned. j - j |
**None showed.
MASTER TABLE OF INDIVIDUAL SUBJECT SCORES: DE SCALE, SPE, AND BUTTERFLY DATA (continued) j
De SPE Bf
Butterfly
Group Group
Ss
Pre Post
Change
Diff Pre Post
Change
Diff Pre
Pre
Int
Post
De
E?
5 16 14 -2 26 36 +10 1 1 2
E2
6 27 22 -5 28 41 +13 1 1 1
7 26 17 -9 27 28 +1 1 1 2
E2
8 17 17 0 39 40 +1 1 1 1 !
E4
13 38 34 -4 26 39 +13 2 1 1
E4
14 36 28 -8 29 34 +5 2 2 2
E4
15 31 27 -4 42 47 +5 2 2 3
(N=16)
E4
16 31 23 -8 30 34 +4 2 1 2
E
E5
17 34 24 -10 28 38 +10 2 2 1
E5
18 28 33 -6 29 35 +4 2 2 2
E5
19 7 8 +1 28 35 +7 1 1 1
E5
20 38 41 +3 30 34 +4 1 1 1
E6
21 8 5 -3 26 46 +20 1 2 1
22 15 12 -3 29 32 +3 1 1
2 s
A
E6
23 39 44 +5 37 39 +2 2 2 2 !
24 31 29 -2
t=-55
29 32 +3
T=+I03
2 2 2 !
Md=2'9.5 Md=Z2.5 M=-3.4 Md=29 Md=35.5 M=+6.5 M=1.5 M=1.33 M=1.5 !
APPENDIX E
SUPERVISORS PERFORMANCE EVALUATION
SUPERVISORS PERFORMANCE EVALUATION
MEMBER____________________ DATE___
SECTION______JOB CLASSIFICATION______DATE EMPLOYED________
INSTRUCTIONS: Open booklet and circle appropriate value
for each Factor. Total these number values and place on
cover page in space reading: Total Rating Score. Rate
each Factor independently and as objectively as possible.
The value of the Performance Scale increases in proportion
to the objectivity of the rating.
TOTAL RATING SCORE: ______________
RATING CLASSIFICATION: __________
Check one
10-15 Poor
16-25 Fair
26-35 Good
36-45 Excellent
46-50 Outstanding
I have discussed each phase of this rating with the member.
Signature of Rating Supervisor_________________Date_________
Signature of Employee____ ______________________ _Date________
Approved for PM&R:______________________________Date_________
Return to Industrial Therapy Office, Room 26, Building 218.
*N0TE: Please use back page for remarks.
Unsatis factory
Satisfactory
120
CIRCLE APPROPRIATE NUMBER FOR EACH EVALUATION
5 4 3_______ 2________ 1
1. Dependability.
Rarely absent.
When absent has
good reason.
Notifies super
visor . Rarely
arrives late.
Loses no time
unnecessarily
on job.
1
Absent fewer
than average. Has
good reason for
absence and noti
fies supervisor.
Generally punc
tual . Rarely
neglects work.
2
Average absence.
Notifies super
visor but some
times lacks
legitimate.
reason. Average
loss of time on
job. Generally
punctual.
3
Absent more than
average number
of days. Does
not always
notify super
visor or to avoid
loss of time on
j ob. Frequently
tardy.
4
Frequently absent;
without suffi
cient reason. 1
Frequently neg
lects to notify
supervisor.
Habitually late.
Needs constant
supervision to
avoid loss of
time.
5 1
2. Reaction to
Criticism.
: Resent company
I policies. Takes
antagonistic
attitude.
Resents orders
or suggestions.
Refuses to help
others. Diffi
cult to work
with.
Reluctant to
accept company
policies and
rules. Cooper
ates when he
has to. Indif
ferent attitude
to supervisor's
suggestions.
Reluctant to
help others.
Accepts company
policies and
adheres to
rules. Accepts
criticisms of
supervisor
without resent
ment . Follows
suggestions.
Generally enthu
siastic about
his job and
company. Follows
orders cheer
fully. Good
teamworker.
Displays whole
some attitude.
Interest in job |
and company is
outstanding.
Offers help to
others on his
own initiative.
Seeks advice j
from supervisor. ;
Offers worth- |
while sugges
tions . 1
I
i
5 4 3 2_______ 1
3. Ability to Learn Adaptable, can do
most jobs he is
trained for.
Capable of learn
ing more. Gener
ally grasps new
ideas.
2
With normal in
struction, can
do related work
satisfactorily.
Has working
knowledge of
job.
3
Slow learner.
Requires consid
erable instruc
tion and check.
Slow to adapt.
4
Very slow
learner. Very
dull memory.
Not adaptable to i
new situations ‘
or assignments.
i
5
and Adjust.
Learns rapidly.
Has capacity for
doing difficult
work. Highly
adaptable.
1
4. Responsibility.
Irresponsible.
Needs constant
supervision.
1
Does assigned
work reluctantly.
Cannot be
depended upon to
see things
through.
2
Accepts but does
not seek respon
sibility. Needs
normal supervi
sion.
3
Willing to ac
cept responsi
bility for job.
Needs little
supervision.
4
Exceptional
sense of duty.
Very willing to
accept responsi- i
bility for job
and see it
through with
minimal supervi
sion.
5
5. Judgment.
Seriously lack
ing in good
judgment. No
sense of propor
tion. Is rash.
Constant super
vision.
Makes immature
decisions. Often
"jumps" to con
clusions . Poor
sense of values.
Ordinarily tries
to reach conclu
sions from con
crete facts.
Generally learns
from experience.
Generally grasps
situation and
draws correct
conclusions.
Uses good sense.
Superior ability j
to reason intel
ligently and
independently.
Uses sound judg- j
ment. Reaches
conclusions j
logically and i
rapidly.
I
ro
t o .
5 4 3________________ 2_________________ 1
6. Follows Instruc
tions. Follows
instructions
exactly. Well
acquainted with
practices and
procedures
related to job.
Alert.
1
Better than aver
age in adherence
to instructions.
Needs little
help. Observes
procedures.
2
Generally
follows instruc
tions . Needs
occasional help
in observance of
practices and
procedures.
3
Frequently fails
to follow
instructions. Not
well acquainted
with practices
and procedures.
4
Ignores instrue- J
tions. Indif
ferent to prac
tices and
procedures. Is
unreliable.
5
7. Volume. Works
very rapidly and
skillfully. Has
great capacity.
Does not hinder
others in their
work. Requires
minimal supervi
sion.
1
Consistently high
in volume of
work. Requires
occasional super
vision to insure
good output.
2
Average volume
of work. Moder
ate speed.
Needs normal
supervision.
3
Below average
in volume of
work. Is "slack”
at times.
Requires close
supervision.
4
Hopelessly slow.
Cannot keep up
with standards.
Requires constant
supervision and
help.
5 i
8. Accuracy of
Work. Consist
ently inaccu
rate. Needs
constant super
vision. Work
produced usually
unacceptable.
Does not meet
standards of
accuracy. Is
not careful to
avoid repetition
of mistakes.
Pays little
attention to
neatness.
Generally
follows instruc
tions . Needs
occasional help
in observance of
practices and
procedures.
Consistent in
doing work neatly
and accurately.
Constantly
strives for
greater accuracy
and neatness.
Exceptional
attention to
details. Con- j
scientious, j
dependable, j
accurate and
neat. Minimum 1
supervision j
necessary. k
ho
oj;
1 2 3________ 4_______________ 5
9. Personal
Qualities. Makes
poor impression.
Indifferent
toward cleanli
ness and good
manners. Not
trustworthy.
5
Often negligent
in manners of
cleanliness.
Lack sociabil
ity.
4
Does not make
definitely bad
impression by
uncleanliness
or ill manners.
Generally
sociable with
others. Can
be trusted.
3
Usually clean in
person. Had
good manners.
Respects feel
ings of others.
2
Pleasant manner, j
Sociable dispo- 1
sition.
Unusually clean ;
in person and
habits. Very
trustworthy. ;
1
10. Initiative.
Seeks and sets
additional
tasks for him
self. Highly
ingenious. Self
starter. ideas
are realistic
and sound.
Resourceful.
Develops assign
ments ably.
Occasionally
contributes
original ideas.
Fairly progres
sive . With
enc our agement
and assistance
offers sugges
tions to improve
job.
Rarely makes
suggestions.
Needs excessive
instructions.
Lacks self
reliance. Needs 1
constant prod
ding.
i
i
i
I-1
K> |
4>\
i
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The Effects Of Two Types Of Experimenter Intervention And Schedules Of Reinforcement On Verbal Operant Conditioning Of Affective Self-References
PDF
The Influence Of The Achievement Motive, The Affiliation Motive, And Incentive Conditions On Roleplaying Ability In Children
PDF
The Effect Of Conditions Of Risk, Internal Versus External Control Of Reinforcement, And Sex On Binary Choice Probability Learning
PDF
Prognostic Expectancy Effects In The Desensitization Of Anxiety Over Invasion Of Body Buffer Zones
PDF
Effects Of Different Treatment Procedures On Reading Ability And Anxiety Level In Children With Learning Difficulties
PDF
Use Of College Students In A Social Therapy Program With Chronic Schizophrenics To Produce Changes In Mood And Social Responsiveness
PDF
Relationship Of Achievement Motivation To Perception Of Degree Of Task Difficulty And Estimate Of Performance
PDF
The Relationship Of Dependency To Verbal Learning Without Awareness
PDF
Human Operant Eye Blink Conditioning, Awareness, And The Extraversion-Introversion Dimension Of Personality
PDF
Human Gsr Classical Conditioning And Awareness Of The Cs-Ucs Relation
PDF
An Experimental Analogue Comparing Two Models Of Psychotherapy: Reciprocal Inhibition And Implosive Therapy
PDF
Effects Of Group Behavior Therapy Imagery On Basketball Performance
PDF
The Modification Of Maladaptive Behavior Of A Class Of Educationally Handicapped Children By Operant Conditioning Techniques
PDF
The Effects Of Anxiety And Threat On Self-Disclosure
PDF
Inhibitory-excitatory interaction of conditioned autonomic responses through interoceptive and exteroceptive conditioning
PDF
Confirmation Of Expectancy And Changes In Teachers' Evaluations Of Student Behaviors
PDF
The Effect Of Discriminability On The Partial Reinforcement Effect In Human Gsr Conditioning
PDF
Differences Between Cues In Effectiveness As Retrieval Aids
PDF
Dynamic Behavior Of Soap Films Comments On Thinning Of Isolated Films And collapsing Of Burst Films
PDF
An Effect Of Verbal Conditioning On Nonverbal Behavior
Asset Metadata
Creator
Orlando, Nita Ethel
(author)
Core Title
Verbal Conditioning Of Dependence-Independence In Domiciled Veterans
Degree
Doctor of Philosophy
Degree Program
Psychology
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,Psychology, clinical
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Navran, Leslie (
committee chair
), Slucki, Henry (
committee member
), Vold, Majorie J. (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c18-637734
Unique identifier
UC11361097
Identifier
6900637.pdf (filename),usctheses-c18-637734 (legacy record id)
Legacy Identifier
6900637.pdf
Dmrecord
637734
Document Type
Dissertation
Rights
Orlando, Nita Ethel
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