Understanding women's depression : the roles of relationship loss and self loss


Material Information

Understanding women's depression : the roles of relationship loss and self loss
Physical Description:
x, 88 leaves : 29 cm.
Drew, Sarah S., 1957-
Publication Date:


Subjects / Keywords:
Psychology thesis, Ph. D
Dissertations, Academic -- Psychology -- UF
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1995.
Includes bibliographical references (leaves 82-87).
Additional Physical Form:
Also available online.
General Note:
General Note:
Statement of Responsibility:
by Sarah S. Drew.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 34403389
System ID:

Table of Contents
    Title Page
        Page i
        Page i-a
        Page ii
        Page iii
        Page iv
        Page v
        Page vi
    Table of Contents
        Page vii
        Page viii
        Page ix
        Page x
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
    Chapter 2. Review of literature
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
    Chapter 3. Methodology
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
    Chapter 4. Results
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
    Chapter 5. Discussion
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
    Appendix A. Informed consent
        Page 72
        Page 73
    Appendix B. Instructions
        Page 74
    Appendix C. Selected items from the beck depression inventory
        Page 75
    Appendix D. Selected items from the dependency scale of the depressive experiences questionnaire
        Page 76
    Appendix E. Selected items for the silencing the self scale
        Page 77
    Appendix F. Relationship loss questionnaire
        Page 78
        Page 79
    Appendix G. Demographic questionnaire
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
    Biographical sketch
        Page 88
Full Text






Copyright 1995


Sarah S. Drew

My Parents and My Grandmother

Who have instilled in me both a love of learning and a compassion for all living beings. I feel blessed by their patience and by the abiding faith they
have shown as I have journeyed along the pathway of becoming.


The opportunity to thank those who have contributed to the
completion of my dissertation is truly a welcome one. I am honoredto have worked with all five members of my doctoral committee. Each is a uniquely gifted individual, and each has made an invaluable contribution both to my dissertation and to my training as a psychologist.
To my chairman, Dr. Harry Grater, I will be eternally grateful for my training as a therapist. In his quiet, provocative way he ignited and nurtured the passions of my mind. I will always remember fondly the countless hours spent discussing cases and theories and more cases. The training I have received from Harry will be the foundation on which I build my career.
Through the entire dissertation process, from conception to final editing, my cochairman, Dr. Martin Heesacker, has been my Rock of Gibraltar. Marty has given generously of both his time and expertise. Even more importantly, he has patiently stood by me as I faced tidal waves of fear and doubt. Without exception, Marty has been available to gently urge me forward. His confidence in me has supported my substantial growth as both a scholar and scientist.

My relationship with Dr. Robert Ziller has been a long and special one. Throughout my graduate school career, Bob has appeared at opportune moments with a warm smile and wise words of iv

encouragement. His singular creativity has also served as an inspiration. In a discipline in which most follow the well beaten path, Bob has been one to consistently blaze his own trail. I salute his courage.
As a female student in counseling psychology, I have been
especially attentive to the women faculty with whom I have studied and worked. I have looked to these women as both mentors and role models-for lessons on professional comportment, for clues on how to integrate the pieces of my life. I feel deeply fortunate to have encountered Dean Phyllis Meek in the course of my education. Over the years, Phyllis and I have interacted in a variety of different environments--classes, conferences, practice. In every instance, I have been impressed with her gentleness and with her unfailing professionalism. Phyllis has both the courage to speak her truth and the sensitivity to deal respectfully with all whom she encounters. We share a special interest in and commitment to the issues and problems of women's lives. I would be proud to follow in her footsteps.

Last, but certainly not least, I wish to thank Dr. Paul Schauble. Paul has been supervisor, friend, and spiritual mentor. He has been at my side through dark times, and valued and believed in me at points when I had lost faith in myself. More than anyone else, he has helped me along the path toward wholeness. I honor his vision, his quiet wisdom and strength. Paul will always have my deep love and respect.
Numerous other friends and &mily members have offered me much needed support through the stressful process of completing my dissertation. My sisters, Lanier and Elizabeth, have always been there with warm words of encouragement. The knowledge of their loving presence I have carried deep within myself. Kerry has been not only my

technical wizard, but my confidante and companion through much of the turmoil and change of graduate education. I thank him for his unceasing love and for teaching me how to seize the moment. Peggy, David, and Clyde, of the "Seminar from Hell," shared ideas and provided a safe place for venting. Roberta and Sandy have stood staunchly by my side, nurtured me through the periods of self-doubt, and respected my need for time and solitude. Amy has been a tireless believer in my potential and a wonderfully encouraging friend. Her tangible gifts have meant more than I can ever say. Mary Beth has been a faithful believer in my abilities, and more importantly in my creative self. I thank her for her patience and unflagging confidence in my ability to spread my wings and soar. Joe joined me late in this process, but has played an invaluable role. I deeply appreciate the help he has given me in tapping the well of my own strength and finding the inner resources to complete this project. He has been earth and fire to my air and virater. And finally, I owe a deep debt of gratitude to Jeff whose irreverent grin and indomitable spirit will live on always in my heart.



ACKNOW LEDGMENTS .................................................................... iv

ABSTRACT ....................................................................................... ix


I INTRODUCTION ................................................................ 1

2 REVIEW OF LITERATURE ................................................. 7

Introduction ......... .... ******** ...... **- **- **- ********- ***** ..... **'*' 7
Definitions and Incidence ................................................... 7
Explanatory Theories for Depression ................................... 13
Psychoanalytic and Traditional Theories ....................... 13
Relational Theories of Developm ent ............................. 18
JaclCs Theory of Depression in Women ......................... 31
Hypotheses ........................................................................ 34

3 METHODOLOGY ............................................................... 37

Participants ....................................................................... 37
Procedures ......................................................................... 38
Independent Variables .................................................. 38
Dependent Variables .................................................... 43
Data Analysis .................................................................... 46

4 RESULTS .......................................................................... 48

Sam ple Characteristics ...................................................... 48
Means and Frequencies ...................................................... 49
Hypothesis Tests ................................................................ 51
Replication of Hypothesis Tests ........................................... 52
Ancillary Analyses .............................................................. 53


5 DISCUSSION ..................................................................... 57

Hypothesis 1 ...................................................................... 57
Hypothesis 2 ...................................................................... 58
Hypothesis 3 ...................................................................... 59
Sex Differences .................................................................. 60
Limitations ........................................................................ 61
Strengths ........................................................................... 63
Suggestions for Future Research ......................................... 65
Implications for Psychotherapy ............................................ 68


A INFORMED CONSENT ....................................................... 72

B INSTRUCTIONS ................................................................. 74

INVENTORY ................................................................. 75

QUESTIONNAIRE .......................................................... 76


G DEMOGRAPHIC QUESTIONNAIRE ...................................... 80

REFERENCES .................................................................................. 82

BIOGRAPHICAL SKETCH .................................................................. 88


Abstract of Dissertation Presented to the Graduate School of the
University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy


Sarah S. Drew

December 1995

Chair: Harry A.Grater
Cochair: Martin Heesacker

The purpose of this study was to explore the role of loss as a
contributory factor in women's depression. In particular, this study evaluates the effects of loss of an intimate partner and loss of self as precipitants to depression. Extending the work of feminist scholarswho have challenged existing theories of women's depression, three models were evaluated. The Traditional Model predicts an association between loss of a significant relationship and depression. The "Self Loss" Model predicts a significant association between self loss and depression; and finally, the Potentiation Model predicts a significant interaction between relationship loss and self loss on depression. Participants in the study were 257 female and male undergraduates who completed a relationship loss questionnaire, a self loss questionnaire, and two standard depression instruments. Results of the study and a replication both showed clear support for the "Self Loss" Model and little support for the alternative models. Loss of self was associated with a significantly greater degree of depression on both depression measures. No association ix

between relationship loss, or the interaction of self and relationship loss was detected. Although the primary intent of this study was to expand knowledge regarding women's depression, the same pattern of results held for both men and women, thus providing insight into men's depression as well. Though fewer men met the criteria for depression, the correlation between self loss and depression still held. The implications of these findings for theory, research and clinical intervention, are discussed.



Man's love is of man's life a thing apart, 'Tis woman's whole existence.
(Byron, 1818 /1958)

There is virtually unanimous agreement among mental health professionals that depression is one of the most commonly occurring psychological disorders. According to a study by Eaton and Kessler R 98 1), approximately one out of six Americans was found to have at least a moderate level of depression. Furthermore, epidemiological studies have repeatedly and consistently shown that women are more likely to become depressed than men (e.g., Boyd & Weissman, 1982; Charney & Weissman, 1988; Eaton & Kessler, 198 1). Although the actual statistics vary, the most conservative estimates suggest that women are twice as vulnerable as men to experiencing some form of depressive illness, at some point in their lives (Weissman & Klerman, 1987). Even more disturbing, recent research points to a steady Increase in the incidence of depression, especially among younger women (NolenHoeksema, 1990; Nolen-Hoeksema & Girgus, 1994; Scarf, 1980).
Explanations for this sex difference in the rate of depression fall into three general categories: those focusing on biological factors, those targeting primarily personality variables, and those highlighting sociocultural influences. The present study will explore only the latter


two categories, both of which seem critical to expanding our understanding of women's depression. (Clearly, any truly complete explanation of the causes of women's depression would have to take into account biological influences on mood and behavior. However, a biological focus was beyond the scope of the present study.)
Predating Freud's more systematic study of the connection between loss and depression (or melancholia) by nearly two thousand years, human beings noted the connection between relationship loss and dysphoria. Though a finer distinction is now drawn between the affects and behaviors that characterize the normal mourning process and the clinical symptoms of depression, the presumed link between loss and the onset of depression has remained strong. At present, several widely accepted, psychoanalytically based theories of depression are built on the assumption that women, in particular, are likely to become depressed in response to interpersonal loss (Blatt & Zuroff, 1992; Lerner, 1988).
Although loss has been shown to be a strong contributor to the onset of depression, the present studywill attempt to expand the field's understanding of the specific role or roles loss may play. By integrating newly emerging theories of women's development with more established work on depression, the opportunity arises to deepen the field's knowledge of a complex disorder and to add a fresh perspective to the continuing dialogue on sex differences.
Over the past decade, feminist theorists, such Carol Gilligan,
Nancy Chodorow, and Jean Baker Miller and her colleagues at the Stone Center of Wellesley College, have challenged the view that healthy female development follows precisely the same pathway as that of healthy male development (Chodorow, 1978; Gilligan, 1982; Miller, 1986; Surrey, 199 1).

Although these theorists and researchers have embraced some aspects of object relations theory, primarily the emphasis on the interpersonal matrix of personality development and the critical role of early parenting, they have questioned the portrayal of development solely as a process of separation from significant others (Chodorow, 1978; Miner, 1991). Instead, beginning with the idea that a girl's identity is formed in the context of identification with, rather than separation from, her mother, women's development is re-envisioned as "growth in connection" or "relationship-differentiation" (Surrey, 1991, p.36). In other words, the construction of a self is hypothesized to be a different process in women and men and to result in a different developmental endpoint. For women, whose developmental pathway includes lifelong identification with the mother, the maintenance of a number of emotionally intimate relationships is seen by these scholars as critical to emotional well-being and to the sustaining of a strong sense of identity over the entire lifespan. In contrast, the developmental pathway of men demands early differentiation from the mother in order for identification with the father to take place. As a consequence of this special demand, men become increasingly more separate and autonomous as they mature, and are assumed to be less invested in relationships, at least with regard to their sense of identity.
If one begins with the proposition that relationships are more
central to a healthy identity or sense of self in women than in men, then clearly a number of possibilities for understanding the high incidence of women's depression follow. It might be plausible, for example, to just accept the proposed link between depression and the loss of a significant relationship, and to understand the sex difference in rates of depression

as a function of the differential importance of high quality relationships to a strong sense of identity in women. Alternatively, Jack (199 1) has offered the provocative conceptualization that depression in women may be more a response to loss of a self, in the context of a relationship, than to the loss of a significant relationship. She has indicated that, .women's vulnerability to depression does not he in their 'dependence' on relationships ... but on what happens to them within their relationships" (p. 2 1). Yet another possibility for understanding women's depression lies in investigating the combined effects of self loss and other loss. The effects of each of these types of loss are investigated in this study.
In 199 1, Jack published a study that employed a simple, but
powerful approach to understanding women's subjective experience of depression, a study that may ultimately also prove useful in understanding the sex difference in this disorder. Rather than relying on exIsting theories of depression, Jack chose to use depressed women as her "guides" and took a stance assuming that these women were fully capable of conveying their own experiences. Based on careful analysis of individual Interviews with 12 women previously diagnosed with major depression or dysthymia; Jack identified a theme common to many of these women, a theme that she labeled "loss of self." Somehow in the context of caring for others, women noted that they had lost a strong sense of themselves and were aware of feeling disconnected from their own needs, wishes, talents, and emotions. For these same women, when a core relationship such as a marriage failed, "the sense of hopelessness and helplessness seemed to stem from despair about the possibility of bringing their own needs and initiative into their relationships," rather


than arising simply as a reaction to the loss of the other (Jack, 199 1, p.28).
Based on this perspective, one of the likely reasons for women's greater vulnerability to depression becomes clear. Without giving up Freud's conception of optimal adult functioning as residing in the capacity to love and to work, one can reasonably speculate that healthy intimacy is more fundamental to a woman's sense of self and competency than to a man's. Because of this potential difference, women's selfevaluation's are strongly tied to their ability to form and maintain highquality attachments. When relationship loss occurs, women become especially prone to varying degrees of depression, depending on the significance of the loss, and the extent to which she has sacrificed her own identity in an attempt to maintain the connection.
Incorporating Jack's idea that loss of a clearly defined sense of self is central to women's depression, the current study will also test two other predictions concerning loss of an intimate relationship. Using young adults who have experienced a recent breakup, three models will be compared. The first model tests the traditional psychoanalytic assumption that loss of a significant relationship is associated with depression. The second model focuses on the association between loss of self and depression. And, the third model looks at the combined effects of loss of a relationship and loss of self on depression.
This is a study about the importance of self loss and relationship loss as they contribute to the development of depression and to its degree of severity. As previously mentioned, depression is a highly debilitating disorder. Though varying in chronicity and severity, from individual to

individual, depression almost always permeates all aspects of functioning, generally disrupting both social and occupational functioning and causing significant suffering Although most theories of depression acknowledge the important role of loss, the present study attempts to replicate, extend, and refine the field's understanding of this connection. An expanded understanding of the disorder could potentially lead to improved prevention and treatment efforts.


In attempting to navigate the delicate balance between the "I" and
the "we," women frequently sacrifice the "I" in the service of
"togetherness," thus assuming a "de-selfed" position in
relationships. Depression may result from the sacrifice of self and
the concomitant loss of self-esteem, which accompanies the unconscious awareness of self-betrayal. (Lerner, 1988, p. 227)


As might be expected based on the scope of the problem, the
existing literature on depression is voluminous. Therefore, this chapter will focus only on the most relevant material. Initially, information on diagnosis and incidence will be discussed. Next relevant theories of depression will be covered, including a look at psychodynamic theories and at newer theories of women's development. Finally, Jacles research and conceptualization of women's depression will be reviewed as it applies to the current study.

Definitions and Incidence

The fourth edition of the Djanostic and Statistical Manual of Mental Disorders 1994), published by the American Psychiatric Association, devotes an entire section exclusively to the "mood disorders" including the Depressive Disorders, the Bipolar Disorders, and two additional disorders that are named according to etiology, i.e., Mood Disorder Due to a General Medical Condition and Substance-Induced Mood Disorder. Also included in the section on mood disorders are


descriptions of mood "episodes." According to the newest classification system, these episodes are defined as the building blocks of mood disorders, but are not diagnosable as independent entities. Included in the newly refined category of mood episodes are the following: Major Depressive Episode, Manic Episode, Mixed Episode, and Hypomanic Episode.
The extensive and expanded coverage of the mood disorders
In the DSM-IV is yet another indicator of the prevalence of mood or affective disorders. In fact, there is widespread agreement within the medical/ psychological community that depressive symptoms are the most common problem encountered in clinical practice ( Keller, 1988; Paykel, 1982). Epidemiologists Charney and Weissman (1988) have estimated that some degree of depression is currently affecting between 10 and 14 million Americans. Furthermore, after reviewing data from several recent large-scale National Institute of Mental Health studies on depressive illness, these same epidemiologists concluded that the lifetime risk for men of experiencing "nonbipolar" depression was 7-12% and for women was 20-25%. A recent article in Time reported that "rates of depression have been doubling in some industrial countries roughly every 10 years" (Wright, 1995, p 52).
Rush (1982; p. 1) noted that approximately 1 in 200 depressed individuals kill themselves and that suicide is now the second leading cause of death in the adolescent population. For young adults, suicide is the third leading cause of death following trafftc accidents and homicides (Wright, 1995). Given the strong connection between depression and suicide, these numbers clearly indicate the profound significance of the problem of affective illnesses, and the pressing need to improve

understanding and treatment for those indivduals experiencing even the mildest forms of these disorders (Hirschfield & Davidson, 1988).

Though bipolar or mixed forms of mood disorder are significant clinical problems, the focus of the present work is solely on the depressive end of the spectrum. Subsumed under the general heading of depressive disorders are Major Depressive Disorder, Dysthymic Disorder, and Depressive Disorder Not Otherwise Specified. In addition, for the purposes this study, the presence of at least several fundamental depressive symptoms such as pervasive sadness or strong feelings of helplessness, will be considered relevant. These types of symptoms are often found in conjunction with adjustment disorders or bereavement reactions. In these instances, symptoms may be may be less intense and enduring, than with major depression, but still cause deep subjective pain or a marked shift downward from typical levels of social or occupational functioning.

According to DSM-IV (1994), Depressive Disorders or Episodes are definable by both inclusion and exclusion criteria. Looking first at exclusion criteria, depressive illnesses are distinguishable by the absence of any history of mania. Second depressive disorders are delineated by the lack of a clear connection between depressive symptomnology and a medical illness, or depressive symptomology and the use or abuse of certain substances. In addition, during the diagnostic process, subtypes of Schizoaffective Disorders, Schizophrenia, Schlzophreniform Disorder, subtypes of Delusional Disorders, and Psychotic Disorders Not Otherwise Specified must be ruled out as the primary syndromes. Finally, uncomplicated bereavement must also be ruled out as an explanation for symptoms.

Turning to inclusion criteria, Depressive Episodes are identifiable by the presence, over at least a two-week period, of either pervasive feelings of sadness or emptiness, or by lack of interest or pleasure in normal activities. In addition to depressed mood or anhedonia, at least four of the following must also be noted : substantial weight loss or gain, or change in appetite; insomnia or hypersomnia; observable psychomotor retardation or agitation; lowered energy levels or chronic fatigue; excessive feelings of worthlessness or guilt; diminished cognitive abilities; recurrent thoughts of suicide or death, a plan to commit suicide, or an actual suicide attempt. To fully meet the criteria for a depressive episode, the core symptoms outlined above must also bring about subjective or objective impairment in critical areas of functioning, such as work and interpersonal relations. Symptoms are most often noted to occur on a daily or near daily basis.
Although the presence of specific symptomotology is necessary to the diagnosis of Depressive Disorders, accurate knowledge of the duration of identified problems is also required. As mentioned previously, in order to meet the criteria for a major depressive episode, five of the symptoms outlined above must have been present during the same two week period. As a further refinement in the the diagnosing of a Major Depressive Disorder, the disorder is designated as a single episode or as recurrent. In the event that all symptom criteria are met for two fun years, an episode of major depression may be labeled chronic. Similarly, Dysthymic Disorder, though generally less debilitating than a major depressive episode or disorder is by definition a chronic form of depression and is in part identified by the persistence of depressed mood over a two year span.

Diagnoses of Depressive Disorders may be further refined by reference to "specifiers describing the most recent episode." These specifiers include level of severity (mild, moderate, or severe) of the episode, and the presence or absence of psychotic features. In cases that no longer meet all criteria for a Depressive Episode, it may be assessed as in either partial or full remission. The final group of specifiers includes criteria for chronicity, catatonic features, melancholic features, atypical features, postpartum onset, longitudinal course, and seasonal pattern of the Depressive Disorders.
Extensive research on risk factors for depression has looked at
such differences as sex, age, social class, race, urban or rural habitation, and marital status. Overall, these studies have produced few clear results. For example, although there has been much speculation on the relationship between life stage and the risk of depression, the results in this area are contradictory and confusing. Only the work on birth cohort (with cohorts defined by decade of birth) and depression seems to indicate a verifiable trend toward an increasing incidence of depression beginning in the 1940s (Goldin & Gershon, 1988; Kessler et al., 1.993). There is some evidence that those from lower socioeconomic classes are more likely to become depressed. Also, some studies have suggested that individuals who are separated or divorced are at greater risk for becoming depressed. However, these findings are far from conclusive. In fact, some of the research on marital status and susceptibility to depression seems to vary with the sex of the participants (Boyd & Weissman, 1982). For example, women and men who have never been married are often found to have approximately equal rates of depressionwhile women who are married, separated, or divorced are found to have higher rates of

depression than men in those same categories. Finally, recently widowed men tend to be more depressed than recently widowed women.
In contrast to the generally inconclusive findings on risk factors for unipolar depression, there is substantial research evidence for a sex difference in the frequency rates of the disorder, with women on the average, being twice as likely as men to become depressed ( Kessler, McGonagle, Swartz, Blazer, & Nelson, 1993 ). Though in comparing rates of mental illness in general neither sex appears to be at greater risk, there are a number of studies showing that specific types of psychopathology occur with greater frequency in one sex or the other (Cox & Radloff, 1984). For example the sex difference in depression has been found consistently over approximately the last forty years, in both clinical and epidemiologic studies, and across the continuum of depressive illness from mild depressive symptoms to clinically diagnosed mpJor depression (American Psychiatric Association, 1994; Boyd & Weissman, 1982; Goodwin & Blehar, 1993; Nolen-Hoeksema, 1990; Scarf, 1980; Weissman & Klerman, 1987). Additionally, the 2:1 female to male ratio has been found to hold fairly constant in international studies on unipolar depression (Weissman et al., 1993). Findings for bipolar forms of mood disorder generally indicate fairly equal incidence rates in men and women, though even in this instance, the actual course of the disorder may vary in sex specific ways.
Depression is a costly disorder. In its most severe forms, an
individual may feel completely incapacitated and unable to function, even at the level of performing basic self-care activities. More commonly, the effects of depression are relatively subtle but may still significantly interfere with interpersonal functioning and with the ability to maintain

adequate work performance. In addition, depressive disorders cause incalculable subjective pain in the form of intense feelings of guilt, worthlessness, hopelessness, and sadness. According to Keller (1988), even though "depression is the most widespread of all the major mental illnesses. .. lIand] despite the potential morbidity and lethality... depression is often underrecognized, underdlagnosed, and undertreated" (p. 280).
In summary, the general category of mood disorders may be broken down into the bipolar disorders, which are defined by the occurrence of at least one manic episode, and the depressive disorders, which are defined by the presence over at least a two-week period of dysphoria. or anhedonla. The present study is solely concerned with the depressive disorders, However, even with this narrowed focus, estimates of lifetime prevalence risks for a major depressive episode range from 2.1-12.3% in men and 4.7-25.8% in women. Prevalence estimates that included milder or more chronic manifestations of depression would swell these figures even further.

Explanatory Theories for Depression

Psychoanalytic or Traditional Theories

Despite the very high incidence of depression, relatively little is known about its etiology. Some types of depression appear to be more endogenous or biologically mediated, while others appear to be more situationally mediated. Preliminary evidence for a genetic component in major depression has been based on the increased risk of contracting the disorder found in first-degree biological relatives. In comparison to the

general population, the risk of experiencing at least one clinically significant depressive episode is increased between 1.5 and three times. There is also some increased risk of dysthymic disorder in the first degree biological relatives of those diagnosed with major depression (American Psychiatric Association, 1994). However, the existence of a familial pattern alone is not sufficient to establish a genetic contribution. Twin and adoption studies which compare the incidence of depression in monozygotic and dizygotic pairs have allowed for a clearer look at genetic contribution. While these studies have established a substantial genetic contribution to bipolar forms of depression, findings regarding unipolar depression are much less definite (Boyd & Weissman, 1982; Charney & Weissman, 1988). As the disciplines of genetics and biological psychiatry become increasingly sophisticated, it is likely that specific research findings will aid in understanding the etiology of some types of depressive disorder. In the interim,personallty and situational factors will continue to be of interest.

While fully acknowledging the input of genetic factors, the current study focuses on the contributions of one type of environmental or situational input to the development of depressive illness. In particular, the higher incidence of depression in women leads to a reconsideration of the presumed effect of inter-personal loss as an adequate explanation for this sex difference. As Kaplan (199 1) has suggested, it is important to look at not just individual personality, but at "personality structures that may be common to the group of people most likely to become depressed, that is, women" (p. 267). The current study attempts to look at and compare the effect of loss on the development of depression in both women and men.

Beginning with Sigmund Freud, the role of loss in depression
became firmly established. In his 1917 work, Mourning and Melancholia, Freud identified both mourning and melancholia as responses to loss. However, he defined only melancholia- what would currently be labeled depression- as pathological. According to Harris (1987). Freud's definition of melancholia, in contrast to mourning, included a lowering of self-esteem. Those individuals most vulnerable to depression were assumed to be "orally fixated," and therefore more needy and dependent. Although Harris denied that the traditional psychoanalytic model is explicitly gender-biased, she does concede that the underlying developmental model, in which girls and women are believed to have impaired self-esteem, due to anatomical inferiority, could imply greater proneness to depression.
Object relations theory also focuses on the connection between loss and depression, not only in terms of the actual loss, but also in terms of the the individual's internal object world and developmental level. As Marcus explains "an object relations theory of depression shifts the emphasis from the external precipitant (e.g., actual loss) to its psychological significance vis-a-vis internalized object relations" (1987, p. 27). For object relations proponents, much of the current understanding of the significance of sex in etiology of depression is based on the work of Sidney Blatt and associates (Blatt, 1974; Blatt, DAfflitti, & Quinlan, 1976; Chevron, Quinlan, & Blatt, 1978).
Though Melanie lGein was the first to elucidate a critical link
between depression and internalized object relations, Blatt among others has built on her theorizing. According to Marcus, he postulates that .object representations initially grow out of repeated experiences of

frustration and gratification with a consistent and need gratifying object" (1987; p. 29). As development proceeds and the child differentiates, internal representations are thought to become more accurate and stable. However, if early parenting is inadequate, the integration process is disrupted, with negative consequences for later functioning.

Based on his particular developmental model, Blatt proposed that there are two types of depression. in addition, he speculated that "not only is the predisposition to depression determined by a failure to establish adequate levels of object representation, but also that the level at which the object is represented is related to the type of depression" (Marcus, 1987, p. 29). In the case of "anaclitic" depression, the object is represented at a very primitive level and the loss of the object is experienced as threatening basic survival. This type of depression can be recognized by the "desperate object seeking behavior of the individual," who may make "exaggerated or distorted attempts to maintain satisfying interpersonal experiences." Introjective depression, by contrast, was defined as indicating a more advanced level of functioning. This second type of depression was thought to reflect "the individual's fears, not about losing the object per se, but about losing the love, acceptance and approval offered by the object." In introjective depression, observers might note a perfectionistic and excessively self-critical style, which by the logic of this theory is based on an attempt to create a stronger and more positive sense of identity.

As their work progressed, Blatt and colleagues (Blatt & Schichmen, 1983; Blatt & Zuroff, 1992) expanded their conception of anaclitic and introjective depression to create a broader theory of psychopathology, that included "two configurations of psychopathology." The first

configuration was based on the anaclitic line of development, that is, primarily on distorted interpersonal relationships. The second configuration, or introjective line of development, involved exaggerated attempts to consolidate a strong sense of self. According to this theoretical position, women were believed to manifest anaclitic pathology more often, and men to manifest introjective pathology. Explanations for this difference were rooted in the differing demands on females and males to find primary objects for affection and identification.
If women are more prone to anaclitic depressions, this would imply, within the object relations framework, that there are sex differences in the developmental process. The phase of development that has been most scrutinized in this regard is the separation-Individuation phase. Differentiation from the primary caretaker begins during this phase and the first steps toward the establishment of identity and object constancy are undertaken. Chodorow (1978) has suggested that the "preoedipal period is more protracted in women" and other researchers have discussed the apparent difficulties that female children seem to encounter in negotiating the rapprochement subphase of separation. As mentioned above, Blatt and Shichman (1983) explained this difference in women and men by pointing to the need of young girls to separate successfully from their mothers. This is considered a more complex process than that required of male children who are biologically different from their mothers and, if not successfully negotiated, blocks the developmental process.

In their most recent work, Blatt and Bers (1993) have focused on a phenomenological approach to understanding both the self and the subtypes of depression. From this perspective, in which the sense of self

is seen as primarily subjective, the self is defined as "the locus of affective experience (I feel), the source of a sense of agency (I do), and the basis for a sense of self-awareness and self-reflectivity (I am)" (p. 173). They propose a revised theory of self-development that integrates object relations theory with phenomenological findings on the self. Three developmental stages are identfed: the rudimentary self, the introjected self, and the mature self reflecting progressively more advanced levels of functioning. Types of depression in this model are distinguished on the basis of precipitating life events and are characterized as either disruption of gratifying interpersonal relationships or disruption of an effective, essentially positive sense of self. Differing symptom clusters or patterns typify the two types of depression and link them to the rudimentary or introjective level of functioning as evidenced in subjective expressions of the self. Those at the mature level of development are seen as less prone to depression.

Segal and Murran (1993), in their response to Blatt and Bers, first point out that the higher rate of depression in women is not completely addressed by the theory-a noteworthy omission. Second, they challenge the strong emphasis on early parenting as the primary influence on the child's self. They draw attention to the work of Gilgan (1982,1989), as an example of research that emphasizes the importance of other factors, such as peer relationships during adolescence and gender role demands in the shaping of "self-identity."

Relational Theories of Development

Gilligan's 1982 book In a Different Voice ushered in a whole new era of thinking about women's development. Though a number of other

theorists had also been instrumental in this newly emerging psychology of women (See for example Chodorow, 1978, Miller, 1976, & Schaef, 198 1), Gilligan's work fully focused the attention of academics, and, to a lesser extent, the general public, on differences between men and women. Although this emphasis on differences was not, in and of itself, unique, Gilligan presented them in such a way that women were characterized, not as weak or pathological but, as having strengths and abilities that had been all but invisible to the field of psychology and undervalued by the culture as a whole. In a study focusing on moral reasoning, women's ethic of care was seen as a valid alternative or counterpart to the traditional, patriarchal ethic of justice, with one ethic emphasizing relationship and the other ethic emphasizing separateness or individuality. In the conclusion to In a Different Voice, Gilligan asserted that "the silence of women in the narrative of adult development distorts the conception of its stages and sequence" (p. 156), and she invites the continued exploration of women's voices and lives.
This exploration of the psychology of women and women's
development has indeed progressed and is typified by both Gilligan's ongoing studies and by the research carried out at the Stone Center of Wellesley College. During the past decade Gilligan and her colleagues at the Harvard Project on Women's Psychology and Girl's Development have carried out several longitudinal studies aimed at mapping female development more accurately (Gilligan, Lyons, & Hamner, 1991; Brown & Gilligan, 1992).
One study was conducted over a three-year period at a private
school for girls in New York and was an attempt to take an in-depth look at female adolescent development and to clarify differences between

males and females in the existence and development of a relationally defined self (Giffigan, Lyons, & Hamner, 199 1). Adolescence was targeted for study as both a critical phase of growth and as a period when girls are at high risk for losing a sense of their own identity.
Twenty-three girls were interviewed once per year over a three year period. These girls were asked to respond to the question, "How would you describe yourself to yourself?" The goal of the interviews was to clarify the conflict between existing theories of a separationindividuation and newer speculations about the centrality of close relationships in the lives of women and girls throughout the lifespan. In previous work feminist scholars had theorized that identity development in girls did not necessarily involve separation from significant others but did reflect a change or differentiation process within the context of key relationships. Study results revealed a number of themes common in the girls' self-descriptions. Themes focusing on a sense of self in relationships were by far the most frequent, with themes of independence occurring only one-fourth as often- suggesting a tendency, though not an exclusive one, for girls to define themselves relationally. Clearly, independence or separation was also valued by the girls, but seemed to be defined more in terms of "renegotiating and reframing of relationships in light of their new ability to consider the other person" (p.85). To reiterate then, if girls defined independence as an increased capacity to meet their own needs, this in turn decreased dependence and allowed for an improved ability to understand others and to build higher quality relationships. These results offer support for the the concept of a relational self and pose a challenge to existing concepts of the self, including the equation of psychological maturity with self-autonomy.

Continuing and expanding the work of the Harvard Project, Brown and Gilligan (1992) conducted yet another longitudinal study at the Laurel School in Cleveland, Ohio. This study culminated in the publication of their book Meetings at the Crossroads, a description of the development of girls between the ages of 7 and 18 years old. Again, the primary goal of this study was to demystify the changes that have been documented to occur in girls during the period between approximately twelve and sixteen years, but to Include the voices of younger girls as a way of pinpointing specific shifts or changes.
Early adolescence can be conceptualized as a crossroads between girlhood and womanhood, a time of great significance in female development. Although there are numerous studies documenting the dramatic changes that commonly occur in teenage girls between ages approximately 12 and 16 years old- including declines in I.Q. scores and poorer academic performance in math and science, decreased risk-taking and self-confidence, and sharp increases in rates of body-image disturbance, depression, and suicidal ideation- there have been no comprehensive explanations for these changes (Pipher, 1994; Uzelac, 1995).
Listening closely to the girls, Brown and Gilligan (1992) have
begun the process of accounting for these changes. In the Laurel School project, for example a striking contrast was noted between the clear, strong voice of girlhood and the confused and fearful voice of the adolescent who all-too-often silenced her own thoughts and feelings. Describing this shift as a relational impasse, the researchers noted a number of commonly occurring disconnections or dissociations "between psyche and body, voice and desire, thoughts and feelings, self and

relationship" (p. 7). These changes as a whole were labeled as the giving up of authentic relationship to an idealized concept of relationship grounded in a feminine ideal of perfection and self-sacrifice. Not surprisingly, the self-silencing behaviors were noted to have profound effects on the adolescent girls' relationships to self and others, and to their ability to function effectively in the world.
Both complementing and extending the work of Gilligan and the Harvard Pro ect is the work of Jean Baker Miller and her colleagues at the Stone Center of Wellesley College. In her provocative work Toward a New EUchology of Women (1976, 2nd ed. 1986), Miller, like Gilligan, questioned the psychological tradition of assuming man as the measure of all things, a tradition that she has indicated probably renders an incomplete vision of human nature. She recommended looking "toward a more accurate understanding of women's psychology as it arises out of women's experience" (1986, p.49) and stressed the necessity of understanding women's behavior, strengths, and difficulties in the context of an oppressive patriarchal culture.
Once again, paralleling Gilligan, Miller's theory focuses on the concepts of self and relationship. She used the idea of an alternative developmental pathway and outcome to de-pathologize women's priorities, needs, and behavior. Similar to Tavris (1992), who has pointedly observed that "when men have problems, society tends to look outward for explanations; when women have problems, society looks inward" (p. 175), Miller did not focus immediately on intrapsychic explanations for problems. Instead, she turned to women's gender role training as a primary source for understanding the prevalence of such difficulties as low-self-esteem and depression.

Comparing male and female gender role training, Miller states that from early in life men are rewarded for being active and rational while women are rewarded for being sensitive and emotionally attuned to others. This leads in turn to the development of differing core identities-doing, for men, and giving, for women- and to differing vulnerabilities and strengths.

For over ten years now, this theme of differences between men and women has been the focus of extensive study at the Stone Center for Developmental Services and Studies. The center is affiliated with Wellesley College and has served as a sort of "think tank" for clinicians and scholars who have interests in the psychology of women. According to Peay, "psychology and feminism, like the twin spirals of a double helix, have interacted to produce new theories in both fields" (1994, p. 22). What has emerged from the Stone Center, and in the work of the Harvard Project, has been labeled difference feminism (Kaniner, 1993). "Difference" feminists, unlike "equal rights" feminists, have tended to acknowledge rather than debate differences between men and women. Having moved beyond their original goal of questioning psychoanalytic formulations that "have tended to portray women as too emotional, too dependent, and lacking clear boundaries" (Jordan, Kaplan Miller, Stiver, & Surrey, 1991, p.v), difference feminists have identified and celebrated what they see as women's superior strengths in the areas of nurturing skills and relational capacities.
Writings from the Stone Center, published in the form of "working papers," began in the early 1980s with a new delineation of female development. Surrey (1991) explicated the construct of the "self-inrelation" as an alternative to the contemporary emphasis on the

autonomous self "as an organizing principle in human development" (p. 51). Drawing on the theories of Miller (1986) and Gilligan (1982), she proposed that women's primary sense of self is relational, hence the term self-in-relation. The working definition of self underlying her approach is of self as "a construct useful in describing the organization of a person's experience and construction of reality that illuminates the purpose and directionality of his or her behavior" (p.52). In contrast to the primary emphasis in the writings of Mahler, Erikson and Levinson on separation as necessary to the establishment of healthy identity, SurTey postulated that capacities such as autonomy and creativity, accepted hallmarks of a mature self, develop within the context primary relationships.
The mother-daughter relationship is offered as the beginning point in the female's alternative developmental pathway. This relationship, in which sharing of feelings between mother and daughter is encouraged and reinforced, is thought to be "the origin of the capacity for empathy and the beginning practice of relational development" (Surrey, 199 1, p-55). In outlining this theory of women's development, Surrey emphasizes three important aspects of the mother-daughter relationship: the child's interest in and desire for emotional connection to the mother, the child's evolving ability for mutual expression of empathy, and finally the development of a capacity for "mutual empowerment" (p.56). Mutual empowerment denotes a feeling in both mother and daughter of interpersonal efficacy, of confidence in perceiving, responding and relating to the feelings of others. This empowerment is based on the open exchange of feelings and the ability to be with or understand the other. "Mothers help to empower their daughters by allowing them to feel successful at understanding and giving support at whatever level is

appropriate" for their age (p. 57). The development of healthy self-esteem is another outgrowth of mutual empowerment. Good relationshipsthose based on mutual understanding and regard-become fundamental to a sense of worth. Women learn to value their ability both to take care of others and to take care of the relationships between themselves and others.

Summarizing the elements of the "core self in women, Surrey emphasized: (1) "an interest in and attention to the other person(s), which form the base for the emotional connection and the ability to empathize with the other(s); (2) the expectation of a mutual empathic process where the sharing of experience leads to a heightened development of self and other; and (3) the expectation of interaction and relationship as a process of mutual sensitivity and mutual responsibility that provides the stimulus for the growth of empowerment and self knowledge" (1991, pp. 58-59). She was careful to distinguish this alternative developmental pathway of women from that of separationindividuation and labels the evolution of the female self as a process of relationship-differentiation" (p. 60). In defining relationship differentiation," Surrey used the analogy of embryological development with the self becoming increasingly complex within the context of connection to significant others. According to this perspective, a two-way model of interaction, identity and the capacity for relationship grow hand in hand from the child's earliest attempts at mutuality to the much richer and more complex forms of adult relating, but the importance of the give and take of intimate sharing or interaction never declines.

In addition to Miller and Surrey, other Stone Center theorists,

such as Jordan, Kaplan, and Stiver (199 1) have elaborated on this model of the self-In-relation, focusing on specific topics such as empathy, dependency, and mutuality, or on the applications of this model to treatment of disorders such as depression, eating problems, and work inhibitions in women. Jordan has reconstrued empathy as a strength arising from the identification process between mother and daughter, rather than as a form of regressive merging indicative of poor selfboundaries, Citing Kohut's 1978 definition of empathy as "a fundamental mode of human relating" (pp. 704- 705), Jordan (199 1) argues that empathy as a relational capacity is far from primitive but is in fact rooted in a well-differentiated self with high levels of ego strength and flexible ego boundaries. Stiver has offered a new view of the concept of dependency, a view that turns more on the idea of mutual support or interdependence as a healthy aspect of adult functioning. Noting the long-held pairing of dependency with primitive oral needs, she has challenged the idea that "to need anything in adulthood is regressive" (199 1, p. 146) and has suggested that the continued growth in adulthood of both women and men is predicated on the ability to share feelings, needs, and vulnerabilities without shame.
Both Jordan and Stiver have addressed the frequency of depression In women. Jordan has linked women's depression to either overidentification with others or to poorly developed self-empathy. Alternatively, StIver called attention to the constraints on women that limit the expression of their inner most needs and the link between the silencing of these needs and depression.
Of the Stone Center Staff, however, it is Kaplan (1991) who has explored the topic of women's depression in the most depth. She began

by reiterating the higher incidence of depression in women, drawing from other writers to describe personality characteristics often associated with depression, such as a need to please others and difficulty in attending to or expressing the self (Arieti & Bemporad, 1978). Agreeing with Arieti and Bemporad that these characteristics are indeed found more often in women, and again noting the much higher incidence of depression in women, Kaplan (199 1) suggested that depression may be "not an 'illness' superimposed on an alien or indifferent personality, but rather may be a distortion-an exaggeration of the normal state of being female in Western society" (p. 207). Building on this idea, Kaplan then moved on to examine "the fundamental overlap between central dynamics of depression and key dimensions in the nature of women's psychological development (p.207). In distinguishing her approach, Kaplan emphasized that whereas others have advocated the necessity of looking at individual personality as a means for understanding depression, she has argued for identifying personality structures that may be common to those most likely to become depressed-women. Furthermore, though maintaining a focus on personality, Kaplan has included the role of the social context as a shaping force.

In a brief survey of factors long held as causes or precipitants of depression, Kaplan (199 1) has included the experience of loss, the inhibition of anger and aggression, the inhibition of action or assertiveness and low self-esteem. She has asserted that these factors have not been adequately studied relative to their influence on depression and their link to sex or gender and goes on to suggest that this is a major oversight in the literature. Expanding on the relevance of loss, Kaplan has suggested that women experience significant feelings of

loss each time they seek, but fail to make, a mutually empathic connection with another person. From the perspective of self-in-relation theory, the experience of this type of loss is critical for two reasons. First, it undermines a woman's "core self-structure as one that can facilitate reciprocity and affective connection in relationships" (Kaplan, 199 1, p.2 11). Second, given the contemporary cultural climate, which does not place a high value on relational capacities, women may fail to value these abilities as strengths. In this framework, depressed women are assumed to have responded especially strongly to loss of an opportunity for mutual understanding and expression of self, rather than the loss of the gratification provided by another. Kaplan has also viewed some inhibition of anger and aggression as normative in women because of the possibility of anger and aggression disrupting valued relationships. For depressed women, this pattern is, again, assumed to be exaggerated, thus leading to disempowerment.
Inhibition of action and of assertiveness is seen as one potential consequence of women's perceived failure to form mutually affirming connections. In a selective way, women who are feeling disconnected from others may blame themselves for this disconnection and may work even harder to improve the relationship or to act on behalf of the other person. At the same time, however, these women may reduce activities on their own behalf, thus narrowing their lives and leading to a greater vulnerability to depression. Low self-esteem in "normal" women can be understood, in part, as a response to the cultural standards against which women measure themselves; standards that support autonomy and individualism, rather than empathy and cooperation. For women who are depressed, Kaplan has hypothesized that numerous or early

experiences of emotional disconnection lead to a damaged sense of self and a woman's fear of her own destructiveness, a fear that arises from taking responsibility for past relationship failures.
Using the self-in-relation theory to understand female

development, Kaplan explicates the points of divergence from existing theories of depression, and in particular from psychoanalytic perspectives. Although both self-in-relation-theory and psychoanalytic theory emphasize the importance of interpersonal loss in depression, psychoanalytic theory focuses only on the loss of "narcissistic supplies" (Kaplan, 199 1, p. 217) and the impact of this loss on self-esteem. In addition to this loss of gratification, self-in-relation theory proposes that another, and perhaps more important, effect of interpersonal loss is on "the chance to take part in a mutually affirming relationship" (p. 217). This type of loss affects a woman's sense of her own worth by calling into question her relational abilities, abilities that are intimately tied to her sense of self, according to Kaplan. To women, relationship loss may be felt primarily as a "failure of the self' (218).
Another feminist theorist, Kaschak (1992) reiterated the view held by many feminist theorists that the most common psychological disorders found in women, such as depression, anxiety disorders, and dissociative disorders can be best understood by reference to the normative demands of the female role. For example, in the case of women's depression, Kaschak cited a link between women's "overriding emphasis on relational aspects of life" and their vulnerability to depression following loss (1992, p. 18 1). She further contended that because of the common developmental experiences of women in American culture, a sense of impending or actual loss becomes embedded in their

expectations and functioning. Expanding the scope of her thinking on loss beyond the usual focus on relationship loss, Kaschak's work parallels aspects of Jack's (1991) conceptualization of loss of self. She has identified loss of self-control, loss of self-definition, loss of a sense of meaningfulness, and loss of the possible, as common to women's experience and as potentially debilitating.

In support of her position on the importance of loss to women's depression, Kaschak (199 1) referred to the literature on sex differences, marriage and depression. This literature shows consistently that married women are more likely than married men to be depressed, and that married women who do not work outside of the home show the highest incidence of depression. Cox and Radloff (1984) reported similar findings stating that marriage and motherhood tend to increase women's risk of some psychiatric disorders, while decreasing rates of these same disorders in men. In her analysis of these findings, Kaschak focused on women's tendency to place the needs of valued others first and points out the potential costs of long-term self-denial-the loss of identity or the capacity for self-definition.

In their discussion of sex differences in unipolar depression,
Hamilton and Jensvold (1992) posed the question of whether there is a depressogenic aspect to the female personality. They offered four possible conceptualizations of the link between personality and depression. First, they suggested that personality type may increase or decrease susceptibility to depression. However, they pointed out that studies attempting to show a connection between certain gender-linked characteristics, such as dependency and assertiveness, and depression have failed to do so. The second conceptualization suggested that some

types of depression may actually alter personality or one's ability to assess it. In the few studies exploring this idea, there is little evidence for the existence of a gender or sex difference. A third conceptualization states that certain personality types and depressive syndromes may "cooccur" in relation to a third variable. In understanding women's high rate of depression for instance, the covariate might be a history of victimization or some unidentified biological factor. The final conceptualization. looks at the possibility that the depressogenic personality may actually constitute an as-yet unidentified. form of milder depression. Hamilton and Jensvold agreed that further research is needed in this area to sort out the tangle of personality types and types of depression. In fact, the diagnosis of depressive personality was considered and rejected for inclusion in the latest edition of the Diagnostic and Statistical Manual of Mental Disorders (APA, 1994).

Jack's Theoly of-Depression in Women

In her recent research efforts, Jack (199 1; Jack & Dill 1992) has undertaken the substantial project of challenging existing theories of women's depression. She joins the growing body of feminists who believe that the field's understanding of women has been seriously limited by the use of primarily male models of health and development, models which have tended to ignore either sex and gender differences or to view female patterns as deviant. Jack's work is an attempt to re-examine the high incidence of depression in women from a less biased perspective than that of traditional psychoanalytic theory and to build on the concept of the relational self as a means to better understanding of women's

strengths and vulnerabilities.
In order to understand clearly women's depression, Jack (199 1) took as her starting point the need to listen to the voices of women as they described their internal worlds. To this end she designed a longitudinal study that involved interviewing 12 women diagnosed with either dysthymia or major depression. These women participated in lengthy, sernistructured interviews shortly after being diagnosed as depressed, and again approximately two years later, to assess for changes in affective, cognitive, and social functioning. Rather than focusing on external factors known to be associated with women's depression- -such as domestic violence and economic oppression- -Jack's interview questions were aimed at bringing to light women's lived experience. She wanted to hear directly from the participants about their lives, their losses and their sadness (199 1, p. 2).
Based on the results of the interview study, Jack developed a scale, the Silencing the Self Scale, designed to measure "gender -specific schemes hypothesized to be associated with depression in women" (Jack & DiII1992, p. 97). In her discussion of the process of "silencing the self," she builds on the idea of a relational self as the normative female self and looks specifically at how women "create and maintain safe intimate relationships" (1992, p. 98). Her basic premise is that the maintenance of high quality relationships is so central to a woman's identity and overall self-evaluation, and that she may at times, both literally and metaphorically silence herself in her efforts to maintain the valued connection. It is this self-silencing that Jack sees as so damaging to women. Over time, the process of shutting down or splitting off parts of the self, in order to better conform to the female role imperative, leads to

a decrease in self-esteem and possibly to feelings of loss of self and ultimately depression.
Jack's theory, however, unlike theories that have tended to
denigrate women for their dependence on relationships, afflrms the need for healthy attachments as natural throughout the lifespan. She states that from the perspective of the relational self, depression does not arise from an inability to detach from an intimate relationship, but from "the inability to make or sustain supportive, authentic connection with a loved person" (1991, p.16). Instead of the traditional focus on developmental deficits as the source of dependency and the cause of depression following relationship loss, Jack has recommended looking at the forms of connection that tend to result in loss of self as a more useful approach to understanding women's depression. "Women's vulnerability to depression does not he in their 'dependence' on relationships or in their depressive response to loss, but in what happens to them within their relationships" (199 1, p. 2 1). The sex difference in depression between married women and married men becomes potentially more explainable in this newer fi-amework.
Jack is careful always to look at women in light of gender norms and sexual Inequality. In contrast to theories that focus exclusively on personality, Jack has looked at the interaction of culture and personality and their combined impact on women's sense of self and behavior. She has pointed out that ultimately, "self-evaluation holds the key to understanding gender differences in the prevalence and dynamics of depression (199 1, p. 90) and that women's sense of "goodness" is based in their ability to make and maintain authentic, Lasting attachments. In carrying out the adult tasks of establishing a strong sense of identity and

securing healthy attachments, some women change or sacrifice the self in order to sustain connections to others. It is this loss of authentic voice that seems to play a significant role in explaining the high incidence of depression in women.


This study attempts to advance the field's understanding of women's depression by comparing predictions based on competing theoretical models discussed previously. Three sets of hypotheses correspond to three competing sets of predictions regarding the relationship between loss of a romantic partner, self loss, and level of depression. (See Figure 1 for a visual depiction of these).
The first hypothesis, the Traditional Model hypothesis, predicts only a main effect of relationship loss on depression. This model does not predict either a self loss main effect, nor an interaction between relationship loss and self loss. The expected direction of the relationship loss main effect is that depression should be significantly higher if a recent relationship loss has occurred, than if such a loss has not recently occurred.
The second hypothesis, the Jack Model hypothesis, predicts only a main effect of self loss on depression. This model does not predict either a main effect of relationship loss, nor an interaction between relationship loss and self loss. The expected direction of the self loss main effect is that more significant levels of depression should be associated with greater degrees of self loss.
The third hypothesisthe Potentiation Model hypothesis, differs from the Traditional and Jack models in that it predicts an interaction

between relationship loss and self loss on depression. This model makes no predictions about main effects because, according to this perspective, the presence or absence of one or both main effects is less crucial conceptually than the expected influence of relationship loss to potentiate self loss and vice versa.
Sex of participant has been included to assess the external validity of the findings. The logical prediction is that the results will be similar in form for men and women, however, the overall frequency of depression in men may be lower, consistent with past research.

Figure 1 36
Predicted Level of Depression. According to the Three Models

Panel A: The Traditional Model Relationship Loss Present Absent Present High Low

Loss of Self

Absent H-uh LOW

Panel B: Jack's (1991) Model Relationship Loss Present Absent Present High High

Loss of Self

Absent LOW Low

Panel C: The Potentiation Model Relationship Loss Present Absent Present Highest Moderate

Loss of Self ___Absent Moderate Lowest


The primary objective of this study is to investigate the role of relationship loss, self loss, and the potentiating effects of combining these two types of loss on the level of depressive affect in women and men.The current study will incorporate the current theorizing on women's relational development as a potential means for better understanding women's greater susceptibility to depression. In a departure from the more traditional models, which link women's depression primarily to the loss of a significant other, the emerging models focus much more on the woman's sense of identity-or loss of this sense of identity-in the context of Intimate relationships. Differences in the early gender identification process and the gender role socialization process are believed to contribute to the likelihood that a woman will experience "feelings of guilt, shame, and depression with the failure of intimate ties" (Jack, 199 1, p. 13).


Participants in this study were 257 undergraduates at the
University of Florida, during the spring and summer semesters of 1994. Students were enrolled in one of two psychology courses: Personal Growth or Psychology of Personality. At a later date, an additional data set was collected and will be separately analyzed. The second data set Included 172 participants. These participants were also University of Florida students in a Personal Growth class.




Data were collected during one portion of a regular class meeting. The purpose of the study was briefly explained, as were the means for protecting the anonymity and confidentiality of all results. The need for thinking through answers, and the importance of completely and honestly answering all questions was stressed. Questionnaire packets were distributed, including the Beck Depression Inventory (BDI), the "Silencing the Self' Scale (STSS), the Depressive Experiences Questionnaire (DEQ), and a number of questions concerning relationship loss generated specifically for this study. Basic demographic data were also collected.The average time to complete the packet was approximately 30 minutes.

Independent Variables

Silencing the Self Scale
One of the primary independent measures of the study was
designed to assess the "loss of self which can occur in the context of an intimate relationship. Jack designed the "Silencing the the Self Scale" (STSS) to "investigate gender-specific schemas hypothesized to be associated with depression in women." The scale is derived from her previous qualitative work, which involved interviewing women diagnosed with major depression. Items for the scale came from the women's own perceptions of their problems and feelings, and are aimed at exploring "the self-negating process that some women experience in their efforts to maintain intimate connection" (Jack & Dill, 1992). Jack's intent was to

expose the means by which learned gender role behavior shapes a woman's interpersonal interactions and her beliefs about her own worth. The emphasis is on cognitive schemes, not on personality.
The STSS consists of thirty-one statements paired with a Likerttype scale, which consists of five choices ranging from strongly disagree on one end to strongly agree on the other. In addition, the final item has an addendum asking participants who scored four or five, on that particular item, to respond in more detail to the question. This more free form question was not included in the current study, due to potential problems in data processing. Four "rationally" derived subscales reflect critical components of Jack's proposed theory of women's depression (Jack, 1991; Jack & Dill, 1992). Each of the subscales was designed to Illustrate a related, but distinct aspect of "loss of self". The first subscale, called "externalized self-perception" is made up of items which indicate a tendency for a woman to judge herself by external standards. Item 6 serves as a representative example, "I tend to judge myself by how I think others see me." The second subscale, labeled "care as selfsacrifice," includes items targeting a specific pattern by which a woman may put the needs of others first, without adequate consideration of her own wishes or needs. For instance, "Caring means choosing to do what the other person wants, even when I want to do something different." Third, comes the Silencing the Self Scale," these items focus on the inhibition self-expression and of actions which might lead to conflict and potentially contribute to relationship loss. An example from this scale is itemel4, "Instead of risking confrontations in close relationships, I would rather not rock the boat." Finally, the fourth subscale called "the divided seW contains items designed to capture some women's experience

of outer compliance to female role imperatives, while the inner self becomes progressively more angry and hostile. Item 16 is a representative example, "Often I look happy enough on the outside, but Inwardly I feel angry and rebellious."
Higher scores on the STSS reflect the degree to which a woman feels the need to conform to the demands of the female gender role, not just the degree of depression. Scores on the scale can range from 31-155. As already mentioned, scores at the upper end would indicate that a woman is struggling with strong internal demands to fulfill Mrceived gender role expectations, while lower scores would indicate a woman who feels significantly less constrained by these role imperatives.
The scale was chosen for this study because of its potential for
expanding our understanding of a significant mental health problem. As an initial step In her research into women's depression, Jack (199 1) did a longitudinal study of twelve women with diagnoses of dysthymia or of major depression. Her extensive interviewing of these women led to the genesis of the first draft of the scale, which was then reviewed by a group of psychologists for face validity. Items were also reviewed by female undergraduates to check for "comprehensibility and clarity." Next, using three diverse samples of women, Jack and Dill (1992) established baseline measures of reliability and validity. Volunteer participants were drawn from a large state university (n=63), from three battered women's shelter (n= 14 1), and from a group of new mothers who had reported alcohol, marijuana, tobacco, or cocaine usage during their pregnancies (n=270). The samples were composed predominantly of Caucasian women in their late adolescence through their thirties. The majority of women were from middle and lower socioeconomic levels, and reported a variety

of current relationship statuses, including single, married, partnered and divorced
Separate measures of the scale's internal consistency were done
for the three different sample groups with alpha coefficients ranging from .86 to .94. The four subscales were also assessed for internal consistency. All subscales except "care as self-sacrifice" demonstrated acceptable internal consistency with results ranging from .74 through .90. At this phase of instrument development, Jack and Dill (1992) recommend against separate use of the "care as self-sacrifice" subscale because it showed showed relatively weak internal consistency. The STSS showed strong test-retest reliability with scores ranging from .88,93, based on Spearman-Brown coefficients.
Jack and Dill (1992) investigated the construct validity of the STSS in several ways. Results of the STSS and BDI were found to have the expected positive correlation, approximately r=.50 within each subgroup of the sample, showing that STSS scores were related to level of depression as measured by a well established self-report instrument for depression. The samples were analyzed independently based on the researchers' contention that members of each sample group would be dealing with specific pressures based on the role demands of their unique social context. This hypothesis was also supported. Analysis of variance indicated significant differences between the means of the three samples, with the college students scoring lowest on both, the STSS and the BDI, the sample of new mothers scoring in the intermediate range, and the battered women scoring highest.
Although additional studies are clearly required to more fully document the reliability and validity of the STSS, the results just

outlined provide a clear indication of the STSS's validity and reliability. Theoretically, the STSS offers an expanded understanding of women's depression. Its focus on, the impact of "loss of self" in a relationship, provides the means for investigating one of the central questions of the current study. The major weakness of the STSS, in terms of the current study, is the complete absence of data on men. Because fundamental assumptions are being made concerning gender differences, men's responses are needed to more thoroughly test Jack's theory and instrument.
The second independent variable in the present study will be the biological sex of participants, female or male. As mentioned above, both Jack's theoretical work and her research have focused exclusively on women. One goal of this study is to test whether relationship loss has a similar effect on members of both sexes, or whether differences actually exist.
RelationshiR Loss Questionnaire
The third independent variable concerns the presence or absence of a relationship loss over the previous twelve months. Since a survey of the current literature failed to locate a suitable instrument, a brief questionnaire was designed to succinctly collect the needed information. Ten questions cover the presence or absence of a romantic loss, the timing of the loss, which partner was the instigator of the breakup, the importance of the relationship, overall satisfaction with the relationship, and the assignment of responsibility for the failure of the relationship. Also included, is a direct question about "giving up the self" as a way of maintaining the former relationship. Though they have less direct

relevance for the present study, three items ask about current involvement in a relationship. It is anticipated that the quality of a new relationship may mitigate the impact of the previous loss, and/or perceptions of the past relationship.

DeRgndent Variables.

Beck Depression Inventory
A central aspect of the present study involves the measurement of level of depression following the breakup of an intimate relationship. The dependent measure chosen for this purpose was the Beck Depression Inventory. According to Shaver and Brennan (1991), the Beck Depression Inventory is "the most prominent and frequently cited self-report measure of depression." Consisting of 21 items, the Beck Depression Inventory (Beck, 1967) (see Appendix C ) may be used either for measuring the intensity of depressive symptomology, or as a brief screening measure. Both of these uses have relevance for the present research. Beck chose his items to represent 21 distinct "symptomattitude" categories falling under the four more general headings of behavioral, affective, cognitive, and physiological manifestations of the disorder. Each item consists of four statements scored from 0-3 which the participant Is asked to choose among. Statements scored "0" represent the nondepressed end of the continuum with increasing scores indicating a greater degree of depression. For example, item A focuses on sadness. It consists of the following four statements: "I do not feel sad" (scored 0), "1 feel sad" (scored 1), "1 am sad all the time and can't snap out of it" (scored 2), and "I am so sad or unhappy that I can't stand it" (scored 3). Total scale scores on the BDI can range from 0 to 63, with

various interim ranges representing different degrees of depression. Following the most widely used criteria, scores from 0 to 4 suggest little or no depression, mild depression is suggested by scores from 5 through 13, moderate depression between 14 and 20, and severe depression by scores 21 and above.
One of the main strengths of the BDI is its proven reliability and validity. According to Shaver and Brennan (1991), the BDI has been shown to have a split-half reliability coefficient of .86, and test-retest reliability with coefficients the in the .70's when repeated testing is done over a span of several weeks. In terms of validity, the BDI is weR known for its construct validity and, in fact, is often used to evaluate new instruments. During the past approximately twenty-five years, the BDI has repeatedly been shown to be positively correlated with both clinically derived, psychosocial, and physiological measures of depression, demonstrating ample evidence of convergent validity. Though depression is often difficult to discriminate from anxiety, the BDI has been used somewhat successfully for this purpose, and also for distinguishing depression from other psychiatric diagnoses. In other words, reasonable discriminant validity been has shown.
With reference to the current study, the BDI has several strengths and weaknesses. Its major weakness is that it was designed for use with clinical populations, not with the normal population which is targeted by the study. However, the test has been used with adults of all ages and many cultural and ethnic groups, and will serve very adequately for present purposes. The major strengths of the BDI are its proven "track record", and the option to derive scores which indicate not just the

presence of depression, but the level or degree of severity of depression, if depression is present at all.

Depressive Experiences Questionnaire.

The second measure of depression was chosen to test the more
traditional model, in which women's depression is seen as arising from unmet dependency needs following the loss of a significant other. The Depressive Experiences Questionnaire was developed by Blatt, D'Afflitti, and Quinlan in 1976. It was designed to tap two primary dimensions of depression, anaclitic and introjective, which Blatt had identified in his earlier work and which reflect his psychoanalytic understanding of the causes of depression (1974). Anaclitic depression is characterized by a longing to be cared for by another and by feelings dependency, fear of abandonment, and helplessness. In contrast, introjective depression Is characterized by feelings of "not measuring up", of the self as a failure, including accompanying guilt and self-denigration (Shaver and Brennan, 1991). Both types of depression are seen as reflecting personality styles and can be applied to the continuum of depression from normal mood fluctuations to the full clinical disorder (Welkowitz, Lish, and Bond, 1985).
The DEQ consisted originally of 66 items, but involved a complex scoring system which Welkowitz et al (1985) revised. The modified DEQ consists of 45 items and now allows for clearer comparisons between genders. Since gender differences are central to the questions raised by the current study, the updated scale and scoring system will be used. Under this system, participants attain a certain score for each of three subscales: dependency (equivalent to the anaclitic dimension), selfcriticism (equivalent to the introjective dimension), and efficacy. Final

scale scores are based on the mean of scores for items loading on each of the specified subscales.

During their preliminary work revising the DEQ, Welkowitz and her colleagues (1985), found internal consistency scores, based on Cronbach's alpha, as follows: dependency .81, self-criticism .86, and efficacy .72. Though no test-retest reliability data was cited, previous work with the original form of the DEQ "indicated considerable stability" over time (Shaver and Brennan, 1985).

Evidence for convergent validity has been shown for both versions of the DEQ, each of which showed predictable correlations with other established measures of depression and self-concept. Klein (1989) also correlated scores derived from the alternative scoring systems and found them to produce very similar results.
The strengths of this scale for the purposes of the present study include its applicability to a nonclinical population of participants, and the theoretical foundations of the scale which posit the exidstence of several types of depression and link these types to personality styles, or possibly, gender or gender role pressures (Blatt, D'Afflitti, & Quinlan, 1976; Welkowitz et al, 1985). According to Shaver and Brennan, the main weakness of the revised scale is the intercorrelation of the dependency and self-criticism subscales, which may indicate that the scales do not actually represent totally distinct factors. However, for present purposes, the level of intercorrelation, at .60 is acceptable.

Data Analysis

All three hypotheses were evaluated with two hierarchical multiple regressions, one for each measure of depression (dependent measures).

The predictive (or independent) variables were sex, relationship loss, and self loss as well as all of their two and three way interactions. Sex will be entered first, then the other two main effects, then the two-way interactions, and finally the three-way interactions.


In order to compare the three models linking types of loss to
depression, two multiple regression analyses were performed. Using a sample of 257 undergraduates, the impact of loss of a partner, self loss (the Jack Model), or the combination of both of these losses was assessed. The first multiple regression involved the use of the BDI as the criterion variable, the second used the dependency subscale of the DEQ. Independent variables included the presence or absence of a recent relationship loss, scores on the STSS as indicators of degree of self loss, and sex of the participants. The various interaction terms: breakup by STSS score, breakup by sex, STSS scores by sex, and finally, breakup by STSS scores by sex were also included as predictor variables. In each multiple regression, the Type III sums of squares were used to calculate. F values.

Smple Characteristics

Of those participants reporting demographic information, 171 or 66.5% were fernale and 86 or 33.5% were male. Ages of participants ranged from 17 to 40, with a mean age of 21.5 years. The participants were predominantly Anglo-American or Euro-American, 77.6%, with the remaining participants being Afro-American, 4.9%, Hispanic American,
6. 1 %, Asian American, 7.3%, Native American (American Indian), 0%, or Other, 4. 1 %. Only 0. 8% of participants were first year students (n = 2) ,
7.4% were sophomores (n 19), 35.8% were juniors (n = 92), 53.3%


were seniors (n =137), and 2.7% were post-baccalaureate or graduate students (n=7). Approximately 53.2 % of participants reported currently being married or partnered (cohabiting or not), 46.7% reported that they were currently single, separated, divorced, or widowed. Ninety-six and one-half percent of the participants described themselves as heterosexual, 0.8% described themselves as homosexual, and 2.7% bisexual.

Means and Freguencies,

The means, standard deviations, minimum and maximum values of participants' responses to key questionnaire items are depicted in Table 1. In addition, two other questionnaire items, which are frequency items, are relevant. First, of the 254 participants who chose to answer the question asking whether they had experienced the breakup of an intimate relationship in the past 12 months, 104 or 59% indicated "yes", and 150 or 4 1 % indicated "no."
This item was of special importance given the study's focus on the possible link between relationship loss and depression. In the end, the most critical analyses could include only those participants who had experienced a recent loss. Second, in response to the question asking participants if they are currently involved in an intimate relationship, 159 or 62% indicated "yes" and 89 or 38% indicated "no". This item was considered to be important for its potentially mitigating effect on depressive symptoms precipitated by past losses.


Table 1
Means, Standard Deviations, Minimum and Maximum Values on
Responses to Key Questionnaire Items

Variable N Minimum Maximum Mean Std Dev

JACK 257 1.1 4.7 2.47 0.66
GAVESELF 254 0.0 8.0 2.13 2.69
BECK 257 0.0 43.0 6.96 6.45
BLATT 257 1.2 6.2 3.70 0.93
WHEN 254 0.0 6.0 1.43 2.04
CHOICE 254 0.0 9.0 2.01 3.02
OLDIMP 253 0.0 9.0 3.04 3.65
RESPONSI 254 0.0 9.0 2.40 3.10
OLDSAT 254 0.0 9.0 2.17 2.81
CURRIMP 254 0.0 9.0 4.93 3.98
CURRSAT 254 0.0 9.0 4.34 3.69

Note. Jack = scores on the Silencing the Self Scale; Gaveself = scores on question 78 of the Relationship Loss Questionnaire (RLQ), range 0 8; Beck = scores on the Beck Depression Inventory; Blatt = scores on the Dependency Scale of the Depressive Experiences Questionnaire; When = scores on question 75 of the RLQ, range 0 6; Choice = scores 76 of the RLQ, range 0 9; Oldimp = scores on question 77 of RLQ, range 0 9; Responsi = scores on question 79 of the RLQ, range 0 9; Oldsat = scores on question 80 of the RLQ, range 0 9; Currlmp = scores on question 82 of the RLQ, range 0 9; Currsat = scores on question 83 of the RLQ, range 0 9.

U)Mthesis Tests

The hypotheses were analyzed in two multiple regression analyses, one each for the BDI and the DEQ, which served as the dependent or criterion variables. The independent or predictor variables were relationship breakup, scores on the STSS, sex, and their interaction terms. The Type III sums of squares were used to calculate F values.
The first hypothesis was the Traditional Model hypothesis, which posited that there should be a main effect for relationship loss on the two measures of depression, and no main effect for self loss, nor a relationship loss by self loss interaction. This hypothesis was not supported. There was no significant main effect for relationship loss either for the BDI, F (1,246) = 1.89, jp < .18, or for the or the Dependency Subscale of the DEQ, F J1,246) = 0.86, p < .36.
The second hypothesis was the Jack Model hypothesis which posited that there should be a main effect for self loss on the two measures of depression, and no main effect for relationship loss, nor a relationship loss by self loss interaction. This hypothesis was clearly supported. First, there was a significant main effect for self loss for the BDI, E(1,246) = 25.78,_R < .000 1, with increased self loss associated with greater depression. Likewise, there was a significant main effect for self loss for the Dependency Subscale of the DEQ, E(1,246) = 45.50, jR < .0001. Again self loss was associated with greater depression.

The third hypothesis was the Potentiation Model hypothesis, which posited that there should be an interaction effect for the interaction of relationship loss and self loss on the two measures of depression. The model made no predictions regarding main effects for relationship loss or

self loss. This hypothesis was not supported. There was no significant interaction effect either for the BDI, F(1,246) = 1.60, p < .2 1, or for the Dependency Subscale of the DEQ, F(1,246) = 1.48, p < .23.

Finally there was little evidence that the general support for the Jack Model was moderated by sex differences. Only one main or interaction effect term, in either multiple regression, even approached statistical significance. The one interaction effect that approached significance was for the interaction of relationship loss, self loss and sex on the Dependency Subscale of the DEQ, F(1,246) = 3.52, p < .07.

Replication of Hv~othesis Tests

As mentioned in the method section, a second data set was collected late in the summer term. At that time an additional 172 University of Florida undergraduates completed the "Life Assessment Questionnaire". Results of this replication using identical multiple regression analyses follow.

Hypothesis 1 or the Traditional Model hypothesis which predicted only a main effect for relationship loss for the two dependent measures was not supported, BDI, f:(1, 148) = 0.00, p < .96; Dependency Scale of the DEQ, F(1, 148) = 0. 11, p < .75. Hypothesis 2 or the Jack Model hypothesis, which predicted a main effect only for self loss was once again clearly supported, BDI, E(1, 148) = 4.80, p < .04 ; Dependency Scale of the DEQ, E(1, 148) = 9.25, p < .003. And finally, the sole prediction for hypothesis 3, or the Potentiation Model hypothesis, was an interaction effect for relationship loss and self loss. No significant

interaction effect was found, BDI, Bl, 148) = 0. 12, p < .73; Dependency Scale of the DEQ, E(l, 148) = 0. 0 1, jR < .94.
As with the Jack Model results from the original data set, there was no compelling support for sex differences in terms of either main effects or interaction effects. This was the case for both the BDI and the DEQ.
On the whole, the results of the two separate analyses were remarkably similar, showing strong support for a link between the presence or degree of self loss and the intensity of depression following a relationship loss. Though fewer men than women reported self loss, when self loss did occur in men, the same pattern of connection between self loss and depressive symptoms was demonstrated; a pattern which had previously been assumed to be applicable only to women.

Ancillmy Analyses

In addition to the results previously described, correlation
coefficients were calculated for each pair of continuous data variables from the original data set only. These correlations are depicted in Table
2. Some of the more notable aspects of this table are discussed below.
In order to address the possibility that absence of support for the traditional model hypothesis might be explained by the insensitivity of the indicator of loss (see Question 74, see Appendix F), the correlations between the item asking how long ago the breakup occurred and scores on the BDI and the Dependency Subscale of the DEQ were inspected. These correlations, as can be seen from Table 1, were small and not

statistically significant, suggesting that recent breakups were not more associated with depression than the ones that occurred longer ago.
Likewise, one could argue that who initiated the breakup would
importantly influence the degree of depressed response, so the correlation between the item asking about who chose the breakup (see Question 76, Appendix F) and scores on the BDI and the Dependency Scale of the DEQ were looked at. Again these correlations were close to 0 and not statistically significant, suggesting that this aspect of breakup was not more indicative of depression than the original measure.
As with the previously discussed correlations, the correlation between the measure of the importance of the recently terminated relationship (see Question 77, Appendix F) and the BDI and DEQ scores were close to 0 and nonsignificant. Two other measures associated with relationship loss were also inspected, one involving which partner was attributed responsibility for the breakup (see Question 79. Appendix F) and, the other how satisfied the client was the past relationship. Again, these correlations were near 0 and nonsignificant. None of these results suggest support for the Traditional Model hypothesis.
Taken together this pattern of correlations underscores that when viewed from the perspectives of more precisely when the breakup occurred, who chose it, how important it was, who was responsible for the breakup or how satisfying the relationship was, the pattern was the same as on the original measure. That is, no support was found for the Traditional Model hypothesis on any of these widely varying measures of the parameters of relationship breakup. Parenthetically, all five of these auxiliary measures associated with breakup, correlated significantly with one another at moderate to high levels of magnitude (See Table 2).

Inspection of the correlation between scores on the BDI and the Dependency Scale of the DEQ reveals only a small to moderate relationship between the two measures. This underscores the importance of selecting more th ain one measure of depression, as well as the potency of self loss in accounting for significant variance in each of these measures.
Although self loss, not relationship loss, clearly accounted for the significant variance in predicting depression scores, there is an interesting set of associations between relationship loss and one aspect of self loss. Even though scores on the STSS did not correlate significantly with any of the measures of relationship loss, another question that directly asked participants whether they had given themselves up in the context of the failed relationship (see Question 78, Appendix F), correlated significantly, positively, and at moderately large magnitudes with each of the breakup associated measures (namely, when the breakup occurred, who chose it, whether it was importantwho was responsible for the breakup or how satisfying the relationship was). This pattern of associations is obviously complex and not amenable to quick interpretations, but may ultimately suggest directions for further exploring the conceptualization. of self loss.



0 o --1 0 4
0 r) 4-4
cq "i 0
C? 9 C? C? C? C? C? 0 LO (OD 4 co

0 4-1
b b ;oZo 0 0
N cq m cq C,4 m 0 co
6 ci 6 6 C5 0 0 cd w
W 00 M Vo
M Lo m cli cq m m 0
0 0 0
Oj 0 cr o o C:) o o C5
> 0
0 0
Cd t'- It to "It Clq to 0 r4
00 t- 11
0 o 0 o
-4 14 Q +0 C5 6 C5 6 C5 ci rj m 0 0 0 Cd
14t 0 r.- Itt -it ca Cd
t- cc t- co u
co 6 C5 d C5
cq 4) U u
co o 0 0 o m cq co t00 0 r- t- Cd V

M 4) %W cd Lo V V C5 0 0 C? 0 64 0 Cd
tl 0 QD
0 Q 0

0 cq 8
-.4 0 r. .4 40
Cd C5 6 6 C5
tow cq
m > o 00
8 000 m 0 o
C5 C5 C5 o
Q to V
w 0
MOO 064
6 0 v
0- 0
cq 0 11
L) V

0 C) LO
0 =) -" P.D 0 &0
U0. uu j 4) M v
I 6- 0 0 w
c c 4 6 6 t- o6 6 Z OD 0 IX


What Nora told Kiki about her romantic past is that she began to see that in each instance, as she molded herself more to her loved
one's life, he turned just a little bit colder, as she, Nora
disappeared piece by piece. (Otto, 1994, p. 83)

All companionship can consist only in the strengthening of two
neighboring solitudes, whereas everything that one is wont to call giving oneself is by nature harmful to companionship: for when a person abandons himself, he is no longer anything, and when two
people both give themselves up in order to come close to each
other, there is no longer any ground beneath them and their being
together is a continual falling. (Rilke, 1975, p. 27-28)

The purpose of this study has been to explore the roles of
relationship loss and self loss as possible triggers of depression in college women and men. This chapter will focus on discussion of the tests of the three principal hypotheses and on other analyses of interest. Also included will be a brief discussion of the strengths and limitatl ns, of the study, and the Implications for both future research and for psychotherapy.

UWgthesis 1

Hypothesis 1 tested the Traditional Model, which predicted a main effect of relationship loss on level of depression. This model of depression was not supported by patterns of scores on either of the two depression measures, the Beck Depression Inventory or the Dependency Scale of the Depressive Experiences Questionnaire.



The lack of support for the Traditional Model of Depression in this study lends additional credence to the calls of feminist theorists in the fields of psychology and psychotherapy to re-examine the core theories that purport to explain the development or causes of psychopathology in women. In particular, the Dependency Scale of the DEQ was designed to target a specific type of depression, which is defined by feelings of abandonment, dependency, neediness, and helplessness. Scores on this scale showed no significant link between the loss of a romantic relationship over the past year and the level of anaclitic depression. This finding poses a challenge to long-accepted clinical wisdom, which has presumed that a connection exists between the onset and severity of depressive symptomology in women and the recent loss of a significant other who was presumably providing for important emotional needs. Although the findings of this study are only one piece of a complex puzzle, the clear lack of support for Hypothesis 1, the Traditional Model, compels further exploration of alternative causative and correlative factors in women's depression. Furthermore, this lack of support for the Traditional Model hypothesis demonstrates the need for a continuing dialogue between research and clinical practice, including a strong argument for the use of well-designed studies that test widely accepted clinical lore.

U)Mthesis 2

Hypothesis 2, the Jack Model hypothesis, predicted a significant
relationship between self loss and level of depression as measured by the BDI and the DEQ Dependency Scale. Using the STSS as the independent


measure, a relationship between the degree of self loss and level of depression was clearly demonstrated on both dependent measures. In short, the participants who reported a greater tendency to have lost a sense of self showed higher levels of depression on two depression instruments. The Jack Model predicted no interaction between self loss and relationship loss and none was found.
The clear support (p < 000 1 in the initial analyses) for this newer model of women's depression is an intriguing outcome. It points to the possibility that, whereas loss is In fact a major factor in the etiology of depression, researchers and clinicians may have been concentrating on the wrong type of loss. During the initial, qualitative phase of her research, Jack's goal was to let go of preconceptions about the causes of depression in women. Outoif the interview data she collected and carefully scrutinized, Jack developed the concept of loss of self, which results from the current study have further validated.

HUgthesis 3

Hypothesis 3, the Potentiation Model, predicted that the
experience of both self loss and the loss of a significant other would interact in their contribution to the development of depression. This model anticipated not just an additive effect of self loss and other loss, but a synergistic or potentiating effect, which would have been manifested in the current study by significantly higher depression scores for those participants reporting both types of loss than for any of the other groups. However, hypothesis three was not supported.


The results in this study were extremely clear: the loss of a significant other, through a relationship breakup, did not predict depressive symptomology; nor did the interaction of relationship loss and self loss. Given this finding, it is not surprising that the Potentiation Model was not supported. Instead, the data clearly point to the newer concept of self loss as a potent predictor of depression.

Sex Differences

The sex of study participants was included in order to look for patterns of difference or similarity between women and men. It was predicted that general scoring patterns would be similar for both sexes, but that the overall frequency of depression among women would be higher. This, two-part prediction was supported and expands to an even greater extent the relevance of the Jack Model for understanding, evaluating and treating depression.
In short, because Jack's theory was thought to apply to women only, the results of the current study may lead to expansion of her thinking. For example, rather than assuming that the most likely source of male depression is task failure, it might be useful to explore the possibility that "loss of self" a similarly powerful contributor to male depression. Also, given these results it will be important to look more thoroughly at the connection between the tendency to give up one's self and the possible sources of this tendency in fernale upbringing or gender roles. Whereas loss of self was more common in women, as was depression in general, some male participants did indeed show self loss as well as the depression that is associated with it.



The sample employed in this study consisted of male and female students enrolled in undergraduate psychology courses at the University of Florida. The average age of subjects was approximately 22 years old, and most subjects fell into the 19-25 year age range. Although there was some ethnic diversity among the participants, overall numbers were not sufftcicnt to perform separate analyses based upon ethnicity. Data on socioeconomic status were not collected, however, college students from this university typically come from middle class backgrounds. Very few participants described themselves as gay or lesbian. Once again not enough data were available to make accurate analyses or statements concerning sexual orientation subgroups. Therefore, based on the sample's demographic parameters, genera-lizability of the findings of the current study would be limited primarily to white, heterosexual, undergraduate college students, between the ages of 19-25. One aim of subsequent studies might be to attempt to replicate these findings with a broader range of age groups, and with a wider array of ethnic groups, and individuals of diverse socioeconomic levels and sexual orientations.

Another limitation of the current study arose as a result of the
retrospective research design. Problems with this design include the need to rely on the recollections of participants,, and also, the possible Interference of intervening events over the time period explored., In addition, memories may be selective or biased. Participants were asked to consider breakups occurring over the last twelve months, which might not ordinarily be viewed a long period of time, but must be considered in


the framework of the rapidness of changes that often occur in young adults during this phase of psychological growth and development.
A second and perhaps more critical limitation imposed by the correlational nature of this design involves its inability to ascertain causal connections. Whereas the data show a correlation between self loss and depression, there is no way of knowing whether self loss brings about depression or vice versa, or whether other unknown factors are at work. This study was merely a first step in extending the discipline's understanding of the role of self loss in depression.
In planning for this study, considerable attention was given to
choosing appropriate instruments to measure the constructs of interest i.e., self loss, relationship loss, and depression. However, there are limitations involved in Instrumentation nevertheless. For examplewith respect to the independent measure, the STSS, used to measure self loss, is a new instrument that has not been fully validated. It was chosen for this study because of its unique measurement contribution in clarifying the role of self loss as a possible contributor to women's depression. The presence of relationship loss was based on a single item that asked for a yes or no response to the question, "Have you experienced the breakup of an intimate relationship in the past 12 months?" Although it is generally preferable to approach such constructs using a multiple question format, a search of the literature revealed no suitable instruments for this purpose. Construct validity may be increased by using multiple items for determining the presence or absence of relationship loss. Alternatively, participants may become frustrated by their perception of redundancy in multiple items asking about relationship loss.


With reference to the dependent measures, the construct of
interest was depression. Two measures of depression, the BDI and the DEQ were used in order to reduce the problem of mono-operation bias. The BDI was used based on its established reputation as a general screening instrument for depression. The DEQ was chosen for its theoretical position which is more traditionally psychoanalytic; one subscale of the DEQ was designed to be especially sensitive to the type of depression that occurs in reaction to loss or abandonment. However, even with multiple measures of depression, the problem of mono-method bias was not eliminated. Both the BDI and the DEQ are self-report measures and do not involve a clinician's or rater's direct observations. This raises the possibility that correlations found between variables may be due to some degree to method variance, rather than true relationships among the constructs of interest.


The major strength of the current study can be found in its attempt to break new ground. Over the past several decades, much attention and interest has focused on the growing discipline of women's studies, to which psychologists, psychoanalysts, and other clinicians have made significant contributions. However, as is often the case with a relatively newer area of investigation, the majority of the work on women's psychological development has been theoretical in nature. In an effort to add to the small number of empirical studies that build on the important work of Gilligan (1982), Jack (199 1), and Miller (1986), the current study has been an investigation of a significant piece of


clinically-relevant theory that re-visits the question of the causes of women's depression. Although the study is not complete in itself, it is a valuable step in the direction of beginning to link empirical testing to what has been primarily a conceptual literature.
A second strength of the study arises out of the clarity of the
results. Support for the Jack Model was unambiguous in the first data set. Further support for the Jack model came from a replication study whose data were collected independently of the first, and showed virtually identical results. Again, the Jack Model predictions were supported, the Traditional Model and the Potentiation Model predictions were not.
A third strength of the current study is the inclusion of male
participants. As mentioned above, the inspiration for this study came from literature on women, and at least some of this work, that of the socalled "difference feminists," assumes that there are fundamental differences in men and women arising from some combination of nature and nurture. One example of these presumed differences, that has been widely written about, is women's stronger relationship orientation. Though from the outset this was intended to be a study exploring women's depression, an interesting opportunity presented itself to look for similarities or differences in men's pattern of responding. In fact, data showed very similar patterns of responding in women and men; though, overall, men showed lower frequencies of self loss and depression.
In order to strengthen the validity of study results, two depression measures were included. This decision was based, in part, on the complex and heterogeneous nature of depressive disorders. As a means of


assessing the potentially wide variations in severity of depression, it was important to measure this key construct from more than one perspective. The use of instruments from two different theoretical perspectives, cognitive and psychodynamic, added. further credibility to study results.

A final strength of the current study lies in its potential clinical value. Whereas much of the psychological literature is only loosely relevant to practitioners, these results have immediate import. If depression is truly the "common cold" of mental health disorders, the concept of self loss, as originated and defined. in the work of Jack, has far reaching implications for both diagnosis and treatment of a range depressive disorders.

Suggestions for Future Research

Because this approach to understanding depression is very new and seems promising, there are myriad possibilities to pursue. The following discussion will touch on only a few of these possibilities and will center primarily on studies that might serve as logical successors to the study outlined here.
One very promising avenue to pursue involves designing a
prospective study to look at shrinkage or expansion of self-concept over the course of a relationship. The work of Hazel Markus and colleagues (Markus & Cross, 1990; Markus & Kitayama, 199 1; Markus & Nurius, 1986; Markus & Osyerman, 1989) on "possible selves," or the multiple roles through which an individual defines himself or herself, would provide a powerful means for tracking changes in identity, selfesteem, and well-being, as the relationship progressed. Although this


type of study will be lengthy and more demanding to complete, it would also avoid the biasing effects of recall inherent in retrospective designs, and would allow for a more detailed analysis of the process of self loss that occurs in some relationships.
To establish the external validity of current results that link self loss and depression, it will be important to employ a similar design with different age groups, ethnic groups, and with individuals from varied socioeconomic and educational levels. Sedney (1987) has looked at the development of identity over the lifespan and her findings suggest that, at midlife, there is some "crossing of paths" of men and women, with women becoming more focused on autonomy and self-expression and men on intimacy and relationships. If this is truly a common pattern, there might be some reversal of the pattern found in younger men and women. During the initial development and testing of the STSS, Jack and Dill (1992) did use her instrument with women from differing ethnic groups and SES levels; however, this diverse sampling should be continued as research progresses. Finally, it would also be interesting to discover whether the results compare across individuals of differing sexual orientations.

Yet another viable approach for extending the line of inquiry on the relevance of self loss as a precipitant of women's depression would involve the use of a known groups design. Two groups of women would be studied, the first would consist of women self-identified as not currently seeking a relationship, the second women who are seriall:y monogamous--moving quickly from one similar relationship to the next. Different predictions would be made for each group. Though members of


thefirst group, women not currently seeking a relationship, would have experienced past relationship loss and depression related to self loss, their growth process would be toward a stronger sense of identity and improved self-esteem. Members of the second group would also have a history of self loss and relationship loss, but would be seeking to relieve the depression and emptiness by immersing themselves in a new relationship. The growth pattern of this group would point more toward an addictive process. Methodology in this study could again involve the use of Markus' theory of possible selves which would allow the exploration of past current and possible selves in these two distinct groups of women (Markus & Wurf, 1987).
A final suggestion for further study would entail investigation into the impact of fathering on an adult daughter's general self-evaluation, and her tendency to silence herself to maintain a relationship or to avoid conflict. This idea comes from the theoretical works of both Jack (199 1) and Lerner (1988) who have used clinical data as the basis for speculating about how a woman learns sex-role behavior, including her concept of what constitutes a "good woman." While Lerner has focused primarily on the broader "family of origin" picture, Jack has shown particular interest in the role of the father. She suggests that women who grew up in families with a particularly dominating or critical father may ultimately be more prone to loss of self in the context of intimate adult relationships. Although a retrospective study comparing women on this dimension would have a number of built in limitations, it night ultimately prove helpful in understanding the integration of sex role into


identity, and in helping to prevent the type of self loss which is correlated with depression.

Implications for PsvchotheraM

The fundamental shift in the understanding of depression
suggested by Jack's work and by the current study, from a focus on relationship loss to a focus on self loss, has special significance for clinicians. Given the present incidence and increasing prevalence of unipolar forms of depression, especially in young people (Charney & Weissman, 1988; Nolen-Hoeksema & Girgus, 1994), certain changes in practice may be required to accommodate the needs of at least some depressed clients-both female and male. As with any good therapy, accurate and sensitive assessment is the first step.

Jack's preliminary study (199 1) serves as a valuable reminder of the need to listen to clients attentively and with as few preconceptions as possible. Rather than basing treatment interventions for depression only on traditionally accepted factors, a thorough investigation of the client's concept of self, primary relationships, family background and dynamics, and guiding cultural or societal norms is needed. Recurrent themes of self-doubt, extreme dependency, shame, "hyper -responsibility", and/or poor boundaries are signals that the possibility of self loss should be explored in greater depth. Questioning the client about gender roles in her family of origin and about her current expectations of herself as a woman worker, and partner may also prove relevant to a fuller understanding of current problems. In addition, if formal testing is to be


included as one aspect of the assessment process, the use of the STSS as an indicator of loss of self may prove useful.

For the female client whose depression may be attributable, at
least in some measure, to self loss, there are a number of issues to keep in mind. First, the primary goal of treatment is to aid the client in the development of a stronger sense of identity or self. Supporting this change may involve a number of different behaviors and interventions, foremost among these, the therapist's ability to contact, acknowledge, and support, even a partially hidden or extremely fragile sense of self. it is imperative that the therapeutic relationship support the client in learning to trust herself, and to openly share perceptions, feelings, needs and abilities (Belenky, Clinchy, Goldberger, & Tarule, 1986). The therapist should remain sensitive to the centrality of relationships in a woman's life, but should not collude with her tendency to silence herself in order to maintain important bonds (Schaef, 1985). As Lerner (1988, 1993) cautions, therapists must be aware of their own subtle biases and of the "pathogenic consequences" of rigid gender roles. Her recommendation is for "reparenting" focused on varied and unique aspects of the client's self, an approach that avoids an exclusive focus on socially sanctioned forms of self-expression. This reparenting process seems particularly well-suited to depressed women, as a foundation for rebuilding the self.
As another step in the recovery process, Bepko and Krestan (1990), have emphasized the necessity of helping the client to find a balance between work and relationships, connection and solitude, giving and receiving. In order to maintain this balance of giving to self and giving to others, they suggest that a conceptual shift must occur within the


women herself--a shift from valuing goodness to valuing wholeness. This change in self-concept, though clearly an internal process, is accomplished through ther-apy that encourages uncensored selfexpression, creativity and the reintegration of discarded or split-off parts of the self.

Only after a sense of identity and differentiation. has been
sufficiently nurtured and strengthened in the crucible of the therapeutic relationship can the focus of treatment shift away from self and back toward the concept of self-in-relationship (Lerner, 1987). In a variation on Martin Buber's (1958) work, the woman's relationship to her "self" must be an I-Thou relationship before it is possible to experience this same type of sacred relatedness in the interpersonal realm- that is, the encounter of whole self with whole self. As Jack has stated, the ultimate goal at this stage of healing is for women to "include themselves--their feelings, initiative, values, perspectives-in intimate relationships" (1991, p. 191). As long as a woman's shame forces the continued silencing of her voice, the possibility of creating a truly mutual and healthy relationship does not exist.
In the simplest terms, this new approach to treating women's

depression would break with tradition by helping women Identify, value, and express their existing skills and abilities. Unlike the more traditional models, which assume that women are inherently deficient, this newer treatment model assumes women have strengths. Practitioners implementing the newer model attempt to build on women's strengths in the treatment process, by challenging cultural attitudes and gender -role


conditioning, and by giving women the opportunity to listen clearly to

their own voices and dream
The well-known poet May Sarton described her own struggling
toward selfhood in the following excerpt from her poem, "Now I Become
MyselrNow I become myself. It's taken Time, many years and places, I have been dissolved and shaken, Worn other people's faces, Run madly, as if Time were there, Terribly old, crying a warning, "Hurry, you will be dead before-" Now to stand still, to be here, Feel my own weight and density! ... Now there is time and Time Is young. 0, in this single hour I live All of myself and do not move L the Pursued, who madly ran, Stand still, stand still, and stop the Sun!
(1991, pp. 28-29)


The primary investigator of this study is Sarah Drew, a doctoral
student in Counseling Psychology. The purpose of the study is to look at the connection between current well-being and the resolution of past romantic relationships.The study requests that you complete a multi. part questionnaire, which will require approximately 30 minutes of your time. Although it will be most helpful if you are able to fully complete the questionnaire, you may chose not to answer selected questions and still receive extra class credit for your participation Your name will never be associated with your answers, so the confidentiality of your responses will be protected.
There is expected to be no risk to you in participating in the study. Also, aside from an opportunity to think about your self and the relationships you have recently been Involved inthere is probably no immediate benefit to you for filling out the questionnaire. The details of the extra credit mentioned above will be arranged and explained by your Instructor. You are free to withdraw your consent to participate at any time during the course of the study. Please ask any questions you may have about the project, either now or later. You may reach me through the Department of Psychology, 114 Psychology Building, 392 -060 1. Leave a message, and I will get back to you as soon as possible.
In the unlikely event that answering certain of the study questions makes you aware of personal problems or concerns, there are several campus resources where you may choose to seek help in resolving these problems. The Counseling Center is located in Peabody Hall, Room 301 72

(Peabody is part of the Criser Complex). The phone number is 392-1575. Student Mental Health is located at the Infirmary Building, and the phone number is 392-1171.

I have read and understood the the procedure described above. I agree to participate and I have received a copy of this form.

Name (Please print) Signature

Witness (P. 1. or Assistant)

Approved for use through April 11, 1995.


1) When all the materials have been distributed, you should have one blue bubble sheet, two copies of the informed consent, and one "Life Assessment Questionnaire".
The questionnaire should consist of nine pages and ninety-one questions. Please note that some copies of the questionnaire are singlesided and some are double-sided.

2) Read the informed consent carefully and ask any questions you may have. If you decide to participate, please complete and sign one copy. Ask someone sitting nearby to witness your signature. The second copy does not need to be filled in. It is yours to keep.

3) Please write only on the bubble sheet, not on the questionnaire. You must use a #2 pencil on the bubble sheet.

4) Since the study is designed in such a way that your privacy is protected, do not put your name or social security number on the bubble sheet. Instead, look for the number in the upper right comer of the "Life Assessment Questionnaire". Write this number in the first four columns of the section called "identification number (social security number); then bubble in the appropriate numbers.

5) Notice on the bubble sheets that beginning with question #1, possible options range from 0-9. Not all questions utilize the entire scale. Just adjust as appropriate.

6) Also, notice that the scale changes from section to section. Please take the time to read the instructions for each section carefully, and look at the length and direction of the scale. (For example, the instructions change from page 4 to page 5.)

7) On some copies of the questionnaire, there is an error on question 86/87. To clarify, use number 86 to bubble in the first digit of your age and number 87 to bubble in the second digit of your age. Ask for help if needed.




Instructions: The Is a questionnaire. On the questionnaire are groups of statements. Please read the entire group of statements in each category. Then pick out the one statement in that group which best describes the way you feel today, that is, right nowl Circle the number beside the statement you have chosen. If several statements in the group seem to apply equally well. circle each one.

0 1 do not feel sad
1 1 feel sad
2 1 am sad all the time and I can't snap out of it
3 1 am so sad or unhappy that I can't stand it

0 1 get as much satisfaction out of things as I used to
I I don't enjoy things the way I used to
2 1 don't get real satisfaction out of anything anymore
3 1 am dissatisfied or bored with everything

0 1 can sleep as well as usual
1 1 don't sleep as well as I used to
2 1 wake up 1-2 hours earlier than usual and I find it hard to get
back to sleep
3 1 wake up several hours earlier than I used to and cannot get back
to sleep



Listed below are a number of statements concerning personal characteristics and baits. Read each item and decide whether you agree or disagree and to what extent. If you strongly disagree, bubble in number 6 on your answer sheet; if you strongly agree, bubble in number 0; if you feel somewhere in between, bubble in one of the numbers between 0 and 6. The midpoint, if you are neutral or undecided, is 3.

0 1 2 3 4 5 6

Strongly Strongly
Agree Disagree

23. Without support from others who are close to me, I would be
34. 1 find it very difficult to say 'No" to the requests of friends. 40. After a fight with a friend, I must make amends as soon as

Please note original numbering from the questionnaire used in data collection has been retained for purposes of clarity.



Please circle the number that best describes how you feel about each of the statements listed below.

0 1 2 3 4
Strongly Somewhat Neither agree Somewhat Strongly
disagree nor disagree agree agree

45.* Caring means putting the other person's needs in front of my own. 50. When my partner's needs and feelings conflict with my own, I
always state mine clearly.
63. My partner loves and appreciates me for who I am. 66. 1 rarely express my anger at those close to me.

Please note original numbering from the questionnaire used in data collection has been retained for purposes of clarity.



Please bubble in the number next to the correct response.

74. Have you experienced the breakup of an intimate relationship in the past 12 months? (If you have been through more than breakup in this time period, please think about the most recent one as you answer questions 74-80.)
No-0 Yes-i

75. Approximately how many months ago did the breakup occur?
does not apply to me-O0
1-2 mos.-1 3-4 mos.-2 5-6 mos.-3 7-8 mos-4 9- 10
mos-5 11-12 mos.--6

76. Was the breakup:
does not apply to me-O
1 2 3 4 5 6 7 8 9
Your choice A mutual choice Your ex-partner's choice

77. How important was the relationship to you?
does not apply to ine-O0
1 2 3 4 5 6 7 8 9
Not at all Extremely

78. In thinking back about the relationship, do you feel that at any time you "gave up your self", in order to keep the relationship going? For example did you change your behavior or hide your feelings in an attempt to maintain a connection to your partner?

does not apply to me --0

1 2 3 4 5 6 7 8
never gave up my self always gave up myself



79. Considering the breakup from your perspective now, do you feel the responsibility for the relationship ending was...

does not apply to me-0

1 2 3 4 5 6 7 8 9
Primarily yours About equally shared Primarily your ex-partner's 80. Overall, how satisfied were you with this relationship?

does not apply to me-0

1 2 3 4 5 6 7 8 9
Not at all Extremely

81. Are you currently involved in an Intimate relationship?

No- 0 Yes- 1

82. How important is this current relationship to you?

does not apply to me-0

1 2 3 4 5 6 7 8 9
Not at all important Extremely important

83. Overall, how satisfied are you with this current relationship?

does not apply to me-0

1 2 3 4 5 6 7 8 9
Not at all satisfied Extremely satisfied


Please bubble in the number next to the correct response.

84.* Are you:

Female-O0 Male-i1
85. Are you:

Heterosexual- 0 Homosexual 1 Bisexual 2

86/87. Your age (Use #84 for the first digit, #85 for the second digit. Example: If your age is 19, answer question #84 1; question #85 9.) 88. Year in school:

Post-Bac. -0
Freshperson -l1
Sophomore -2
Junior -3
Senior -4
Grad or Professional Student -5 89. Race:

Anglo or Euro-American -0
Afro-American 1
Hispanic American -2
Asian American -3
Native American (American Indian) -4 Other -5

*Please note original numbering from the questionnaire used in data collection has been retained for purposes of clarity.

90. Relationship Status:

Single --0

Partnered (but not living together) -1 81
Living Together -2
Married -3
Separated -4
Divorced --5
Widowed --6

91. Number of Children:

0 1 2 4 5, Etc.


American Psychiatric Association. (1994). Diagnostic and statistical
manual of mental disorders (4th ed.). Washington, DC: Author.

Arieti, S., & Bemporad, J. (1978). Severe and mild depression:The
psychotherapeutic approach. New York: Basic Books.

Beck, A. T. (1967). Depression: Causes and treatment. Philadelphia:
University of Pennsylvania Press.

Bepko, C., & Krestan, J. (1990). Too good for her own good: Searching for
self and intimacy in important relationships. New York:

Belenky, M. F., Clinchy, B. V., Goldberger, N. R., & Tarule, J. M. (1986).
Women's ways of knowing: The development of self voice and mind.
New York: Basic Books.

Blatt, S. J. (1974). Levels of object representation in anaclitic and
introjective depression. Psychoanalytic Study of the Child, 29,107157.

Blatt, S. J., & Bers, S. A. (1993). The sense of self in depression: A
psychodynamic perspective. In Z.V. Segal & S.J. Blatt (Eds.), The
emotional self in distress: Cognitive and psychodynamic
perspectives (pp.171-210). New York: Guilford.

Blatt, S., D'Afflltti, J. P., & Quinlan, D. M. (1976). Experiences of
depression in normal young adults. Journal of Abnormal
Psychology. 85. 383-389.

Blatt, S. J., & Shichman, S. (1983). Two primary configurations of
psychopathology. Psychoanalysis and Contemporary Thought, 6,

Blatt, S. J., & Zuroff, D. C. (1992). Interpersonal relatedness and selfdefinition: Two prototypes for depression. Clinical Psychology
Review, 12, 527-562.

Boyd, J. H. & Weissman, M. M. (1982). Epidemiology. In E.S. Paykel
(Ed.), Handbook of affective disorders (pp.109-125). New York:


Brown, L. M., & GiUigan, C. (1992). Meeting at the crossroads: Women's
psvchology and girls' developmen. Cambridge, MA: Harvard
University Press.
Buber, M. (1958). I and Thou (2nd ed., R.G. Smith, Trans.). New York:

Byron, G.G. (1958). Don Juan. Boston: Houghton Mifflin.

Charney, E. A., & Weissman, M. M. (1988). Epidemiology of depressive
illness. In J. J. Mann (Ed.), Phenomenology of depressive illness
(pp. 45-74). New York: Human Sciences Press.

Chevron, E. S., Quinlan, D. M., & Blatt, S. J. (1978). Sex roles and
gender differences in the experience of depression. Journal of
Abnormal Psychology, 8, 680-683.

Chodorow, N. (1978). The reproduction of mothering: Psychoanalysis and
the sociology of gender. Berkeley, CA: University of California
Cox, S., & Radloff, L.S. (1984). Depression in relation to sex roles:
Differences in learned susceptibility and precipitating factors. In C.
S. Widom (Ed.), Sex roles and Psychopathology (pp. 123-143). New
York: Plenum.

Eaton, W. W., & Kessler, L. J. (1981). Rates of symptoms of depression
in a national sample. American Journal of Epidemiology, 114, 528538.

Freud, S. (1917). Mourning and melancholia. Standard Edition, Vol. 14.
London: Hogarth.

Gilligan, C. (1982). In a different voice: Psychological theory and
women's development. Cambridge, MA: Harvard University Press.

Gilligan, C., Lyons, N. P., & Hanmer, T. J. (1990). Making connections:
The relational worlds of girls at Emma Willard School. Cambridge,
MA: Harvard University Press.

Goldin, L. R., & Gershon, E. S. (1988). The genetic epidemiology of major
depressive illness. In A. J. Frances & R. E. Hales (Eds.), Review of
Psychiatry (Vol. 7), (pp.149-168). Washington, DC: American
Psychiatric Press.

Goodwin, F. K., & Blehar, M. C. (1993). Introduction: Special edition:
Toward a new psychobiology of depression in women. Journal of
Affective Disorders, 29, 75-76.


Hare-Mustin, R. T., & Maracek, J. (1986). Autonomy and gender: Some
questions for therapists. Psychotherapy, 23, 205-212.

Hamilton, J. A., & Jensvold, M. (1992). Personality, psychopathology,
and depressions in women. In L.S. Brown & M. Ballou (Eds.),
Personality and psychopathology: Feminist reappraisals (pp. 116143). New York: Guilford.

Harris, H. (1987). Psychoanalytic theory and depression. In R. Formanek
& A. Gurian (Eds.), Women and depression: A lifespan perspective
(pp. 19-26). New York: Springer.

Hirschfield, M. A., & Davidson, L. (1988). Risk factors for suicide. In A.J.
Frances & R.E. Hales (Eds.), Review of psychiatry (Vol.7), (pp.307333). Washington, DC: American Psychiatric Press.

Jack, D. C. (1991). Silencing the self: Women and depression. Cambridge,
MA: Harvard University Press.

Jack, D. C., & Dill D. (1992). The Silencing the Self Scale: Schemas of
intimacy associated with depression in women. Psychology of
Women Quarterly, 16, 97-106.

Jordan, J. V., Kaplan, A. G., Miller, J. B., Stiver, I. P., & Surrey J. L.
(1991). Women's growth in connection: Writings from the Stone
Center. New York: Guildford Press.

Kamlner, W. (1993, October). Feminism's identity crisis. The Atlantic,
272 (4). 51-68.

Kaplan, A. (1991). The "self-in-relation": Implications for depression in
women. In J. V. Jordan, A. G. Kaplan, J. B. Miller, I. P. Stiver, & J.
L. Surey (Eds.), Women's growth in connection: Writings from the
Stone Center (pp.206-222). New York: Guilford.

Kaschak, E. (1992). Engendered lives: A new psychology of women's
experience. New York: Basic Books.

Keller, M. B. (1988).Unipolar depression: Afterword. In A..J. Frances &
R.E. Hales (Eds.), Review of sychiatry (Vol.7, pp. 147-148).
Washington, DC: American Psychiatric Press.

Kessler, R. C., McGonagle, K. A., Nelson, C. B., Hughes, M., Swartz, M.,
& Blazer, D. G. (1993). Sex and depression in the National Comorbidity Survey. II: Cohort Effects. Journal of Affective
Disorders, 30, 15-26.

Kessler, R. C., Mcgonagle, K. A., Swartz, M., Blazer, D. G., & Nelson, C.
B. (1993). Sex and depression in the National Comorbidity Survey
I: Lifetime prevalence, chronicity and recurrence. Journal of
Affective Disorders, 29, 85-96.

Kohut, H. (1978). The psychoanalyst in the community of scholars. In P.
Ornstein (Ed.), The search for the self: Selected writings of Heinz Kohut, (Vol.2, pp. 685-724). New York: International Universities

Lerner, H. G. (1987). Female depression: Self-sacrifice and self-betrayal
in relationships. In R. Formanek & A. Gurian (Eds.), Women and
depression: A lifespan perspective (pp.200-22 1). New York:

Lemer, H. G. (1988). Women in therapy. Northvale, NJ: Jason Aronson.

Lerner, H. G. (1993). The dance of deception: Pretending and truthtelling in women's lives. New York: HarperCollins.

Levinson, D. (1978). The seasons of a man's life. New York: Alfred A.

Mahler, M. (1972). On the first three subphases of the separationindividuation process. International Journal of Psychoanalysis, 53.,

Marcus, B. F. (1987). Object relations theory. In R. Formanek & A.
Gurian (Eds.), Women and depression: A lfespan perspective
(pp.27-40). New York: Springer.

Markus, H., & Cross, S. (1990). The interpersonal self. In L.A. Pervin
(Ed.), Handbook of personality theory and research (pp. 576-608).
New York: GuFldford.

Markus, H., & Kitayama, S. (1991) Culture and the self: Implications for
cognition, emotion, and motivation. Psychological Review, 98,

Markus, H., & Oyserman, D. (1989). Gender and thought: The role of
self-concept. In M. Crawford & M. Gentry (Eds.), Gender and
thought: Psychological Perspctives (pp. 100-12 7). New York:
Springer -Verlag.

Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist,

Markus H., & Wurf E. (1987). The dynamic self- concept: A social
psychological perspective. Annual Review of Psychology, 38, 299337.

Miller, J. B. (1976). Toward a new psychology of women. Boston: Beacon

Miller, J. B. (1986). Toward a new psychology of women (2nd ed.).
Boston: Beacon Press.

Miller, J.B. (1991). The development of women's sense of self. In J. V.
Jordan, A. G. Kaplan, J. B. Miller, I. P. Stiver, & J. L.Surrey (Eds.),
Women's growth in connection: Writings from the Stone Center
(pp.11-26). New York: Guilford.

Nolen-Hoeksema, S. (1990). Sex differences in depression. Stanford, CA:
Stanford University Press.

Nolen-Hoeksema, S. & Girgus, J. S. (1994). The emergence of gender
differences in depression during adolescence. Psychological
Bulletin, 115, 424-443.

Otto, W. (1994). Now you see her. New York: Ballantine.

Paykel, E. S. (Eds.). (1982). Handbook of affective disorders. New York:

Peay, P. (1994, July /August). What do women want? Common
Boundary, 12, 22-34.

Pipher, M. (1994). Reviving Ophelia: Saving the selves of adolescent girls.
New York: Ballantine.

Rilke, R.M. (1975). Rilke on love and other difficulties: Translations and
considerations (J. J. Mood, Trans.). New York: W.W. Norton.

Rush, A. J. (Ed.). (1982). Short-term psychotherapies for depression. New
York: Guilford Press.

Sarton, M. (1991). Now I become myself. In M. Sewell (Ed.), Cries of the
spirit (pp.28-29). Boston: Beacon Press.

Scarf, M. (1980). Unfinished business: Pressure points in the lives of
women. New York: Ballantine.

Schaef, A. W. (1985). Women's reality: An emerging female system in a
white male society. New York: HarperPaperbacks.

Schafer, R. (1959). Generative empathy in the treatment situation.
Psychoanalytic Quarterly. 28. 342-373.

Sedney, M. A. (1987). Development of androgyny: Parental influences.
Psychology of Women Quarterly, 11, 311-326.

Segal, Z. V., & Muran, J. C. (1993). Commentary. In Z. V. Segal & S. J.
Blatt (Eds.), The emotional self in distress: Cognitive and
psychodynamic perspectives (pp.211-217). New York: Guilford.

Shaver, P .R., & Brennan, K. A. (1991). Measures of depression and
loneliness. In J. P. Robinson, P. R. Shaver, & L. S. Wrightman
(Eds.), Measures of personality and social psychological attitudes
(pp. 195-289). San Diego: Academic Press.

Stiver, I.P. (1991). The meaning of care: Reframing treatment models. In
J. V. Jordan, A. G. Kaplan, J. B.Miller, I. P. Stiver & J. L. Surrey (Eds.), Women's growth in connection: Writings from the Stone
Center (pp.250-267). New York: Guilford.

Surrey, J. L. (1991). The "self-in-relation": A theory of women's
development. In J. V. Jordan, A. G. Kaplan, J. B.Miller, I. P. Stiver
& J. L. Surrey (Eds.), Women's growth in connection: Writings
from the Stone Center (pp.51-66). New York: Guilford.

Tavris, C. (1992). The mismeasure of woman. New York: Simon and

Uzelac, E. (1995, September / October). Innovations: In a daughter's
voice. Common Boundary, 13, 49-53.

Weissman, M. M., Bland, R., Joyce, P. R., Newman, S.,Wells, J. E., &
Wittchen, H. (1993). Sex differences in rates of depression: Crossnational perspectives. Journal ofAffective Disorders,29, 77-84.

Weissman, M. M., & Klerman, G. L. (1987). Gender and depression. In R.
Formanek & A. Gurian (Eds.), Women and depression: A lifespan
Perspective (pp.3-15). New York: Springer.

Welkowitz, J., Lish, J. D., & Bond R. N. (1985). The Depressive
Experiences Questionnaire: Revision and validation. Journal of
Personality Assessment, 49, 89-94.

Wright, R. (1995, August 28). The evolution of despair. Time. 146, 50-57.


Sarah Sevier Drew was born on January 30, 1957, in Tallahassee, Florida, to Edward deBelle Drew and Elizabeth Lanier Drew. She grew up in Jacksonville, Florida, and attended Episcopal High School. Inl978, Sarah became a Phi Beta Kappa graduate of the University of Florida.
Following a period of exploration, Sarah began graduate studies in counseling psychology and developed a particular interest In object relations theories and the psychology of women. Sarah was awarded a Master of Science degree in 1991 and completed her internship at the University of Maine Counseling Center during 1992-1993.
Since her return from Maine, Sarah has worked in two
professional positions, one involving research the other focusing on clinical Intervention. She Is planning to pursue her Interest In treating the psychological problems of women.