31295005501662

DEVELOPING A SCALE TO MEASURE CODEPENDENCY
by
LYNDA ANNE SPANN, B.A.
A THESIS
IN
FAMILY STUDIES
Submitted to the Graduate Faculty

of Texas Tech University in
Partial Fulfillment of
the Requirements for
the Degree of
MASTER OF SCIENCE
' IN
HOME ECONOMICS
Approved
May, 1989
13
Copyright 1989, Lynda A. Spann

ACKNOWLEDGMENTS
I want to express my great appreciation to Judith L. Fischer, chairperson of my
committee, for her expert guidance and gentie support from the inception of this research
project I also appreciate the significant contributions of the other committee members. Dr.
Jean Scott and Dr. Nancy Bell. I am grateful to Carol Jean Schoenrock for her expert help
in reviewing the original scale, for her assistance in collecting the student sample data, and
mostly for her understanding and inspiration throughout this research experience.
I am grateful to Leigh Mires for her belief in me and encouragement throughout this
project. I am deeply grateful to my parents, Jimmie and Norma Spann, for the undying
faith and acceptance they have given me throughout my life and throughout this latest
educational endeavor. Finally, I thank my friend Anna Wegleitner, my siblings, and many
other supportive friends who have been truly interested and helpful in many ways during
this experience.
11
CONTENTS
ACKNOWLEDGMENTS ri
LIST OF TABLES v
LIST OF FIGURES vi
L INTRODUCTION 1
Statement of the Problem 1
Codependency 2
Research Objectives 11
Limitations of the Study 12
Conclusion 12
n. REVIEW OF THE LITERATURE 13
Introduction 13
Theoretical Foundations for Codependency 13
Empirical Foundations for Codependency 19
Hypotheses 27
in. METHODOLOGY 30
Subjects 30
Measures 31
Procedures 38
IV. RESULTS 39
Descriptive Statistics 39
Preliminary Analysis 48
Main Analysis 49
Additional Analysis 54
ui
V. DISCUSSION 56
Findings 57
Conclusions and Recommendations 61
REFERENCES 64
APPENDICES 69
A. BEM SEX ROLE INVENTORY 69
B. MARLOWE-CROWNE SOCIAL DESIRABILITY SCALE 72
C. SHORTENED VERSION OF THE INTERNAL VS.
EXTERNAL CONTROL OF REINFORCEMENT SCALE 75
D. ROSENBERG SELF-ESTEEM SCALE 78
E. ORIGINAL 38-ITEM CODEPENDENCY SCALE 80
F. DEMOGRAPHIC QUESTIONNAIRE 83
IV
LIST OF TABLES
1. Descriptive Statistics of Total Sample (N=236) 40

2. Descriptive Statistics of Student Sample (N=192) 42
3. Descriptive Statistics of Al-Anon Sample (N=30) 44
4. Descriptive Statistics of Codependent Sample (N=14) 46
5. Unrotated Factor Loadings for First Factor Codependency Scale 50
6. Correlation Matrix of Codependency Scale with Other Scales 51
LIST OF HGURES
1. Characteristics of Codependency 9
2. 15-Item Codependency Scale 37
VI
CHAPTER I
INTRODUCTION
Statement of the Problem
In the recent past, there has been a surge of discussion and writing on codependency
(Beattie, 1987; Bepko & Krestan, 1985; Capell-Sowder, 1984; Cermak, 1986b; Gorski &
Miller, 1984; Greenleaf, 1984; Schaef, 1986; Subby, 1984; Subby & Friel, 1984;
Wegscheider-Cruse, 1985; Whitfield, 1984; Woititz, 1984). However, this discussion has
remained theoretical with no empirical research to support it. The present study is
designed to address the need to strengthen the concept of codependency by developing a
reliable and valid scale to measure codependency.
In order to come up with a codependency scale, existing theories on codependency
were drawn upon in developing scale items. These items were then reviewed by experts in
the field and further refined. Therefinedversion of the scale was piloted by giving it twice
to the same group of students with a two-week period in between each administration.
Test-retest reliability and internal consistencyreliabihtywere computed and scale
modifications were carried out as guided by these findings.
In the present study, the scale was administered twice, once to known groups
thought to be high or low in codependency, second, college students were assessed to
establish convergent and discriminant validity. The codependency scale was correlated
with other scales which were thought to be related and unrelated to codependency.
Developing a scale to measure codependency should serve the trifold purpose of
bridging theory, research, and practice. First, the development of a codepencency scale
should clarify and make more concrete the meaning of the concept of codependency, thus
providing a specific operational definition of the term. Second, areliableand valid scale
1
should facilitateresearchin the field. Third, measurement of codependency could aid in
the pursuit of self-awareness and self-determined change. Clinicians are increasingly
using the term codependency in their practices. To date these clinicians do not have a
standardized tool to aid them in assessing codependency. An assessment tool for
codependency is needed, particularly in view of the fact that several treatment programs are
cropping up in the United States designed exclusively to treat codependency. These
treatment programs need to have at their disposal an objective test to measure
codependency which can be scored quickly and accurately.
The purpose of the study is to move beyond the theories of codependency to the
areas of research and practice through developing a measure for codependency. The
significance of the present study lies not only in the development of a scale, but in the
futureresearchand clinical contributions that mayresultfrom the availability of a reUable
and valid measure of codependency.
Codependencv
In order to develop a complete conceptual definition of the tenn codependency, it is
important to review some of the definitions articulated by pioneers in this field. When
defining codependency, theorists in the field often cite symptoms,reactions,problems,
and coping mechanisms that codependents have in common. There is quite a bit of overlap
and agreement among these theorists about the characteristics of codependency.
Definitions of Codependency
Several issues have emerged withrespectto the definition of codependency. One
major issue that can serve as an artificial divider of theorists' definitions of codependency
is the issue of whether or not they view codependency as being linked to alcoholism or
drug addiction. Several theorists use the term co-alcoholism when describing the
phenomenon of codependency.
In "Co-Dependency: An Emerging Problem Among Professionals" (1984), Charles
Whitfield states, "Co-alcoholism can be defined as ill health, or maladaptive or problematic
behavior that is associated with living, working with, or otherwise being close to a person
with alcoholism" (p. 45). Although he limits himself to the context of alcoholism,
Whitfield maintains a broader perspective on co-alcoholism or codependency than do some
authors in saying, "It affects not only individuals, but families, communities, businesses
and other institutions, and states and countries" (p. 45). Gorski and Miller (1984) have a
narrower perspective on the term co-alcoholism. They state, "Co-alcoholism is a primary
condition thatresultsfrom the debilitating physiological stress produced by living in a
committed relationship with an alcoholic or drug dependent person" (p. 77). In "The
Responsibility Trap: A Blueprint for Treating the Alcoholic Family" (1985), Bepko and
ICrestan describe co-alcoholics as "parents and children of alcoholics and nondrinking
spouses" (p. 33). Woititz (1984) also addresses codependency by using the tenn
co-alcoholic, but does not offer a definition of co-alcoholism.
Jael Greenleaf (1984) draws a distinction between the teirns co-alcoholic and
para-alcoholic. Greenleaf examines the "differences in the derivation of behavior of an
adult in a volitionalrelationshipwith an alcoholic spouse and of a child in a compulsory
relationship with a parent" (p. 5). She states, "The prefix co- means with or necessary for
the functioning of. The adult who assists in maintaining the social and economic
equilibrium of the alcoholic person is indeed co-alcohoUc... The prefix para- means like
or resembling. The child who grows up in a family with the alcoholism syndrome leams
behavior from both parents and becomes para-alcoholic" (p. 6).
Kathy Capell-Sowder (1984) uses the term codependency and describes a
codependent, saying, "the person involved in a primary loverelationshipwith someone

addicted, frequentiy displays symptoms of addiction him/herself in the ways that he/she
relates to the relationship" (p. 19). Capell-Sowder goes on to parallel the symptomatology
of codependent addiction to alcohol and drug addiction. She writes that the alcohohc
experiences a growing preoccupation with alcohol, an increasingly higher tolerance of
alcohol, a loss of control, denial of problem, major life areas spiralling downward (e.g.,
family relationships, friends, job, health, finances), and a loss of power of choice.
Similarly, the codependent experiences an increasing focus on the alcoholic, an increase in
tolerance to unacceptable behavior from the drinker, loss of control of drinker and of own
emotions and reactions, denial that she is affected by the drinking lifestyle, decline in major
life areas (e.g., family friends, job, health, spiritual life), and loss of ability to see options
(Capell-Sowder, 1984).
As demonstrated above, many theorists view codependency as being solely related
to alcoholism or drug addiction. However, there are an equal number of theorists who
have a broader perspective on codependency, viewing it as a condition that is not
necessarily linked to chemical dependency.
Although Sharon Wegscheider-Cmse (1985) places an emphasis on alcoholism, she
defines codependency as "a specific condition that is characterized by preoccupation and
extreme dependence (emotionally, socially, and sometimes physically), on a person or
object. Eventually, this dependence on another person becomes a pathological condition
that affects the codependent in all other relationships" (p. 2). Similarly, Beattie (1987)
states, "A codependent person is one who has let another person's behavior affect him or
her, and who is obsessed with controlling that person's behavior."
In "Co-Dependency~A Paradoxical Dependency" (1984), Subby and Friel write,
"Let us define co-dependency, then, as a dysfunctional pattern of living and
problem-solving which is nurtured by a set of rules within the family system. These rules
make healthy growth and change very difficult" (p.32). Subby (1984), who views
codependency asrelatedto many unhealthyrealitiesother than alcoholism, defines
codependency as "An emotional, psychological and behavioral condition that develops as a
result of an individual's prolonged exposure to, and practice of, a set of oppressive
rules—rules which prevent the open expression of feeling as well as the direct discussion of
personal and interpersonal problems" (26). In "Co-Dependence: Misunderstood -
Mistreated" (1986), Schaef does not offer a concise definition of codependency, but rather
describes it as being a sub-disease of what she calls the addictive process. Schaef (1986)
writes, "The addictive process is an unhealthy and abnormal disease process, whose
assumptions, beliefs, behaviors, and lack of spirituality lead to a process of nonliving that
is progressively death-oriented" (p.21).
In "Diagnosing and Treating Co-Dependence" (1986a), Cermak defines
codependence as "arecognizablepattem of personality traits, predictably found within
most members of chemically dependent families, which are capable of creating sufficient
dysfunction to warrant the diagnosis of Mixed Personality Disorder as outiined in DSM
ni" (p. 1). Cermak also writes, "There is no reason to expect exportation of the concept
to the rest of the mental health field until it can be packaged into well-defined bodies of data
and not merely described by eloquent metaphors" (1986b, p. 18).
Characteristics of Codependency
In "Co-Dependency: An Emerging Issue" (1984), several authors (Greenleaf;
Subby & Friel; and Woititz) write of common characteristics of codependency. In
describing co-alcoholics and para-alcoholics, Jael Greenleaf (1984) lists behavioral
symptoms that are often exhibited by these individuals. Those behaviors that are common
among codependent people (co-alcoholics and para-alcoholics) are: grandiosity, lack of
trust, blaming/projecting, judgmentalness, lying, depression,flattenedaffect, solitariness.

and disniptiveness (Greenleaf, 1984). Subby and Friel (1984) also list some of the
common characteristics of codependency:
1. Difficulty in identifying feelings.

2. Difficulty in expressing feelings.
3. Difficulty in forming or maintaining close relationships.
4. Perfectionism.
5. Rigidity in attitudes and behavior.
6. Difficulty adjusting to change.
7. Feeling overlyresponsiblefor other people's behavior or
feelings.
8. Constant need for other's approval in order to feel good about
self.
9. Difficulty making decisions.
10. General feelings of powerlessness over one's life.
11. A basic sense of shame and low self-esteem over perceived
failures in one's life.
In "The Co-Dependent Spouse: What Happens To You When Your Husband Is An
Alcoholic" (1984), Woititz provides a Ust of symptoms of co-alcoholism, and then
expounds upon each symptom. According to Woititz, "The following are the symptoms of
co-alcoholism: denial; protectiveness, pity-concern about the drinker; embarrassment,
avoiding drinking occasions; shift in relationship; guilt; obsession, continual worry; fear;
lying; false hope, disappointment, euphoria; confusion; sex problems; anger, lethargy,
hopelessness, self-pity, remorse, and despair" (p. 84).
In "Choice-Making" (1985), Sharon Wegscheider-Cruse writes, "Co-dependency is
a lifestyle, a patterned way of relating to others. It's a way of interpreting experience.
And it's a lifestyle with low self-esteem at the core" (p. 3). Wegscheider-Cmse suggests
that codependency is characterized by:
An inability to have spontaneous fun, and inability to let go.

Problems with intimacy.
Inability to know what normal behavior is.
An exaggerated need for the approval of others.
Confusion about making decisions.
Anxiety about making changes.
Black and white judgments.
Fear and denial of anger.
Lies and exaggeration, when it would be easy to tell the truth.
Fear of abandonment.
Tendency to look for people to take care of.
Need to control self and others.
Wegscheider-Cruse (1985) also delineates four progressive stages of codependency:
dependent bonding, fear, emotional paralysis, and behavioral stuckness.
In "Diagnosing and Treating Co-Dependence" (1986b), Cermak suggests the
following outiine as a "Diagnostic Criteria for Co-Dependent Personality Disorder" (p.l 1):
A. Continued investment of self-esteem in the ability to control

both oneself and others in the face of serious adverse
consequences.
B. Assumption of responsibility for meeting others' needs to the
exclusion of acknowledging one's own.
C. Anxiety and boundary distortions around intimacy and
separation.
D. Enmeshment in relationships with personahty disordered,
chemically dependent, other co-dependent, and/or impulse
disordered individuals.
E. Three or more of the following:
1. Excessive reliance on denial
2. Constriction of emotions (with or without dramatic
outbursts)
3. Depression
4. Hypervigilance
5. Compulsions
6. Anxiety
7. Substance abuse
8. Has been (or is) the victim of recurrent physical or sexual
abuse
9. Stress related medical illnesses
10. Hasremainedin a primaryrelationshipwith an active
substance abuser for at least two years without seeking
outside help.
Ann Wilson Schaef dedicates an entire chapter to the characteristics of codependence
in her book entitled "Co-Dependence: Misunderstood-Miso-eated" (1986). Schaef outlines
eleven major codependency issues:

1. Extemal Referenting
Relationship addiction
Cling-clung relationships
Lack of boundaries
Impression management
Not trusting own perceptions
2. Caretaking
Making yourself indispensable
Being a martyr
8
3. Physical Illness
4. Self-Centeredness
5. Control Issues
6. Feelings
Being out of touch with feelings
Distorted feelings
7. Dishonesty
8. Being Central
9. GuUibiUty
10. Loss of morality
11. Fear,rigidity,judgmentalism
In "Codependent No More" (1987), Melody Beattie offers tiie most comprehensive
list of characteristics of codependency. Although it is not feasible to reproduce the list in
its entirety here, it is important to note that Beattie divides the characteristics mto several
major groupings: caretaking, low self-worth, repression, obsession, controlling, denial,
dependency, poor communication, weak boundaries, lack of trust, anger, sex problems,
and progressive issues such as emotional, mental, or physical illnesses.
Inreviewingthe existing literature on codependency, it is apparent that there exists
some commonality and agreement among theorists about the characteristics of
codependency. Although theorists vary on descriptions and wordings, several key
elements of codependency emerge in common. Figure 1 presents a box score of common
descriptors of codependency. Thus, the characteristics of codependency that have been
identified as important for the development of a codependency scale are: dependency,
obsession, caretaking, repressed feelings, lack of trust, poor communication, low
self-worth, control, denial, and rigidity.
It was the purpose of the present study to provide a clear conceptual definition of
codependency and to back that definition up with a body of valid statistical data. Taking
into consideration the existing definitions, the common characteristics noted above, and
this author's own understanding of codependency, the following has emerged as the
working definition of codependency for the proposed study: Codependency is a
psychosocial condition that is manifested through a dysfunctional pattem of relating to

Characteristics of Codependencv Authors Citing Characteristic
1. Lack of trust in self or others. Beattie (1987); Greenleaf

(1984); Schaef (1986)
2. Judgmentalness. Greenleaf (1984); Schaef

(1986)
3. Lying, dishonesty. Greenleaf (1984); Schaef

(1986); Wegscheider- Cmse
(1985); Woititz (1984)
4. Difficulty identifying feelings. Beattie (1987); Schaef (1986);
Subby & Friel (1984)
5. Difficulty expressing feelings/ Beattie (1987); Greenleaf
poor communications. (1984); Schaef (1986);
6. Rigidity; difficulty adjusting to Scheaf (1986); Subby & Friel
change. (1984); Wegscheider-Cruse
(1985)
7. Difficulty making decisions. Subby & Friel (1984);
Wegscheider-Cmse (1985)
8. Low self-esteem. Beattie(1987); Schaef (1986);
9. Denial. Beattie (1987); Schaef (1986);
Wegscheider-Cmse (1985);
Woititz (1984)
10. Obsession. Beattie (1987); Woititz
(1984);
11. Fear. Schaef (1986); Wegscheider-
Cmse (1985); Woititz (1984)
12. Sex problems. Beattie (1987); Woititz

(1984)
HGUREl
Characteristics of Codependency
10
13. Anger. Beattie (1987); Wegscheider-

Cruse (1985); Woititz (1984)
14. Progressive issues to Beattie (1987); Schaef (1986);
include illlness. Woititz (1984)
15. Caretaking; taking responsibility Beattie (1987); Cermak
for meeting others' needs to the (1986); Schaef (1986);
exclusion of self. Wegscheider-Cmse (1985)
16. Need to control. Beattie (1987); Cermak
(1986); Schaef (1986);
17. Boundary distortions; problems Beattie (1987); Cermak
with intimacy. (1986); Schaef (1986);
18. Dependency; need for approval Beattie (1987); Schaef (1986);
Subby & Friel (1984);
HGURE 1 - Continued
11
otiiers. This dysfunctional pattem ofrelatingis characterized by: extreme focus outside of
self (dependency, obsession, and caretaking); lack of open expression of feelings,
(repressed feelings, lack of trust, and poor communication); and because of low
self-worth, attempts to derive a sense of purpose and meaning through relationships with
others (control, denial, and rigidity).
Research Objectives
The purpose of the present study is to develop a scale that will reliably and validly
measure codependency. Developing such a scale is a multi-faceted process. This process
includes reviewing theories on codependency, developing a clear working definition of the
term, writing scale items that reflect the characteristics of codependency, and finally testing
the scale forreliabilityand validity. In the study the codependency scale was developed,
reviewed, and critiqued by experts in order to establish face validity. Test-retest
correlations and internal consistency reliability coefficients were calculated to estimate the
reliability of the scale. The next steps involved giving the codependency scale to people
who were thought to be high or low in codependency. In addition, a large group of^
college students were given the scale. Both of these samples were also given several other
scales that are either somewhat related or completely unrelated to codependency.
Convergent and discriminant validity was assessed through correlating these other scales
that should and should not be related to codependency with the codependency scale. With
the completion of the study, there should be a new and useful measure for codependency
that will help begin to fill the need for empirical work on codependency. This assessment
tool will also aid in future research on codependency as well as in the clinical treatment of
codependents.
12
Limitations Qfthg Study
Limitations of the present study should be noted. The pilot study was conducted
with a class of sophomores from a southwestern university located in a politically and
religiously conservative region. Therefore, generalizabiUty to other populations is limited.
That sample was alsorelativelysmall (40 cases). The sample used for the remainder of the
study was taken from the same southwestern conservative community, making
generalizability to other populations limited. The known group samples have a
disproportionately higher number of women than men, because there are many more
women than men that attend Al-Anon and Codependents Anonymous in this community.
The polling of the codependent sample did not meet the total number of participants sought
for the research. The scale is a paper-and-pencil, self-report questionnaire, which is
limited in its ability to measure interactional patterns. A last limitation of the study is that
many of the subjects in the known group that is thought to be high in codependency are
seeking some form of treatment for their codependency, such as attendance at Al-Anon or
Codependents Anonymous.
Conclusion
Along with the growing discussions and writings on codependency there is a need to
strengthen theories with research and good cUnical assessment tools. To date littie research
has been done on codependency, and there is no standardized test to measure
codependency. The present smdy has addressed this need by developing areliableand
valid measure for codependency.

CHAPTER n
REVIEW OF THE LITERATURE
Introduction
Codependency is a term which has only been a part of treatment jargon for less than
a decade. This term seemed to have emerged simultaneously in several different treatment
centers in Minnesota in the late seventies (Beattie, 1987). Originating in the chemical
dependency field, the word codependency was first "used to describe the persons whose
lives were affected as a result of their being involved with someone who was chemically
dependent. The 'co-dependent' spouse or lover of someone who was chemically
dependent was seen as having developed a pattem of coping with Ufe that was not healthy,
as a reaction to someone else's dmg or alcohol abuse" (Subby & Friel, 1984). In more
recent years, several authors have expanded their definition of codependency to include
persons who are involved in relationships with persons other than those who are
chemically addicted (Beattie, 1987; Schaef, 1986; Subby, 1984; Subby & Friel, 1984).
The literature review begins with an overview of the theoretical foundations for
codependency followed by comments on the empirical foundations for codependency. The
Uterature review continues with an in-depth description of other variables thought to be
related to codependency. The chapter concludes with the hypotheses for the proposed
study.
Theoretical Foundations for Codependencv
In reviewing the literature on codependency, it is apparent that several issues seem
to recur among authors widi respect to the definition of codependency. The issue of
whether or not codependency is Umited to chemical dependency was discussed in the
13
14
previous chapter. Several other key issues debated by theorists are whether or not
codependency is: considered an addiction or a pathological or dysfunctional pattem;
characterized by behaviors/symptoms, or by attitudes (mle orientation); andrelatedto lack
of spirituality. Before examining these key codependency issues let usrestatethe working
definition and characteristics of codependency identified above as important for the present
study: Codependency is a psychosocial condition that is manifested through a
dysfunctional pattem ofrelatingto others. This dysfunctional pattem ofrelatingis
characterized by: extreme focus outside of self (dependency, obsession , and caretaking);
lack of open expression of feelings, (repressed feelings, lack of tmst, and poor
communication); and because of low self-worth, attempts to derive a sense of purpose and
meaning through relationships with others (control, denial, and rigidity).
Codependency and Chemical Dependency
As noted in the previous chapter, one of the major issues that divide theorists'
definitions of codependency is the issue of whether or not they view codependency as only
being linked to alcoholism or drug addiction. Most of the authors that view codependency
in this way (Bepko & Krestan, 1985; Gorski & Miller, 1984; Greenleaf, 1984; Whitfield,
1984; and Woititz, 1984) even use die term co-alcoholism when describing the
phenomenon of codependency. These authors have a constricted context from which they
view codependency, and tiiey seem to "come dangerously close to saying that
co-alcoholism or co-dependence is something that is 'caught' from the alcoholic or is part
of the contagion of alcoholism" (Schaef, 1986, p. 17).
There is an equal and growing number of theorists (Beattie, 1987; Cermak, 1986b;
Schaef, 1986; Subby, 1984, 1987; Subby & Friel, 1984; and Wegscheider-Cmse, 1985)
who do not solely link the condition of codependency to chemical dependency. These
theorists acknowledge that codependency is fostered in chemically dependent homes, but

15
do not limit it to that type of family system. Beattie (1987) suggests that other high risk
groups for codependency are people who "love, care about, or work with troubled people"
(p.46). Other authors (Beattie, 1987; Schaef, 1986; Subby, 1984; Subby & Friel, 1984;
and Wegscheider-Cmse, 1985) maintain that codependents come from famihes that are
very rigid and emotionally repressive. Subby (1987) writes that "co-dependency can
emergefiromany family system where certain overt (spoken) and covert (unspoken) rules
exist — mles that interfere with the normal process of emotional, psychological, behavioral
and spiritual development" (p. 15). Although it is important torecognizethat the notion of
codependency first emerged in the arena^f chemical dependency, the term is no longer
Umited to that field and should also be used to help people who are not connected to an
alcoholic/addict.
Addiction, Pathology, or Dysfunction

A few authors view codependency itself as a disease or primary addiction. Both
Capell-Sowder (1984) and Schaef (1986) write of codependency in terms of "addiction" or
an "addictive process." Gorski and Miller (1984) state that co-alcohoUsm is a "primary
condition" (p. 77), and Wegscheider-Cmse (1985) writes that the dependency of a
codependent eventually "becomes a pathological condition" (p. 2). In "Co-Dependency:
The Therapeutic Void," Wegscheider-Cmse (1984) states that "Co-dependency is a
primary disease and a disease within every member of an alcoholic family. It is what
happens to family members when they try to adapt to a sick family system that seeks to
protect and enable the alcoholic" (p. 1).
Several theorists describe codependency in terms of pathological or dysfunctional
pattems. Whitfield (1984) uses the term "ill health" when defining co-alcoholism, and
describes it as a primary illness with a recognizable, diagnosable, and treatable range of
symptoms and a chronic and progressive prognosis. Wegscheider-Cmse (1985) and

16
Schaef (1986) write of codependency as being a pathological condition. While Cermak
(1986a) describes codependency as creating a dysfunction within the individual, he also
describes it as a disease of relationships (1986a, 1986b). Beattie (1987), and Subby and
Friel (1984) write of codependency asrelatedto a dysfunctional pattem of living.
According to several authors, there are drawbacks to identifying codependency as
a character disorder or pathology (Asher & Brissett, 1988; Cermak, 1986b; Gierymski &
Williams, 1986). In "Diagnosing and Treating Co-Dependence," Cermak (1986a) raises
the issue that "the speed of improvement seen in some recovering co-dependents argues
against conceptualizing their condition as a character disorder." He then argues that "the
disease of co-dependence, like the disease of chemical dependence, is tmly a disease of
denial ~ much more so than the traditional personality disorders," and "once a client's
denial system has started to crumble, there is a critical period of vulnerability during which
education is most effective" (p. 101). Both of the articles written by Gierymski and
Williams (1986) and Asher and Brissett (1988) suggest that the concept of codependency
is an incidence of medicalization of behavior that has been labeled deviant; where
codependents are considered to be sick and not bad. These authors caution that such
medicalizations may have negative consequences. Asher and Brissett wam that the
medicalization of codependency is reminiscent of the old model of viewing women married
to alcoholics as psychopathological. They go on to quote Sontag (1978) in saying that
"psychological theories of illness are a powerful means of placing the blame on the ill.
Patients who are instmcted that they have, unwittingly, caused their disease are also made
to feel that they have deserved it" (p.55-56). I prefer the notion of dysfunctional pattem
over the concept of pathology when describing codependency.

17
Behaviors or Attitudes
As noted in the previous chapter, there has been some variability across definitions
of codependency. Cermak states his belief that "definitions of co-dependence have failed
to coalesce because we have been trying to simplify the concept" (1986a, p. 102). He
goes on to explain that codependency is a very complex psychological concept, whose
"complexity arises from the fact that co-dependence simultaneously refers both to
intrapsvchic and to interpersonal dynamics" (p. 102).
Some theorists limit their description of codependency to terms of individual
behaviors or behavioral symptoms (intrapsychic), while others write of attitudes or rule
orientations that arereinforcedin the family (interpersonal). Whitfield (1984) defines
co-alcoholism in terms of "maladaptive or problematic behavior" (p.79). Greenleaf also
writes of CO- and para-alcoholism in terms of behavior. Both Capell-Sowder (1984) and
Woititz (1984) define codependency in terms of behavioral symptoms, while Cermak
(1986a) defines it in terms of personality traits.
It would be difficult to speak of codependency without mentioning characteristic
behavior or behavioral symptoms. However, limiting the discussion of codependency to
behavior suggests that codependency is solely an intraindividual phenomenon.
Codependency is a systemic phenomenon, involving both intraindividual and interpersonal
or interactional dynamics. The definition and discussion introduced above includes the
concept of interpersonalrelatingas a key element in codependency.
Several authors who use behavior to describe codependency do not limit
themselves to behaviors, but include attimdes and family mle orientations as descriptors
and predictors of codependency (Beattie, 1987; Bepko & Krestan, 1985; Schaef, 1986;
Subby, 1984, 1987; Subby & Friel, 1984; and Wegscheider-Cmse, 1985). Attimdes and
family mle orientations develop from an interactional process. Wegscheider-Cmse (1985)
writes, "The bizarre mles and distorted communications in dysfunctional, or troubled,

18
family systems may stunt or impede normal emotional and behavioral development
Anyone who lives in a family of denial, compulsive behavior, and emotionalrepressionis
vulnerable to co-dependency" (p. 5). Subby (1984,1987) and Subby and Friel (1984)
define codependency in terms of oppressive family mles. "Codependency is the product
of delayed or intermpted identity brought about by die practice of dysfunctional rules"
(Subby, 1987, p. 55). It is interesting to note that most of the theorists who do not limit
their description of codependency to individual behavior do not link codependency solely
to chemical dependency.
Spirituality
In "Co-Dependence: Misunderstood-Mistreated," Schaef (1986) includes the
notion of lack of spirituality as an indicator or symptom of codependency.
Wegscheider-Cmse (1985) defines spirituality as "a quahty that infuses life with meaning.
It energizes our lives and ultimately provides us with a rationale for life and death" (p.
195). Schaef (1986) gives several examples of spiritual deterioration, and writes that
spiritual deterioration "is a characteristic of co-dependence. You cannotrecoverfrom this
disease . . . withoutreconnectingwith your spirituality" (p. 62). When addressing the
issue of recovery from codependency, Beattie (1987) and Wegscheider-Cmse (1985) also
include spirituality as an important element of recovery. Since, as defined in the previous
chapter, codependents misguidedly attempt to derive a sense of purpose and meaning
throughrelationshipswith others, it is my opinion that at some level spirituaUty (not
rehgiosity) is essential to the healing or recovery from codependency.

19
Empirical Foundations for Codependencv
Characteristics of Codependency and Related Research
Several of the individual components of codependency are related to other concepts
that have undergonerigorousempirical research. Four of tiiese constmcts are: sex-role
orientation, social desirability, locus of control, and self-esteem. A review of these
concepts and theirrelationto codependency follows.
Sex-Role Orientation
Muchresearchhas been carried out surrounding the sex-role orientation variables
of masculinity and femininity (Bem, 1974, 1977; Lubinski, Tellegen, & Butcher, 1981,
1983; Payne, 1987; Spence & Helmreich, 1978; Spence, Helmreich, & Strapp, 1975;
Whitley, 1983). To assess these variables, researchers typically have used the Masculinity
and Femininity scales of die Bem Sex-Role Inventory (BSRI; Bem, 1974) or the Personal
Attributes Questionnaire (PAQ; Spence, Helmreich, & Stapp, 1974). Bem (1974)
originally wrote that the BSRI characterizes a person as masculine or feminine as a
function of the difference between his or her endorsement of masculine and feminine
personality characteristics, and that "in general, masculinity has been associated with an
instmmental orientation, a cognitive focus on 'getting the job done' and femininity has
been associated with an expressive orientation, an affective concem for the welfare of
others" (Bem, 1974, p. 156).
Spence (1983) suggests that the BSRI and PAQ primarily measure only what their
manifest content indicates, and she cautions investigators to use the BSRI and PAQ only as
measures of particular trait clusters. Payne (1987) writes:
From a content point of view, the Mascuhnity scales would be better

labeled as "indices of dominance" (Wiggins & Holzmuller, 1981),
"dominance-poise" (Tellegen & Lubinski, 1983), "assertiveness," or
"instrumentality" (Spence, 1983)... The Femininity scales, on the other
hand, would be better described as measures of nurturance (Wiggins &
20
HozmuUer, 1981), nurturance-warmth (Tellegen & Lubinski, 1983), or
"expressiveness" (Spence, 1983). (p. 360)
Individuals who score high on masculinity describe themselves as being self-reliant,
assertive, dominant, and forceful on die BSRI and as feeling superior, self-confident, able
to stand up well under pressure, and competitive on the PAQ. Individuals who score high
on femininity describe themselves as being yielding, compassionate, loyal, eager to sootiie
hurt feelings, and sensitive to the needs of others on the BSRI and understanding of
others, warm in relation with others, helpful to others, and able to devote self completely
to others on the PAQ (Spence, 1983). Some of the fundamental characteristics of
codependency noted in the previous chapter, such as extreme focus outside of self through
dependency, obsession, and caretaking, arereflectedin many of these femininity scale
items.
There have been several research studies conducted in the past to assess the notions
of masculinity and femininity and theirrelationto self-esteem (Bem, 1977; Lubinski,
Tellegen, & Butcher; Payne, 1987; Spence, Helmreich, & Stapp, 1975; and Whitley,
1983). Through their 1975 study using the PAQ, Spence et al. found feminine and
undifferentiated subjects to score low in self-esteem and masculine and androgynous
subjects to score high in self-esteem. Bem's (1977) pattem of results using the BSRI
closely matched that reported by Spence et al. (1975). However, Bem found that the
pattem may actually be somewhat different for the two sexes. The men who were high in
masculinity were also high in self-esteem, regardless of where they stood on femininity.
In contrast, women who were high in both masculinity and femininity were highest in
self-esteem, women who were low in both masculinity and femininity were lowest in
self-esteem, and women who were high in one and low in the other fell in between, with
the masculine women perhaps having a bit of an edge.
Whitley (1983) analyzed the relation between sex role orientation and self-esteem, an
indicator of psychological well-being most widely used in sex role studies, through a
21
meta-analysis. His final sample consisted of 35 studies which included a total of 6,424
women and 5,692 men. The results of the meta-analysis indicated that diere is a relation
between masculinity and self-esteem in bodi sexes, and provided die strongest support for
the masculinity model, "in which one's psychological well-being is seen to be a function of
the extent to which one has a masculine sex role orientation, irrespective of one's gender"
(Whitiey, 1983, p. 766).
Payne (1987) conducted a study to investigate the issue of how masculinity or
femininity contribute to adjustment. For this study, 92 male and 92 female undergraduates
were given the short BSRI, the PAQ, and a broad range of self-report adjustment
measures. In his smdy, Payne (1987) found that femininity or expressiveness correlated
weakly with traditional adjustment indicies (global and social self-esteem, and lack of
anxiety), but also weakly to moderately with more socially oriented aspects of adjustment,
including loneliness, lower social distmst, and lower aggression. Instmmentality
correlated moderately to highly with self-esteem, lower anxiety, and lower loneliness. As
evidenced through the above mentioned studies, not only is sex-role orientation related to
the codependency characteristic of extreme focus outside of self, but also to the
codependency element of low self-worth.
Social Desirahilitv
Several theorists suggest that one of the characteristics of codependency is the need
for approval (Beattie, 1987; Schaef, 1986; Subby & Friel, 1984; and Wegscheider-Cmse,
1985). Studies that assess need for approval often use social desirability scales. There are
a few attitudinal instmments diat measure acquiescence: the Marlowe-Crowne Social
Desirability Scale (MC-SDS; Crowne & Marlowe, 1960), the Edwards Social Desirability
Scale (E-SDS; Edwards, 1957), and the Agreement Response Scale (Couch & Keniston,
1960). "In die development of the M-C SDS, social desirability was defined more broadly
22
toreferto the need of Ss to obtain approval by responding in a culturally appropriate and
acceptable manner" (Crowne & Marlowe, 1960, p. 353). Although, according to the
audiors, the socially undesirable responses on the M-C SDS do not imply maladjustment,
high scores imply a great need for die approval of others (Crowne & Marlowe, 1960). It
is precisely tiiis aspect of the constmct of social desirability that is shared by the constmct
of codependency, for codependent individuals often seek the approval of others.
Strickland and Crowne (1963) conducted a study assessing the relationship between
the need for approval and the premature termination of psychotherapy. In diis study, the
Marlowe-Crowne Social Desirability Scale was used as the measure of need for approval.
The test sample consisted of 85 patients in treatment during 1960 and 1961, whose age
range was from 20 to 54, with a mean of 33.7 years. The major hypothesis was diat
patients characterized by a high need for approval are more likely to terminate
psychotherapy early.
As hypothesized, Strickland and Crowne (1963) found that high and low need for
approval patients differ strikingly in the length of time theyremainin psychotherapy, with
approval-dependent patients terminating much earlier than those less approval-motivated.
"These findings are consonant with recent studies on the approval motive (Conn, 1962;
Crowne & Liverant, 1963) and support the conceptualization of the approval-dependent
person as one who is concerned with protecting and maintaining a vuberable self-image"
(Strickland & Crowne, 1963, p. 99). Based on diese studies, social desirability should be
related to codependency.
Locus of Control
The concept of intemal versus extemal (I-E) control of reinforcement, or "locus of
control," holds diat people acquire a generalized expectation about the source of
reinforcement for their actions (Ashkanasy, 1985; Rotter, 1954; Rotter, Chance, & Phares,
23
1972). "Rotter proposed that this expectation can berepresentedon an internal-external
control continuum. At one end are 'internals' who believe that reinforcement derives from
within their own personalities at the other end are 'extemals' who see reinforcement to be
determined by fate, chance, or powerful others" (Ashkanasy, 1985, p. 1328). Rotter's
(1966) own definition that has guided much of the early work on the development of an
I-E scale and variable is the following:
When a reinforcement is perceived by the subject as following some action

of his own but not being entirely contingent upon his action, then, in our
culture, it is typically perceived as theresultof luck, chance, fate, as under
the control of powerful others, or as unpredictable because of the great
complexity of the forces surrounding him. When the event is interpreted in
this way by an individual, we have labeled this a belief in external control.
If the person perceives that the event is contingent upon his own behavior
or his own relatively permanent characteristics, we have termed this a
belief in intemal control, (p.l)
Rotter's I-E scale has found extensive application in the psychological literature.
Many have described the application of the constmct in clinical, social, educational, and
organizatonal psychology, finding in general that intemals are better adjusted than
extemals, and more able to cope with life (Joe, 1971; Lefcourt, 1972, 1976, 1980; Phares,
1973; Prociuk & Lussier, 1975).
Among odier variables, in his "Review of the Intemal-Extemal Control Constmct as
a Personality Variable," Joe (1971) reviewed the relation between the locus of control
constmct and attempts to control the environment, risk-taking, and therelationof the I-E
control measure to adjustment Joe (1971) found that intemals showed more initiative and
effort in controlling dieir environment, and could control their own impulses better than
extemals. He found evidence to support Liverant and Scodel's (1960) hypothesis that
intemals would be more cautious and conservative than extemals inrisk-takingsituations
in an attempt to control events. Finally, Joe (1971) reported a linear relationship between
intemal-extemal control and adjustment, with extemals tending to be more maladjusted
than intemals.
24
Research has also shown extemal locus of control to be associated with
hopelessness and depression (Joe, 1971; Lefcourt, 1976; Miller & Seligman, 1976;
Prociuk, Breen, & Lussier, 1976). In their study, Prociuck, Breen, and Lussier (1976)
examined therelationshipbetween negative expectancies about the future (hopelessness)
"and two theoretically relevant constmcts: generalized expectancies forreinforcement,and
depression" (p. 299). There were two samples employed in this study: 67 freshman
university students, and 44 senior psychology smdents. The smdents were given the
Hopelessness Scale (Beck, Weissman, & Trexler, 1974), die Rotter I-E Scale, and the
Beck Depression Inventory (Beck, 1967). Results of die smdy supported the predicted
relationship between hopelessness and intemal-extemal locus of control and depression.
Hopelessness was found to be significandy related to perceived extemal control and
depression in both samples.
The notion of locus of contol isrelevantto codependency in diat codependent people
try to have and believe they can have control over odier individuals, suggesting diat
extemal forces can have an effect on odiers' behaviors. Codependents also tend torelyon
extemal sources of reinforcement for themselves. Some of the characteristics of
codependency, as defined in die previous chapter, diat are indicative of a tendency for
extemal locus of control are lack of tmst, low self-worth, and controlling behavior.
Self-Esteem
As noted in die previous chapter, low self-esteem is one of the major characteristics
of codependency. According to Whidey (1983), global self-esteem "comprises a person's
self-evaluation across a number of areas, such as feehngs of adequacy and worth, feelings
of being a 'good' or 'bad' person, physical appearance, personal skills, and sexuality" (p.
767). The constmct of self-esteem has been smdied by many people, and reviews of the
25
literature (Wylie, 1961) suggest that self-esteem has beenrelatedto almost every variable at
one time or another.
Gauthier, Pellerin, and Renaud (1983) conducted a study diat evaluated die
effectiveness of two cognitive strategies for the enhancement of self-esteem. The results
indicated that cognitiverestmcmringand cognitverehearsalwere equally effective in
improving self-esteem and reducing depressive feelings. Gauthier et al. (1983) state that
the results of their smdy "concord with those of previous studies (e.g., Coleman,
1975;Wilson & Krane, 1980) that show levels of depression to vary inversely as a
function of self-esteem, and poor self-esteem to be a determinant of depressive feeUngs"
(p. 396).
Jaques and Chason (1977) studied the relationship between self-esteem and low
stams groups. In their smdy, 972 subjects completed questionnaires which included the
Rosenberg Self-Esteem Scale. The results of this smdy indicated that persons with lower
ascribed status did not, as a group, have lower self-esteem scores than persons with higher
ascribed status. However, die study showed that persons with lower achieved status in the
areas of educational attainment and occupational prestige did have lower self-esteem scores
than persons of higher stams.
Bodi the definition of codependency and the relation of sex roles to self-esteem
suggest diat there will be an inverserelationshipbetween codependency and self-esteem.
Codependency Research
The need forresearchin die area of codependency is great. According to Cermak
(1986b):
To date, no criteria have been agreed upon for assessing whether a client is
co-dependent. Without such criteria, no standards exist for assessing die
presence and depdi of padiology, for developing appropriate treatment
plans, or for evaluating die effectiveness of therapy. Treatment team
members are hindered in dieir efforts to communicate clearly and
26
understandably about specific clients, and comparison smdies of
co-dependence are not possible. Unless we can begin gathering reliable
and validresearchdata, co-dependence willremainconfined to clinical
impression and anecdote, (p.3)
It is the purpose of die present smdy to gadierreliableand valid data on codependency.
Few smdies exist which research codependency. One examined codependency in
lesbianrelationships,die odier interviewed wives of alcoholics. Smalley (1987) writes of
codependency pattems in lesbian relationships: "codependency is a learned pattem of
behaviors, feelings, and beliefs. It often results in self-neglect and is characterized by an
extemal locus of control... As widi lesbian and heterosexual women, Ifindthe
differences in co-dependent pattems and interdependency to be one of degree or emphasis,
not one of kind" (p. 126). Smalley (1987) illustrates with case smdies treatment strategies
diat she found to be successful in overcoming these codependent issues. The main goal of
her treatment process is to help women have more comfortablerelationshipsthat are
self-expanding. The particular skills Smalley (1987) deems important are "to encourage
women: (a) gendy but firmly to observe and intervene on their ownrelationshippattems,
(b) leam to lower their stress levels in both self-intimacy as well as intimacy with others,
and (c) leam aboutrelationshipdevelopment and develop additional interaction skills"
(p. 130). Smalley does not provide additionalresearchdata to go along with the case
illustrations.
Asher and Brissett (1988) conducted a longimdinal study of women married to
alcoholics. The primary data for their smdy consisted of taped interviews with 52 wives of
men diagnosed and treated for alcoholism. Each woman was interviewed diree times using
a standardized interview guide: when she initially entered a family program for spouses of
alcoholics, at 7 months following her program participation, and 15 months following her
program participaton. Asher and Brissett (1988) write:
Our interest in the notion of codependency serendipitously emerged during

die course of diis broad field study. Our curiosity was piqued by die
frequency with which the women in our sample spontaneously used die
27
term 'codependency' to describe diemselves or dieir behavior. Their use of
die term was very matter of fact, yet dieyflounderedfor definition and
explanation when probed by die interviewer as to its meaning in dieir lives.
(p.336)
Asher and Brissett (1988) found a great deal of diversity and conceptual and behavioral
ambiguity between and widiin dierespondents'characteriaztions of codependency. In
spite of diis ambiguity, they found two common dimensions of codependency in die
responses of die women: (1) notions of caretaking and pleasing odiers, and (2) affliction
by association widi a chemically dependent person. On the odier hand, they:
isolated a number of radier pronounced differences of opinion regarding the

nature and substance of codependency: (1) whether codependency involves
a substantial alteration of one's self, (2) whether codependency is an innate
personal characteristic or a learned social role, (3) whether codependency is
unique to alcohol-complicatedrelationships,(4) whether codependency is a
disease, and (5) whether codependency is a temporary or permanent
condition, (pp.346-347)
The definitional ambiguity of codependency that Asher and Brissett (1988) found in
dieir study points to the need for areliableand valid conceptualization and
operationalization of the term. Asher and Brissett's (1988) very last statement is, "If we
would draw any one point from these questions and our own initial inquiry, it would be
that more thorough articulation and examination of the social constmction of codependency
are called for" (p.349). The present smdy addresses the need for a working definition of
codependency, an operationalization of the term, and further articulation and examination
of the constmct of codependency.
Hypotheses
Hypothesis 1
Femininity as measured by the BSRI will be related positively to codependency.

28
Hypothesis 2
Masculinity as measured by the BSRI will berelatednegatively to codependency.
Hypotheses 1 and 2 are suggested in light or previous research with the BSRI (Bern, 1974;
Spence & Helmreich, 1978) and the theoretical constmcts of codependency.
Hypothesis 3
Social desirability as measured by Marlowe-Crowne Social Desirability Scale will be
related positively to codependency. As indicated above, codependents are often
characterized by a great need for approval of others.
Hypothesis 4
Locus of control as measured by Rotter's Intemal-Extemal Locus of Control Scale
will berelatedpositively to codependency. (Higher score=More extemal.)
Hypothesis 5
Self-esteem as measured by Rosenberg's Self-Esteem Scale will be related
negatively to codependency. As noted above, one of die characteristics of codependency is
low self-esteem.
Hypothesis 6
Age, race, occupation, and income will be unrelated to codependency.
Hypothesis 7
Males will score lower dian females on codependency. Many of the characteristics
of codependency appear to be related to what are traditionally viewed as feminine type
characteristics.
29
Hypodiesis 8
The known sample group of codependents will score higher on the codependency
scale than will therecoveringgroup. Individuals who have been active in self-help groups
for three or more years should be recovering from many aspects of their previous
codependent behavior.
CHAPTER m
METHODOLOGY
The purpose of die present study is to develop areliableand valid scale to measure
codependency. The selection of subjects, measures, procedures, hypodieses, and method
of data analysis are discussed in diis chapter.
Subjects
A large number of subjects were needed for die present study to help establish
concurrent validity (hypodieses 1-7). This sample was taken from undergraduate
Sociology classes, and totaled 192 subjects. In order to estimate predictive validity
(hypothesis 8) a known group of people who are thought to be codependent, and should
therefore score high on a codependency scale, was needed. A control group of otherwise
similar people who are not thought to be codependent was also needed.
For this smdy, the known group of codependents consisted of a sample of people
from Al-Anon and CoDependents Anonymous groups who had been involved in these
groups for less than one month. With this limited time in a recovery group, these
individual should score higher on the codependency scale. The control group of
non-codependents was a sample of people that had been actively involved in Al-Anon for
more than three years. Although persons who came to Al-Anon might be thought
codependent, dieir long-term involvement in diis group with its emphasis on working a
12-step program should lead to recovery from codependency and lower scores on die
codependency scale. Both of these samples were drawnfromLubbock, Texas. The
number of subjects in the Al-Anon group was 30 and the number of subjects in die
codependent group was 14.
30
31
Measures
Concurrent validity can be established by correlating the codependency scale to

scales diat have been identified as sharing similar constmcts. It is important diat die
correlations are found to be moderate, as Campbell and Fiske (1959) state, "Tests can be
invalidated by too high correlations widi odier tests from which diey were intended to
differ" (p. 81). Participants in die present smdy completed a questionnaire packet
consisting of demographic information, die codependency scale, and odier related
self-report inventories.
Three constmcts that were believed to share similarities with die codependency scale
were feminine sex-role orientation as measured by die Femininity scale of die Bem
Sex-Role Inventory (Bem, 1974), social desirability as measured by the Marlowe-Crowne
Social Desirability Scale (Crowne & Marlowe, 1960), extemal locus of control as
measured by a shortened version of Rotter's (1966) Intemal versus Extemal Control of
Reinforcement Scale. Conversely, the codependency scale was believed to have no
significant cortelation to particular subject demographics such as age, race, income, and
occupation, and was predicted to be negatively correlated to masculine sex-role orientation
as measured by the Masculinity scale of the Bem Sex-Role Inventory and to self-esteem as
measured by the Rosenberg Self-Esteem scale (Rosenberg, 1965).
Bem Sex-Role Inventory

The Bem Sex-Role Inventory (Appendix A), or BSRI (Bem, 1974) is a self-report
measure of therespondent'sendorsement of culmral definitions of masculine and feminine
characteristics. The BSRI is a 60-item scale containing 20 items of masculine traits, 20
items of feminine traits, and 20 items of neutral social desirability widi respect to sex. The
Masculinity and Femininity scales are independent of each other (Bem, 1974: Spence &
Helmreich, 1978). A trait appears on the masculine Ust if it was independendy determined
32
by bodi male and female judges to be significandy more desirable for males dian for
females in American society. Personality characteristics diat were independendy judged by
bodi males and females to be significandy more desirable for females than males in
American society are die traits that appear on die feminine list Similarly, traits appearing
on die neutral Ust were independendy judged by males and females to be no more desirable
for one sex dian for die odier. The individual taking die BSRI indicates on a seven-point
scale (ranging from never or almost never tme to always or almost always tme), the extent
to which die trait characterizes him/herself. The scoring of the BSRI can yield four distinct
categories: mascuUne (high mascuUne-low feminine), feminine (high feminine-low
mascuUne), androgynous (high masculine-high feminine), and undifferentiated (low
masculine-low feminine) (Bem, 1977). These categories are not used in the present study,
but rather separate mascuUnity and femininity scores.
The subjects in Bem's (1974) original smdy consisted of 44 male and 279 female
students in introductory psychology at Stanford Univeristy, and 117 male and 77 female
paid volunteers at FoothiU Junior College. In order to estabUsh intemal consistency of die
BSRI, coefficient alpha was computed separately for the Masculinity and Femininity
scores of the subjects in each of the two normative samples. The results showed scores to
be highly reUable, both in the Stanford sample (MascuUnity = .86; Femininity = .80) and
in the Foothdl sample (Masculinity = .86; Femininity = .82) (p. 158). Test-retest
correlations over 4 weeks were r = .90 for both MascuUnity and Femininity (n=28 males
and 28 females). Bem (1974) reported a moderate correlation between die California
Psychological Inventory and the BSRI. The BSRI Masculinity correlation with CPI Males
was -.42 and with CPI Females was -.25. The BSRI Femininity correlation with CPI
Males was .27 and with CPI Females was .25. Note that the CPI scale is keyed in the
feminine direction.
33
In Ught of previousresearchwidi die BSRI and die dieoretical constmcts of
codependency, it was proposed diat die codependency scale would correlate moderately
widi die Femininity scale, and would have a negative correlation widi die MascuUnity scale
of die BSRI.
Marlowe-Crown Social Desirability Scale
The Marlowe-Crown Social DesirabiUty Scale (Appendix B), or M-C SDS (Crowne
& Marlowe, 1960), consists of 33 items which met die criterion of being defined by
behaviors which are culturally approved but which are improbable of occurrence, and have
minimal pathological or abnormal implications if responded to in either die sociaUy
desirable or undesirable directions. Although, according to the authors, die sociaUy
undesirable responses on die M-C SDS do not imply maladjustment, high scores imply a
great need for die approval of others (Crowne & Marlowe, 1960). It is diis aspect of the
constmct of social desirability diat is shared by the constmct of codependency.
Of the 33 items, 18 are worded such that they are culmrally acceptable but probably
untme (e.g., "I never hesitate to go out of my way to help someone in trouble") and 15 are
probably tme but culturally undesirable (e.g., "I sometimes feel resentful when I don't get
my way") (Crowne & Marlowe, 1960, p. 351). Testees are instmcted to read each item
and decide whether the statement is tme or false as it pertains to themselves personally.
One point is scored for each response in the socially desirable direction widi scores varying
between 0 (no social desirabiUty) to 33 (highest social desirabUity). Crowne and Marlowe
(1960) reported that die intemal consistency coefficient for the final form of die M-C SDS,
using Kuder-Richardson formula 20, was .88, and a test-retest correlation was .89.
Crowne and Marlowe also reported the correlation of the M-C SDS and die Edwards SDS
to be .35, which was significant at the .01 level.

34
Intemal Versus Extemal Control of Reinforcement Scale
The Intemal Versus Extemal Control of Reinforcement Scale, or I-E scale (Rotter,
1966), is considered to be a measure of generaUzed expectancy for intemal-extemal
control. The I-E scale is a 29-item forced-choice instrument including six fiUer items
intended to make the purpose of the test somewhat more ambiguous. Each of the 29 items
consists of a pair of alternatives containing a statementreflectingintemal control and a
statement reflecting extemal control. Sample items are: "a. Many of the unhappy things
in people's lives are partly due to bad luck. b. People's misfortunes result from the
mistakes they make," and "a. In the long run people get dierespectthey deserve in this
world, b. Unfortunately, an individual's worth often passes unrecognized no matter how
hard he tries" (Rotter, 1966, p. 11). Subjects are instmcted to select die one statement of
each pair which diey more strongly believe to be the case as far as diey are concemed.
According to Rotter (1966,1975) die forced-choice format was adopted specifically to
balance out social desirability. The score of die I-E scale is the total number of extemal
choices. A high score represents more extemal contt-ol choices and a low score represents
more intemal control choices. A shortened 14-item version of the I-E scale containing
direefiUeritems was used for this study (Appendix C).
Rotter (1966) reports diat a sample of 400 Ohio State University Elementary
psychology smdents had an intemal consistency reliability of .70 as estimated by the
Kuder-Richardson formula. Rotter argues diat diisreUabilityis somewhat limited since
diis is a forced-choice scale. Rotter suggests diat discriminant validity is indicated by die
lowrelationshipswith such variables as intelligence, social desirability, and political
Uberalness. Rotter (1966) also writes, "Most significant evidence of the constmct validity
of die I-E scale comes from predicted differences in behavior for individuals above and
below die median of the scale orfromcon-elarions with behavioral criteria" (p. 25).
35
Rosenberg Self-Esteem Scale
The Rosenberg Self-Esteem scale (Appendix D) (Rosenberg, 1965) is a
multidimensional global self-esteem scale. Rosenberg (1965) designed his measure of
self-esteem for use with adolescents. The original scale contained 10 items, and was
structured as a Guttman scale. A six-item scale was also formed by combining some of the
ten responses. The Rosenberg Self-Esteem scale has been used with adults as both a six-
and ten-item instrument. However, Jacques and Chason (1977) and Dobson et al. (1979)
report low reUability for the six-item scale.
In theirresearch.Carmines and Mclver (1981) found that one of ten items of the
Rosenberg Self-Esteem scale relates only weakly to the others. Subsequendy (Payne,
1987), the instmment has been used as a nine-item scale. In addition, rather than scoring it
as a Guttman scale, a five-point response scale has been used with the responses scored in
a simple additive fashion (Carmines & Mclver, 1981; Payne, 1987). hi the present study
the Rosenberg Self-Esteem scale diat was used had nine items and afive-pointLiken-type
response scale scored by summation. Some examples of the items are: I feel that I have a
number of good qualities; I am able to do things as well as most other people; and I feel I
do not have much to be proud of.
Rosenberg's (1965) main sample consisted of 5,024 high school juniors and seniors
from ten randomly selected New York schools, and a Guttman scale reproducibility
coefficient of .92 was obtained. Silber and Tippett (1965) found a test-retest correlation
over two weeks of .85 (N=28). Silber and Tippett (1965) also found that the scale
cortelated from .56 to .83 with several similar measures and clinical assessment (N=44).
Codependency Scale
In order to develop a reUable and valid scale to measure codependency, several steps
had to be taken. First, test items had to be selected and written. Items were designed to
36
reflect one of die nine characteristics of codependency: dependency, obsession,
caretaking,repressedfeeUngs, lack of tmst, poor communication, control, denial, and
rigidity. The initial version of die scale had 35 items. This version of die scale was dien
given to four experts on codependency. These experts were people who worked in the
field of treating codependency and chemical dependency. The experts were asked to
review the scale and judge whether or not the Ust of items would assess codependency.
Taking into account die judges' feedback, some changes were made and die scale was
revised to contain 38 items. The revised 38-item scale (Appendix E) was used for a pilot
smdy.
As a pilot smdy, die revised scale was administered twice to the same sample after a
two week interval. The sample used for the pilot smdy consisted of 40 undergraduate
Family Studies students at Texas Tech University. In order to ascertain the reliabdity of
the codependency scale the pilot smdyresultswere used to assess test-retest reUabiUty, as
weU as intemal consistency reUabiUty by means of Cronbach's alpha. The test-retest
reUability for the 38-item scale was r= .86 and the intemal consistency reUability was
alpha= .83 on the first administration and .85 on the second administration.
The data for both administrations of this scale were further studied, and items that
were equal to or less than .30 on corrected item-total correlation were omitted, leaving a
15-item scale. The test-retest correlation for the 15-item scale was r= .87 and die intemal
consistency reUabiUty was alpha= .86 on both administrations. The mean score on each
item for the undergraduates on die first administration of die 15-item scale was 3.30 and
the mean score on the second administration of the 15-item scale was 3.24 on each item.
Figure 2 contains a copy of the 15 item codependency scale. AU nine areas identified as
characteristics of codependency were fairly equaUyrepresentedon the 15 item scale.
Review of the items showed that those dropped had conceptual overlap widi those kept.
37
Read the foUowing statements and place die number in die spaces provided diat best
descnbes you according to die following Ust
1 Strongly Disagree
2 Moderately Disagree
3 SUghdy Disagree
4 SUghdy Agree
5 Moderately Agree
6 Strongly Agree
— 1. It is hard for me to make decisions.

2. It is hard for me to say "no."
3. It is hard for me to accept compUments graciously.
— 4. Sometimes, I almost feel bored or empty if I don't have problems to

focus on.
— 5. I usually do not do things for other people diat diey are capable
of doing for themselves.
6. When I do something nice for myself I usuaUy feel guilty.
7. I do not worry very much.
8. I tell myself that diings wiU get better when die people in my Ufe
change what they are doing.
9. I seem to haverelationshipswhere I am always there for them, but
they are rarely there for me.
10. Sometimes I get focused on one person, to the extent of neglecting
otherrelationshipsand responsibdities.
11. I seem to get intorelationshipsthat are painful for me.
12. I don't usually let others see the "real" me.
13. When someone upsets me, I wiU hold it in for a long time, but once
in a while I explode.
14. I will usually go to any lengths to avoid open conflict.
15. I often have a sense of dread or impending doom.
nGURE2
15-Item Codependency Scale

38
Procedures
For the present smdy, subjects were given a questionnaire packet containing die
codependency scale, die BSRI, die Marlowe-Crowne Social Desirability Scale, a short
version of Rotter's Intemal-Extemal Locus of Control Scale, Rosenberg's Self-Esteem
Scale, and a short demographic questionnaire (Appendix F). Hypodieses 1, 3, and 4,
which predicted a positive correlation between codependency and femininity, social
desirability, and extemal locus of control, were tested by Pearson's r. Hypotheses 2 ans
5 which predicted a negative correlation between codependency and mascuUnity and
self-esteem, were tested by Pearson's r. For hypothesis 6, dierelationshipof age and
income to codependency was tested with Pearson's r, the relationship of race was tested
with a t-test, and relationship of occupation to codependency was tested with a one-way
analysis of variance. Hypotheses 7 and 8, which predicted that the known group of
codependents would score higher on the codependency scale than would the comparison
sample and that males would score lower on codependency than females was tested with
t-tests. For the participants of this smdy, all scales were tested for Cronbach's alpha
reUabiUty.
CHAPTER IV
RESULTS
The results of die statistical analysis for diis smdy are divided into four sections: (a)
descriptive statistics, (b) preliminary analysis, (c) main analysis, and (d) additional
analysis. Descriptive statistics are reported for die total sample as weU as for die smdent,
Al-Anon, and codependent groups separately. Intemal consistencyreUabiUtymeasures for
aU of the scales used in die smdy and a factor analysis of die codependency scale
constimted preliminary analyses. The main analysis section tests die research hypodieses
as presented in Chapter m. Finally, additional analyses explored the relation of
codependency to quality of life questions.
Descriptive Statistics
Table 1 highUghts demographic characteristics of the 236 participants in the study.
The characteristics of the smdent sample (n=192), Al-Anon sample (n=30), and
codependent sample (n=14) appear in Tables 2, 3, and 4, respectively. The mean age of
the respondents in the total sample was 24.72 years with a standard deviation of .72. The
average age of the student sample was 20.06 years with a standard deviation of 3.94, the
average age of the Al-Anon sample was 47.48 years with a standard deviation of 11.80,
and the average age of the codependent sample was 39.93 years with a standard deviation
of 9.70. The total sample was largely Caucasian (88.6%). The tables provide a complete
breakdown of the race/ethnic origin of the sample. There were 156 females and 80 males
in the total sample. In the student sample there were 118 females and 74 males. There
were 26 females and 4 males in the Al-Anon sample, and 12 females and 2 males in die
codependent sample.
39
40
TABLE 1
Descriptive Statistics of Total Sample (N=236)
Variable Frequency Percent Mean

Age of Respondent 24.72
Race/Ethnic Origin
Caucasian 209 (88.6%)
Black 6 ( 2.5%)
Hispanic 16 ( 6.8%)
Native American 4 ( 1.7%)
Missing 1 ( -4%)
Gender
Female 156 (66.1%)
Male 80 (33.9%)
Marital Stams ^_^
Single 179 (75.8%)
Martied 35 (14.8%)
Divorced 15 ( 6.4%)
Widowed 3 ( 1-3%)
Remanied 4 ( 1.7%)
ReUgious AffiUation ,^ ^ ^ ^ s
Catholic 48 (20.3%
Prot/Christian 181 ^?^'lls\
Jewish 1 -^^
Other 3 ( 1.3%)
Not Religious 3 ( 1-3%)
Residencê Growing up ^^ ^^^^^^^
Small City 101 (42.8%)
Large City 83 35.2%
Missing 1 ( -^^^^
^Xfes^'sional 26 (11.0%)
Blue Collar 21 8.9%
Sales 24 (10.2%)
Cferical 18 ( 7.6%)
Manager 6 ( ^.^/^
Not Employed 136 (57.6%)
41
TABLE 1 - Continued
Income
$1- $5,000 48 (20.3%)
$5,000-$10,000 25 (10.6%)
$10,000-$15,000 5 ( 2.1%)
$15,000-$20,000 4 ( 1.7%)
$20,000-$25,000 9 ( 3.8%)
$25,000-$30,000 6 ( 2.5%)
$30,000-$40,000 3 ( 1.3%)
Over $40,000 3 ( 1.3%)
No earned income 132 (55.9%)
Missing 1 ( -4%)
Satisfied Widi Life
Not At AU 9 ( 3.8%)
Mildly 31 (13.1%)
Moderately 102 (43.2%)
Strongly 94 (39.8%)
Adjusted
Poorly 5 ( 2.1%)
Mildly 31 (13.1%)
Well 139 (58.9%)
Very WeU 61 (25.8%)
42
TABLE 2
Descriptive Statistics of Smdent Sample (N=192)
Race/Ethnic Origin
Black 6 ( 3.1%)
Hispanic 16 ( 8.3%)
Missing 1 ( .5%)
Gender
Female 118 (61.5%)
Male 74 (38.5%)
Marital Stams
Single 178 (92.7%)
Married 6 ( 3.1%)
Divorced 8 ( 4.2%)
ReUgious AffiUation
Catholic 46 (24.0%)
Prot./Christian 140 (72.9%)
Jewish 1 ( .5%)
Other 2 ( 1.0%)
Not Religious 3 ( 1.6%)
Residence Growing up ^ ^^ ^X ^ \ ^># V
Rural 39 (20.3%)
Large City 74 (38.5%)
Missing 1 ( .5%)
Occupation
Professional 10 ( 5.2%)
Blue CoUar 18 ( 9.4%)
Sales 22 (11.5%)
Clerical 15 ( 7.8%)
Manager 2 ( 1.0%)
Farmer 5 ( 2.6%)
Income
$1- $5,000 46 (24.0%)
$5,000-$10,000 21 (10.9%)
$10,000-$15,000 3 ( 1.6%)
$15,000-$20,000 0 ( .0%)
43
TABLE 2 - Continued
$20,000-$25,000 1 ( .5%)
$25,000-$30,000 1 ( .5%)
$30,000-$40,000 0 ( .0%)
Over $40,000 1 ( .5%)
Satisfied Widi life
Not At AU 6 ( 3.1%)
Mildly 24 (12.5%)
Strongly 76 (39.6%)
Adjusted
Poorly 4 ( 2.1%)
Mildly 22 (11.5%)
WeU 116 (50.4%)
Very WeU 50 (26.0%)
44
TABLE 3
Descriptive Statistics of Al-Anon Sample (N=30)
Race/Ethnic Origin
Gender
Female 26 (86.7%)
Male 4 (13.3%)
Marital Stams
Married 19 (63.3%)
Divorced 5 (16.7%)
Widowed 2 ( 6.7%)
Remarried 4 (13.3%)
ReUgious AffiUation
CathoUc 1 ( 3.3%)
Residence Growing up
Rural 8 (26.7%)
Large City 8 (26.7%)
Occupation
Professional 9 (30.0%)
Blue CoUar 3 (10.9%)
Sales 1 ( 3.3%)
Clerical 1 ( 7.6%)
Manager 6 (10.0%)
Farmer 0 ( .0%)
Income
$1- $5,000 1 ( 3.3%)
$5,000-$ 10,000 2 ( 6.7%)
$10,000-$15,000 2 ( 6.7%)
$15,000-$20,000 3 (10.0%)
$20,000-$25,000 5 (16.7%)
$25,000-$30,000 3 (10.0%)
$30,000-$40,000 3 (10.0%)
Over $40,000 1 ( 3.3%)
45
TABLE 3 - Continued
Satisfied Widi life

NotAtAU 0 ( .0%)
Mildly 1 ( 3.3%)
Strongly 18 (60.0%)
Adjusted
Pooriy 0 ( .0%)
Mildly 2 ( 6.7%)
Well 18 (60.0%)
Very WeU 10 (33.3%)
Number of Years in Al-Anon 8.9
46
TABLE 4
Descriptive Statistics of Codependent Sample (N=14)
Race/Ethnic Origin
Caucasian 14 (100%)
Gender
Female 12 (85.7%)
Male 2 (14.3%)
Marital Stams
Single 1 ( 7.1%)
Married 10 (71.4%)
Divorced 2 (14.3%)
Widowed 1 ( 7.1%)
ReUgious AffiUation
Catholic 1 ( 7.1%)
Other 1 ( 7.1%)
Residence Growing up
Rural 4 (28.6%)
Large City 1 ( 7.1%)
Occupation
Professional 7 (50.0%)
Blue CoUar 0 ( .0%)
Sales 1 ( 7.1%)
Clerical 2 (14.3%)
Manager 1 ( 7.1%)
Farmer 0 ( .0%)
Income
$1- $5,000 1 ( 7.1%)
$5,000-$10,000 2 (14.3%)
$10,000-$15,000 0 ( .0%)
$15,000-$20,000 1 ( 7.1%)
$20,000-$25,000 3 (21.4%)
$25,000-$30,000 2 (14.3%)
$30,000-$40,000 0 ( .0%)
Over $40,000 1 ( 7.1%)
Missing 1 ( 7.1%)
47
TABLE 4 - Continued
Satisfied Widi Life

Not At AU 3 (21.4%)
Mildly 5 (35.7%)
Strongly 0 ( .0%)
Adjusted
Poorly 1 ( 7.1%)
Mildly 7 (50.0%)
WeU 5 (35.7%)
Very WeU 1 ( 7.1%)
48
The tables also provide a complete breakdown of die marital stams, reUgious
affiUation, and residence when growing up of die sample. WhUe most of die subjects in
die smdent sample were single (92.7%), aU of the Al-Anon sample was or had been
married, and there was only one subject from die die codependent sample diat was single.
Most of die total sample (76.7%) was Protestant/Christian in their religious affiUation. In
die smdent sample 72.9% of the subjects were Protestant/Christian, 96.7% of die Al-Anon
sample were Protestant/Christian, and 85.7% of die codependent sample were
Protestant/Christian.
Most of the subjects in the smdent sample were unemployed (62.5%). In die
Al-Anon sample 43.4% were unemployed, and 30.0% were professionals. Half of die
codependency sample subjects were professionals. The tables provide a breakdown of die
subjects' yearly income.
The tables also include the results of two questions asked on the demographic
questionnaire: (a) "In general, how satisfied are you with your Ufe?" and (b) "Compared
to others, in general, how weU adjusted do you think you are?" Only a few subjects in the
smdent and Al-Anon samples indicated that they were not satisfied with their life or poorly
adjusted. However, there was no subject from the codependency group that felt strongly
satisfied with their Ufe, and oidy one that felt very well adjusted. The tables provide a
further breakdown of the results to these questions.
FVeliminarv Analvsis
The total sample (n=236) results were used to assess the intemal consistency
reUabiUty by means of Cronbach's alpha for each of the scales used in the smdy. The
intemal consistency reUabUity for the femininity scale on the Bem Sex Role Inventory was
alpha= .85, and for the mascuUnity scale on the BSRI alpha= .87. For the
Marlowe-Crowne Social DesirabiUty Scale Cronbach's alpha= .76. The intemal

49
consistency reUabUity for the short version of Rotter's Intemal vs. Extemal Control of
Reinforcement Scale was .44. This lowreliabdityscore may be a result of the forced
choice format or of the shortened version of the scale. For Rosenberg's Self-Esteem Scale
the intemal consistency reliabdity was alpha= .86. The Codependency Scale had an
intemal consistencyreUabiUtyof .77.
Codependency Scale items were factor analyzed. There were four factors with
eigenvalues greater than one extracted using the varimax procedure of orthogonal rotation.
The first factor accounted for 24.2% of the variance. With the exception of one item, all
items loaded (.37 or better) on the first unrotated factor. This provides support for the
unidimensionaUty of the codependency scale. The unrotated factor scores for the first
factor are presented in Table 5.
Main Analvsis
The correlation results for hypodiesis one through five arereportedin Table 6 and
are described below. The entire sample was used in testing hypotheses 1 through 5.
Hypothesis 1
Thefirsthypodiesis stated diat femininity as measured by the BSRI would be related
positively to codependency. For die total sample (n=236) the mean score for each item on
die femininity scale of the BSRI was 5.01. The conrelation of femininity scores widi
codependency scores was not significant [r(234)= .04, £= .28, n.s.]. Thus, the
hypothesis was not supported in die correlation analysis. An alternative exploratory
analysis was conducted to address the lack of support for this hypothesis, and wiU be
reported in the last section of this chapter.

50
TABLE5
Unrotated Factor Loadings for First Factor

Codependency Scale
Codependency Factor Loading

Scale Question
1 .59034
2 .52993
3 .56331
4 .46589
5 .15736
6 .62819
7 .36509
8 .41618
9 .53969
10 .55329
11 .53385
12 -44673
13 .44432
14 .50136
15 -46312
51
TABLE6
Correlation Matrix of Codependency Scale widi Odier Scales
FEM. MASC. MC-SDS IE LOG S-E

Hypothesis 1 2 3 4 5
Total Sample .0385 -.3049 -.1513 .1926 -.5368

CODEP. (236) (236) (236) (234) (233)
N.S. p=.000 p= .010 p=.002 p=.000
Females -.0004 -.2433 -.1741 .1838 -.5556
CODEP. (156) (156) (156) (154) (153)
N.S. p=.001 p= .015 p= .011 p=.000
Males -.0267 -.4235 -.0915 .2102 -.4826
CODEP. (80) (80) (80) (80) (80)
N.S. p=.000 N.S. p=.031 p=.000

52
Hypodiesis 2
The second hypodiesis stated that mascuUnity as measured by die BSRI would be
negativelyrelatedto codependency. The mean score on die mascuUnity scale for die total
sample (n=236) was 4.90. The mascuUnity scale had a significant negative correlation
widi die codependency scale [i(234)= -.30, ji< .001]. This hypothesis was strongly
supported in the smdy.
Hypothesis 3
The third hypothesis stated diat social desirabiUty as measured by die
Marlowe-Crowne Social DesirabiUty Scale would berelatedpositively to codependency.
For die total sample (n=236), die mean score on die MC-SDS was 14.18. The correlation
analysis did not support this hypodiesis, but indicated diat there is a significant negative
correlation between the MC-SDS and die codependency scale [1(234)= -.15, j2= .01].
Therefore, the hypothesis was rejected.
Hypothesis 4
The fourth hypothesis stated that extemal locus of control as measured by the
shortened version of Rotter's Intemal-Extemal Locus of Control Scale would be related
positively to codependency. The mean extemal locus of control score for the total sample
(n=234) was 4.80. Extemal locus of control had a significant positive correlation with
codependency [i(232)= .19, c< .01]. The hypothesis was supported in this correlation
analysis.
Hypothesis 5
The fifth hypothesis stated diat self-esteem as measured by Rosenberg's Self-Esteem
Scale would be negatively related to codependency. For the total sample (n=233) the mean
53
score on die self-esteem scale was 3.19. The correlation analysis showed self-esteem to
have a significant negative relationship to codependency [r(231)= -.54, j2< .001].
Hypothesis 6
The sixth hypothesis stated diat age, race, occupation, and income would be
unrelated to codependency. The relationship of age and income to codependency was
tested for the smdent sample with Pearson's r. Age was not significantlyrelatedto
codependency [r(190)= .10, i2=.08, n.s.]. Income was also found to be nonsignificandy
related to codependency [r(190)= -.02, j2=.38, n.s.]. The relationship of race to
codependency was tested for die smdent sample with a t-test. No significant relationship
was found between race (Caucasian versus noncaucasian) and codependency IT(189)=
-.43, J2= .66, n.s.]. The relationship of occupation to codependency was tested in a one
way analysis of variance for the total sample, and no significant relationship was found
[F(7, 228)= .63, |2= .73, n.s.]. As predicted, variables thought to be unrelated to
codependency were unrelated. However, as might be expected in a smdent sample, there
was little variation on income and occupation. Fumre research on more general
populations should attempt to repUcate these findings to insure die independence of
codependencyfi-omage, income, and occupation.
Hypothesis 7
The seventh hypothesis stated diat males would score lower than females on
codependency. This hypodiesis was tested for die student sample using a one tail t-test.
The mean codependency score for females (n=l 18) was 3.38, and the mean codependency
score for males (n=74) was 3.11. The t-test showed a significance in die direction
predicted [l(190)=2.71, £< .01]. Thus, the hypothesis was supported. In view of support
for this hypothesis, cortelations were remn separately for males and females. These
54
findings are presented in Table 6. As may be seen, die pattems of conflations remain
similar for men and women to diose found for die total sample.
Hypothesis 8
The eighdi hypodiesis stated diat die known sample group of codependents would
score higher on the codependency scale dian die known recovering group sample. This
hypodiesis was tested using a one tail t-test. The mean codependency score for die known
recovering group sample (n=30) was 2.48 and the mean score on codependency for die
known sample group of codependents (n=14) was 4.00. The t-test showed a significance
in die direction predicted [1(42)= -4.93, £< .001]. This hypothesis was supported. The
t-test was recomputed for females only because of so few males, and showed diat t(36)=
-5.82, p< .001.
Additional Analvsis
As indicated above, hypothesis 1, which stated that femininity should correlate
positively with codependency was not significant. Inreviewingthe femininity scale of the
BSRI, 7 items were described in the literature review as having the highestrelationshipto
codependency as defined in this smdy. These seven feminine characteristics are: yielding,
sympathetic, sensitive to the needs of others, compassionate, eager to soothe hurt feelings,
and guUible. An intemal consistancy reUabiUty measure was mn using these seven items
to form a new femininity scale, and Cronbach's alpha was found to be .73. When this
condensed 7-item femininity scale was correlated with the codependency scale a significant
correlation was found [i(234)= .12, £< .05]. This analysis is suggestive of a possible
place to look in the future. For men and women separately die correlations were simdar
but not significant. For men r(78)= .09, N.S., and for women r(154)= .09, N.S. The
lack of significance can be attributed to the lower number of subjects.

55
The "satisfied widi life" and "adjusted to Ufe" questions were also correlated widi die
codependency scale. Bodi of diese items were negatively correlated widi codependency.
The correlation for satisfaction was r(234)= -.39, £< .001, and for adjustment r(234)=
-.40, ii< .001. For men die correlation for satisfaction was r(78)= -.48, £< .001, and die
con-elation for adjustment was r(78)= -49, i2< .001. For women, die correlation for
satisfaction was r(154)= -.36, £< .001, and for adjustment, r(154)= -.37, u< -001.
In addition four questions were asked about having either a practicing alcoholic
parent, a recovering alcoholic parent, a practicing alcohoUc spouse, or a recovering
alcohoUc spouse. The results of diese items were correlated with die codependency scale.
The only correlation with significantresidtswasrecoveringalcoholic spouse with
codependency, where r(33)= -.41, £< .01. This indicates that the more time their spouse
has had in recovery from alcoholism, the less the subject's codependency. The questions
concerning parental and spouse active alcohoUsm were notrelatedto codependency scores.
The assessments of parental and spouse alcohoUsm were one-item questions with a "yes"
or "no" response, and have questionablereUabiUty.In light of the many authors who link
addictions in the famUy with codependency (Capell-Sowder, 1984; Wegscheider-Cmse,
1985), future research should investigate diis more closely with reUable scales of parental
and/or spouse alcohoUsm.

CHAPTER V
DISCUSSION
The purpose of die present smdy was to develop a scale diat would reliably and
validly measure codependency. Through developing such a scale, die dieoretical concept
of codependency can be strengdiened widi supportive empirical research. To date, diere is
Utde research to be found in die codependency literamre, and no empiricaUy tested
assessment tool for codependency. The purpose of this smdy was to address the need for
research as well as die need for areUableand vaUd cUnical assessment tool dirough a
rigorous development and testing of a proposed scale to measure codependency. This
measurement of codependency can also aid in the pursuit of self-awareness and
self-determined change.
The different stages of this smdy consisted of reviewing theories on codependency,
developing a clear working definition of the term, writing scale items thatreflectthe
characteristics of codependency, andfinaUytesting die scale for reliabdity and validity.
Intemal consistency reUabiUty was calculated for the codependency scale, and convergent
and discriminant vaUdity was assessed dirough correlating the codependency scale with
other scales thought to be somewhat related or completely unrelated to codependency. The
preceding chapters have detaded therelatedtheoretical and empirical Uterature, the
methodology employed in the smdy, and the results of the analysis. The present chapter
includes a discussion of the findings, conclusions, and recommendations for future
codependency research.
56
57
Findings
The first hypodiesis, diat femininity as measured by die BSRI would be related
positively to codependency, was not supported Aldiough it was not significant, die result
was in die predicted direction. This hypodiesis was suggested in Ught of previous research
widi die BSRI (Bem, 1974; Spence & HeUnrcich, 1978) and die dieoretical constmcts of
codependency. In die Sex-Role Orientation section of Chapter 2, several femininity scale
items on the BSRI were Usted (yielding, compassionate, loyal, eager to soothe hurt
feelings, and sensitive to the needs of odiers), and said to bereflectiveof some of die
fundamental characteristics of codependency, such as extreme focus outside of self
through dependency, obsession, and caretaking.
In reviewing thisfirsthypothesis and the femininity scale of the BSRI it seems diat
although some of the items, as noted above, arereflectiveof codependency, there are many
more items that appear to be unrelated to codependency. Some of these unrelated items
are: cheerfid, shy,flatterable,soft-spoken, tender, childlike, does not use harsh language,
and loves children. The five codependencyrelevantitems on the femininity scale
mentioned above and two others (sympathetic and gulUble) were extracted as a short
version of the femininity scale. When these were correlated with codpendency, diere was
a modest but significant correlation in support of the hypothesis. However, diis analysis is
inconclusive and only provides a suggestion for where to look in the future.
As predicted in the second hypothesis, masculinity as measured by the BSRI was
related negatively to codependency. For the total sample, the higher the masculinity score,
the lower the codependency score. The support for this hypothesis indicates diat
individuals who see themselves as having more mascuUne, "dominance" (Wiggins &
Holzmuller, 1981), or "instmmental" (Spence, 1983) characteristics also describe
themselves as less codependent. Some of die descriptors on the masculinity scale of the
BSRI are: self-reliant, independent, assertive, wdling to take risks, makes decisions
58
easily, and wiUing to take a stand. These characteristics as part of die mascuUnity scale
were negatively related to codependency. Theresultsof this hypodiesis help to estabUsh
discriminant vaUdity.
Contradictory to die diird hypodiesis, social desirability as measured by die
Marlowe-Crowne Social DesirabiUty Scale was not positivelyrelatedto codependency, but
rather was found to be related negatively to codependency. This hypodiesis was based on
die idea diat one of die characteristics of codependency is the need for approval (Beattie,
1987; Schaef, 1986; Subby & Friel, 1984; and Wegscheider-Cmse; 1985), and that
studies that assess need for approval often use social desirability scales. Crowne and
Marlowe (1960) defined social desirabiUty as "die need of Ss to obtain approval by
responding in a culturaUy appropriate and acceptable manner" (p. 353).
The results of this hypothesis suggested that either codependents do not seek
approval or diat die M-C SDS does notreflectdiat need for approval. Possibly, as die
name of the scale indicates, the social desirability scale simply assesses a person's need to
be seen as culturaUy appropriate or sociaUy desirable, and does not tap into a person's need
to gain approval from people with whom they interact For example, some of the items are
"I have never intensely disliked anyone," "I Uke to gossip at times," and "I can remember
'playing sick' to get out of something." Another possible explanation for the negative
correlation found between the codependency scale and the social desirabiUty scale is that
the wording of the items on the codependency scale is such that these are sociaUy
undesirable attributes to admit about onself. Future research should explore this issue in
more depth.
Certain contexmal issues may have influenced the negative correlation found
between social desirabiUty and codependency. First, die smdent sample data was coUected
on the last day of a sociology class and theresponseswere anonymous. Both factors may
have contributed toreducingtendencies to answer in a socially desirable direction.

59
Second, die participants who were in Al-Anon were aware of dieir program's demands for
rigorous honesty. People who have attended Al-Anon for a whde may be more honest, or
diink diat diey should be. FinaUy, by virtue of being in a self-help group, die codependent
sample may have viewed honesty as being more important than culmral acceptabiUty. This
group may have answered in die non-sociaUy desirable direction in an effort to gain
approval dirough being honest These characteristics of die groups making up die sample
may account for thereversedresults of diis hypodiesis.
The fourth hypodiesis stated that extemal locus of control as measured by a
shortened version of Rotter's Intemal-Extemal Locus of Control Scale would be related
positively to codependency. Despite low reUability of die locus of control scale diis
hypothesis was supported by the results of the analysis, which indicated that people who
were more codependent had more of an extemal locus of control. Three characteristics of
codependency that are reflected in extemal locus of control are lack of tmst, low
self-worth, and controUing behavior. This finding helps to validate the working definition
of codependency as weU as the codependency scale.
The fifth hypothesis, which stated that self-esteem as measured by Rosenberg's
Self-Esteem Scale would berelatednegatively to codependency, was strongly supported.
The higher a person scored on codependency, the lower they scored on the self-esteem
scale. This finding helps to validate die codependency scale and supports the oft-cited
contention that low self-esteem is a characteristic of codependency (Beattie, 1987; Scheaf,
1986; Subby & Friel, 1984). Not only is diis useful information for codependency theory
and research, but also for cUnicians working with codependent cUents.
There was significant support found for the sbcth hypothesis which stated that age,
race, occupation, and income would be unrelated to codependency. The correlation of age
widi codependency was tested for die student sample only because diere were some
Al-Anon's over the age of 60. These outUers had die potential for distorting the
60
conflation. The results of die sixth hypodiesis suggests diat codependency is not partial
to age, race, occupation, or income. A codependent individual would be just as likely to
be in dieir twenties or dieir fifties, be Caucasian or of anodier race, be a blue collar worker
or a professional, or earn $5,000 per year or $40,000 per year. As might be expected in a
smdent sample, diere was Utde variation on income and occupation. Future research
should attempt to repUcate diesefindingsto insure the independence of codependency from
age, income, and occupation.
The seventh hypothesis which stated diat males would score lower dian females on
codependency was supported This seems to berelatedto the fact diat many of die
characteristics of codependency appear to be negativelyrelatedto what are traditionaUy
viewed as masculine type characteristics. AdditionaUy, it is important to note that there
were far fewer men dian women in the Al-Anon (13.3% men, 86.7% women) and
codependent samples (14.3% men and 85.7% women). This may indicate that in our
society fewer men deal with codependency if it is present than do women.
In hypothesis eight, the known group sample of codependents were predicted to
score higher on the codependency scale than the recovering Al-Anon group. This
hypothesis was very stongly supported, as the codependent group's mean was
considerably higher than the recovering group's mean. The codependent group's mean
placed them in the sUghdy agree range of the scale and the Al-Anon group's scores were in
the moderately disagree range. The mean for college students was 3.27, placing them in
between die two other groups in the sUghdy disagree range. This provides a very strong
indication that die codependency scale is acmaUy measuring codependency. It is worth
noting that die number of subjects in die codependency group was quite low (n=14). This
group's response rate was much lower than the Al-Anon group and much lower than
hoped for. A possible explanation for the lowerresponserate for the codependents is that
the codependents who did not remm the questionnaire packet may have initially taken it out
61
of dieir own codependent tendencies such as need for approval and difficulty making
decisions, but did not foUow dirough as it was an anonymous questionnaire.
Aldiough die intemal consistencyreUabdityscore was not as large in die present
smdy as it was for the same 15 items in the pilot smdy (alpha= .77; and alpha= .86
respectively), dus may berelatedto context In die pilot, die 15 items were embedded in a
38-item scale while in die present smdy, die 15 items constimted the only items of die
scale. It is also possible that smdents in die present sample who took die test on die last
day of die semester were less careful in responding to die items than were smdents in die
pilot who fUled out the items earlier in dieir semester. The carelessness hypothesis
receives some support in that the only tworeversescored items on the scale had the lowest
correlation widi the total scale score (r= . 12, and r= 29 respectively). Additional research
is needed on this issue. However, it is suggested that in fumre use of this scale, the words
"do not" appear underlined for items 5 and 7: "5. I usuaUy do not do things for other
people that they are capable of doing for themselves." and "7. I do not worry very much."
Tuming to other issues, fhe factor analysis supported the unidimensionaUty of the
codependency scale. Items loaded on the unrotated first factor, a factor which accounted
for 24.2% of the variance.
Conclusions and Recommendations
The presentresearchsmdy provides some support for thereUabiUtyand vaUdity of
die proposed codependency scale. Aldiough two of die hypodieses were not significandy
supported, hindsight aUows for possible explanations for their lack of support. These
expanations have been detailed in the previous section. Furthermore, most of the
hypodieses were supported, and these results suggest that this smdy accompUshed what it
set out to accompUsh: to develop a reliable and valid scale to measure codependency.
62
Thisresearchhas the potential to bridge dieory, practice, and research. As a result
of diis study, die theoretical concept of codependency should now be somewhat more
clear and concrete. The operational definition of codependency diat emerged in this smdy
has been largely supported by die findings, and can be useful to cUnicians as weU as
researchers. The 15-item codependency scale diat has emerged as areUableand vaUd
assessment tool for codependency has considerable value for clinicians who are treating
codependency in their practices, and who need a quick diagnostic instrument to point to a
possible area of intervention. This scale meets die need for a standaixUzed objective test to
measure codependency which can be scored quickly and accurately. AdditionaUy, this
measurement of codependency can serve as an aid in die pursuit of self-awareness and
self-determined change.
A valid and reliable measure of codependency also provides possibiUties for future
research. Such a tool could be used to compare different populations, as well as broaden
understanding of interpersonal relationships. It could also be used to assess progress for
a person who is codependent and involved in some sort of treatment for codependency.
Some recommendations for futureresearchsmdies on codependency include: (a)
looking for generational pattems of codependency, (b) smdying interpersonal dynamics
surrounding codependency, arid (c) examining communication pattems that codependents
engage in with their families or significant others. These studies should come from a
family systems point of view since they arerelatedto how a system affects an individual
and how an individual influences a system. Furtherresearchcould be done to study
therapists' own issues with codependency and how that influences their dierapy. It also
remains for fumreresearchto investigate therelationshipof codependency to such
intrapersonal variables as anxiety, depression, and general adjustment, as well as to such
interpersonal variables as loneliness, intimacy, and equity inrelationships.It could be
aigued that the characteristics of codependency and items from the codependency scale
63
represent general lack of adjustment. Although single items used to correlate
codependency with adjustment showed moderate correlations, further research is needed
to estabUsh the ways in which codependency is similar to and different from adjustment in
general.
In conclusion, the present smdy marks thefirststep in providing areUableand valid
measure of codependency. As the first empirical smdy of codependency, it can begin to
serve as a bridge to existing theory and practice, and a catalyst for future research.
REFERENCES
Asher, R & Brissett D. (1988). Codependency: A view from women manied to

alcoholics. Thg International Journal of th^ Addictions. r\. ^Î-'^SO
Ashkanasy, N. W. (1985). Rotter's Intemal-Extemal Scale: Confirmatory factor analysis

and correlation widi social desirabiUty for alternative formats. Journal of PersonaUtv
and Social Psychology. 4« 1328-1341.
Beattie, M. (1987). Codependent no mnry New York: Harper/Hazelden.
Beck, A. T. (1967). Det>ression! CUnical. experimental and dieoretical aspects. New

York: Harper & Row.
Beck, A. T., Weissman, A., Lester, D., & Trexler, L. (1974). The measurement of
pessimism: The hopelessness scale. Journal of Consulting and Clinical Psvchologv.
42,736.
Bem, S. L. (1974). The measurement of psychological androgyny. Journal of

Consulting and Clinical Psychology. 42. 155-162.
Bem, S. L. (1977). On the utiUty of altemative procedures for assessing psychological

androgyny. Journal of Consulting and Clinical Psvchologv. 45.196-205.
Bepko, C , & Krestan, J. A. (1985). The responsibilitvttâp:A blueprint for treating the
alcoholic family. New York: MacmiUan.
Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by

multitrait-multimethod matrix. Psvchological Bulletin. 56. 81-105.
CapeU-Sowder, K. (1984). On being addicted to the addict: Co-dependent relationships.

In Co-dependencv: An emerging issue (pp. 19-24). Pompano Beach, FL: Healdi
Communications.
Carmines, E. G., & Mclver, J. P. (1981). Analyzing models widi unobserved

variables. In G. W- Bohmstedt & E. F. Borgatta (Eds.), Social measurement Current
issues. Beverly Hills, CA: Sage.
Cermak, T. (1986a). Diagnostic criteria for codependency. Joumal of Psychoactive
Drugs, j l , 15-20.
Cermak, T. (1986b). Diagnosing andtteatingco-dependence. MinneapoUs: Johnson
Institute Books.
Coleman, R.F. (1975). Manipulation of self-esteem as a determinant of mood of elated
and depressed women. Joumal of Abnormal Psvchologv. 84. 693-700.
64
65
«*Ill*;'î^^'^^^^?'l:i^* J^^^^^- Yeasayers and naysayers: Agreeing response set as
a personality vanable. Joumal of Ahnnr^î ,nd s'nH.l Psyrhnin^y^^ i s i - i 7 4
Crowne, D. P., & liverant,, S. (1963). Confonnity under vaiying conditons of

personal committnent Joumal of Ahnormal and Snrial Psychology. 66. 547-555.
Crowne,p. P., & Marlowe, D. (1960). A new scale of social desirabUity independent of
psychopadiology. Joumal of Consulriny Psyrholnyy, 74 349-354.
Crowne, D. P., & Marlowe, D. (1964). The anprnyal mnriyp New York: Wiley, 1964.
EdwMxis,A. (1957). The social desirahility vapahle in personality assessment New
York: Dryden.
Gaudiier, J., Pellerin, D., & Renaud, P. (1983). The enhancement of self-esteem: A
companson of two cognitive stratedies. Cognitive Therapy and Research. 7. 389-398.
Gierymski, T., & Wdliams, T. (1986). Codependency. Joumal of Psychoactive
Drugs, I&, 7-13.
Gorski, T., & MiUer, M. (1984). Co-alcoholic relapse: Family factors and warning
signs. In Co-dependencv: An emerging issue rpp. 77-«2V Pompano Beach, FL:
Health Communications.
Greenleaf, J. (1984). Co-alcohoUc/para-alcoholic: Who's who and what's the

difference. In Co-dependencv: An emerging issue (pp. 5-18). Pompano Beach, FL:
Health Communications.
Jaques, J. M., & Chason, K. J. (1977). Self-esteem and low stams groups: A changing
scene? The Sociological Ouarteriv. 18. 399-412.
Joe, V. C. (1971). Review of the intemal-extemal control constmct as a personality

variable. Psychological Reports. 28. 619-640.
Lefcourt, H. M. (1972). Recent developments in the smdy of locus of control. In B. A.

Maher (Ed.), Progress in experimental personalityresearch(pp. 1-39). New York:
Academic Press.
Lefcourt, H.M. (1976). Locus of control: Current trends in theory and research. New
York: Wiley.
Lefcourt, H.M. (1980). Personality and locus of contt-ol. In J. Garber & M. E. P.

SeUgman (Eds.), Human helplessness: Theory and applications (PP. 245-259). New
York: Academic Press.
Liverant, S., & Scodel, A. (1960). Intemal and extemal conttx)! as determinants of
decision making under conditions of risk. Psychological Reports. 7. 59-67.
Lubinski, D., Tellegen, A., & Butcher, J. N. (1981). Therelationshipbetween

androgyny and subjective indicatiors of emotional well-being. Joumal of Personality
and Swial PsyghQlogy. 4Q, 722-730.
66
Lubinski, D., Tellegen, A., & Butcher, J. N. (1983). Masculinity, femininity and
^ g y n y viewed and assessed as distinct concepts. Joumal of Personality and
Social Psypl]oloyy. dd 428-439.
Miller, W. R., & SeUgman, M. E. P. (1976). Learned helplessness, depression and die
perception ofreinforcementBehaviour Research and Therapy. 14.7-17.
Payne, F. D. (1987). "MascuUnity," "Femininity," and die complex consttaict of
adjusttnent Sex Roles. 17 359-374.
Phares, E. J. (1973). Locus of control: A personality determinant of behavior.
Momstown, NJ: General Learning Press.
Prociuk, T. J., Breen, L. J., & Lussier, R. J. (1976). Hoplessness, intemal-extemal
locus of conttol, and depression. Joumal of ninical Psychology. 32. 299-300.
Prociuk, T. J., & Lussier, R. J. (1975). Intemal-extemal locus of conttol: An analysis
and bibliography of tworesearchyears (1973-1974). Psychological Reports. 37.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton Univ. Press,
Princeton, NJ.
Rotter, J. B. (1954). Social leaming and clinical psychology. Englewood Cliffs, NJ:
Prentice-HaU.
Rotter, J. B. (1966). Intemal-extemal conttol ofreinforcementPsychological
Monographs. 80 (1, Whole No. 609).
Rotter, J. B. (1975). Some problems and misconceptions related to the constmct of
intemal versus extemal control ofreinforcementJoumal of Consulting and CUnical
Psychology. 43. 56-67.
Rotter, J. B., Chance, J., & Phares, E. J. (Eds.) (1972). Applications of a social
leaming theory of personalitv. New York: Holt, Rinehart & Winston.
Schaef, A. W. (1986). Co-Dependence: Misunderstood - Mistteated. San Fransisco:
Harper & Row.
SUber, E., & Tippett, J. (1965). Self-esteem: Clinical assessment and measurement
validation. Psychological Reports. 16.1017-1071.
Smalley, S. (1987). Dependency issues in lesbian relationships. Joumal of
Homosexuality. 14,125-135.
Sontag, S. (1978). Illness as metaphor. London: Penguin.
Spence, J. T. (1983). Comment on Lubinski, Tellegen, and Butcher's "MascuUnity,
femininity, and androgyny viewed and assessed as distinct concepts." Joumal of
Personality and Social Psvchologv. 44.440-446.
67
^^^"r^VcLL'g.1^^^ ^l^^^- Ma.SCulinityanHfen.ininify. Th.^r
i'xl!" g^mensions mmelates anH nntrrrdrnr^ Austin: University of Texas
^ ^ T ^ J ' 7 * ' ^ j^H.^'Fcich, R. L. (1979). The many faces of androgyny: A reply to

LOCKsiey and Colten. Joumal of Personality and .Sncial Psychology. 37. 1032-1046.
Spence, J. T., Helnireich, R. L., & Stt^pp, J. (1974).The Personal Attributes

^^lestionnaire: A measure of sex role stereotypes and mascuUnity-femininity. Joumal
^tiPPlgPient Abstract Service Catalog ^f Selecte/i Dnruments in Psychology. 4. 43.
Spence, J. T., Helmreich, R. L., & Sttapp, J. (1975). Ratings of self and peers in
sex-role attnbutes and dieir relation to self-esteem and conceptions of masculinity and
feminimty. Joumal of Personality and fSocial Psycholo^ -M 9Q-^Q
Strickland, B. R., & Crowne, D. P. (1963). Need for approval and die premattu-e
^Î^JJ^^tionofpsychodierapy, Joumal of Consulting and Clinical Psvchologv. 27.
Subby, R. (1984). Inside die chemically dependent marriage: Denial & manipulation. In
Co-Dependencv: An Rmerging Issue Tpp 25-30). Pompano Beach, FL: Health
Communications.
Subby, R. (1987). Lost in the Shuffle: The Co-Dependent Reality. Pompano Beach,
FL: Health Communications.
Subby, R., & Friel, J. (1984). Co-dependency: A paradoxical dependency. In

Co-Dependencv: An Emerging Issue (pp. 31-44). Pompano Beach, FL: Healdi
Communications.
Tellegen, A., & Lubinski, D. (1983). Some methodological comments on labels, ttaits,
interaction, and types in the smdy of "Femininity" and "Masculmity": Reply to
Spence. Joumal of Personalitv and Social Psychology. 44. 447-455.
Wegscheider-Cmse, S. (1984). Co-dependency: The dierapeutic void. In

Co-Dependency: An Emerging Issue (pp. 1-4). Pompano Beach, FL: Health
Communications.
Wegscheider-Cmse, S. (1985). Choice-Making. Pompano Beach, FL: Health
Communications.
Whitfield, C. (1984). Co-dependency: An emerging problem among professionals. In

Co-Dependency: An Emerging Issue (pp. 45-54). Pompano Beach, FL: Health
Communications.
Whidey, B.E. (1983). Sex role orientation and self-esteem: A critical meta-analytic
review. Joumal of Personality and Social Psychology. 44. 765-778.
Wiggins, L. S., & Holzmuller, A. (1981). Further evidence on androgyny and

interpersonal fexibility. Joumal of Research in PersonaUtv. 15. 67-80.
Wilson, A. R., & Krane, R. V. (1980). Change in self-esteem and its effects on
symptoms of depression. Cognitive Therapy and Research. 4. 419-421.
68
Woititz, J. (1984). Chemical dependency: Insidious invader of intimacy. In

Co-Dependencv: An Emerging Issue (pp. 55-60). Pompano Beach, FL: Health
Coimnunications.
Wylie, R. (1961). The self-concept Lincoln, Nebraska: University of Nebraska Press.
APPENDIX A
BEM SEX ROLE INVENTORY
69
70
Describe yourself using the following:
1 Never or almost never tme

2 UsuaUy not tme
3 Sometimes but infiequendy tme
4 OccasionaUy ttne
5 Often tme
6 UsuaUy tme
7 Always or almost always tme
1. Self reUant, can count on myself.

2. Yielding or give in to odiers.
3. Helpful.
4. Defend my own beliefs.
5. Cheerful.
6. Moody.
7. Independent.
8. Shy.
9. Conscientious, do what I should.
10. Athletic.
11. Affectionate.
12. Theatrical or dramatic.
13. Assertive, stand up for myself.
14. Flatterable, can be flattered.
15. Happy.
16. Sttong personality.
17. Loyal.
18. Unpredictable.
19. Forceful.
20. Feminine.
21. ReUable, others can count on me.
22. Analytical or logical.
23. Sympathetic.
24. Jealous.
25. Leadership abiUty.
26. Sensitive to the needs of others.
27. Tmdiful.
71
28. WilUng to take risks.
29. Understanding.
30. Secretive, keep things to myself.
31. Make decisions easily.
32. Compassionate or feeling for odiers.
33. Sincere.
34. Self-sufficient.
35. Eager to soodie hurt feeUngs.
36. Conceited.
37. Dominant or in charge.
38. Soft-spoken.
39. Likable.
40. MascuUne.
41. Warm.
42. Solemn or serious.
43. WiUing to take a stand
44. Tender.
45. Friendly.
46. Aggressive.
47. Gullible, fall for things easily.
48. Inefficient, don't have good work habits.
49. Act as a leader.
50. Childlike.
51. Adaptable.
52. IndividuaUstic, particidar to me.
53. Do not use harsh language.
54. Unsystematic or unorganized.
55. Competitive.
56. Love children.
57. Tactful, earful about what is said.
58. Ambitious.
59. Gentle.
60. Conventional, go along with other's expectations.
APPENDDC B
MARLOWE-CROWNE SOCIAL DESIRABILITY SCALE
72
73
Listed below are a number of statements conceming personal attimdes and traits. Read
each Item and decide whedier die statement is true (J) or false (F) as it pertains to you
personally. ^ / ./ f
1. Before voting I dioroughly investigate die quaUfications of aU die candidates.

2. I never hesitate to go out of my way to help someone in trouble.
3. It is sometimes hard for me to go on widi my work if I am not encouraged.

4. I have never intensely disUked anyone.
5. On occasion I have had doubts about my abUity to succeed in Ufe.
6. I sometimes feel resentful when I don't get my way.
7. I am always careful about my manner of dress.
8. My table manners at home are as good as when I eat out in a restaurant
9. If I could get into a movie without paying and be sure I was not seen I would
probably do it
10. On a few occasions, I have given up doing something because I thought too
UtdeofmyabiUty.
11. I like to gossip at times.

12. There have been times when I felt like rebeUing against people in authority even
though I knew they were right
13 No matter who I'm taUdng to, I'm always a good Ustener.
14. I canremember"playing sick" to get out of something.
15. There have been occasions when I took advantage of someone.
16. I'm always wiUing to admit it when I make a mistake.
17. I always try to practice what I preach.

18. I don't find it particularly difficult to get along widi loud mouthed, obnoxious
people.
19. I sometimes try to get even radier dian forgive and forget.
20. When I don't know something I don't at all mind admitting it.
21. I am always courteous, even to people who are disagreeable.
22. At times I have really insisted on having diings my own way.

74
23. There have been occasions when I felt Uke smashing diings.
24. I would never think of letting someone else be punished for my wrongdoings.
25. I never resent being asked toretuma favor.

26. I have never been irked when people expressed ideas very different from my
own.
27. I never make a long trip without checking die safety of my car.
28. There have been times when I was quite jealous of the good fortune of others.
29. I have almost never felt the urge to teU someone off.
30. I am sometimes irritated by people who ask favors of me.
31. I have never felt that I was punished without cause.
32. I sometimes diink when people have a misfortune they only got what diey
deserved.
33. I have never deUberately said something that hurt someone's feelings.
APPENDDC C
THE SHORTENED VERSION OF THE INTERNAL VERSUS
EXTERNAL CONTROL OF REINFORCEMENT SCALE
75
76
Please select the one statement of each of die following pairs which you more sttongly
^ w l ! i K • , • ^^^^^ ^^ ^^ y^"" ^ concemed Indicate the statement diat you have
selected by circUng die letter diat precedes it
1. a. Chddren get intottioublebecause dieir parents punish them too much.

D. ihe ttouble widi most chddren nowadays is diat dieir parents are too easy widi
2. a. Many of die unhappy diings in people's lives are partly due to bad luck,
b. People s misfortunes result from die mistakes diey make.
3. a. One of die majorreasonswhy we have wars is because people don't take enough
interest in poUtics.
b. There wiU always be wars, no matter how hard people tty to prevent diem.
4. a. In die long run people get the respect diey deserve in diis worid.
b. Unfortunately, an individual's worth often passes unrecognized no matter how
hard he tties.
5. a. The idea that teachers are unfair to smdents is nonsense.

b. Most smdents don't reaUze die extent to which dieir grades are influenced by
accidental happenings.
6. a. Without therightbreaks one cannot be an effective leader.

b. Capable people who fad to become leaders have not taken advantage of their
opportunities.
7. a. No matter how hard you try some people just don't like you.
b. People who can't get others to like them don't understand how to get along with
others.
8. a. Heredity plays the major role in determining one's personality,

b. It is one's experiences in Ufe which determine what they're like.
9. a. I have often found that what is going to happen will happen.

b. Tmsting to fate has never tumed out as well for me as making a decision to take a
definite course of action.
10. a. In the case of the weU prepared student there is rarely if ever such a thing as an
unfair test
b. Many times exam questions tend to be so unrelated to course work that studying
is really useless.
11. a. Becoming a success is a matter of hard work, luck has Utde or nothing to do with
it.
b. Getting a good job depends mainly on being in therightplace at therighttime.
12. a. The average citizen can have an influence in government decisions.

b. This world is run by the few people in power, and there is not much the Utde guy
can do about it.
77
13. a. When I make plans, I am almost certain that I can make them work.
b. It is not always wise to plan too far ahead because many things mm out to be a
matter of good or bad formne anyhow.
14. a. There are certain people who are just no good,

b. There is some good in everybody.
APPENDIX D
ROSENBERG SELF-ESTEEM SCALE
78
79
Describe yourself using the foUowing:
1 Almost always
2 Often
3 At times
4 Seldom
5 Never
1. I feel that I'm a person of worth, at least on an equal basis with others.
2. I feel that I have a number of good quaUties.
3. AU in all, I am incUned to feel that I am a failure.
4. I am able to do things as weU as most other people.
5. I feel I do not have much to be proud of.
6. I take a positive attimde toward myself.
7. On the whole, I am satisfied with myself.
8. I wish I could have morerespectfor myself.
9. I certainly feel useless at times. At times I diink I am no good at aU.

APPENDDC E
ORIGINAL 38-ITEM CODEPENDENCY SCALE
80
81
d^^h^J^^^'^'^^^^^^'^^T
descnbes you according to the ffoUowing
JJ P^^^ Ust:
*^ " ^ ^ in * e spaces provided diat best
1 Strongly Disagree
2 Moderately Disagree
3 SUghdy Disagree
4 SUghdy Agree
5 Moderately Agree
6 Strongly Agree
1. I have a hard time knowing what I am feeling.
2. I feel like I should not burden odier people widi my problems.
3. I Idee to focus on my problems more than on odier people's problems.
_ 4. When I don't say what I want to say it is often in order to avoid anodier person's
anger.
_ 5. It is easy for me to taUc about what I am feeUng at die moment

_ 6. I often find myself tmsting the wrong people.
_ 7. It is hard for me to make decisions.
_ 8. It is easy for me to tmst people.
_ 9. I usually pay more attention to another person's needs than to my own.

10. It is hard for me to say "no."
11. It is hard for me to accept compUments graciously.
12. Several of my relationships are with people who seem to have serious problems
13. Sometimes, I almost feel bored or empty if I don't have problems to focus on.
14. I usuaUy do not do things for other people that they are capable of doing for
themselves.
15. It is harder for me to identify and solve someone else's problem than it is to
identify and solve my own.
16. It is easy for me to let other people do things for me.
17. When I do something nice for myself I usually feel gudty.
18. I diink a lot about what other people think of me.
19. I focus as much energy on myself as I do on other people.

82
20. Sometimes I am unable to quit diinking about another person or event.
21. I do not worry very much.
22. I do not usuaUy abandon myroutinewhen I'm upset about somebody or

something.
23. People very close to me say I'm always telUng diem what to do.
. 24. I seem to haverelationshipswidi people where we spend a lot of time sttiiggling

for power.
. 25.1 usuaUy face my problems head on and tty to solve them as soon as they come
up.
. 26. I teU myself that things wdl get better when the people in my life change what they
are doing.
. 27. I seem to haverelationshipswhere I am always there for them, but they are rarely
there for me.
. 28. I don't have time to do what I need to do for myself.

_ 29. Sometimes I get focused on one person, to the extent of neglecting other
relationships and responsibilities.
_ 30. The way that I feel about myself often depends on die way diat odiers feel about
me.
_ 31. When I'm alone, I rarely get bored or resdess.
_ 32. I seem to get intorelationshipsdiat are painful for me.
_ 33.1 don't usually let others see die "real" me.

_ 34. I feel as if I have to waUc on eggshells when I'm around some people.
35. When someone upsets me, I wdl hold it in for a long time, but once in a while I
explode.
36. I wdl usuaUy go to any lengths to avoid open conflict.
37. I often have a sense of dread or impending doom.
38. I remain loyal even after people hurt me.

APPENDDC F
DEMOGRAPHIC QUESTIONNAIRE
83
Please provide the foUowing information about yourself
Age:
Race/Ethinc:
Caucasian/White
^Black
Hispanic
__Oriental/Asian
^_Native American
Sex: Female, Male
Number of brothers:
Number of sisters:
Marital stams closest to how you are now:

single
married
divorced
widowed
remarried
If married, how long? yrs.
If divorced, how long? yrs.
PresentreUgiousaffiliation or preference:
CadioUc
Protestant/Christian
Jewish
Other (specify)
When you were growing up, were you primarily a resident of:
Rural community or farm
Small city/Suburb
Large city
Are you a student? yes, no. If so, what is your G.P.A?
Major area of smdy:

Social /Behavioral Science
Physical Science
Arts/Humanity
Engenering
Agricidture
Home Economics
Are you employed? If so, what is your major employment?

85
___Professional
Blue Collar
Sales
_Qerical
^Housewife
^__Manager
^Farmer
How many hours per week do you work?
Income: under $5,000
$5,000 to $10,000
$10,000 to $15,000
$15,000 to $20,000
$20,000 to $25,000
$25,000 to $30,000
$30,000 to $40,000
$40,000 to $60,000
over $60,000
One of my parents is/was a practicing alcoholic.
One of my parents is/was arecoveringalcoholic.
My spouse/significant other is/was a practicing alcoholic.
My spouse/significant other is/was a recovering alcohoUc.
From birth to age 18 how many years did you live with a practicing alcoholic? ,A
recovering alcoholic?
In general, how satisfied are you with your life?

1 Not at aU satisfied
2 MUdly satisfied
3 Moderately satisfied
4 Strongly sarisfied
Compared to others, in general, how well adjusted do you think
you are?
1 poorly adjusted
2 MUdly adjusted
3 WeU adjusted4 Very weU adjusted
PERMISSION TO COPY
In presenting this thesis in partial fulfillment of the
requirements for a master's degree at Texas Tech University, I agree
that the Library and my major department shall make it freely avail-
able for research purposes. Permission to copy this thesis for
scholarly purposes may be granted by the Director of the Library or
my major professor. It is understood that any copying or publication
of this thesis for financial gain shall not be allowed without my
further written permission and that any user may be liable for copy-
right infringement.
Disagree (Permission not granted) Agree (Permission granted)
Student's signature Student6s signature
Date Date ^

31295005501662

Uploaded by

Copyright:

Available Formats

31295005501662

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

31295005501662

Uploaded by

Copyright:

Available Formats

DEVELOPING A SCALE TO MEASURE CODEPENDENCY

Submitted to the Graduate Faculty

Copyright 1989, Lynda A. Spann

I want to express my great appreciation to Judith L. Fischer, chairperson of my

Conclusions and Recommendations 61

1. Descriptive Statistics of Total Sample (N=236) 40

Statement of the Problem

designed to address the need to strengthen the concept of codependency by developing a

reliable and valid scale to measure codependency.

In order to come up with a codependency scale, existing theories on codependency

Test-retest reliability and internal consistencyreliabihtywere computed and scale

modifications were carried out as guided by these findings.

thought to be high or low in codependency, second, college students were assessed to

Developing a scale to measure codependency should serve the trifold purpose of

the pursuit of self-awareness and self-determined change. Clinicians are increasingly

standardized tool to aid them in assessing codependency. An assessment tool for

cropping up in the United States designed exclusively to treat codependency. These

treatment programs need to have at their disposal an objective test to measure

codependency which can be scored quickly and accurately.

futureresearchand clinical contributions that mayresultfrom the availability of a reUable

and valid measure of codependency.

In order to develop a complete conceptual definition of the tenn codependency, it is

defining codependency, theorists in the field often cite symptoms,reactions,problems,

and agreement among these theorists about the characteristics of codependency.

Several issues have emerged withrespectto the definition of codependency. One

In "Co-Dependency: An Emerging Problem Among Professionals" (1984), Charles

Whitfield states, "Co-alcoholism can be defined as ill health, or maladaptive or problematic

Whitfield maintains a broader perspective on co-alcoholism or codependency than do some

narrower perspective on the term co-alcoholism. They state, "Co-alcoholism is a primary

condition thatresultsfrom the debilitating physiological stress produced by living in a

ICrestan describe co-alcoholics as "parents and children of alcoholics and nondrinking

co-alcoholic, but does not offer a definition of co-alcoholism.

para-alcoholic. Greenleaf examines the "differences in the derivation of behavior of an

adult in a volitionalrelationshipwith an alcoholic spouse and of a child in a compulsory

behavior from both parents and becomes para-alcoholic" (p. 6).

Kathy Capell-Sowder (1984) uses the term codependency and describes a

codependent, saying, "the person involved in a primary loverelationshipwith someone

experiences a growing preoccupation with alcohol, an increasingly higher tolerance of

Similarly, the codependent experiences an increasing focus on the alcoholic, an increase in

As demonstrated above, many theorists view codependency as being solely related

have a broader perspective on codependency, viewing it as a condition that is not

necessarily linked to chemical dependency.

Although Sharon Wegscheider-Cmse (1985) places an emphasis on alcoholism, she

defines codependency as "a specific condition that is characterized by preoccupation and

extreme dependence (emotionally, socially, and sometimes physically), on a person or

object. Eventually, this dependence on another person becomes a pathological condition

her, and who is obsessed with controlling that person's behavior."

In "Co-Dependency~A Paradoxical Dependency" (1984), Subby and Friel write,

"Let us define co-dependency, then, as a dysfunctional pattern of living and

codependency as "An emotional, psychological and behavioral condition that develops as a

personal and interpersonal problems" (26). In "Co-Dependence: Misunderstood -

is progressively death-oriented" (p.21).

In "Diagnosing and Treating Co-Dependence" (1986a), Cermak defines

codependence as "arecognizablepattem of personality traits, predictably found within

dysfunction to warrant the diagnosis of Mixed Personality Disorder as outiined in DSM

and not merely described by eloquent metaphors" (1986b, p. 18).

In "Co-Dependency: An Emerging Issue" (1984), several authors (Greenleaf;

Subby & Friel; and Woititz) write of common characteristics of codependency. In

describing co-alcoholics and para-alcoholics, Jael Greenleaf (1984) lists behavioral

among codependent people (co-alcoholics and para-alcoholics) are: grandiosity, lack of