The Dependent Patient: A Practitioner's Guide

Citation preview

The Dependent Patient A Practitioner's Guide

R O B E R T F. B O R N S T E I N

American Psychological Association • Washington, DC

Copyright © 2005 by the American Psychological Association. All rights reserved. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written permission of the publisher. Published by American Psychological Association 750 First Street, NE Washington, DC 20002 www.apa.org To order APA Order Department P.O. Box 92984 Washington, DC 20090-2984 Tel: (800) 374-2721; Direct: (202) 336-5510 Fax: (202) 336-5502; TDD/TTY: (202) 336-6123 Online: www.apa.org/books/ E-mail: [email protected] In the U.K., Europe, Africa, and the Middle East, copies may be ordered from American Psychological Association 3 Henrietta Street Covent Garden, London WC2E 8LU England Typeset in Goudy by Stephen McDougal, Mechanicsville, MD Printer: Sheridan Books, Ann Arbor, MI Cover Designer: Naylor Design, Washington, DC Technical/Production Editor: Emily Leonard The opinions and statements published are the responsibility of the authors, and such opinions and statements do not necessarily represent the policies of the American Psychological Association. Library of Congress Cataloging-in-Publication Data Bornstein, Robert F. The dependent patient: a practitioner's guide / Robert F. Bornstein.—1st ed. p. cm. Includes bibliographical references and index. ISBN 1-59147-203-2 (alk. paper) 1. Dependency (Psychology) I. Title. RC569.5.D47B669 2005 616.85'81—dc22 British Library Cataloguing-in-Publication Data A CIP record is available from the British Library. Printed in the United States of America First Edition

2004019485

To my students, whose insights encourage me to revisit important ideas and question longstanding assumptions. Without realizing it, they have taught me far more than I could ever teach them.

CONTENTS

Preface

xi

I.

1

Conceptual and Empirical Foundations Chapter 1. Conceptualizing Dependency Psychodynamic Models Behavioral and Social Learning Models Culture and Gender Role Cognitive Models Trait Models An Interactionist Model Integrating the Contributions of Extant Theoretical Frameworks

3 4 6 8 11 13 15

Chapter 2.

21 21 22

Chapter 3.

18

Quantifying Dependency Classifying Measures of Dependency Stand-Alone Self-Report Measures Self-Report Scales Embedded in Longer Tests Projective Tests Interviews Combining and Contrasting Test Results....

27 28 31 33

Dependency Across the Life Span Dependency as a Pervasive Life Theme The Evolution of Dependency

39 40 45

vn

Life-Span Dependency: Toward an Integrated View Chapter 4- Context-Specific Deficits and Strengths From Pervasive Passivity to Situational Variability An Interactionist Perspective on Dependency Dependency-Related Problems and Deficits.. Dependency-Related Skills and Strengths ... Clinical Implications of the Interactionist Perspective Chapter 5.

II.

Healthy and Unhealthy Dependency Distinguishing Healthy and Unhealthy Dependency Unhealthy Dependency Patterns From Laboratory to Consulting Room: Diagnosis, Assessment, and Treatment of Dependent Patients

57 58 61 63 67 70 73 74 82

86

Clinical Applications

89

Chapter 6.

91

Diagnosis

Dependent Personality Disorder in the DSM Epidemiology Differential Diagnosis and Comorbidity Effective Use of Diagnostic Information: A Framework for the Practitioner Chapter 7. Assessment Testing Versus Assessment Measuring Implicit and Self-Attributed Dependency Needs Dependency in Context: Perceptions, Defenses, and Social Support Assessing Dependency Subtypes Assessing Dependency's Impact Effective Use of Assessment Information: A Framework for the Practitioner

Vlll

54

CONTENTS

93 96 99 104 Ill 112 112 116 120 121 123

Chapter 8.

Approaches to Treatment The Psychodynamic Perspective The Behavioral Perspective The Cognitive Perspective The Humanistic-Experiential Perspective ... Effective Use of Traditional Treatment Models: A Framework for the Practitioner

Chapter 9. An Integrated Treatment Model Assimilative Integration of Dependency Treatment Models Integrating Strategies Across Domains of Patient Functioning Effective Use of Therapeutic Integration Strategies: A Framework for the Practitioner Chapter 10. Specialized Treatment Issues Contexts and Settings Alternative Treatment Modalities Special Treatment Challenges Effective Use of Alternative Treatment Strategies: A Framework for the Practitioner

131 132 135 138 141

144 151 154 156

164 173 174 176 183

185

References

193

Index

235

About the Author

243

CONTENTS

IX

PREFACE

Problematic dependency takes many forms. Some people are obviously dependent, unable to make even the smallest decision without an inordinate amount of advice and reassurance from others. But many people who do not appear dependent still bring dependency-related issues to therapy. The depressed patient who can no longer manage her life, the anxious patient who cannot leave home on his own, the borderline patient with deep-seated abandonment fears, and the elderly patient who is unable to carry out basic activities of daily living—all these people (and countless others) have difficulties involving dependency. This book is about working effectively with dependent patients. It discusses cutting-edge treatment techniques, outlines strategies for diagnosing dependency, and reviews procedures for assessing dependency-related personality dynamics that are not captured by formal diagnostic criteria. The central premise of this book is straightforward: Dependency is an important issue in clinical practice, but it is also a ubiquitous feature of human experience. We are, in the end, social creatures, bound to each other from our first days to our last. Thus, effective clinical work with dependent patients does not involve quashing dependency in all its forms, but replacing unhealthy dependency with healthy connectedness. Two events that took place nearly 20 years ago laid the groundwork for this book, and they illustrate two themes that characterize my approach to the dependent patient. The first event occurred in the mid-1980s while I was working in a psychiatric inpatient unit in upstate New York. There I met a 28-year-old bulimic woman named Marta, who—despite being highly dependent—showed considerable variability in behavior. Around her siblings Marta was clingy and insecure, but at work she was confident and assertive— completely in command. Marta's behavior did not square well with clinical writings on dependency at that time, and her unpredictable behavior was XI

difficult to comprehend. Now I understand Marta better. Her dependent behavior was unpredictable, but it was unpredictable in a very predictable way. Marta—like many dependent patients—was displaying an array of selfpresentation strategies that, while superficially different, had the common goal of pleasing those closest to her, drawing others in, and minimizing the possibility of relationship disruption. From Marta 1 learned that dependent behavior is far more diverse (and far more deliberate) than clinicians once believed. Although dependent people are motivated to seek protection and help from others, they express this need in many different ways—sometimes acting meek and passive, at other times becoming active and even quite aggressive. The second eye-opening event took place a year later, and it involved a research study that refused to come out the way we had hoped. As so often happens in the world of science, however, our "failed" investigation proved far more informative than it would have had the results turned out as expected. Joseph Masling, Frederick Poynton, and I were assessing the relationship between self-report and projective dependency scores in college students. We used well-established questionnaire and projective measures of dependency—the Depressive Experiences Questionnaire and the Rorschach Oral Dependency Scale—and we expected that scores on the two measures would show strong positive correlations. They did not. No matter how many samples we ran, the result was always the same: A very modest link between self-report and projective dependency scores. Because we had been taught to expect different measures of the same trait to correlate strongly, we assumed that something had to be wrong with one or both tests, and it took us a long time to understand what we had found. Now we know these results are but one example of a more general pattern that emerges in clinical assessment: Self-report and projective measures tap different aspects of dependency. Knowing this, clinicians must treat these assessment instruments as unique sources of data about a patient's personality and coping style. Divergences between scores on different indices of dependency can be as important and informative as convergences—sometimes more so. For the past 20 years I have studied the dynamics of dependency in clinical and research settings, and the results of these investigations form the foundation of this book. I am not alone in my interest in dependency, of course: There have been hundreds of clinical and empirical papers on this topic during the past two decades. Effective treatment of the dependent patient begins with understanding the results of these research programs, but it does not end there. Self-awareness is important as well. The ubiquitous, "experience-near" nature of dependency demands that knowledge of research and clinical evidence go hand-in-hand with self-reflection and self-scrutiny. To work effectively with dependent patients, clinicians must understand their own beliefs about dependency and their reflexive responses to dependent behavior in others. I discuss this challenge—and ways to overcome it—in the second half of the book. xii

PREFACE

To set the stage for effective clinical work with dependent patients, I have drawn upon as broad an array of sources as possible, including: • Laboratory and clinical data. Clinical studies are ideal for understanding the antecedents, correlates, and consequences of dependency, but laboratory experiments provide a unique method for manipulating dependency-related psychological processes and disentangling complex relationships among these processes. • Studies of nonclinical participants and studies of patients. Since the early 1950s, there have been more than 600 empirical studies of interpersonal dependency. At least half involved nonclinical participants, and the results of these investigations tell us much about the inter- and intrapersonal dynamics of dependency in vivo. • Studies of dependent personality traits and studies of dependent personality disorder. Some investigators operationalize dependency using dependent personality disorder symptoms; others use selfreport, projective, or behavioral indices to quantify patients' dependency levels. Because different assessment methods tap different aspects of functioning, these contrasting findings can teach us a great deal about the various ways dependency is expressed. The Dependent Patient is intended not to be a cookbook but a framework each clinician will use differently. Thus, I review a broad array of theories, assessment methods, and treatment approaches. I make recommendations in each area—useful diagnostic techniques, assessment tools, and psychotherapeutic interventions—but these recommendations are merely guidelines to be adapted to each patient and each situation. This book is structured so that research evidence informs every clinical decision confronting the practitioner. Thus, in the first half of the book I review the conceptual and empirical foundations of dependency: Conceptualizing Dependency (chap. 1), Quantifying Dependency (chap. 2), Dependency Across the Life Span (chap. 3), Context-Specific Deficits and Strengths (chap. 4), and Healthy and Unhealthy Dependency (chap. 5). In the second half of the book, I discuss clinical applications of these findings: Diagnosis (chap. 6), Assessment (chap. 7), Approaches to Treatment (chap. 8), An Integrated Treatment Model (chap. 9), and Specialized Treatment Issues (chap. 10). The Dependent Patient is based on the scientist—practitioner model but not constrained by it. It extends this model while applying the model's core principles to integrate dependency-related research and clinical data. Recent changes within the mental health professions have compelled us to reconceptualize the traditional role of the scientist-practitioner. Managed care, a changing professional landscape, prescription and hospital admitting PREFACE

Xlii

privileges, the emergence of new categories of mental health professionals— all have reshaped the life and work of the clinician. Add to this the fact that clinicians—like patients—are imperfect processors of information, subject to all sorts of perceptual and memory biases that limit our ability to apply knowledge in an objective, dispassionate way, and it is easy to see that bringing clinical science to the consulting room is fraught with pitfalls, but replete with opportunities as well. I would like to thank the many people who contributed to this book, without whom I could not have written it. I am indebted to Susan Reynolds and Mary Lynn Skutley of APA Books, whose enthusiasm for this project provided the spark that enabled me to complete it, and to Emily Welsh and Emily Leonard, whose editorial expertise improved The Dependent Patient in countless ways. I am grateful to the library staff of Gettysburg College—most especially Susan Roach and Linda Isenberger—for going beyond the call of duty to help me locate and obtain books and articles on dependency. As always, Susan and Linda saw my many requests as an opportunity—not a burden—and they tracked down every piece of information I requested, no matter how obscure. I am indebted to Carolyn Tuckey for her help in constructing readerfriendly graphs and figures. Without her creative efforts, my ideas would not have been expressed as clearly or as well. Many colleagues gave willingly of their time and energy, commenting on drafts of The Dependent Patient, providing advice and support, and improving the book tremendously. I am grateful to Kathleen Cain, Kristen Eyssell, Fritz Gaenslen, Jeffrey Johnson, Mary Languirand, Joseph Masling, Anne Sauve, and Joel Weinberger for their very helpful feedback regarding preliminary versions of various chapters. Their insightful comments and challenging questions helped clarify my thinking and enabled me to communicate these ideas more effectively.

XIV

PREFACE

I CONCEPTUAL AND EMPIRICAL FOUNDATIONS

1 CONCEPTUALIZING DEPENDENCY

Virtually every mental health professional has encountered patients who are overly dependent—patients who seem unable to make a decision on their own and alienate those around them with insatiable neediness and clinging insecurity. Studies show that rigid, inflexible dependency has myriad negative effects on a person's social, career, and romantic relationships. As every clinician knows, it can undermine treatment as well. The problem, in its most basic form, is straightforward: Many dependent patients become so comfortable in therapy's protective cocoon that they resist change to perpetuate the relationship. Impending termination brings forth an array of responses—some conscious, some unconscious—aimed at undoing progress and obviating gain. The therapist's emotional reactions to the overdependent patient's resistance run the gamut from infantilizing overprotectiveness to thinly-veiled hostility—even outright anger and resentment. There is no doubt about it: Dependent patients present unique challenges for practitioners. In this chapter, I review theoretical models of interpersonal dependency. These models conceptualize dependent traits from a variety of perspectives, and each model informs the clinician in important ways about some key features of dependency. The theoretical frameworks described in this chapter—along with the dependency assessment tools discussed in chapter 2—

serve as context for a review of empirical research in chapters 3 through 5. Ultimately, the theoretical frameworks described here help set the stage for an in-depth consideration of clinical strategies for diagnosing (chap. 6), assessing (chap. 7), and treating dependent patients (chaps. 8-10). PSYCHODYNAMIC MODELS Although the most widely studied psychoanalytic perspective on dependency is Freud's well-known "oral fixation" model, there are actually several psychodynamic frameworks relevant to this issue. These models differ with respect to the intra- and interpersonal dynamics that are presumed to underlie a dependent personality orientation, but they share a common emphasis on early relationships as the building blocks of dependency. Classical Psychoanalytic Theory In classical psychoanalytic theory, dependency is inextricably linked to events that occur during the first months of life—the period Freud termed the "oral" stage of development. In Freud's (1905/1953) model, frustration or overgratification during the infantile, oral stage was thought to result in oral fixation and an inability to resolve the developmental issues that characterize this period (i.e., conflicts regarding dependency and autonomy). As Freud (1908/1959, p. 167) noted, "one very often meets with a type of character in which certain traits are very strongly marked while at the same time one's attention is arrested by the behavior of these persons in regard to certain bodily functions." Thus, classical psychoanalytic theory postulates that the orally fixated (or "oral dependent") person will (a) remain dependent on others for nurturance, guidance, protection, and support; and (b) continue to exhibit behaviors in adulthood that mirror those of the oral stage (e.g., preoccupation with activities of the mouth, reliance on food and eating as a strategy for coping with anxiety). Empirical support for Freud's (1905/1953, 1908/1959) psychoanalytic model of dependency has been mixed. On the positive side, it is clear that early relationships with parents and other caregivers play a role in the etiology of dependent personality traits (Fisher & Greenberg, 1996; Masling, 1986). On the negative side, however, there is no evidence that oral fixation as Freud (1905/1953) described it affects dependency-related behaviors during adolescence and adulthood. Efforts to link the development of dependent traits to infantile feeding and weaning variables have produced uniformly negative results. Moreover, studies indicate that dependent children, adolescents, and adults do not show greater preoccupation than their nondependent counterparts with food- and mouth-related activities (Bornstein, 1992,1996a). Despite the theory's limitations, it would be a mistake to reject Freud's classical psychoanalytic model of dependency outright. Aside from calling 4

CONCEPTUAL AND EMPIRICAL FOUNDATIONS

attention to the importance of early relationships in shaping dependent attitudes and behaviors, the psychoanalytic model is unique in its emphasis on unconscious determinants of dependency. As we will see in chapter 2—and again when we consider assessment and treatment issues in the second half of the book—the distinction between unconscious (or "implicit") dependency needs and conscious dependency strivings is critical in understanding the complex inter- and intrapersonal dynamics of dependency. Object Relations Theory and Self Psychology Beginning in the 1920s, the focus of psychodynamic metapsychology shifted from Freud's (1905/1953) drive-based framework to a more personcentered approach that came to be known as object relations theory (Greenberg & Mitchell, 1983). Within this framework, personality development and dynamics are conceptualized in terms of (a) self-other interactions (both real and imagined) and (b) internalized mental representations of self and significant figures (Bornstein, 2003; Galatzer-Levy & Cohler, 1993; Huprich, 2001). Although there is considerable overlap between object relations theory and self psychology—with respect to underlying assumptions as well as formal terminology—there are some noteworthy differences as well. Object relations models generally emphasize the interpersonal dynamics and enduring relationship patterns that foster and maintain dependent behavior (e.g., Fairbairn, 1952; Kernberg, 1975), whereas models derived from self psychology emphasize the role of the dependent person's self-concept and internalized self-representation in the etiology and dynamics of dependency (e.g.,Kohut, 1971, 1977).1 As Greenberg and Mitchell (1983) and others (e.g., Galatzer-Levy & Cohler, 1993) have noted, the reconceptualization of psychoanalytic concepts in relational terms introduced a fundamentally new paradigm for understanding continuity and change in personality development and dynamics. Instead of being understood solely in terms of a dynamic balance among id, ego, and superego, stability in personality was now seen as stemming from continuity in the core features of key object representations, including the self-representation (Bornstein, 1996a, 2003). Conversely, personality change was presumed to occur in part because internalized representations of self and other people evolve in response to changing life circumstances. In recent years, Blatt's (1974, 1991) theoretical framework has been the most influential object relations model of dependency. Integrating psychoanalytic principles with research on cognitive and social development, Blatt and his colleagues (e.g., Blatt & Schichman, 1983; Blatt & Zuroff, 1992) have argued that dependent personality traits result from a mental represen'Among the seminal object relations and self psychology contributions to the etiology and dynamics of dependency are those by Guntrip (1961), Jacobson (1964), Bowlby (1969, 1980), Ainsworth (1969), and Sandier and Dare (1970).

CONCEPTUALIZING DEPENDENCY

tation of the self as weak and ineffectual. Retrospective and prospective studies of parent-child interactions confirm that those parenting styles that cause children to perceive themselves as powerless and vulnerable are in fact associated with high levels of dependency later in life. As children internalize a mental representation of the self as weak, they (a) look to others to provide nurturance and support; (b) become preoccupied with fears of abandonment; (c) behave in an overtly dependent manner; and (d) show increased risk for depression and other "anaclitic" (i.e., dependency-related) psychopathologies (see Blatt & Homann, 1992, for a review of research in this area).

BEHAVIORAL AND SOCIAL LEARNING MODELS The basic premise of the behavioral perspective is straightforward: People exhibit dependent behaviors because those behaviors are rewarded, were rewarded, or—at the very least—are perceived by the individual as likely to elicit rewards. Early behavioral and social learning models of dependency were strongly influenced by the work of Hull (1943) and Mowrer (1950). Thus, dependency was initially conceptualized as an acquired drive, the impetus for which was the reduction of basic, primary drives (e.g., hunger) within the context of the infant-caregiver relationship (see Dollard & Miller, 1950, for a detailed discussion of this view). Ainsworth (1969, p. 970) provided a succinct summary of the behavioral-social learning perspective, noting that within this framework dependency is regarded as "a class of behaviors, learned in the context of the infant's dependency relationship with his mother . . . although the first dependency relationship is a specific one, dependency is viewed as generalizing to subsequent interpersonal relationships." Instrumental and Emotional Dependency As the behavioral approach gained influence during the early 1950s, researchers delineated separate categories of dependent behavior (e.g., helpseeking, reassurance-seeking, etc.). Although several frameworks were developed to address this issue, the mo'st influential early subtype model was that of Heathers (1955), who hypothesized that dependent behaviors could be usefully divided into instrumental and emotional categories. Heathers argued that in instrumental dependency, other peoples' responses serve as tools that help the individual meet some goal. Thus, the instrumentally dependent person's actions are directed primarily toward task-oriented helpseeking. In emotional dependency, other peoples' responses are reinforcing in and of themselves—merely eliciting the desired response (e.g., reassurance, support) is the dependent person's goal. Thus, the emotionally dependent person's actions tend to be focused on obtaining succorance and nurturance rather than instrumental help. Within Heathers's framework in6

CONCEPTUAL AND EMPIRICAL FOUNDATIONS

strumental and emotional dependency were thought to have different antecedents, correlates, and interpersonal consequences. Heathers's (1955) instrumental-emotional dependency distinction was influential throughout the late 1950s and early 1960s, but it was eventually criticized by Walters and Parke (1964), Gewirtz (1972), and others. Although Heathers had conceptualized instrumental and emotional dependency as orthogonal constructs, studies showed that the behaviors associated with these two dependency subtypes were in fact strongly linked, and scores on observational measures of instrumental and emotional dependency were highly intercorrelated in children (Kagan & Mussen, 1956) and adults (Baltes, 1996). Other investigations suggested that instrumental and emotional dependency had common antecedents and parallel acquisition, maintenance, and extinction patterns (Ainsworth, 1969). Most clinicians and researchers now regard instrumental and emotional dependency as two facets of a broader dependent personality style, with most (but not all) people who show high levels of one dependency facet also showing high levels of the other.2 Social Reinforcement of Dependent Behavior A natural outgrowth of the behavioral view was the notion that dependent behaviors are shaped in social settings. Even if dependent behavior was first acquired in the child's early interactions with parents and other caregivers, this behavior must be reinforced (at least occasionally) in later relationships, or it will eventually be replaced by other social influence strategies. Because many children are rewarded for exhibiting dependent behavior in some relationships but not others, an intermittent reinforcement pattern is common— a pattern that renders dependent behavior highly resistant to extinction (Bhogle, 1978; Turkat, 1990). As children learn which behaviors are effective in eliciting the desired responses, and in which relationships these behaviors are (and are not) successful, they gradually adjust their help- and reassurance-seeking efforts to maximize rewards in different contexts. Studies confirm that intermittent reinforcement of dependent behavior plays a key role in the interpersonal dynamics of dependency—not only in children, but in adults as well. Such intermittent reinforcement patterns have been identified in a broad array of settings, including classrooms, hospitals, rehabilitation centers, and nursing homes (Baltes, 1996; Kilbourne & Kilbourne, 1983; Sroufe, Fox, & Pancake, 1983; Turkat & Carlson, 1984). As Bandura and Walters (1963) and others (e.g., Walters & Parke, 1964) pointed out, however, dependent behavior need not be reinforced directly in 2

Although Heathers's (1955) subtype model has fallen out of favor, other subtype models have proven useful in understanding the dynamics of dependency and the contrasting behaviors and relationship patterns of different dependent patients. In contrast to Heathers's behavioral framework, most contemporary dependency subtype models are derived from trait theory, using factor-analytic, clusteranalytic, and circumplex techniques (e.g., Pincus & Gurtman, 1995; Pincus & Wilson, 2001). CONCEPTUALIZING DEPENDENCY

order to be maintained. Observational learning is also important, and two processes are key in this regard: • Vicarious reinforcement. To the extent that a child notices other children (e.g., siblings, peers) being rewarded for exhibiting dependent behaviors, she is more likely to exhibit those behaviors herself. The influence of vicarious reinforcement will be particularly strong if the child (a) observes dependent behaviors being rewarded consistently and (b) admires or identifies with the person exhibiting dependent behavior (see Bandura, 1978). • Modeling. As Bandura (1977) noted, modeling—including symbolic modeling (i.e., imitation of persons pictured on film and in other media)—can play a powerful role in shaping and maintaining the developing child's help- and reassurance-seeking tendencies, even in the absence of direct or vicarious reinforcement. Symbolic modeling may be particularly important in the early acquisition of dependent traits, because many fictional characters in children's literature—especially female characters—exhibit high levels of stereotypic "helpless" dependency.3 CULTURE AND GENDER ROLE Social reinforcement of dependent behavior does not occur randomly, but is influenced by aspects of an individual's culture and his or her place within that culture. Individualistic societies typically emphasize autonomy and independence at the expense of social connectedness, and these societies tend to be particularly intolerant of dependency, especially in adults. Moreover, some cultures have rigid expectations regarding appropriate gender role-related behavior, whereas other cultures are more flexible in this domain (Cross, Bacon, & Morris, 2000; Cross & Madson, 1997). Studies indicate that gender role norms can have a powerful impact on women's and men's willingness to acknowledge underlying dependency needs (Bornstein, 1995c; Cadbury, 1991). The Impact of Cultural Context Dependent behavior is tolerated more readily in communitarian (or sociocentric) cultures than in individualistic ones (Doi, 1973; Johnson, 1993; Quantitative analyses of dependency-related behavior in children's stories were conducted by Fischer and Torney (1976) and White (1986). In both investigations female characters were portrayed as (a) more dependent and helpless than their male counterparts, (b) more likely to receive help, and (c) more receptive to receiving help. Fischer and Torney further found that exposure to these storybook characterizations had significant, measurable effects on 5-year-old children's dependencyrelated behavior in the classroom. Such results strongly support the role of symbolic modeling in shaping children's dependent behavior—at least in the short term.

CONCEPTUAL AND EMPIRICAL FOUNDATIONS

Neki, 1976). Thus, adolescents and young adults in Japan and India have traditionally shown higher levels of self-reported dependency than those raised in North America, a pattern that continues today (Bornstein & Languirand, 2003). To a great extent, these sociocentric—individualistic cultural differences reflect prevailing norms regarding the relative importance of family ties and maintenance of group harmony. Even within a given culture, members of subgroups that value connectedness most strongly tend to score highest on self-report measures of dependency (Yamaguchi, 2004). As cultures change, attitudes regarding dependency evolve in predictable ways. Thus, as a sociocentric society becomes increasingly Westernized, population-wide dependency levels in that society tend to decrease. A second, subtler dynamic occurs as well: As individualistic values and norms are introduced into a communitarian society, people experience dependency con' flicts—discontinuities between longstanding sociocentric norms and newer individualistic values. Only recently have Japanese policymakers recognized the deleterious effects that such dependency conflicts have on their citizens' psychological adjustment. Feeling compelled to connect and compete at the same time (and with many of the same individuals), Japanese businesspeople report increased stress, frustration, and alienation (Sato, 2001). A similar process has occurred as the Indian economy has modernized and become Westernized during the past several decades (Bhogle, 1983; Singh & Ojha, 1987).4 Gender and Gender Role Gender—like culture—affects an individual's attitude regarding dependency, and societal norms are important in this context as well. Most investigations of gender differences in dependency have been carried out in Western societies (e.g., the United States, Canada, Great Britain), and in virtually every one of these investigations, women obtained significantly higher scores than men did on self-report tests of dependency (Bornstein, 1995c). Moreover, in both women and men, high self-report dependency scores were associated with high femininity and low masculinity scores on measures of gender role. Table 1.1 summarizes the results of extant studies in this area, and as this table shows, highly consistent findings have been obtained in these investigations: In almost every analysis to date, there has been a positive correlation between dependency and femininity, and a negative correlation between dependency and masculinity. 4

Although recent cultural shifts have exacerbated dependency conflicts among Japanese citizens, the tensions between individualism and collectivism actually have a long history in many communitarian societies. Moreover, even as individualistic values become firmly entrenched, underlying motivations remain stable: Strengthening social ties and maintaining group cohesion continue to shape certain behavior patterns (e.g., competitive business practices) that appear on the surface to reflect individualistic concerns.

CONCEPTUALIZING DEPENDENCY

TABLE 1.1 Dependency and Gender Role DependencyDependencymasculinity correlation femininity correlation Dependency measure

Study Anderson (1986) Bornstein, Bowers, &Bonner(1996b) Chevron, Quinlan, & Blatt(1978) Golding & Singer (1983) Klonsky, Jane, Turkheimer, & Oltmanns (2002) Sanfilipo(1994) Watson, Biderman, &Boyd(1989) Welkowitz, Lish, & Bond (1985) Zuroff, Moskowitz, Wielgus, Powers, &Franko(1983)

Women

Men

Women

Men .52*

PZ

-.34*

-.02

IDI

-.64"

-.55**

.30*

.37*

DEQ

-.48"

-.36*

.07

.42*

DEQ

-.38*

-.36*

.52*

.06

PIPD

-.08 -.36*

.07

-.28*

.14" .44**

.40*

-.29**

-.29**

.09

.09

DEQ

-.11

-.22

.20

.43*

DEQ

-.48**

-.27*

.11

.27*

DEQ

PGDS

-.01

.07

Note. PZ = Dependency subscale of the Pensacola Z Scale (Jones, 1957); IDI = Interpersonal Dependency Inventory (Hirschfeld et al., 1977); DEQ = Depressive Experiences Questionnaire (Blatt, D'Afflitti, & Quinlan, 1976); PIPD = Peer Inventory for Personality Disorders (Klonsky et al., 2002); PGDS = Peer Group Dependency Scale (Lapan & Patton, 1986). All studies except Chevron et al. (1978), Klonsky et al. (2002), and Sanfilipo (1994) used the Bern Sex Role Inventory (Bern, 1974) to assess gender role.

*p < .05. "p < .005.

A very different pattern emerges when projective measures of dependency are used. Table 1.2 summarizes the results of a large-scale metaanalysis of gender differences in dependency based on 97 published studies (Bornstein, 1995c). As the top portion of Table 1.2 shows, women scored higher than men on every questionnaire measure of dependency (and all but one of these gender difference effect sizes was statistically significant). As the bottom portion of Table 1.2 shows, however, men scored higher than women on every projective dependency test. Although these projective-test gender differences were not statistically significant for most individual measures, when the results were pooled using meta-analytic techniques, men showed a small—but statistically reliable—elevation in projective dependency test scores (d = .11, Combined Z = 1.93, p < .05).5

5 A d of. 11 suggests that, on average, men obtain projective dependency scores that are . 11 standard deviations higher than those of women. While a d of. 11 is considered a modest effect size, it is substantially larger than many well-established, widely accepted effect sizes in psychology and medicine (see Meyer, Pilkonis, Proietti, Heape, & Egan, 2001, for a summary of these effect sizes).

10

CONCEPTUAL AND EMPIRICAL FOUNDATIONS

TABLE 1.2 Gender Differences in Dependency: Self-Report Versus Projective Measures Measure

Number of effect sizes

DEQ IDI Dy Scale MCMI DP Scale EPPS SAS LKDOS PDQ-R Other

18 16 12 8 7 5 4 3 2 20

ROD TAT HIT Other

17 3 2 4

Combined effect size (d)

Combined Z

P

8.13 6.20 10.91 5.28 5.03 3.98 4.42 3.38 0.90 7.17

fflCQ)CD-?:S

"2 g - o « l S 8 ^ ^ f ^ g S g f f i ^ c

II S g-

S

5

?S

">co.&

S§-^

S -Q. c

>°£

I =

.§

-°

i

f1 CO "5

D)° CD X

!E w" — C/5 ^

fI

s| c: T3

1 8O -c^ E§C D|-^S ; .

M-^-o

s a. ;S.-D »- i§=-o •55 r c D•§ . ± a. : S® - QS ) ^x0 -S - s ^3T•* OC ' - £Sc o I Q-«og§°-^c?>-i°-^ogcoS S"a5c^-^ajo(DcOQ.cDEgg75 o S = £ g-.c o c > - ^ - o ( D S c o S

< CD

0)

*. E

CO

£ CD

CO O)

CD JS

LU

D-

I-

c

E 8 °