AN EMPIRICAL AND EXISTENTIAL EXAMINATION OF NARCISSISTIC FUNCTIONING Except where reference is made to the work of others, the work described in this thesis is my own or was done in collaboration with my advisory committee. This dissertation does not include proprietary or classified information. ________________________________________ Scott Allen Peebles Certificate of Approval: _____________________________ _____________________________ Anthony J. Guarino John C. Dagley, Chair Associate Professor Associate Professor Educational Foundations, Counselor Education, Counseling Leadership, and Technology Psychology, and School Psychology _____________________________ _____________________________ Holly A. Stadler Stephen L. McFarland Professor Acting Dean Counselor Education, Counseling Graduate School Psychology, and School Psychology AN EMPIRICAL AND EXISTENTIAL EXAMINATION OF NARCISSISTIC FUNCTIONING Scott Allen Peebles A Dissertation Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy Auburn, Alabama August 7, 2006 iii AN EMPIRICAL AND EXISTENTIAL EXAMINATION OF NARCISSISTIC FUNCTIONING Scott Allen Peebles Permission is granted to Auburn University to make copies of this dissertation at its discretion, upon request of individuals or institutions and at their expense. The author reserves all publication rights. ______________________________ Signature of Author ______________________________ Date of Graduation iv VITA Scott Allen Peebles, son of Glen Charles and Paula (Small) Peebles, was born June 20, 1977, in Warner Robins, Georgia. He graduated from Orangeburg- Wilkinson High School in 1995. He was accepted into the Bell Honors Program at Georgia Southern University in the Fall of 1995, and he graduated magna cum laude with a Bachelor of Science degree in Psychology in May, 1999. He returned to Georgia Southern University in the Fall of 1999 to pursue a Master of Science degree in Clinical Psychology. He completed requirements for this degree in May, 2001. After working as a Mental Health Professional at Ogeechee Behavioral Health Services for one year, he entered Graduate School, Auburn University, in August, 2002. He married Dawn Savannah Simpkins, daughter of Glenn Simpkins and Janet (Kennedy) Davis, on May 17, 2003. He and his wife are expecting their second child (due in October 2006) after losing their first, Kylie Brooke Peebles, to anencephaly on October 28, 2003. v DISSERTATION ABSTRACT AN EMPIRICAL AND EXISTENTIAL EXAMINATION OF NARCISSISTIC FUNCTIONING Scott Allen Peebles Doctor of Philosophy, August 7, 2006 (M.S., Georgia Southern University, 2001) (B.S., Georgia Southern University, 1999) 137 Typed Pages Directed by John C. Dagley The Narcissistic Personality Inventory (NPI) and its factors have been shown to simultaneously predict both psychological adjustment and lack thereof. Results from a host of empirical studies suggest that although one particular NPI factor predicts poorer adjustment and psychological functioning on a consistent basis, the other three factors predict varying levels of psychological health and adjustment. The present study sought to examine the predictive ability of the NPI and its factors in relation to existentially- oriented measures of psychological functioning. One hundred and ninety individuals vi completed the NPI and several existential construct measures (e.g., purpose in life, self- actualization, and death anxiety). The regression analyses demonstrate that the NPI and its factors indeed differentially predict psychological health and adjustment on these existential measures. These findings support an existential conceptualization of narcissism that focuses on the relative adaptivity of a narcissistic character structure and a dimensional view of personality functioning. This conceptualization is consistent with traditional counseling psychology values and beliefs. vii ACKNOWLEDGMENTS The author would like to thank Dr. John Dagley for his support, encouragement, and patience throughout this dissertation process. The author also expresses his deep appreciation to those researchers and clinicians who assisted him in tracking down some of the existential scales used in this study. Also, it is appropriate and fitting that the author thank his wife, Dawn Savannah Peebles, for her support and love during the course of this investigation. viii This dissertation study was prepared in accordance with the Publication Manual of the American Psychological Association, fifth edition (2001). This dissertation manuscript was prepared using Microsoft Word, 2003 Edition. The data analysis for this dissertation study was conducted through use of SPSS 11.0.0 for Windows, Standard Version. ix TABLE OF CONTENTS LIST OF TABLES xi CHAPTER 1 ? INTRODUCTION Purpose of the Study 1 Introduction and Brief Review of Relevant Literature 2 Statement of the Problem 6 A Brief Summary of Existential Ideas on Narcissism 7 Significance of the Study: A Counseling Psychology Perspective 10 Brief Overview of the Present Study, Research Questions and 12 Derived Hypotheses CHAPTER 2 ? LITERATURE REVIEW Introduction 15 Early Definitions of the Narcissism Construct 15 Empirical Examinations of Narcissism 17 The Narcissistic Personality Inventory (NPI) 19 Factor Analysis of the NPI 24 Initial Applications of the NPI Factor Models 26 Further Applications and Extensions of the NPI Factor Models 32 Summary of NPI-driven Conceptualizations of Narcissism 45 Existential Conceptualization of Narcissism 46 Key Existential Constructs: Relationships to Narcissistic Functioning 52 Synthesis and Summary of Existential Conceptualization of Narcissism 56 Conceptual Measures of Existential Constructs 57 CHAPTER 3 ? METHOD Participants 61 Measures 62 Procedure 67 Hypotheses Examined by this Study 67 CHAPTER 4 ? RESULTS Overview 69 Hypothesis 1: Total NPI as Predictor of Existential Functioning 70 Hypothesis 2: The Exploitativeness/Entitlement Subscale and its 72 Predictive Abilities x Hypothesis 3: The Leadership/Authority, Superiority/Arrogance, and 75 Self-Absorption/Self-Admiration Subscales and their Predictive Abilities Reliability and Validity of Measures Used in this Study 78 Preliminary Analyses of Demographic Variables in Relation to the 87 Measures Employed CHAPTER 5 ? DISCUSSION Review of the Empirical Findings 92 Meaning and Interpretation of Results 95 Implications for Research 106 Review of the Goals of this Dissertation Study: Implications for Theory 107 Treatment, and the Field of Counseling Psychology Limitations of this Dissertation Study 111 REFERENCES 113 APPENDIX A ? DEMOGRAPHIC SHEET 126 xi LIST OF TABLES TABLE 1: Total NPI Correlation Analysis with all Existential Measures 71 TABLE 2: Zero-order and Partial Correlations of Existential Constructs 74 with Exploitativeness/Entitlement (EE), Leadership/Authority (LA), Superiority/Arrogance (SA) AND Self-Absorption/Self-Admiration (SS) NPI Factors TABLE 3: Mean Scores, Scale Range, and Standard Deviations for Narcissistic 79 Personality Inventory (Including Subscales) and Existential Scales TABLE 4: Intercorrelations of NPI Total and Subscale Scores 83 TABLE 5: Zero-order Correlations of Existential Content Scales 85 TABLE 6: Correlations of Age with Scores from the Narcissistic Personality 89 Inventory, NPI Subscales, and all Existential Scales 1 CHAPTER 1 INTRODUCTION Purpose of the Study This dissertation study sought to (1) reinterpret the construct called narcissism (and functioning that results from a narcissistic way of being) from an existential perspective, (2) examine the construct of narcissism in relation to key existential constructs (such as self-actualization, purpose/meaning in life, the search for meaning in life, authenticity, and death anxiety), and (3) reaffirm that narcissism is indeed a complex and multidimensional construct that relates to a spectrum of functioning through use of a number of existential construct measures. Previous studies of narcissism (which will be discussed within) demonstrated that narcissism is indeed a multifaceted and multidimensional construct. Researchers conducting these studies generally interpreted their results through the lenses of Kohutian self-psychology, cognitive-behavioral theory, or through social psychological theory (e.g., attribution theory). However, no study examined the narcissism construct from an existential stance. At least one group of authors (Watson, Little, Sawrie, & Biderman, 1992) suggested that the narcissism construct be studied through more diverse theoretical lenses to promote further clarity in defining this important psychological construct. Toward this end, an existential and empirical examination of narcissism was conducted by choosing valid and reliable measures that assessed narcissism and key existential constructs, using the 2 narcissism measure (and its subscale measures) to predict scores on the existential measures, and drawing conclusions (consistent with existential theory) based on the findings. The following portions of this first chapter will provide, in turn, an introduction and brief review of the relevant literature that provided the rationale for this dissertation study. The review will include a discussion of the narcissism construct and its conceptual basis, information regarding advances in the empirical study of the narcissism construct, a brief conceptualization of narcissism from an existential theoretical frame, an explanation of the problem to be studied within this dissertation, and a section regarding the significance of this study to counseling psychology as a discipline. Introduction and Brief Review of Relevant Literature The personality construct known as narcissism has been considered and defined time and again by philosophers, clinicians, researchers, and lay persons. Early philosophical ideas regarding narcissism and those who demonstrate narcissistic characteristics informed modern conceptualizations of the construct. Traditionally, narcissism has carried a strongly negative connotation in both clinical and everyday parlance. Early clinical conceptualizations of narcissism focused on these assumed negative aspects (e.g., a lack of empathy for others, a sense of entitlement). Few of these early ideas garnered significant research support, as many of the conceptualizations were derived from psychoanalytic and psychodynamic theoretical bases (Freud, 1914/1957; Kernberg, 1985; Kohut, 1971; Mahler, 1972). Empirical research was not traditionally a focus of such perspectives at the time many of these ideas were developed. However, 3 these ideas became entrenched as definitive of narcissism, and the conceptualizations spawned from them dominated and guided clinical interventions for a number of years. Clinical interest in the construct of narcissism remained strong for many decades after Sigmund Freud?s (1914/1957) first musings on the subject. However, as Raskin and Terry (1988) noted, empirical study of the narcissism construct lagged far behind clinical observations and discussions regarding treatment. The first solid empirical studies of narcissism were fostered by the inclusion of the Narcissistic Personality Disorder in the American Psychiatric Association?s Diagnostic and Statistical Manual of Mental Disorders (DSM ?III; American Psychiatric Association, 1980). Since the inclusion of the narcissism construct in the ubiquitous DSM categorical scheme, clinical and research scales alike that measure narcissism improved markedly. Early studies of narcissism using these clinical and research scales appeared to indicate that as the number of narcissistic characteristics an individual possessed increased (i.e., the greater the narcissism score), the more likely the individual experienced difficulty with general psychological functioning (Ashby, Lee, & Duke, 1979; Millon, 1994, 1997; Richman & Flaherty, 1990; Soyer, Rovenpor, Kopelman, Mullins, & Watson, 2001; Watson & Biderman, 1993). These initial findings seemed to provide strong support for traditional conceptualizations of narcissism, in that they suggested that the effect of narcissistic characteristics on one?s psychological health is uniformly negative. Based on review of these studies alone, one might argue that nothing adaptive or positive comes from display of narcissistic characteristics. 4 However, findings using one particular narcissism measure began to suggest otherwise. The original Narcissistic Personality Inventory (NPI; Raskin & Hall, 1979) was developed via a direct review of the DSM-III (DSM ?III; American Psychiatric Association, 1980) criteria for the Narcissistic Personality Disorder, as were many of the other narcissism measures. Interestingly, findings from initial studies using the NPI showed that narcissism scores correlated positively with psychological characteristics generally associated with psychological health and adjustment (such as self-esteem, creativity, and extroversion), while also correlating positively with other characteristics usually associated with psychological dysfunction (such as excessive sensation-seeking and psychoticism) (Emmons, 1981; Raskin, 1980; Raskin & Hall, 1981). These findings were difficult to explain initially, as most researchers (and clinicians alike) still considered narcissism unhealthy and damaging to an individual?s level of functioning. Given these contradictory findings using the original NPI, researchers began to question whether narcissism was truly a unidimensional construct. The availability of factor analytic statistical procedures provided these researchers with an opportunity to test this hypothesis. Three factor analyses were conducted on the NPI (two by the same researcher, both providing the same factor structure). Two of the analyses yielded a 54- item scale comprised of four subscales (Exploitativeness/Entitlement, Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration) (Emmons, 1984, 1987); the other (Raskin & Terry, 1988) resulted in a 40-item scale made up of seven subscales (Authority, Exhibitionism, Superiority, Entitlement, Exploitativeness, Self-Sufficiency, and Vanity). 5 Both NPI versions have been employed in subsequent research studies, while providing similar patterns of results. While total NPI scores using both factor-analyzed versions continued to correlate with both adaptive and maladaptive psychological characteristics, certain subscale scores (for the Emmons, 1984, 1987 version: Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration; for the Raskin & Terry, 1988 version: Authority, Superiority, and Self-Sufficiency) consistently correlated with measures of psychological well-being. On the other hand, particular NPI subscale scores (Emmons, 1984, 1987: Exploitativeness/Entitlement; for Raskin & Terry, 1988: Exhibitionism, Entitlement, and Exploitativeness) consistently correlated with measures of psychological distress (or impaired psychological functioning). Furthermore, the use of partial regression procedures has served to clarify these findings to an even greater degree. Removing the variance associated with the seemingly more adaptive aspects of narcissism provides even more convincing evidence that factors such as Exploitativeness/Entitlement do not bode well for one?s psychological health, while partialing out the effects of the more maladaptive aspects suggests even more strongly that factors such as Leadership/Authority are healthy and adaptive (e.g., Raskin, Novacek, & Hogan, 1991;Watson, Little, Sawrie, & Biderman, 1992; Watson & Morris, 1991). These findings suggest that narcissism facets/traits are related in complex and rich ways. One might view narcissism as one views cholesterol levels in the body; levels of ?good? narcissism may serve as a protective factor and promote psychological health, 6 while levels of ?bad? narcissism will override these protective factors and lead to unhealthy self functioning. Statement of the Problem Many studies have served to elucidate the complexities of the narcissism construct. However, questions still remain unanswered. Authors of one study (Watson, Little, Sawrie, & Biderman, 1992, p.447) suggested that additional research focus on use of ?diverse theoretical perspectives? when considering the narcissism construct. Many of the aforementioned studies were conducted to support a continuum hypothesis (Watson, Hickman, & Morris, 1996) of narcissistic functioning, which was derived primarily from study of a Kohutian self-psychological perspective. This continuum hypothesis has gained much research support, as the previously discussed studies attest. Narcissism is better conceptualized as a complex, multifaceted, and rich construct. It is also better considered as influencing a spectrum of functioning ranging from solidly healthy to trenchantly unhealthy. This dimensional perspective on the narcissism construct is crucial to the present narcissism study as well. In addition to the Kohutian self-psychology approach, narcissism findings have been examined and explained from a cognitive-behavioral stance and from knowledge of social psychological theory. At least one group of authors (Watson, Little, Sawrie, & Biderman, 1992) suggests that the narcissism construct be examined through more diverse theoretical lenses to promote further clarity in defining this construct. To this end, the present author has chosen to conceptualize and examine narcissism from an existential/humanistic perspective. 7 This particular theoretical perspective was chosen for several reasons, including (1) that the researcher prefers this approach as a clinical intervention conceptual framework, (2) existential/humanistic conceptual frameworks were absent from empirical studies of narcissism, (3) employment of an existential theoretical frame (and use of measures of existential constructs) would help to further clarify the narcissism construct, (4) existential ideas are consistent with the assertion that narcissism (as well as other personality constructs) must be viewed dimensionally (and not categorically), and (5) findings from this study would help to depathologize individuals diagnosed with personality ?disorders? and reframe their struggles in a manner more consistent with traditional counseling psychology values (e.g., focusing on psychological health, resilience, and emphasizing personal strengths and abilities). A Brief Summary of Existential Ideas on Narcissism Existential perspectives have been applied to the study of narcissism. However, existential practitioners have rarely chosen to test their conceptualizations empirically. Though some might argue that existential constructs are not amenable to empirical study, many measures of such constructs have been developed and evaluated. Examples include the Purpose-in-Life Test (PIL; Crumbaugh & Maholick, 1969), the Seeking of Noetic Goals Test (SONG, Crumbaugh, 1977a), the Death Anxiety Scale (DAS; Templer, 1970), the Authenticity Inventory (AI:3; Goldman & Kernis, 2004), and the Short Index of Self- Actualization (SISA; Jones & Crandall, 1986). Furthermore, the psychometric properties of these measures, for the most part, are considerably sound. However, prior to 8 discussing the empirical existential facets of the study, a brief description of existential theoretical ideas upon which these instruments are based is warranted. Frankl (1984) and Yalom (1980) have argued that humans are meaning-seeking (and meaning-making) creatures by nature. These activities are considered primary life tasks by existentialist theorists. Should individuals fail to continue to create this purpose and meaning in life, then they are apt to experience an existential vacuum state characterized by boredom and a deep sense of emptiness (Frankl, 1984). Frankl (1984) then argued that individuals will, if subjected to this existential vacuum state for a lengthy period of time, react to it with existential frustration (which will then engender a noogenic neurosis, or serious psychological decompensation). Yalom (1980) argued that a sense of meaning and purpose must emerge from our honest and genuine engagement in interactions with our surrounding environment. Following this premise, Shaw (2000) argued that narcissism is a motivational structure created (at both a conscious and unconscious level) to resolve the very human problem of personal insignificance. Shaw posited that narcissism is a personal myth system created to shield individuals from the painful awareness of their own mortality and personal limitations. He also asserted that this creation of a personal narcissistic myth is a common, normal human reaction to their existential condition. Therefore, it is logical to assert that those who adopt a narcissistic motivational structure to deal with the harsh realities of life fall along a continuum of healthy and unhealthy functioning. Some individuals find a way to make this myth work for them in a meaning-creating way; 9 others will find this motivational structure lacking and may be prone to a noogenic neurosis. Taking this argument of narcissism as a motivational psychological structure one step further, van Deurzen-Smith (2000, p. 232) argues that ?as organisms, we are primarily self-centered, and it is in our nature and best interest to orientate ourselves towards the world with the assumption of our right and entitlement.? In line with Dawkins? (1976) arguments, van Deurzen-Smith (2000) asserts that even our genetic material is selfish, in that genes must express themselves selfishly (e.g., at the expense of other genes) to insure their survival. Hence, it seems that van Deurzen-Smith (2000) is arguing that narcissism is ingrained in our biology, psychology, sociology, culture, and spiritual life. However, she also points out that in taking care of our own individual needs, we also must learn to balance these needs with those of the community around us (van Deurzen-Smith, 2000). In this view, some narcissistic meaning structures are adaptive, while others are clearly not. Hence, an existential theoretical perspective fits well with extant NPI data collected to date. From an existential theoretical viewpoint, adoption of a narcissistic motivational structure certainly may lead to a range (or continuum) of psychological functioning, depending on the form of the motivational structure chosen. Should one choose a motivational structure based on the more adaptive narcissistic characteristics (e.g., Leadership/Authority), while basing the structure less on the not-so-adaptive narcissistic characteristics (e.g., Exploitativeness/Entitlement), then one is likely to reap the benefits of greater self-esteem, personal growth, and openness toward change. 10 However, should an individual choose the opposing narcissistic structure, one is likely to suffer difficulties in maintaining a sense of purpose and meaning in life. Significance of the Study: A Counseling Psychology Perspective As discussed previously, conceptualizations of personality constructs such as narcissism have been plagued by biases in interpretation and worldview. Though the advent of the DSM-III and its categorical system of personality disorder classification has led to an explosion of personality research, this research has great potential to mislead clinicians who are asked to diagnose and treat those with personality disorders. Though the most recent version of the manual (DSM-IV-TR; APA, 2000) has become the gold standard of categorical diagnostic classification and conceptualization, many researchers and clinicians have begun to question the integrity of the DSM classification system (especially in reference to the personality disorders). Widiger and Frances (2002) noted that the DSM system forces clinicians to spend an inordinate amount of time performing assessments to clarify personality disorder diagnoses. Categorical approaches also have a difficult time explaining away the frequent comorbidity of personality disorders (Widiger & Frances, 2002). Critics have overlooked another major drawback of a categorical system of classification. Such systems force clinicians who diagnose and treat clients to focus on the negative and maladaptive aspects of a particular personality configuration. Reliance on the DSM-IV-TR (APA, 2000) criteria for conceptualization of narcissistic personality configurations (as well as others) runs the risk of pathologizing a style of operating in- the-world that might be highly adaptive for the individual client. Treatment plans based 11 on such a categorical approach would ignore reliance on (and enhancement of) client strengths that could be useful in treatment. Use of a richer, more complex, dimensional approach allows the clinician to view certain aspects of a narcissistic motivational structure as adaptive and useful to the individual. Counseling psychology, as a field, has attempted to balance concerns regarding pathology and mental dysfunction with attempts to promote wellness, personal growth, and preventative interventions (Hoffman & Driscoll, 2000). Counseling psychologists are expected to be proficient in assisting individuals and groups that fall along a continuous dimension of psychological health and wellness (Hoffman & Driscoll, 2000). Even when working with individuals that a medical or categorical model of treatment may label ?pathological,? counseling psychologists endeavor to consider such an individual?s strengths and abilities in their treatment plans and goals. In fact, many in the field have suggested that the promotion of wellness and psychological health is an emerging role for counselors and counseling psychologists (Gladding, 2004; Myers, Sweeney, & Witmer, 2001; Witmer & Sweeney, 1999). This focus on wellness and health, according to these authors, should apply to all individuals with whom we work (Myers, Sweeney, & Witmer, 2000). Hence, the present study sought to support a view of narcissism (and psychological functioning in general) based on traditional counseling psychology values and beliefs. Instead of spending inordinate amounts of time clarifying personality disorder diagnoses (such that these problems can be treated and/or fixed), this study promotes greater clinician focus on assessing the adaptive strengths that all clients (no 12 matter how pathological) possess. This study advocates for focus on using the adaptive facets of narcissism as strengths upon which to build in therapy/counseling, rather than maintaining sole focus on the less adaptive facets of such a motivational structure. Such a focus helps to reduce the stigma associated with a personality disorder diagnosis and shifts the focus to a consideration of how best to help the individual client achieve wellness and personal growth. Brief Overview of the Present Study, Research Questions, and Derived Hypotheses The present study sought to examine narcissism from an existential viewpoint. Previous studies conducted by various researchers provided evidence that (1) narcissism is a multidimensional, rich, and complex personality construct, (2) that measures of narcissism (and, more importantly, its components or facets) predicted a range of psychological functioning (healthy to unhealthy), (3) that certain components of narcissism consistently predicted varying degrees of psychological health (such as Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration), and (4) that particular components of narcissism consistently predicted psychological dysfunction (e.g., Exploitativeness/Entitlement). However, none of these studies were conducted from an existential theoretical viewpoint. Theoretically speaking, existential conceptualizations of the narcissism construct are basically consistent with narcissism research findings (as discussed previously in this chapter). However, these theoretical ideas have not been tested empirically to date, despite the existence of several solid existential construct measures. The present study 13 sought to clarify existential thinking in regard to narcissism by addressing the following broad research questions: 1. Would overall (or total) narcissism scores predict scores on existential construct measures designed to assess psychological health/functioning and psychological distress/dysfunction in a similar manner as they do with measures of other psychological constructs? 2. Would components of narcissism (such as Leadership/Authority and Exploitativeness/Entitlement) differentially predict psychological health or dysfunction when assessed from an existential view (as they do with measures of psychological health/dysfunction developed from other theoretical viewpoints)? Toward this end, the present study sought to examine the predictive ability of the Narcissistic Personality Inventory (through use of zero-order regression and partial regression analyses) as related to measures of traditionally existential constructs thought to signify psychological health or lack thereof. This study also examined the predictive ability of the NPI subscale scores in relation to the existential constructs. The existential measures chosen included the Purpose-in-Life Test (PIL; Crumbaugh & Maholick, 1969), the Seeking of Noetic Goals Test (SONG, Crumbaugh, 1977a), the Death Anxiety Scale (DAS; Templer, 1970), the Authenticity Inventory (AI:3; Goldman & Kernis, 2004), and the Short Index of Self-Actualization (SISA; Jones & Crandall, 1986). The Rosenberg Self-Esteem Scale (RSE; Rosenberg, 1965) was also employed (as has been common practice in NPI research, as NPI scores display complex relationships with RSE scores). 14 The following specific research hypotheses were developed through consideration of existential theory and examination of NPI literature. 1. The total NPI score will predict greater self-esteem, a greater sense of purpose in life, greater capacity for self-actualization, greater authenticity, and reduced death anxiety. No prediction was made regarding the SONG score. 2. The Exploitativeness/Entitlement factor score will predict a lack of purpose in life, avoidance of searching for a sense of purpose, greater death anxiety, lowered capacity for self-actualization, less authenticity, and lowered self-esteem. This prediction will be augmented by statistically partialing out the effects of the proposed healthier facets of narcissism (Leadership/Authority, Superiority/Arrogance, and Self- Absorption/Self-Admiration). 3. The Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self- Admiration NPI subscale scores (to varying degrees) will predict a greater sense of purpose in life, less death anxiety, greater capacity for self-actualization, greater authenticity, and greater self-esteem. These predictions should strengthen when the effects of Exploitativeness/Entitlement are partialled out of the predictive equation. No prediction of the SONG score (indicative of a search for meaning and purpose) was made, as one may or may not continue searching if one already has a sense of meaning and purpose in life. 15 CHAPTER 2 LITERATURE REVIEW Introduction The previous chapter presented the theoretical rationale for engaging in this dissertation study by providing a cursory review of the related psychological literature, summarizing and synthesizing this information, providing a case for the necessity of this study, and by specifying the research questions and empirical hypotheses examined by the study. This second chapter will provide a more in-depth review of the literature related to (1) the history of the narcissism construct, (2) empirical study of narcissism, (3) specific findings from studies employing the Narcissistic Personality Inventory, (4) examination of existential theoretical ideas related to narcissism, (5) discussion of key existential theoretical constructs, and (6) a discussion of research employing various empirically-validated existential measures. Early Definitions of the Narcissism Construct The concept referred to as narcissism has been considered and defined time and again by philosophers, clinicians, researchers, and lay persons. The origin of this concept may be traced to Greek mythology (Lowen, 1985). The myth of Narcissus is the tale of a young man who was made to fall in love with his own image by the Greek gods. After seeing his reflection in a fountain of water, Narcissus was unable to move from that spot 16 and soon perished. He was a person unable to love others, while in essence losing the ability to see himself clearly and objectively. These early philosophical ideas regarding narcissism and those who demonstrate narcissistic traits informed modern conceptualizations of the construct. Traditionally, narcissism has carried a strongly negative connotation in both clinical and everyday parlance. Early clinical conceptualizations of narcissism focused on these assumed negative aspects. Few of these early ideas garnered significant research support, as many of the conceptualizations were derived from psychoanalytic and psychodynamic theoretical bases (and empirical research was not traditionally a focus of such perspectives at the time). However, these ideas became entrenched as definitive of narcissism, and the ideas dominated clinical conceptualization and guided clinical interventions for a number of years. Psychoanalytic and psychodynamic clinicians have traditionally viewed narcissism as a psychological disorder to be cured (Kernberg, 1985; Lowen, 1985). Narcissism was hardly viewed as adaptive in any form or fashion. The following statement is characteristic of many psychoanalytic views proffered by these eminent therapists: Are we all narcissistic then? Does it mean that narcissism is a normal aspect of the human personality? No. In my opinion, narcissism is a pathological condition. I draw a distinction between a healthy concern for one?s appearance, based on a sense of self, and the displacement of identity from the self to the image, which is characteristic of the narcissistic state (Lowen, 1985, p. 26). 17 Therefore, narcissism was viewed as a strictly pathological condition that is easily distinguished from healthy functioning. Like Narcissus, all individuals with narcissistic characteristics were assumed to have fallen in love with a grandiose self-image. This image was assumed to have been created by the individual to compensate for early parenting failures (Kernberg, 1985; Lowen, 1985). Empirical Examinations of Narcissism As noted in the previous dissertation chapter, such psychoanalytic and psychodynamic ideas regarding narcissism held sway for many years, despite a lack of empirical evidence supporting their validity. Raskin and Terry (1988) noted that empirical study of narcissism lagged far behind clinical observation until the inclusion of the Narcissistic Personality Disorder diagnosis in the Diagnostic and Statistical Manual of Mental Disorders, third edition (DSM ? III; American Psychiatric Association, 1980). The recognition of extreme narcissism as a personality disorder sparked clinical research interest, leading to the development and validation of various scales intended to measure this personality construct. Some of these scales were developed to measure the personality construct itself, while others were developed to measure narcissism as a form of personality pathology. The DSM-III (American Psychiatric Association, 1980) described the narcissistic personality as being characterized by a grandiose sense of self-importance or uniqueness; a preoccupation with fantasies of unlimited success, power, brilliance, beauty, or ideal love; exhibitionism; an inability to tolerate criticism, indifference from others, or defeat; entitlement or the expectation of special favors without assuming reciprocal 18 responsibilities; interpersonal exploitativeness, relationships that alternate between extremes of overidealization and devaluation; and a lack of empathy. This description has not changed markedly in more recent versions of the DSM nosological system. While the DSM-III description was intended as a set of diagnostic criteria for practitioners in clinical settings, this same description has also been used as the basis for development of important scales for measuring the narcissism construct in individuals who have not evidenced pathological levels of narcissism. Of these instruments, many appear to describe narcissism as a uniformly negative indicator for mental health. Scale scores derived from such instruments as the Narcissistic Personality Disorder Scale (NPDS; Ashby, Lee, & Duke, 1979), the Millon Clinical Multiaxial Inventory III - Narcissism Scale (MCMI-III; Millon, 1994, 1997), and the Narcissistic Traits Scale (NTS; Richman & Flaherty, 1990) consistently correlate with indicators of unhealthy psychological and social functioning (Millon, 1994, 1997; Soyer, et al., 2001; Watson & Biderman, 1993). Findings from studies using these scales continue to build the case that narcissism is a personality disorder characterized by maladaptive functioning in a broad sense (e.g., in many areas of an individual?s life). Many of these narcissism scales were developed based on DSM-III criteria (while others were simply influenced by those criteria). If one looked only at these studies, one might be easily convinced that there is nothing adaptive at all about narcissism or its component traits. However, another narcissism scale developed directly from DSM-III diagnostic criteria consistently yielded different results. The Narcissistic Personality Inventory (NPI; 19 Raskin & Hall, 1979; Raskin & Terry, 1988), though based on criteria for an assumed personality disorder, appeared to measure both adaptive and maladaptive aspects of the narcissistic construct. In early work using the NPI, narcissism was considered a unidimensional construct (and was compared to other psychological characteristics as such), as opposed to a complex and multidimensional one. Early findings, though enlightening and thought-provoking, were also confusing. Total NPI scores showed unexpected patterns of correlation with indicators of healthy and unhealthy psychological functioning. The Narcissistic Personality Inventory (NPI) Raskin and Hall (1979) developed the original version of the Narcissistic Personality Inventory (NPI). The NPI is a self-report instrument designed to measure the personality construct of narcissism (Raskin & Hall, 1979). Though the items were developed by examining the DSM-III criteria for Narcissistic Personality Disorder (NPD) (and the authors state that the stimulus for the development of the NPI was the inclusion of the NPD in the DSM-III), the NPI was not intended as a measure of a personality disorder per se (Raskin & Hall, 1979). The authors note that the inventory should be viewed as ?a measure of the degree to which individuals differ in the trait we have labeled ?narcissism?? (Raskin & Hall, 1979, p. 590). The original scale contained 223 items. All items were comprised of a pair of statements (participants chose A or B), one signifying a narcissistic response and the other a non-narcissistic one. Raskin and Hall (1979) then trimmed the scale down to two separate forms (with 40 items per scale) based on findings from relevant statistical procedures. 20 After development of the original version of the NPI, Raskin (1980) and others sought to explore the relationships between narcissism and other psychological characteristics (while also supporting the validity of the NPI as a narcissism measure). Raskin undertook a study to examine the relationship between NPI scores and creativity, for example. He reasoned that many of the DSM-III descriptors for the Narcissistic Personality Disorder were similar to descriptions of the personalities of highly creative people (e.g, creative people have been described in the literature as self-absorbed, self- oriented, have a rich inner life, are impulsive, are autonomous, are self-assertive, are dominant, can be exploitative, etc.). Raskin administered the original, 40 item NPI and Part II of the Barron Symbolic Equivalents Test (Barron, 1974) to an undergraduate student sample. Participants were also asked to self-evaluate their own level of creativity. Students were then grouped for analysis according to their self-reported level of creativity and by results of their Barron tests (Raskin, 1980). Raskin?s (1980) findings were generally supportive of his hypotheses. Significant differences in NPI scores were found according to creativity group. Specifically, the high creativity/high self-report group displayed significant differences in NPI scores with all three other groups (the high creativity/low self-report, the low creativity, high self-report, and the low creativity/low self-report groups). As expected, the high creativity/high self- report group had the highest average NPI scores (mean of 32.8), while the low creativity/low self-report group displayed the lowest (mean = 22.5) (Raskin, 1980). Scores on the Barron test and self-reported creativity were positively correlated with NPI scores, and both scores predicted higher NPI scores when taken together. 21 Hence, narcissism had been systematically correlated with another significant psychological characteristic (and, oddly, one that is not generally considered a maladaptive one). Raskin (1980) suggested that researchers continue to search for evidence that narcissism is ?a core trait of personality? by demonstrating that narcissism shares commonalities with many other behavioral domains (p. 59). His hope was that such suggested studies would eventually lead to an empirically-based theory of narcissism (as opposed to clinical theorizing unsupported by research). Indeed, further work would be done using the NPI in an attempt to substantiate conceptualization of narcissism as a personality construct. Raskin and Hall (1981) correlated NPI scores with scores from the Eysenck Personality Inventory (EPQ) (Eysenck & Eysenck, 1968). In particular, Raskin and Hall (1981) were interested in EPQ Extroversion, Psychoticism, and Lie scale scores. Working from the ?assumption that abnormality is continuous with normality,? Raskin and Hall (1981, p. 159) assessed behavioral and psychological correlates of narcissism as ?manifested to a lesser extent in normal individuals? (p. 159). They made the following predictions: (1) NPI scores would correlate moderately (.20 to .40) with EPQ Extroversion and Psychoticism scores, (2) EPQ Extroversion and Psychoticism scores taken together would predict higher NPI scores than either EPQ scale alone, and (3) no prediction was made in relation to the EPQ Lie scale. The participants were undergraduate students. Results of the study were predominately as expected, supporting the assertion that narcissism is a valid and important personality characteristic. EPQ Extroversion and Psychoticism scores were 22 both significantly and positively correlated with NPI scores (both at .23, in the .20-.40 range predicted). The correlation between NPI scores and Neuroticism scores was not significant. Also, as expected, Extroversion and Psychoticism scores taken together better predicted NPI scores than when examined alone (R=.30, p<.05). Interestingly, NPI scores were found to correlate negatively with EPQ Lie scale scores (Raskin & Hall, 1981). Thus far, Raskin and Hall?s findings support the assertion that narcissism is related in an important way to other significant personality constructs. Note that some of the characteristics listed for Extroversion (e.g., outgoing, taking chances, large social network) are often viewed as positive characteristics. Also, no correlation was noted between NPI scores and Neuroticism, suggesting that many who have narcissistic characteristics are likely functioning at least an average level. Thus, early evidence exists that narcissism is not a unitary and negative construct; adaptive elements exist. Other researchers correlated total NPI scores with logically related behavioral correlates or personality constructs during the early period of empirically-oriented narcissism research. Emmons (1981) examined the relationship between narcissism and sensation-seeking. Through examination of descriptions of high sensation-seekers in the literature, Emmons noted that many behaviors/personality variables correlate with those attributed to narcissistic individuals (such as need for dominance, extroversion, autonomy, and exhibitionism). Therefore, Emmons suggested that an overall positive correlation should be observed between narcissism and sensation seeking. He also suggested that narcissism scores would correlate positively with all subscales (General, Thrill and Adventure Seeking, Experience Seeking, Boredom Susceptibility, and 23 Disinhibition) of the Sensation Seeking Scale (Zuckerman, 1971). Finally, Emmons asserted that narcissism scores should correlate highest with the Disinhibition scale. Emmons (1981) administered both the Sensation Seeking Scale (Form IV) and a newly revised Narcissistic Personality Inventory with only 54 items (which would be further described in work Emmons had not yet published). Findings were reported with respect to the sex of the participants, as significant sex differences were found in the correlation matrix. For males, NPI scores were significantly correlated (positively) with Disinhibition and Boredom Susceptibility (Emmons, 1981). For females, NPI scores were correlated with General Sensation Seeking, Experience Seeking, and Disinhibition. Emmons reported anecdotally, without supporting empirical evidence, that high NPI scores tend to misidentify the purpose of the NPI. While low scorers tended to view the NPI as a measure of conceit or self-love, high scorers tended to refer to it as a measure of self-concept or self-esteem. This finding supported the idea that a spectrum or continuum of narcissistic functioning exists, as higher functioning narcissists are more likely to have insight into their personality traits. The aforementioned findings, while interesting and instructional for researchers and clinicians, are a bit confusing. Narcissistic Personality Disorder, by definition, is a maladaptive form of functioning in the world (APA; DSM-IV, 1994). One would expect that, when using an instrument that was developed out of criteria for a personality disorder (such as the NPI), the instrument would correlate only with indicators of maladaptive functioning. However, total NPI scores had been shown to correlate with such personality variables as extroversion, experience seeking, and creativity. Also, NPI 24 scores had not been shown to correlate with EPQ Neuroticism scores (Raskin & Hall, 1981). The question of how an instrument designed to assess a negative personality trait could correlate with measures of solid psychological functioning was an enigma. For those researchers who used the NPI to assess narcissism, the answers came through use of burgeoning factor analysis methods to more accurately assess the nature of higher- order personality characteristics (one of which happens to be narcissism). The NPI was subjected to two factor-analytic studies (Emmons, 1984; Raskin & Terry, 1988), which rendered two versions of the instrument. Though both have accumulated validity and reliability evidence, the Emmons (1984) version will be the primary one used in this study. Both studies will be described, and both instruments will be examined. Factor Analysis of the NPI The initial factor analysis studies performed on the NPI were Emmons? (1984). A refined 54-item NPI was administered to 451 undergraduate students, along with instruments measuring a wealth of other psychological constructs to conduct three separate studies (which will be discussed later). The first of the three studies concerned the factor analysis itself. Interitem correlations were assessed. The resultant correlation matrix was then submitted to a principle components factor analysis (Emmons, 1984). Four orthogonal factors emerged from this analysis. One factor (comprised of 11 items) seemed to suggest heavy interpersonal manipulation within the narcissism construct (such as exploitation of favors, exploitation of others, etc.). This factor was hence entitled Exploitativeness/Entitlement (EE). The second factor (9 items) seemed to assess enjoyment of leadership positions and the need to be seen as an authority, and it was 25 hence called Leadership/Authority (LA). A third factor (12 items) suggested arrogance, superiority, and grandiosity, so it was called Superiority/Arrogance (SA). The final factor (10 items) included items such as those that indicate that individuals enjoy looking at their reflections. It was entitled Self-absorption/Self-admiration (SS) (Emmons, 1984). This four factor solution again emerged in a follow-up study using a sample of 388 participants (Emmons, 1987). Raskin and Terry (1988), using the same 54-item NPI, arrived at a different factor solution (due to application of a different set of statistical procedures). The NPI was administered to 1,018 participants. Through a series of carefully applied statistical procedures, Raskin and Terry (1988) trimmed the NPI item count down to 40. This 40- item version of the NPI yielded seven factors. While some of the factors appear to overlap with those found by Emmons (1984), some factors do not. In the Raskin and Terry (1988) NPI, Exploitativeness and Entitlement are separated as factors. Authority and Superiority remained significant factors as well. However, the new factors Self- Sufficiency, Vanity, and Exhibitionism emerged (Raskin & Terry, 1988). Though the two NPI versions may seem to cloud the emergent picture of this narcissism construct, the two versions are not as divergent as they might appear at first blush. Both instrument versions have been employed by researchers in multiple studies, and a few important studies have incorporated both versions of the instrument. Overall, the 54-item, Emmons (1984, 1987) version of the NPI has been the preferred one for research purposes. The factor structure of the Emmons (1984, 1987) version appears easier to replicate than the Raskin and Terry (1988) version. Most studies discussed in 26 this review will center on the Emmons (1984, 1987) version, though a few will mention the Raskin and Terry (1988) version for purposes of highlighting similarities between the two instrument versions. The important point made here is that both versions appear to tap both adaptive and maladaptive aspects of narcissism. Initial Applications of the NPI Factor Models Emmons (1984), in an attempt to support the construct validity of his version of the NPI, correlated NPI scores with a host of personality dimension measures. These measures included the Edwards Personal Preference Schedule (EPPS, Edwards, 1959), the Sixteen Personality Factor Questionnaire (16PF, Cattell, Eber, & Tatsuoka, 1970), the Eysenck Personality Inventory (EPI, Eysenck & Eysenck, 1968), the Self-Perception Inventory (Soares & Soares, 1965), the Self-Consciousness Scale (Fenigstein, Scheier, & Buss, 1975), the Body Self-Consciousness Scale (Miller, Murphy, & Buss, 1981), the Uniqueness-Seeking Scale (Snyder & Fromkin, 1977), the Self-Monitoring Scale (Snyder, 1972), and the narcissistic-competitive items from the Interpersonal Checklist (ICL, LaForge & Suczek, 1955). Emmons (1984) also correlated NPI scores with peer ratings of participants? narcissistic traits. Emmons (1984) correlated both total NPI scores and each factor score with indicators on the aforementioned personality instruments. Total NPI scores correlated positively (and significantly) with such personality characteristics as need for achievement, need for aggression, need for dominance, exhibitionism, need for surgency, social boldness, extroversion, adoption of a tough pose, independence, self-esteem, body consciousness, body competence, need for uniqueness, and self-monitoring (Emmons, 27 1984). Note that this list includes a mixture of ?adaptive? and ?maladaptive? personality characteristics. These findings are not surprising, given the findings of previous studies already mentioned. Correlations with NPI factor scores (Exploitativeness/Entitlement, Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration) yielded more important and instructive findings. Generally speaking, a distinct pattern emerged when NPI factor scores were assessed in relation to measures of other pertinent personality characteristics. The Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration factors correlated positively with measures of psychological health and negatively with measures of psychological distress, while the Exploitativeness/Entitlement factor displayed the opposite pattern (Emmons, 1984). When isolated, it seemed that three of the NPI factors tapped levels of relatively healthy, adaptive functioning, while one of the factors assessed unhealthy, maladaptive functioning The Leadership/Authority factor will be discussed first, followed by Superiority/Arrogance and Self-Absorption/Self-Admiration. Leadership/Authority scores correlated positively with need for achievement, need for dominance, exhibitionism, warmth, surgency, social boldness, self-control, extroversion, adoption of a tough pose, self-esteem, body consciousness, body competence, need for uniqueness, and self-monitoring (Emmons, 1984). These factor scores also correlated negatively with need for abasement, guilt proneness, self-sufficiency, neuroticism, self-ideal discrepancy, social anxiety, and other-directedness (Emmons, 1984). The Leadership/Authority factor, on the whole, appeared to tap the most aspects of narcissism. While correlating with such 28 negative indicators as display of a tough pose, exhibitionism, body consciousness, and need for uniqueness, most of the Leadership/Authority correlates are indicators of positive functioning (Emmons, 1984). This finding will be replicated and extended in studies yet to be discussed. The Superiority/Arrogance and Self-Absorption/Self-Admiration subscales yielded a similar, but more mixed, pattern. The Superiority/Arrogance factor correlated positively with such characteristics as need for achievement, need for aggression, need for autonomy, need for dominance, surgency, social boldness, radical thinking, extroversion, tough pose, independence, self-esteem, body consciousness, need for uniqueness, and self-monitoring, while correlating negatively with need for abasement, need for succorance, guilt proneness, anxiety, self-ideal discrepancy, and social anxiety (Emmons, 1984). The Self-Absorption/Self-Admiration factor correlated positively with need for abasement, need for dominance, surgency, radical thinking, extroversion, independence, self-esteem, body consciousness, body competence, need for uniqueness, and self-monitoring, while correlating negatively with neuroticism, self-ideal discrepancy, social anxiety, and other-directedness (Emmons, 1984). The Exploitativeness/Entitlement factor, however, yielded a consistent pattern of correlations with measures of maladaptive functioning. This particular factor correlated positively with need for achievement, need for aggression, need for dominance, surgency, suspiciousness, tension, anxiety, tough pose, neuroticism, self-consciousness, body competence, need for uniqueness, and self-monitoring (Emmons, 1984). This factor also 29 correlated negatively with need for abasement and tender-mindedness. The findings from Emmons? (1984) study set the precedent and tone for future studies with the NPI. Raskin and Terry (1988) performed similar analyses on their alternate version of the NPI. Total NPI and factor scores were compared to trait rankings developed by narcissism experts, California Q-Sort (CQ; Block, 1961) responses, the Adjective Check List (ACL; Gough & Heilbrun, 1965), the California Psychological Inventory (CPI; Gough, 1956), and behavioral rankings drawn from a leaderless group discussion (Raskin & Terry, 1988). Raskin and Terry (1988) also compared NPI scores to those drawn from the Leary circumplex model of interpersonal behavior through comparison with the Interpersonal Check List (ICL; Leary, 1956). Raskin and Terry (1988) first examined NPI total score correlates. NPI scores were found to correlate positively and significantly with traits such as sensation seeking, extroversion, dominance, energy level, exhibitionism, criticality, assertiveness, leadership, self-satisfaction, self-confidence, ambitiousness, and self-centeredness. NPI scores were found to correlate negatively with such characteristics as submissiveness, patience, timidity, sensitivity, modesty, deference, abasement, and succorance. These findings speak to the solid validity of the NPI as a measure of narcissism, as this description is very close to that provided in other studies. However, the most interesting findings from Raskin and Terry?s (1988) work concern their factor scores. A pattern similar to that found with the Emmons (1984, 1987) version of the NPI emerged, as some NPI factors appeared to tap more adaptive aspects of functioning, while others tapped maladaptive ones (Raskin & Terry, 1988). For 30 example, the factors Authority, Superiority, and Self-Sufficiency correlated directly with such personality traits as assertiveness, self-confidence, leadership, sociability, ambition, independence, self-satisfaction, and achievement-oriented (Raskin & Terry, 1988). These descriptors would generally be viewed as healthy characteristics in most individuals; they also come very close to the correlates of Emmons? (1984, 1987) Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration scales. However, the Exhibitionism, Exploitativeness, and Entitlement factors correlated with such characteristics as dramatic, impatient, aggressive, hostile, tactless, power-oriented, dominant, and self-indulgent (Raskin & Terry, 1988). These characteristics are some of the very same that would apply to those who score high on Emmons? (1984, 1987) EE scale. Thus, though the factor structures differ between the two versions, both appear to tap both healthy and unhealthy components of narcissistic functioning. However, puzzling findings with the NPI would ensue. In a follow-up study, Emmons (1987) confirmed the validity of his NPI version by conducting yet another factor analysis and by comparing NPI total and factor scores to those gleaned from other measures of narcissism/Narcissistic Personality Disorder. As mentioned previously, the factor structure that emerged from this second study was identical to the first (Emmons, 1984; 1987). Emmons (1987) then compared NPI scores to scores obtained on the Narcissistic Personality subscale of the MCMI (Millon, 1982), the Narcissistic Personality Disorder Scale (NPDS, Solomon, 1982), and the Selfism scale (Phares & Erskine, 1984). Finally, Emmons (1987) compared NPI scores to various measures of mood variability and emotional intensity. 31 Emmons (1987) found that total NPI scores correlated positively and significantly with MCMI and Selfism scores, but not with NPDS scores. At the factor level, only Exploitativeness/Entitlement correlated positively with all three objective measures of narcissism. Leadership/Authority correlated positively with MCMI scores, Self- Absorption/Self-Admiration correlated positively with Selfism scores, and Superiority/Arrogance correlated positively with MCMI and Selfism scores. Regarding measures of affective intensity and mood variability, total NPI scores correlated directly with variability in negative affect. At the factor level, Exploitativeness/Entitlement correlated positively with variability of positive affect, variability of negative affect, and affective intensity. Leadership/Authority and Superiority/Arrogance displayed no correlations with any of these measures, while Self-Absorption/Self-Admiration correlated directly with variability of positive and negative affect (Emmons, 1987). These are puzzling findings, indeed. One would expect NPI scores to correlate positively and significantly with any measure of narcissism. However, NPI scores did not correlate with NPDS scores (save the lone Exploitativeness/Entitlement factor correlation). This finding suggests that either (1) narcissism scales are not measuring the same construct, (2) that different types or forms of narcissism exist (and that each scale is measuring a separate form), or (3) some narcissism scales are not completely capturing the breadth and depth of the narcissism construct. Further work with the NPI served to clarify the narcissism construct and demonstrate that the relationships between narcissism and other psychological constructs (such as self-esteem) are complex and rich. The NPI 32 appears, at this stage, to be the most comprehensive, valid, and reliable narcissism instrument. Further Applications and Extensions of the NPI Factor Models In an enlightening series of studies, a group of researchers has employed findings from the NPI factor scores to explain some of the contradictory findings within narcissism research. The findings to be discussed in this section of the manuscript help to resolve some of the major remaining questions regarding the construct of narcissism, which are: (1) How can a narcissism scale measure both healthy and unhealthy individual functioning?, (2) What are the specific relationships among and between the different factors within the larger construct of narcissism?, and (3) How do these relationships predict levels of adaptive/maladaptive functioning within individuals? The primary mode of exploration in these studies will be the use of partial correlation techniques to parcel out the effects of one factor of narcissism (say, Exploitativeness/Entitlement), while examining correlates of the other factors (Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration). Previously discussed research provided clues that the Exploitativeness/Entitlement scale of the Emmons (1984, 1987) NPI correlated with unhealthy functioning, while the other three scales correlated with healthy functioning (or, at worst, with a mixture of healthy and unhealthy functioning). The studies that will now be discussed clearly support the hypothesis that narcissistic functioning falls along a continuum from solidly healthy to trenchantly unhealthy (e.g., Watson, Hickman, & Morris, 1996; Watson, Little, Sawrie, & Biderman, 1992). 33 One of the hallmark symptoms of Narcissistic Personality Disorder is a lack of empathy (APA, DSM-IV, 1994). Watson and Morris (1991) explored the relationship between NPI total and factor scores and measures of empathy, as well as with measures of social responsibility and social desirability. Previous studies (Watson, Grisham, Trotter, & Biderman, 1984; Watson, Hood, Morris, & Hall, 1987) found contradictory evidence regarding narcissism and empathy. Total NPI scores were found to correlate inversely with some empathy scales, but not others (Watson, et al., 1984). Adding to the confusion, Exploitativeness/Entitlement correlated negatively with all empathy measures; Leadership/Authority correlated negatively with one but positively with another; Superiority/Arrogance correlated negatively with two measures, but positively with the third; and Self-Absorption/Self-Admiration displayed a negative relationship to one scale but none with the other two (Watson, et al., 1984). A similar pattern was discovered when evaluating two social desirability instruments (Watson, et al., 1984; Watson, et al., 1987). The Watson and Morris (1991) study clarified these relationships through the comparison of zero-order and partial factor score correlations. Zero-order correlations between the NPI factor scores and the other instruments suggested that Exploitativeness/Entitlement was clearly indicative of unhealthy functioning (negative correlations with Empathic Concern, Perspective Taking, Social Responsibility, and Social Desirability), Leadership/Authority was primarily indicative of healthy functioning (negative correlation with Personal Distress; no correlations with other factors), Superiority/Arrogance was an ambiguous indicator (negative correlation with 34 Empathic Concern, but also negative relationship with Personal Distress), and Self-Absorption/Self-Admiration was not indicative of either healthy or unhealthy functioning (no correlations with any factor). However, the patterns changed when the effects of certain factors were parceled out. Watson and Morris (1991) found that, when controlling for the effects of Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration, Exploitativeness/Entitlement displayed negative correlations with Empathic Concern, Perspective Taking, Social Responsibility, and Social Desirability, while also yielding a positive correlation with Personal Distress. Hence, partialing out the effect of the other three scales appears to make the negative effects of higher Exploitativeness/Entitlement even more unhealthy (Watson & Morris, 1991). When controlling for the effects of Exploitativeness/Entitlement, the Leadership/Authority scale scores correlated negatively with Personal Distress and positively with Social Responsibility, the Superiority/Arrogance scale scores correlated negatively with Personal Distress (with no relation now to Empathic Concern), and the Self-Absorption/Self-Admiration scale scores still displaying no correlations whatsoever (Watson & Morris, 1991). Hence, partialing out the negative effects of Exploitativess/Entitlement seemed to boost the positive correlations of at least two of the NPI factors with measures of healthy functioning (while also eliminating correlations with indicators of negative functioning). Watson and Morris (1991) used the aforementioned findings to support the contention that narcissism is a complex, multidimensional construct and that narcissistic functioning falls along a continuum of functioning (from healthy to unhealthy). Thus, the 35 narcissistic personality construct has foundation in both adaptive and maladaptive functioning (Watson & Morris, 1991). Those narcissists that display primarily exploitative behaviors and have an attitude of entitlement are the most likely to suffer psychological and interpersonal consequences (especially if they lack the ?good? aspects of narcissism, such as leadership ability, authority, and a solid belief in their own self- worth). However, narcissists who are good leaders, who use their authority in a healthy manner, and who believe themselves capable are unlikely to suffer personal distress (especially if they do not possess an attitude of entitlement and do not look to exploit others) (Watson & Morris, 1991). Another study conducted by Raskin, Novacek, and Hogan (1991) seems to support this theoretical viewpoint as well. They sought to examine the question of whether narcissism is truly positively correlated with self-esteem. These researchers wished to test the hypothesis that narcissism is related to a defensive form of self-esteem. If narcissism were truly related to a defensive form of self-esteem, then the argument that narcissism can be adaptive and functional in certain forms would be weakened. Correlations between NPI scores and self-esteem scores would be inflated due to the operation of this defensive ego state. However, Raskin, et al., (1991) found evidence more supportive of the continuum hypothesis instead. Their findings indicated that defensive self-esteem is composed of two primary components (Social Desirability and Grandiosity), while healthy self-esteem is composed of three factors (unbiased self-evaluation, self-regard based on others? appraisals, and positive comparison to the ideal self). Narcissism was, again, shown to 36 positively correlate with self-esteem. However, the nature of this self-esteem depended on the foundation of that self-evaluation. Raskin, et al., (1991) found that defensive self- esteem is driven by grandiosity, not social desirability. Recall that, in the previously discussed study, only one narcissism factor correlated negatively with Social Desirability (Exploitativeness/Entitlement) (Watson & Morris, 1991). Those who exploit others and have an air of entitlement are also the most likely to boost their self-esteem defensively to protect their grandiose self-image (John & Robins, 1994). They are also the most unlikely to possess a healthy level of self-esteem (Watson & Morris, 1991). Hence, Raskin et al., (1991) provided further evidence that a continuum of narcissistic functioning exists. Those narcissists who create grandiose self-images, who feel entitled, and who exploit others as a result are more likely to suffer lower self-esteem when the adaptive aspects of narcissism are factored out (Watson & Morris, 1991). Similar findings were consistently borne out in future studies. Watson, Little, Sawrie, and Biderman (1992) examined the complex relationships among narcissism, self-esteem, and empathy. Measures used in this study included the Emmons (1984, 1987) NPI, the O?Brien Multiphasic Narcissism Inventory (OMNI; O?Brien, 1987), the Interpersonal Reactivity Index (which served as the empathy measure; Davis, 1980), the Superiority and Goal Instability Scales (Robbins & Patton, 1985), and the Rosenberg Self-Esteem scale (Rosenberg, 1965). Zero-order correlations with the NPI factors were examined first, then partial correlation techniques were again used to demonstrate the importance of attending to the NPI factor scores carefully (Watson, et al., 1992). 37 Zero-order correlations displayed the following pattern: (1) Exploitativeness/Entitlement correlated negatively with Empathic Concern and Perspective Taking, (2) Leadership/Authority, Superiority/Arrogance, and Self- Absorption/Self-Admiration correlated positively with self-esteem, (3) Leadership/Authority and Superiority/Arrogance correlated inversely with Personal Distress, and (4) Superiority/Arrogance and Self-Absorption/Self-Admiration displayed an inverse correlation with Empathic Concern (Watson, et al., 1992). Note that Exploitativeness/Entitlement again appears unquestionably maladaptive, Leadership/Authority appears a healthy aspect, and Superiority/Arrogance and Self- Absorption/Self-Admiration suggest a mixture of adaptive and maladaptive functioning. However, partial correlations paint a clearer and more comprehensive picture. When removing the variance associated with Exploitativeness/Entitlement, Superiority/Arrogance and Self-Absorption/Self-Admiration no longer correlated inversely with Empathic Concern. Leadership/Authority displayed a positive relationship with Empathic Concern and Perspective Taking, and Self-Absorption/Self-Admiration showed a negative relationship with Personal Distress (Watson, et al., 1992). When the effects of the three other subscales were partialed out, Exploitativeness/Entitlement predicted lower self-esteem, greater personal distress, greater Goal Instability, reduced Empathic Concern, and reduced capacity for Perspective Taking (Watson, et al., 1992). Factoring out the effects of Exploitativeness/Entitlement seemed to make the other three subscales appear more adaptive and healthy, while factoring out the effects of Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration 38 appeared to suggest Exploitativeness/Entitlement is an even more unhealthy facet (Watson, et al., 1992). Obviously, assessment of only the total NPI scores leaves much of the picture unclear. Watson and Morris (1990) examined narcissism in relation to a disparate theoretical view from their own (which tended to center on a Kohutian self-psychology perspective). Through use of the NPI and Irrational Beliefs Test (IBT; Jones, 1969), Watson and Morris (1990) examined the relationship of NPI factors to endorsement of a host of irrational beliefs, which are assumed to translate into maladaptive and unhealthy functioning in the Rational-Emotive theory of human functioning. Even when using a different theoretical stance, the complex NPI findings discussed in previous studies were borne out again. Zero-order data suggested that (1) high Exploitativeness/Entitlement factor scores predicted greater Demand for Approval, greater Self-Expectations, greater Blame Proneness, greater Problem Avoidance, less Dependency, and greater Helplessness for Change; (2) high Leadership/Authority factor scores predicted lower Demand for Approval, lower Emotional Irresponsibility, lower Anxious Overconcern, lower Problem Avoidance, and less Dependency; (3) high Superiority/Arrogance scores predicted lower Demand for Approval, lower Frustration Reactivity, lower Anxious Overconcern, lower Dependency, and lower Helplessness for Change; and (4) high Self-Absorption/Self- Admiration scores predicted lower Emotional Irresponsibility and lower Dependency (Watson & Morris, 1990). Partialing out the effects of Exploitativeness/Entitlement again made the other three factors appear more healthy, as doing so increased the number of 39 inverse correlations among Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration scores and a host of irrational beliefs (including those aforementioned). Partialing out the effects of the three healthier factors also revealed Exploitativeness/Entitlement as an even more maladaptive factor, as this factor correlated directly with no fewer than eight separate irrational beliefs (Watson & Morris, 1990). Another study examined narcissism in relation to various measures of self- consciousness and the clinical construct of splitting. Watson and Biderman (1993) used both the Emmons (1984, 1987) and the Raskin and Terry (1988) versions of the NPI, the Gerson (1984) Splitting Scale, the Private Self-Consciousness Scale (Fenigstein, Scheier, & Buss, 1975), the Social Anxiety Scale (Costello & Comrey, 1967), and the Public Self- Consciousness scale (Fenigstein, Scheier, & Buss, 1975) as basis of comparison. The authors also included a measure of Depression and Anxiety (Costello & Comrey, 1967) and Rosenberg?s (1965) Self-Esteem Scale. Watson and Biderman (1993) reported findings largely consistent with the pattern previously discussed. Analysis of the zero-order data suggested that the Exploitativeness/Entitlement factor of the Emmons NPI was related to greater self- reflection, greater style consciousness, greater appearance consciousness, and a greater tendency to engage in splitting. The other three subscales displayed no relation to splitting; a positive relation to self-awareness, insight, and self-esteem; and a negative relationship to measures of anxiety and depression (Watson & Biderman, 1993). Partialing out Exploitativenss/Entitlement again made Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration appear even more adaptive, 40 as a negative relationship appeared between Leadership/Authority, Self-Absorption/Self-Admiration, and splitting (while the zero-order findings remained stable or increased in strength). Partialing out the assumed healthier narcissism subscales made Exploitativeness/Entitlement appear even more unhealthy as well, as a greater score was now also related to higher levels of depression and anxiety (Watson & Biderman, 1993). The Raskin and Terry (1988) version of the NPI yielded similar results. Based on previous results, Watson and Biderman (1993) treated the Exhibitionism, Exploitativeness, and Entitlement factors as negative indicators, while the Authority, Superiority, and Self-Sufficiency factors were considered positive and healthy aspects. When partialing out the negative narcissism factors, the positive factors correlated negatively with splitting, depression, and anxiety while correlating positively with self- esteem, self-awareness, insight, and the ability to self-reflect (Watson & Biderman, 1993). When the positive factors were partialled out, the negative narcissism factors were related to greater splitting, greater self-reflection, greater appearance consciousness, and greater anxiety (Watson & Biderman, 1993). Continuing their work, Hickman, Watson, and Morris (1996) employed the Emmons NPI, the OMNI, the Life Orientation Test (Scheier & Carver, 1985), the Hopelessness Scale (Beck, Weissman, Lester, & Trexler, 1974) and the Expected Balance Scale (Staats, 1989) to further clarify the interactions of the narcissism factors. Their primary interest centered on the differential relationship of NPI factors to tendencies toward optimism and pessimism. Initial zero-order findings revealed that 41 higher Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self- Admiration predicted greater optimism and less pessimism, though no relationships between Exploitativeness/Entitlement and either construct were revealed. However, when the positive and negative aspects of narcissism were partialed out, greater Exploitativeness/Entitlement was indeed shown to predict less optimism and greater pessimism (Hickman, et al., 1996). Watson, Hickman, and Morris (1996) also examined the relationship between narcissism and felt experience of shame. Measures used included the NPI, the Shame Scale of the Personal Feelings Questionnaire ? 2 (Harder & Zalma, 1990), the Adapted Shame and Guilt Scale (ASGS; Hoblitzelle as quoted by Harder & Zalma, 1990), and the Rosenberg Self-Esteem scale. Findings suggested that greater self-esteem was related to higher levels of Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self- Admiration, while also related to lower Exploitativeness/Entitlement scores (Watson, et al., 1996). The effect of narcissism on shame was assessed in two steps. During the first step, the self-esteem scores were considered first, while the narcissism data were entered into the regression equation second. Total NPI, Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration scores had no relation to shame measures. Total NPI and Exploitativeness/Entitlement scores were weak predictors of shame, as the regression coefficients obtained were well below the levels required for statistical significance (Watson, et al., 1996). During the second step, the narcissism scores were entered first, while the self- esteem scores were accounted for afterwards. Total NPI, Leadership/Authority, 42 Superiority/Arrogance, and Self-Absorption/Self-Admiration scores became inverse predictors of shame, while total NPI and Exploitativeness/Entitlement scores were direct predictors of shame (Watson, et al., 1996). These findings refute the suggestion that narcissism and self-esteem correlate due to a defensive ego configuration. Instead, narcissism and self-esteem relate and interact along a continuum to predict indicators of functioning, such as shame (Watson, et al., 1996). Watson, Varnell, and Morris (1999) employed the Multidimensional Perfectionism Scale (Frost, Marten, Lahart, & Rosenblate, 1990), the NPI, and the Rosenberg Self-Esteem scale to examine the relationship between narcissism and perfectionism. Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self- Admiration scores were shown to predict high self-esteem, higher self and other-oriented perfectionism, but less socially prescribed perfectionism (Watson, et al., 1999). Exploitativeness/Entitlement scores held no relationship to self-esteem, while predicting high self, other, and socially prescribed perfectionism (Watson, et al., 1999). Interestingly, when self-esteem was controlled for, the three healthier subscales also predicted high socially prescribed perfectionism (Watson, et al., 1999). These findings again suggest a complex relationship between narcissism and self-esteem. Those who score highly on the suggested adaptive aspects of narcissism seemingly must possess a high level of self-esteem as a buffer against pressure to perform. Also, when the negative effects of Exploitativeness/Entitlement were controlled for, the relationship between Superiority/Arrogance and self-oriented perfectionism disappeared (Watson, et al., 1999). Leadership/Authority and Superiority/Arrogance also showed no relationship to 43 other-oriented perfectionism. When the three healthy NPI subscales were controlled for, greater self-esteem no longer predicted self and other-oriented perfectionism (Watson, et al., 1999). In short, healthy narcissism and healthy self-esteem predict less perfectionism, while unhealthy narcissism and poor self-esteem predict greater levels of perfectionism. Variations on this theme continue throughout the literature. Watson, Morris, and Miller (1997) found that zero-order NPI factor and total scores correlated directly with self-esteem, measures of assertiveness, and measures of hypercompetitiveness. However, other findings (again, arrived at through partial correlation procedures) revealed that partialing out adaptive NPI factors led to an inverse relationship between self-esteem and hypercompetitiveness, a direct relationship between self-esteem and assertiveness, no correlation between Exploitativeness/Entitlement score and assertiveness, and a direct correlation between Exploitativeness/Entitlement scores and measures of hypercompetitiveness (Watson, et al., 1997). Furthermore, partialing out Exploitativeness/Entitlement led to a loss of association between NPI scores and hypercompetitiveness, a small inverse relationship between self-esteem and hypercompetitiveness, and a significant positive relationship among total NPI scores (and Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration scores specifically), self-esteem, and assertiveness (Watson, et al., 1997). Watson, Hickman, Morris, Milliron, and Whiting (2001) examined relationships between narcissism, self-esteem, and parental nurturance. Findings suggested that, when the effects of Exploitativeness/Entitlement are accounted for, Leadership/Authority scores were positively correlated with high levels of father nurturance, while higher 44 Superiority/Arrogance scores were associated with greater mother nurturance. When the adaptive aspects of narcissism were accounted for, Exploitativeness/Entitlement scores had relationships to lower self-esteem and lower father nurturance. Also, when the effects of self-esteem were partialed out, Superiority/Arrogance displayed a negative relationship with mother nurturance, while Self-Absorption/Self-Admiration related negatively to father nurturance (Watson, et al., 2001). These findings add weight to the assertion that healthy self-esteem is an important moderating variable when assessing for the impact of narcissism factors. Even the healthy narcissism factors (especially Superiority/Arrogance and Self-Absorption/Self-Admiration) reveal themselves as less healthy without a solid level of self-esteem present. Soyer, Rovenpor, Hopelman, Mullins, and Watson (2001) compared Raskin and Terry (1988) NPI scores with measures of Machiavellianism (MACH-IV; Christie & Geis, 1970), autonomy, need for achievement, and self, family, job, and global life satisfaction. Total NPI scores were found to predict greater Machiavellianism (which replicated a finding from McHoskey, 1995), greater autonomy, greater self-satisfaction). However, NPI scores displayed no relationship to family, job, or overall life satisfaction (Soyer, et al., 2001). When assessing for the effects of the Raskin and Terry (1988) factors, evidence for the continuum hypothesis of narcissistic functioning was found. Authority, consistent with findings from Emmons NPI studies (1984, 1987), was the best predictor of adjustment and healthy functioning. Authority predicted greater need for achievement, life satisfaction, self-satisfaction, family satisfaction, and self-sufficiency. 45 Self-sufficiency was the next most reliable predictor of psychological health, as those scores predicted greater need for achievement, need for autonomy, and life and self- satisfaction (Soyer, et al., 2001). Superiority findings were mixed; greater Superiority predicted greater Machiavellianism and lower job satisfaction, but also predicted greater self-satisfaction. The other four narcissism factors were primarily either neutral or negative predictors of health. Vanity and Exhibitionism predicted greater Machiavellianism, but held no relation to any measure of satisfaction. The Entitlement and Exploitation factors predicted greater Machiavellianism, greater self-satisfaction, and poor job satisfaction. Removal of the effects of Exploitation and Entitlement eliminated NPI score correlations with Machiavellianism, need for achievement, and need for autonomy while strengthening the link between NPI score and self-satisfaction. Furthermore, removal of the effects of Authority and Self-Sufficiency led to an increased association between NPI score and Machiavellianism, lower levels of life, job, and family satisfaction, and no prediction of self-satisfaction, need for autonomy, and need for achievement based on NPI score (Soyer, et al., 2001). Summary of NPI-Driven Conceptualizations of Narcissism The aforementioned body of research provided strong evidence that the construct of narcissism is a complex one composed of many key factors. Some of these factors are reliable predictors of healthy, adaptive functioning (such as Leadership/Authority), while others are predictors of unhealthy, maladaptive functioning (such as Exploitativeness/Entitlement). These research findings also show that when the adaptive aspects of narcissism are accounted for, the maladaptive ones appear even more 46 unhealthy (and vice versa). Narcissism has also been shown to possess a complex and rich relationship with self-esteem, which appears to moderate the relative level of health for those scoring high in narcissism. The healthy aspects of narcissism have been associated with such indicators of healthy functioning as greater empathic concern (Watson, et al., 1992), greater ability to engage in perspective taking (Watson, et al., 1992), greater self-esteem (e.g., Raskin, Novacek, & Hogan, 1991; Watson, et al., 1992), less personal distress (Watson, et al., 1992), fewer irrational beliefs (Watson & Morris, 1990), greater assertiveness and less hypercompetitiveness (Watson, Morris, & Miller, 1997), greater social responsibility (Watson & Morris, 1991), and social desirability (Watson & Morris, 1991). The unhealthy aspects of narcissism have been linked with poorer job satisfaction (Soyer, et al., 2001), lower self-esteem (e.g., Watson, et al., 2001; Watson, et al., 1992), greater perfectionism (Watson, et al., 1999), a more pessimistic and less optimistic life outlook (Hickman, et al., 1996), greater self-absorption without insight (Watson & Biderman, 1993), lower social desirability (Watson & Morris, 1991), less empathic concern and diminished ability for perspective taking (Watson & Morris, 1991), and higher levels of anxiety and depression (Watson & Biderman, 1993; Watson, Sawrie, Greene, & Arredondo, 2002). Existential Conceptualization of Narcissism Existential perspectives have been applied to the psychological construct known as narcissism. However, existential practitioners have rarely chosen to put their conceptualizations to empirical test. Though some might argue that existential constructs 47 are not amenable to empirical study, many measures of such constructs have been developed and evaluated. Major existential theoretical stances on narcissism will be discussed, synthesized, and combined to produce a coherent existential view, and then used to facilitate the interpretation of the present study. Finally, salient measures of key existential constructs will be discussed. Frankl (1984, p. 105) believed that each individual?s primary motive in life is a ?will to meaning.? Hence, seeking meaning and purpose in one?s life is the central task of each individual. People pursue awareness of the spiritual importance of life and awareness of their place in relation to the larger spiritual design. Should individuals fail to discover this purpose and meaning in life, then they are apt to experience an existential vacuum state characterized by boredom and a deep sense of emptiness. Recall that individuals who evidence very high levels of narcissistic traits often are prone to experiencing boredom (Emmons, 1981). Frankl argued that individuals will, if subjected to this existential vacuum state for a lengthy period of time, react to it with existential frustration. Existential frustration will then likely engender a ?noogenic neurosis? in the individual, which is characterized by decompensation due to failure to find life meaning (Frankl, 1984, p. 106). Yalom (1980) agrees that we are meaning-seeking creatures. However, he emphasized that we are meaning-making creatures as well). In fact, Yalom argues that humans are incapable of finding meaning when directly seeking it; we must find meaning through participation in life. Meaning and purpose must emerge from our honest and genuine engagement in interactions with our surrounding environment (Yalom, 1980). 48 Following this premise, Shaw (2000) argued that narcissism is a motivational structure created (at both a conscious and unconscious level) to resolve the very human problem of personal insignificance. Shaw (2000) posited that narcissism is a personal myth system created to shield individuals from the painful awareness of their own mortality and personal limitations. He also asserted that this creation of a personal narcissistic myth is a common, normal human reaction to their existential condition (Shaw, 2000). Therefore, it is logical to assert that those who adopt a narcissistic motivational structure to deal with the harsh realities of life fall along a continuum of healthy and unhealthy functioning. Some individuals find a way to make this myth work for them in a meaning-creating way; others will find this motivational structure lacking and may be prone to a noogenic neurosis. Taking this argument of narcissism as a motivational psychological structure one step further, van Deurzen-Smith (2000, p. 232) argues that ?as organisms, we are primarily self-centered, and it is in our nature and best interest to orientate ourselves towards the world with the assumption of our right and entitlement.? She also asserts that even our genetic material is selfish, in that genes must express themselves selfishly (e.g., at the expense of other genes) to insure their survival (Dawkins, 1976, as cited in van Deurzen-Smith, 2000). Hence, it seems that narcissism may be ingrained in our biology, psychology, socialization, culture, and spiritual life. Narcissism is an integral part of being human and alive in the world. We are all prone to self-centeredness, a sense of 49 entitlement, the capability to exploit others, to lack empathy at times, and to see ourselves in a grandiose manner. However, van Deurzen-Smith (2000) continues on to argue that this self-centered, selfish tendency that humans naturally possess is counterbalanced with a sense of the needs of our world and environment as a whole. She notes that as one becomes more aware of one?s individual role in the overall life context into which we are thrown, one gains a greater appreciation of the need for a sense of kinship and community (van Deurzen-Smith, 2000). She notes that we must stand in the tension of supporting and defending our own individual interests, while also deriving a deep sense of meaning and purpose from awareness of our ?role in the global scheme of things? (van Deurzen-Smith, 2000, p. 233). No individual can survive (or, at least, function in a healthy manner) without attending to both their own needs and the needs of those around them. A certain degree and form of narcissistic expression, in this view, is healthy and adaptive, as situations will emerge in which one?s own needs are of the utmost importance in the moment. van Deurzen-Smith (2000) asserts that our mode of functioning in the world affects us holistically. She argues that we must consider the physical, psychological, social, and spiritual aspects of life and of our own functioning. We must also attend to the interactions among these factors. Her basic premise is that we all, to varying extents, buy into the narcissistic motivational structure posited by Shaw (2000). Individuals may choose to adopt a positive, meaning-creating narcissism, a negative, meaning-obscuring narcissism, or some variant in-between. Individuals may also change their narcissistic 50 style due to internal or external factors over time. Regardless of the style of the individual?s motivational structure, it will inevitably have an effect on the physical, psychological, social, and spiritual functioning of the individual (van Deurzen-Smith, 2000). Physically speaking, narcissism becomes unhealthy and maladaptive if the motivational structure leads a person to view the body as a machine or computer, rather than an alive, feeling instrument. If we lose sight of the limitations of our own bodies long-term with a grandiose, mechanical view of them, we ignore a given aspect of being- in-the-world (van Deurzen-Smith, 2000). Should a person lose a sense of solid physical contact with the Earth, one runs the risk of using natural resources in a manner that is selfish and destructive. It is also possible that by denying death through a pervasive, self- centered, grandiose physical relationship with the world that one will forget to live fully and completely. However, our health will improve if we adopt those narcissistic aspects that do not preclude viewing ourselves as fallible and limited. If we accept our limitations physically while testing them in a careful, measured manner, we are less likely to experience injury and illness. Psychologically, narcissism?s effect on one?s relationship to self is the key. On the positive side, one may best protect one?s own psychological well-being with the ability to seek attention from others, exhibit a bit of grandiosity, exhibit charm, and enhance one?s self without experiencing attendant guilt (Cooper, 1987; Cooper & Ronningstam, 1992; Rose, 2002; Wink, 1991). Individuals need strokes from others at times to maintain our positive self-relationship; we need reinforcement that our internal being is worthy and 51 valuable. However, if narcissism is taken too far, individuals may lose respect and lose touch with their sense of self. Socially, narcissism may fall on a continuum from boon to bane as well. Those who motivate themselves in a healthy manner will make responsible attempts to engage with others and explore new social territory with confidence (van Deurzen-Smith, 2000). Healthy narcissism allows one to maintain knowledge and awareness of one?s social needs and provides one with the confidence and energy to meet them with assertiveness. It allows for a balance between one?s own social desires and the desires of others. Unhealthy narcissism will lead to exploitation and objectification of others. Lack of empathy will lead to unsuccessful encounters with others in the long term. Unhealthy narcissism would be characterized by a lack of attendance to the needs of others, while seeking social approval only for oneself. Finally, spiritual functioning must be considered in van Deurzen-Smith?s (2000) view. The most obvious risk of adopting an unhealthy narcissistic structure is a loss of meaning and purpose in life. Divorcing oneself from one?s own feelings, from the feelings of others, substituting a grandiose self-image for the real thing, failing to recognize personal limitations, and viewing the world as subject to one?s own needs and wants will eventually lead to a very unhappy existence. To maintain a balanced sense of ourselves and our place in the world, ?we [must] discover the secret of the ?I? as a channel rather than an entity? (van Deurzen-Smith, 2000, p. 127). Healthy narcissism exhibits itself in an ability to maintain awe and reverence for the world at large and other 52 people in it, while still maintaining a sense of self-importance despite personal limitations (van Deurzen-Smith, 2000). Key Existential Constructs: Relationships to Narcissistic Functioning For purposes of the present study, key existential constructs were chosen for examination in relationship to Narcissistic Personality Inventory scores. The existential constructs selected are purpose in life/search for meaning, authenticity, death anxiety, and self-actualization. The relationship between narcissism and a search for meaning and purpose has been discussed. However, authenticity, death anxiety, and self-actualization have yet to be defined, and their proposed relationships to the narcissistic motivational structure have yet to be discussed. It is important to emphasize that existential theorists believe all of these concepts to be intertwined and related in a complex manner. Bugental (1965) defined the concept of authenticity as ?both a hypothesized ultimate state of atoneness with the cosmos and the immense continuum leading toward that ultimate ideal? (p. 32). To add clarity, he stated that authentic living is both a goal state, which will never be completely achieved, and a process of attempting to achieve that goal). Therefore, Bugental viewed authenticity as a dimensional concept; one can move toward and away from authenticity moment to moment and day to day. Authenticity requires acceptance of certain given aspects of life, such as awareness (finiteness, potential to act, choice, and separateness). Given awareness of responsibility, one must make a commitment. Basically, this means making choices in the moment and standing by the consequences of these choices, whatever they may be. 53 Bugental (1965) also set forth three key attributes of authentic being. The first is being as fully aware as one can be at any given moment (with the awareness that complete awareness is impossible for humans), the second is choosing what possibility one will invest in at a given moment in time, and the third is taking responsibility for the choice made and the implications for said choice. Therefore, authentic living involves awareness and acceptance of both the limitations and the possibilities of human existence in all dimensions of existence (physical, social, psychological, and spiritual). Anxiety is a given, as striving for authenticity demands that one make choices and take responsibility for those choices (and their consequences). Bugental (1965) argued that problems in living stem from denials or distortions of authenticity and authentic living. Given Bugental?s (1965) explications of the concept of authenticity and his argument that problems in living stem from denials or distortions of authenticity, one might argue that adoption of certain aspects of a narcissistic motivational construct (e.g., Exploitativeness/Entitlement) are modes of inauthentic living (and therefore are likely to lead to psychological dysfunction and a loss of health). Existentialists would surely agree that treating others as objects and believing that the world exists only to serve the self are obvious signs of inauthentic existence. The NPI studies mentioned earlier certainly support this notion, as those who score highly on Exploitativeness/Entitlement are the most likely to suffer from depression and anxiety (among other difficulties). However, not every expression of narcissism is an inauthentic one. Awareness is key to authentic living (Bugental, 1965). Research evidence supports the claim that those who adopt an unhealthy style of narcissism are least likely to identify the purpose of the 54 NPI (Raskin, 1980; Raskin & Terry, 1988). It could be argued that, while those who adopt the more adaptive aspects of narcissism are aware of their narcissism, those who score highly on factors such as Exploitativeness/Entitlement are unaware of their narcissistic motivational construct. Furthering the argument, it may be posited that healthy narcissists are relatively well aware of their personality structure and its operation in the world, whereas unhealthy narcissists lack awareness and insight into their thoughts, feelings, and behavior. Of course, this awareness would exist along a continuum, and empirical research using the NPI seems to support this assumption. Death anxiety and attitudes toward death are also key concepts in existential perspectives. Yalom (1980) has written at length about the impact of both conscious and unconscious awareness of one?s own mortality. Yalom (1980) stated that becoming truly aware of one?s own mortality is a terrifying and frightening experience for most. As humans who possess self-awareness, we are motivated to come to terms with our own mortality (along with coming to terms with the awareness that our lives are meaningless unless we provide that meaning ourselves). As a result, Yalom argued that humans construct psychological protections against this realization of our own frailty, such as belief in the ultimate rescuer. One might argue, as Shaw (2000) and others have, that narcissism is one such psychologically protective mechanism. Yalom (1980) proposed (and frequently supported the following assertion through research) that those who live fully, authentically, and with solid awareness often are the least likely to experience high levels of death anxiety. He noted that these people generally feel so fulfilled by their lives (and create meaning structures with well-defined 55 purposes) that they are able to accept their own eventual deaths without great levels of anxiety. Simply put, life fulfillment predicts less anxiety regarding one?s own demise. One might argue that narcissists who adopt a more healthy protective mechanism would suffer less death anxiety. As studies have suggested that healthy narcissism is related to indicators of psychological health and happiness, perhaps these narcissists are also less likely to be overly concerned about their own deaths. Those who adopt a less healthy and more maladaptive style of narcissism (e.g., one characterized by exploitation of others and/or a sense of entitlement) would be posited to experience greater death anxiety and more negative attitudes toward death. The final existential construct to be discussed is self-actualization. Maslow (1968, 1970) was the first to discuss self-actualization in detail. He argued that humans have an innate need to move toward personal growth, and that all human needs fall into a hierarchy (Maslow, 1968, 1970). The highest level on the need hierarchy is self- actualization. Maslow (1968, 1970) defined self-actualization as the need to fulfill one?s own potential, to fully use one?s own talents and gifts, to pursue one?s own true interests, and to seek growth beyond one?s current level of functioning. Self-actualization is not a goal state to be reached; it is a process of personal growth and fulfillment of potential (Maslow, 1968, 1970). Maslow (1982) described characteristics of self-actualizing people such as clear and efficient perceptions of reality; comfort with reality; spontaneity, simplicity, and naturalness; detachment and need for privacy; autonomy; peak experiences; feelings of attachment and identification with humanity; strong friendships 56 (though limited in number); lack of hostility in sense of humor; and a balance in polarities in personality structure. Given this picture of self-actualization and the self-actualizing person, parallels with narcissistic functioning become apparent. The most obvious connections lie in mentions of autonomy and relationships with others. High scorers on the adaptive aspects of the NPI often score highly on measures of autonomy and social responsibility/social desirability. Thus, those who possess the adaptive narcissistic traits (without possessing the less healthy ones) are more likely to seek fulfillment of their personal potential. Those who are high in the less healthy aspects, however, are more likely to attend to more basic needs instead. However, researchers have not yet directly assessed the relationship between narcissism (and its factors) and self-actualization, so these hypotheses have received only indirect support to date. Synthesis and Summary of Existential Conceptualization of Narcissism For purposes of the present study, narcissism is conceptualized as a personal motivational structure adopted (with and without conscious awareness) to cope with the very human experience of coming to terms with one?s own mortality and personal limitations. Humans naturally possess narcissistic traits and the capacity to act narcissistically; narcissism is in our genes, our socialization, our cognitive structures, our emotional experiences, and our spiritual lives. The question is one of which narcissism we choose. Narcissism falls along a spectrum of healthy to unhealthy functioning in all realms of existence (the biological, psychological, social, and spiritual). This motivational structure has consequences for every aspect of human functioning. 57 Should we choose narcissism composed of healthy, adaptive narcissistic facets, while also developing a healthy level of self-esteem, we are more likely to work toward self-actualization, experience less death anxiety, have more positive attitudes toward death, and live in an authentic, genuine manner. Should we choose narcissism composed of unhealthy, maladaptive facets, without attendant protective traits (e.g., healthy self- esteem), we are likely to need to meet more basic needs, experience more death anxiety, have more negative attitudes toward death, and live less authentically. Furthermore, the more adaptive our narcissistic motivational structures, the greater sense of meaning and purpose we will experience. Conceptual Measures of Existential Constructs Many tests and scales have been developed to measure existentially-oriented constructs. Examples include (but at not limited to) the Personal Orientations Inventory (Shostrom, 1966), the Personal Orientations Dimensions Scale (Shostrom, 1975), and the Death Anxiety Scale (Templer, 1970). Some of the more psychometrically sound instruments will be used in this study to examine the proposed complex relationships between narcissism and these constructs. Each of these instruments either directly or indirectly measures purpose in life/meaning in life, death anxiety, authenticity, or death anxiety/attitudes toward death. Each scale will also be discussed in terms of its proposed relationship to the construct of narcissism from an existential perspective. The Purpose- in-Life Test (PIL; Crumbaugh & Maholick, 1969) was developed from the theoretical stance taken by noted existential psychotherapist Viktor Frankl. The purpose of this test ?is to detect existential vacuum,? which, as mentioned earlier, can lead to psychological 58 distress (Crumbaugh & Maholick, 1969, p. 1). This scale has been used for two primary purposes: To serve as a clinical instrument to provide diagnostic and treatment guidelines, and to serve as a group-administered research instrument. This instrument will serve the latter purpose in the present study. The PIL is sometimes utilized in conjunction with a companion scale developed by Crumbaugh (1977a) called The Seeking of Noetic Goals Test (SONG). This test, derived from the same theoretical background as the PIL, is intended to measure the strength of an individual?s motivation to seek life meaning/purpose (Crumbaugh, 1977a). The author suggested that researchers pay close attention to the complex relationship between PIL and SONG scores. For example, a high PIL score and a low SONG score would suggest that an individual maintains a solid sense of purpose and is no longer seeking life meaning (Crumbaugh, 1977a). However, a low PIL score and a high SONG score suggests a lack of life purpose but a strong will to find it (Crumbaugh, 1977a). Individuals who possess high PIL and low SONG scores would likely exhibit healthier forms of narcissism, while those who display the opposite pattern would be likely to express unhealthier forms of narcissism. The PIL and the SONG are the measures of life meaning and purpose in this study. The second existential construct of interest is death anxiety. The Death Anxiety Scale (DAS; Templer, 1970) was developed out of the author?s frustration with the poor validity and reliability of measures of death anxiety. Items include such statements as ?I am very much afraid to die? and ?I often think about how short life really is? (Templer, 1970). High levels of death anxiety have consistently been shown to have a negative 59 effect on psychological health (Templer, 1970; Dougherty, Templer, & Brown, 1986; Lonetto & Templer, 1986). Healthier narcissists would likely exhibit less death anxiety, while unhealthier ones would show greater levels. The Authenticity Inventory (AI:3; Goldman & Kernis, 2004) was developed as a measure ?to assess the unobstructed operation of one?s true or core self in one?s daily enterprise? (Goldman & Kernis, 2002, p. 18). The theoretical basis of this instrument was derived from the ideas of Maslow (1968) and Rogers (1961). In essence, this scale attempts to tap the degree to which a person is able to behave in a consistent manner with their felt motives, feelings, desires, and thoughts (without succumbing to internal or external pressure to act in an inconsistent manner) (Goldman & Kernis, 2002). The AI:3 yields a total composite score and four factor scores, including Awareness, Unbiased Processing, Behavior Action, and Relational Orientation (Goldman & Kernis, 2004). One major study indicated that greater authenticity is related positively to indicators of healthy psychological functioning and negatively to indicators of unhealthy functioning. Inventory scores have been positively correlated with self-esteem and life satisfaction, while showing an inverse relationship to measures of contingent self-esteem and net negative affect (Goldman & Kernis, 2002). However, more studies using this scale are needed. Healthy narcissistic forms of expression should be positively related to authenticity scores. The final scale of interest measures the existential construct labeled self- actualization. The Short Index of Self-Actualization (SISA; Jones & Crandall, 1986) is a 15-item scale derived from Shostrom?s (1965, 1966) Personal Orientation Inventory. To 60 date, few studies have employed the SISA as a measure of self-actualization. Jones (1980) found that the SISA is effective in differentiating self-actualized and non-self- actualized individuals on the basis of score agreement with ratings from clinical and counseling psychologists. Jones (1980) also demonstrated that the SISA scores correlate positively and significantly with POI scores. Furthermore, Crandall, McCown, and Robb (1988) found that participation in assertiveness training groups led to an increase in SISA scores, and that this change continued to last for at least one year. As autonomy is purported to be a key characteristic of those high in self-actualization, the finding that greater assertiveness is positively related to self-actualization does not seem surprising. The present study would expect unhealthy narcissists to exhibit lower self-actualization scores, while healthy narcissists would display higher ones. 61 CHAPTER 3 METHOD Participants A sample of 190 students was recruited through requests to faculty members within counseling and psychology departments at Auburn University and Southern Union State Community College. Students were provided extra credit for participation in the study, as well as eligibility for a drawing (which would provide funds for books). The sample was comprised of 122 women (64.2%) and 68 men (35.8) with a mean age of 22 years (ranging from 19 to 54 years, SD = 5.21). Auburn University students comprised 101 of the total 190 (53.2%) students, while Southern Union students comprised the remaining 89 (46.8%). Regarding racial and ethnic considerations, 164 (86.3%) students identified themselves as White/Caucasian, 19 (10%) identified as African-American, 2 (1.1%) as Latino/Hispanic, 1 (0.5%) as Asian/Pacific Islander, 1 (0.5%) as Native American, and 3 (1.6%) as mixed race. In terms of sexual orientation, 168 students (88.4%) identified themselves as straight/heterosexual, 1 (.5%) as gay/homosexual/lesbian, 2 (1.1%) as bisexual, and 19 (10%) provided no response. Sixty (31.6%) of the students were freshmen, 41 (21.6%) were sophomores, 50 (26.3%) were juniors, 31 (16.3%) were seniors, and 7 (3.7%) were classified as graduate students (one student did not provide a classification). 62 Measures The measures used in this study included the Rosenberg Self-Esteem Scale (RSE; Rosenberg, 1965), the Emmons (1984, 1987) version of the Narcissistic Personality Inventory (NPI), the Purpose-in-Life test (PIL; Crumbaugh & Maholick, 1969), the Seeking of Noetic Goals Test (SONG; Crumbaugh, 1977b), the Death Anxiety Scale (DAS; Templer, 1970), the Authenticity Inventory (AI:3; Goldman & Kernis, 2004), and the Short Index of Self-Actualization (SISA; Jones & Crandall, 1986). Measures were chosen for various reasons, including solid psychometric properties, theoretical relevance and congruence with targeted constructs, and practical factors. The Rosenberg Self-Esteem Scale (RSE) (Rosenberg, 1965). This scale has been employed in numerous studies as a measure of global self-esteem, and it has been employed routinely in studies that involve narcissism as a primary variable of interest (e.g., Watson, Hickman, & Morris, 1996; Watson, Little, Sawrie, & Biderman, 1992). The scale is composed of ten items (e.g., ?I feel that I have a number of good qualities.?); half are positively worded, while the other half are negatively worded (and, therefore, are reverse scored). Each item is rated on a four-point scale (from ?strongly agree? to ?strongly disagree?). Responses are summed to yield a single scale score. Scores range from 10 to 40, with higher scores indicating higher levels of self-esteem. In terms of reliability, alpha coefficients generally range from .72 to .90 (Gray-Little, Williams, & Hancock, 1997; Robins, Hendin, & Trzesniewski, 2001), while test-retest reliability estimates have been reported at .88 (Robins, et. al., 2001). 63 The Emmons (1984, 1987) NPI. The Emmons (1984, 1987) version of the NPI appears to be the most widely employed and best validated measure of narcissism to date. This version of the NPI contains 54 items. Responding to each item involves choosing between two statements; one statement reflects a narcissistic orientation (e.g., ?If I ruled the world it would be a much better place.?), while the other does not (?The thought of ruling the world frightens the hell out of me.?). Scores are calculated by totaling the number of chosen narcissistic responses (for total scores and for each subscale). The Emmons (1984, 1987) NPI yields a total score and four subscale scores (Exploitativeness/Entitlement, Leadership/Authority, Superiority/Arrogance, and Self- Absorption/Self-Admiration). Regarding reliability for the scale, Cronbach?s alpha for the total NPI has been measured at .87, while the factor score alphas generally range from .68 to .81 (Emmons, 1984, 1987). Validity of the scale has been supported by two separate factor analyses (Emmons, 1984, 1987) that provided the same subscale structure (with subscale intercorrelations falling in a range from .13 to .57), by correlation with the MCMI Narcissistic scale at .66 (Prifitera & Ryan, 1984), through demonstration that the NPI and MCMI agreed in classifying psychiatric patients on the basis of diagnosis 74% of the time (Prifitera & Ryan, 1984), and by the finding that peer ratings of personality correlated with total NPI scores at .64 (Emmons, 1984). The Purpose-in-Life Test (PIL) (Crumbaugh & Maholick, 1969) and Seeking of Noetic Goals Test (SONG) (Crumbaugh, 1977a). These tests were examined together, as they were created as companion assessment instruments. The PIL is an instrument designed to assess the presence (or absence) of a sense of meaning and purpose in life. 64 The PIL is composed of three parts. Only Part A of the PIL was used in this study. Parts A and B of the PIL required short-answer or essay responses and were not as amenable to quantitative study. The SONG is a companion scale to the PIL designed to measure a person?s strength of motivation to find a sense of purpose and meaning in life. Both scales consist of 20 statements which respondents rate on a seven point, Likert-type scale. An example of a typical PIL item is ?My personal existence is . . . (rated from ?utterly meaningless and without purpose? to ?very purposeful and meaningful?), whereas a typical SONG item would be ?I think about the ultimate meaning of life . . . (rated from ?never? to ?constantly?). Scores for the PIL and SONG are totaled separately to yield two single scale scores ranging from 20 to 140. Norms have been developed for both instruments, though these norms were published some time ago and have yet to be updated (Crumbaugh & Maholick, 1969; Crumbaugh, 1977). Split-half reliability of the PIL has been estimated at .81, with a Spearman-Brown correction producing a .90 estimate (Crumbaugh & Maholick, 1964). Reliability of the SONG has been estimated at .71 using a Pearson Product-Moment analysis, with a Spearman-Brown correction showing an estimate of .83 (Crumbaugh, 1977b). Crumbaugh (1968, as cited in Crumbaugh & Maholick, 1969) correctly predicted the order of PIL means for four separate samples. Crumbaugh (1977a) found that patients in an alcoholism treatment unit demonstrated a significant decrease in SONG scores when treated using logotherapy techniques, while patients exposed to general treatment regimens displayed a non-significant increase in SONG scores. Crumbaugh (1977a) 65 found that SONG scores generally correlate in a range from .27 to -.52 with PIL scores, which indicated that PIL and SONG score correlations are unpredictable. The Death Anxiety Scale (DAS) (Templer, 1970). The original form of the DAS was a 15-item, true-false, forced choice scale intended to measure a wide range of life experiences with death (Templer, 1970). A representative item reads: ?I often think about how short life really is.? Scores ranged from zero to fifteen, with greater scores indicating greater death anxiety. This scale has not changed much since its original inception, though McMordie (1979) modified the scale such that items were rated on a Likert-type scale (from one, ?Strongly Disagree? to five, ?Strongly Agree?), with total scores ranging from fifteen to seventy-five. The McMordie (1979) scale revision was employed for purposes of this study. A solid body of reliability evidence has been amassed for the DAS. Templer (1970) reported a test-retest reliability estimate of .83, with an internal consistency estimate of .76 (using the Kuder-Richardson Formula 20). Construct validity studies have shown that psychiatric patients who verbalized greater death anxiety scored significantly higher than those patients who did not verbalize high death anxiety, and that college students? DAS scores correlated positively with scores on the Fear of Death Scale (.74), the Si scale of the MMPI (.25), the Welsh Anxiety Scale (.39), the Manifest Anxiety Scale (.36), and with endorsement of a high number of emotionally loaded words (.25) (Templer, 1970). DAS scores correlated negatively with the following MMPI scales: K (- .43) and Pd (-.24) (Templer, 1970). 66 The Authenticity Inventory (AI:3) (Goldman & Kernis, 2004). The version of the AI used in this study will be the most recently developed and validated form, the AI:3. The AI:3 provides a composite score, as well as four subscale scores (Awareness, Unbiased Processing, Behavior Action, and Relational Orientation). As the subscales were not employed in this study (due to the author?s interest in comparing NPI scales to the full scale), information regarding their scoring, validity, and reliability has been omitted. The scale is composed of 45 items, rated on a five-point scale from Strongly Disagree to Strongly Agree. Approximately half of the items are reverse-scored. Higher scores indicate greater levels of authenticity. A sample item would be ?I find it very difficult to critically assess myself.? Goldman and Kernis (2004) reported an internal consistency of .90 for the entire scale. Scores from the original version of the AI were shown to correlate positively with scores on a Life Satisfaction measure (.40) and Rosenberg SE scores (.33), while correlating negatively with scores from a measure of contingent SE (-.27) and an assessment of net negative affect (-.31). Hence, initial evidence exists that high AI:3 scores correlate positively with measures of psychological health and negatively with measures of psychological disturbance. The Short Index of Self-Actualization (SISA) (Jones, 1980). The SISA was developed as a shorter, more research-friendly alternative to the well-validated Shostrom (1965, 1966) Personal Orientation Inventory (POI) and Personal Orientation Dimensions (POD; Shostrom, 1975). The most recently developed SISA is a 15-item, six-point rating scale ranging from ?Strongly Agree? to ?Strongly Disagree.? A representative item reads 67 as ?it is better to be yourself than to be popular.? Reverse scoring must be employed for approximately half of the items. Items are totaled to yield a single scale score ranging from 15 to 90, with greater scores indicating greater levels of achieved self-actualization. In terms of reliability, Cronbach?s alpha for the SISA has ranged from .63 to .68 (Flett, Blankenstein, & Hewitt, 1991; Jones, 1980; McLeod & Vodanovich, 1991), while test-retest reliability has been measured at .69 (Jones, 1980). Validity evidence has been provided through successful classification (according to SISA score) of self-actualized vs. non-self-actualized individuals (after categorical nomination by clinical and counseling psychologists) and correlation with POI scores at .67 (Jones, 1980). Procedure Participants were provided with an informed consent document to review and sign. Then, they were given a packet containing a demographic questionnaire and each of the aforementioned scales. The demographic questionnaire was provided first, while the order of the included scales was counterbalanced to prevent order effects. Aside from the demographic information, no other personal information was collected from the participants. All measures were group administered in a classroom setting, and the packet generally took from 40 to 60 minutes to complete. Hypotheses Examined by This Study 1. The total NPI score will predict greater self-esteem, a greater sense of purpose in life, greater capacity for self-actualization, greater authenticity, and reduced death anxiety. No prediction was made regarding the SONG score. 68 2. The Exploitativeness/Entitlement factor score will predict a lack of purpose in life, avoidance of searching for a sense of purpose, greater death anxiety, lowered capacity for self-actualization, less authenticity, and lowered self-esteem. This prediction will be augmented by statistically partialing out the effects of the proposed healthier facets of narcissism (Leadership/Authority, Superiority/Arrogance, and Self- Absorption/Self-Admiration). 3. The Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self- Admiration NPI subscale scores (to varying degrees) will predict a greater sense of purpose in life, less death anxiety, greater capacity for self-actualization, greater authenticity, and greater self-esteem. These predictions should strengthen when the effects of Exploitativeness/Entitlement are partialled out of the predictive equation. No prediction of the SONG score (indicative of a search for meaning and purpose) was made, as one may or may not continue searching if one already has a sense of meaning and purpose in life. 69 CHAPTER 4 RESULTS Overview This chapter provides detailed descriptions of the statistical results of analyses conducted to test the hypotheses developed for this dissertation study. The first three sections of this chapter center on the findings from analyses that directly examined the three study hypotheses discussed in Chapters 1 and 3 of this manuscript. These hypotheses, which guided the statistical analyses used and interpretation of the study results, are as follows: 1. The total NPI score will predict greater self-esteem, a greater sense of purpose in life, greater capacity for self-actualization, greater authenticity, and reduced death anxiety. No prediction was made regarding the SONG score. 2. The Exploitativeness/Entitlement factor score will predict a lack of purpose in life, avoidance of searching for a sense of purpose, greater death anxiety, lowered capacity for self-actualization, less authenticity, and lowered self-esteem. This prediction will be augmented by statistically partialing out the effects of the proposed healthier facets of narcissism (Leadership/Authority, Superiority/Arrogance, and Self- Absorption/Self-Admiration). 3. The Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self- Admiration NPI subscale scores (to varying degrees) will predict a greater sense of 70 purpose in life, less death anxiety, greater capacity for self-actualization, greater authenticity, and greater self-esteem. These predictions should strengthen when the effects of Exploitativeness/Entitlement are partialled out of the predictive equation. No prediction of the SONG score (indicative of a search for meaning and purpose) was made, as one may or may not continue searching if one already has a sense of meaning and purpose in life. The remaining sections of this chapter focus on (1) statistical analyses of the reliability and validity of the measures used in the study, and (2) findings related to the analysis of demographic variables assessed during the study. Hypothesis 1: Total NPI as Predictor of Existential Functioning It was hypothesized that higher total Narcissistic Personality Inventory scores would predict higher scores on the measures of self-esteem, purpose-in-life, self- actualization, and authenticity. It was also hypothesized that higher total NPI scores would predict lower scores on the death anxiety measure. No prediction was made regarding the relationship between the NPI scores and scores from the measure designed to assess the degree to which a person was actively seeking a sense of meaning/purpose in life (the Seeking of Noetic Goals scale). These hypotheses were examined through zero-order correlational analyses, in which total NPI scores were entered as the predictor variables. The findings from these analyses may be reviewed in Table 1. Before engaging in the zero-order correlational analyses, statistical analyses were performed to evaluate key assumptions underlying the use of zero-order regression procedures. These analyses included examination of outliers, skewness, kurtosis, and 71 those related to linearity and homoscedasticity of residuals for all scale scores included in the analysis. These initial analyses indicated that no key assumptions were violated in the performance of the zero-order correlation procedures. Therefore, no adjustments were made in analyzing the data entered. Data for two participants was excluded due to missing data points, leaving 188 cases included in the final analyses. Table 1 Total NPI Correlation Analysis with All Existential Measures Variable r p value (2-tailed) Self-Esteem .28 < .0001 Purpose-in-Life .20 < .05 Seeking of Noetic Goals .16 < .05 Self-Actualization .10 NS Death Anxiety .01 NS Authenticity .13 NS Overall, the hypotheses for the total NPI scores in relation to the existential measure scores obtained mixed statistical support. As hypothesized, higher total NPI scores did predict greater self-esteem and a greater sense of purpose-in-life. These predictions were statistically significant at the .0001 and .05 levels, respectively. However, though statistically significant, these predictions remained relatively weak to moderate (with R values ranging from .20 to .28). Counter to the hypothesized predictive ability of the NPI, NPI scores demonstrated no significant predictive ability in relation to scores from the self-actualization, authenticity, or death anxiety measures. Though no 72 hypothesized prediction was made for the NPI and Seeking of Noetic Goals measure, a relatively weak but statistically significant predictive effect was obtained. Greater total NPI scores weakly predicted a greater search for sense of purpose/meaning in life (R = .16). The hypothesis that greater total NPI scores would predict greater self-esteem, purpose-in-life, authenticity, and self-actualization (while also predicting reduced death anxiety) obtained mixed support from this data. The total NPI score did not significantly predict scores on either the self-actualization or the death anxiety measure. Secondly, even on those instruments that the NPI total score did predict scores in a statistically significant manner, the resulting correlations were relatively small (ranging from .16 to .28). Therefore, though NPI total scores did predict scores on many of the existential scales in a statistically significant manner, the NPI score did not seem to predict scores on these scales in a meaningful way. Hypothesis 2: The Exploitativeness/Entitlement Subscale and its Predictive Abilities The second hypothesis centered on the ability of the Exploitativeness/Entitlement subscale score to predict levels of functioning on each existential measure. It was hypothesized that higher Exploitativeness/Entitlement (EE) scores would predict lower scores on the purpose-in-life, seeking of meaning/purpose in life, self-actualization, authenticity, and self-esteem measures. It was also hypothesized that higher EE scores would predict higher scores on the death anxiety measure. Furthermore, it was hypothesized that each of the aforementioned predictions would increase in magnitude and statistical significance when the effects of the other NPI subscales 73 (Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration) were statistically partialed out of the regression equation. The hypothesized predictions were examined first through examination of the zero-order (or simple) regression analyses of the EE subscale score with each existential scale score, in which the only predictive factor entered into the regression equation was EE score alone. After these analyses were conducted, the EE scores were then subjected to a partial regression analysis. This analysis was conducted by entering the EE score into the regression equation alone in the first step, then by statistically accounting for the effects of the other three NPI subscales in the second step of the regression equation. Table 2 contains the results of both the zero-order and partial regression analyses related to the Exploitativeness/Entitlement subscale. In regard to the zero-order findings (on the left side of Table 2) related to the NPI Exploitativeness/Entitlement (EE) subscale (e.g., without the effects of the other subscales factored out), the EE subscale score was found to significantly predict scores on all of the existential scales except the Rosenberg Self-Esteem scale (which was counter to the hypothesized predictive relationship). Greater levels of Exploitativeness/Entitlement were found to predict a more intense search for meaning in life (seeking of noetic goals) and greater death anxiety, while also predicting lower levels of authenticity, self-actualization, and achieved sense of purpose- in-life. These findings are consistent with those expected from review of previous studies using the NPI, as the Exploitativeness/Entitlement subscale score often predicts greater psychological dysfunction and less psychological health (even when the effects of the other, usually healthier, NPI factors remain in play statistically). 74 Table 2 Zero-order and Partial Correlations of Existential Constructs with Exploitativeness/Entitlement (EE), Leadership/Authority (LA), Superiority/Arrogance (SA), and Self-Absorption/Self-Admiration (SS) NPI Factors Narcissistic Personality Inventory ________________Zero-order_______________ _________________Partial________________ Existential Meas. EE LA SA SS EE LA SA SS Self-Esteem -.07 .37*** .23*** .34*** -.30*** .43*** .28*** .38*** Purpose-in-Life -.18** .38*** .21** .26*** -.41*** .49*** .30*** .35*** Seek Noetic Goals .31*** .04 .03 .07 .32*** -.09 -.10 -.04 Self-Actualization -.19** .36*** .19** .15* -.37*** .47*** .28*** .23** Death Anxiety .18* -.17* -.20** .06 .27*** -.26*** -.29*** .00 Authenticity -.24*** .34*** .19** .20** -.44*** .47*** .31*** .31*** Note: * P <.05; ** P < .01; *** P < .001. Partial correlations involve examining the effects of EE while controlling for LA, SA, and SS NPI factors and vice versa. 75 The findings regarding the predictive validity of the Exploitativeness/Entitlement subscale score when the effects of the other three subscales were partialed out of the regression equation (right side of Table 2) also were consistent with findings from previous studies using the NPI. When the effects of Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration were factored out of the regression equation, the Exploitativeness/Entitlement subscale score predicted an even greater level of psychological dysfunction. Hence, without the protective effects of the other facets of narcissism, increased Exploitativeness/Entitlement predicted even greater death anxiety (.18 zero-order vs. .27 partial) and even more reduced levels of authenticity (-.24 vs. -.44), self-actualization (-.19 vs. -.37), and purpose-in-life (-.18 vs. -.41). Also, whereas the Exploitativeness/Entitlement score did not significantly predict self-esteem scores (-.07) before the effects of the other subscales were partialed out, the EE score did predict decreased self-esteem (-.30) after these effects were statistically removed. The only score prediction relatively unaffected by removing the effects of the other subscales was the seeking of noetic goals (or search for purpose in life). This regression coefficient did not change significantly, despite the statistical removal of the effects of the other subscales (though it did remain a moderately positive correlation of the same statistical significance level). Hypothesis 3: The Leadership/Authority, Superiority/Arrogance, and Self- Absorption/Self-Admiration Subscales and their Predictive Abilities The final hypothesis examined in this study centered on the abilities of the other remaining NPI subscale scores to predict scores on the existential measures. It was 76 hypothesized that higher Leadership/Authority (LA), Superiority/Arrogance (SA), and Self-Absorption/Self-Admiration (SS) NPI subscale scores (to varying degrees) would predict higher scores on the purpose-in-life, self-actualization, authenticity, and self- esteem measures. It was also hypothesized that higher LA, SA, and SS subscale scores would predict lower scores on the death anxiety measure. No prediction was made regarding these subscale scores in relation to the Seeking of Noetic Goals score. Finally, it was hypothesized that these predictive relationships would increase in both magnitude and statistical significance when the presumed psychologically harmful effects of the Exploitativeness/Entitlement subscale were removed from the regression equation. The zero-order findings for these NPI subscales may be found on the left side of Table 2. These findings were consistent with findings from previous studies using the NPI and its subscales, in that the Leadership/Authority (LA), Superiority/Arrogance (SA), and Self-Absorption/Self-Admiration (SS) scores predicted better psychological functioning (and, in turn, lower levels of dysfunction). The Leadership/Authority subscale was, by far, the best predictor of psychological health, in that it predicted greater self-esteem (.37), greater sense of achieved purpose-in-life (.38), greater self- actualization (.36), and greater authenticity (.34). This subscale also predicted reduced death anxiety (-.17). In terms of magnitude, the Superiority/Arrogance score was the next best predictor of psychological health. This subscale score also predicted greater self-esteem (.23), greater sense of purpose-in-life (.21), greater self-actualization (.19), and greater authenticity (.19) (though to a lesser degree than LA scores). However, the SA score also 77 predicted a reduction in death anxiety similar in magnitude to that of the LA subscale (-.20). Finally, though the Self-Absorption/Self-Admiration subscale score seemed the least solid predictor of the three in terms of psychological health, its findings remained consistent with those from previous studies as well. The SS subscale score predicted greater self-esteem (.34), sense of purpose-in-life (.26), self-actualization (.19), and authenticity (.20). However, the SS subscale score did not predict death anxiety (.06) as the other two assumed healthy subscales did. It is worthy of note that none of the healthier narcissism subscales predicted in either direction the search for purpose in life (or seeking of noetic goals). When the unhealthy effects of the Exploitativeness/Entitlement NPI subscale were partialed out of the regression equation, the healthy effects of the Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration narcissism facets became even statistically clearer and stronger. For the Leadership/Authority subscale, removal of the effects of EE predicted even greater self- esteem (.37 vs. .43), sense of achieved purpose-in-life (.38 vs. .49), self-actualization (.36 vs. .47), and authenticity (.34 vs. .47), while also predicting even more reduced levels of death anxiety (-.17 vs. -.26). For the Superiority/Arrogance subscale, removal of EE effects also predicted significant increases in self-esteem (.23 vs. .28), purpose-in-life (.21 vs. .30), self-actualization (.19 vs. .28), and authenticity (.19 vs. .31), while also predicting even more reduced levels of death anxiety (-.20 vs. -.29). Finally, removal of EE effects from the Self-absorption/Self-admiration subscale predicted greater self- esteem (.34 vs. .38), sense of achieved purpose-in-life (.26 vs. .35), self-actualization (.15 78 vs. .23), and authenticity (.20 vs. .31). However, removal of EE effects did not lead to a prediction of decreased death anxiety with increasing levels of SS (.06 vs. .00). In summary, the findings from the zero-order and partial regression analyses involving the Leadership/Authority (LA), Superiority/Arrogance (SA), and Self- Absorption/Self-Admiration (SS) NPI subscales were consistent with previously conducted studies. Zero-order regression analyses provided strong indications that increased levels of LA, SA, and SS predicted greater levels of psychologically healthy characteristics (and, therefore, improved psychological functioning) from an existential view. These increased levels of psychologically healthy characteristics and healthy functioning increased even further when the effects of the harmful Exploitativeness/Entitlement NPI subscale were removed statistically. On the other hand, regression analyses demonstrated that the Exploitativeness/Entitlement subscale score predicted reduced levels of psychologically healthy characteristics, and that these predictions were strengthened when the effects of the LA, SA, and SS subscales were removed statistically. Reliability and Validity of Measures Used in This Study To further reinforce the validity of the findings discussed above, reliability and validity data were collected on each of the measures used in this dissertation study. Means and standard deviations of the scales obtained with this sample are presented in Table 3. In some cases, difficulties were encountered in locating scale norms for the instruments used in this study, as many of these scales were not developed for use as 79 clinical instruments. However, these data will be discussed in relation to the available scale norms whenever possible. Table 3 Mean Scores, Scale Range, and Standard Deviations for Narcissistic Personality Inventory (Including Subscales) and Existential Scales Scale Mean Scale Range SD Narcissistic Personality Inventory (NPI) 20.51 0 - 54 8.12 NPI Exploitativeness/Entitlement 2.81 0 - 11 2.11 NPI Leadership/Authority 4.62 0 - 9 2.50 NPI Superiority/Arrogance 4.13 0 - 11 2.48 NPI Self-Absorption/Self-Admiration 4.04 0 - 9 2.27 Rosenberg Self-Esteem Scale 30.88 10 - 40 4.93 Purpose-in-Life Scale 105.52 20 - 140 16.23 Seeking of Noetic Goals Scale 76.07 20 ? 140 15.49 Short Index of Self-Actualization 61.55 15 - 90 7.04 Death Anxiety Scale 43.32 15 - 75 10.17 Authenticity Inventory ? 3 159.13 45 - 225 16.65 Raskin and Terry (1988) reported normative data for 1,018 college students in regard to total NPI scores. The mean for the Raskin and Terry (1988) sample was 20.59 (SD = 7.93). The sample in the present study produced a mean score of 20.51 (SD = 8.12). Therefore, the present sample did not differ significantly from a much larger sample of college students in regard to total NPI score. Unfortunately, Raskin and Terry (1988) did not provide norms for the NPI subscales with their sample. It is difficult to 80 know how this particular sample would compare to others without such norms. However, Watson, Taylor, and Morris (1987) provided mean scores NPI factor scores for their sample of 203 college students, which serve as a crude basis of comparison in the absence of solid normative data. Watson, Taylor, and Morris (1987) reported means of 2.91 (Exploitativeness/Entitlement), 5.12 (Leadership/Authority), 3.62 (Superiority/Arrogance), and 3.37 (Self-Absorption/Self-Admiration) for the four NPI subscales. The means produced from this sample seem comparable to these previously reported means. Most striking is the replication of a relatively low Exploitativeness/Entitlement score as compared to the other subscales for a college student population. Normative data for the Rosenberg Self-Esteem (RSE, 1965) scale were difficult to locate. The RSE is conventionally scored on a Likert-type scale. However, the Likert scale used varies depending on the purpose of the study using the RSE and the researcher?s preference. Gray-Little, Williams, and Hancock (1997) reported that the scale can be scored on a 4-point, 5-point, or 7-point Likert scale. Hence, the lack of uniformity in scoring the RSE makes mean comparisons difficult. For example, Watson, Varnell, and Morris (1999) reported a mean score of 22.84 (SD = 4.68) for their sample of 400 undergraduate students. This would seem a significant mean difference from the 30.88 mean score obtained with the present sample. However, Watson, Varnell, and Morris (1999) anchored the RSE scale from zero to three (which would provide a score range from 0 ? 30), whereas the present study anchored the scale from one to four (with a possible score range from 10 ? 40). If one uses the upper end of the scale range as a 81 reference point for comparing the two samples, it would seem that the present sample?s RSE mean score would be comparable to the mean obtained in the Watson, Varnell, and Morris (1999) study. Both the Purpose-in-Life Test Manual (Crumbaugh & Maholick, 1969) and an article by Crumbaugh and Henrion (1988) reported that the normative mean score for individuals not identified as psychiatric patients was 112.42 (based on a sample of 1,151 individuals). The obtained mean for this sample was 105.52, somewhat lower than the expected mean. Crumbaugh and Maholick (1969) suggested that 102 be considered the cutoff score between ostensibly normal populations and those with diagnosable psychiatry illnesses. The mean for this sample approached this boundary. However, it would appear that no published study has assessed what would be a normative score for a college student population. Perhaps, given the developmental concerns of most college- aged students, their achieved sense of purpose-in-life needs to be assessed differently than for middle-aged adults who are considered psychologically healthy. In regard to the Seeking of Noetic Goals Test (SONG; Crumbaugh, 1977), the test manual reported a mean score of 73 for adults without diagnosable mental illness (as opposed to a mean of 85 for those who did meet criteria for a psychiatric diagnosis). The mean for the present sample was 76.07, which fell much closer to the expected norm for healthy adults. Normative data for the Short Index of Self-Actualization (Crandall, McCown, & Robb, 1988) have not yet been reported, as this particular instrument has not been widely used in research. Based on where the mean score for this particular sample (61.55) fell in the possible range of scores (15 ? 90), one might tentatively conclude that this sample fell 82 in the moderate range of self-actualization (as the score was higher than the mid-point, but considerably lower than the extreme upper range). Similarly, norms for the McMordie (1979) version (scored on a Likert-type scale) of the Death Anxiety Scale (Templer, 1970) have not yet been developed. This college student sample evidenced a mean of 43.32, which is 5.82 points higher than the midpoint of the scale?s range (37.5). For the Authenticity Inventory ? 3 (Goldman & Kernis, 2004), no group norms have been published to date. However, Goldman and Kernis (2002) presented mean scores for 79 college students who were given an earlier version of the authenticity measure. Scoring for the earlier, 44-item (one fewer than the most recent version) version is similar. The mean for the Goldman and Kernis (2002) sample was 154.52 (SD = 13.02). The present sample obtained a mean of 159.13, which did not appear statistically different from the 154.52 mean for the Goldman and Kernis (2002) sample of students. The reliability of the Narcissistic Personality Inventory (NPI) and its subscales was assessed through use of Cronbach?s Alpha (Cronbach, 1951). The alphas obtained for the total NPI and all subscales may be found within Table 4. The reliability findings were generally consistent with those obtained by other researchers. However, the alpha level for the Exploitativeness/Entitlement scale was slightly lower than the .68 level reported in previous work (Emmons, 1984, 1987). Item removal, according to the full reliability analysis, would not have improved the internal consistency of this particular scale. Validity of the Narcissistic Personality Inventory was assessed through examination of the intercorrelations between the total scale score and each subscale 83 score. Table 4 presents the intercorrelations of the NPI Total Scale score and all four subscale scores. It should be noted that all four subscale scores correlated significantly and positively with the NPI total score, with a range from .67 (Exploitativeness/Entitlement) to .76 (Superiority/Arrogance and Self-Absorption/Self- Admiration). In addition, each of the subscales correlated significantly and positively with one another, though not as highly as with the total NPI score. These correlations ranged from .62 (between Superiority/Arrogance and Leadership/Authority) to .34 (Self- Absorption/Self-Admiration and Exploitativeness/Entitlement). The average correlation between subscales was .43. These findings suggest that the NPI is a psychometrically sound instrument, a finding that supports previous studies using the NPI. The NPI subscales did appear to measure separate facets of narcissism, while also each correlating acceptably with the full scale score. Table 4 Intercorrelations of NPI Total and Subscale Scores Scale/Subscale NPI Total NPI EE NPI LA NPI SA NPI SS NPI Total (.85) .67 .74 .76 .76 NPI EE (.60) .38 .37 .34 NPI LA (.77) .62 .44 NPI SA (.67) .44 NPI SS (.68) Note: NPI = Narcissistic Personality Inventory; EE = NPI Exploitativeness/Entitlement Subscale; LA = NPI Leadership/Authority Subscale; SA = NPI Superiority/Arrogance Subscale; SS = NPI Self- Absorption/Self-Admiration Subscale. All correlations were statistically significant at p<.0001 level. Alpha coefficients displayed on the main diagonal. 84 Reliability and validity evidence for the existential scales was assessed in a similar manner to that used for the Narcissistic Personality Inventory. Table 5 presents the Cronbach?s coefficient alpha level found with the present sample when examining each of the remaining existentially-oriented scales. For the most part, the coefficient alpha levels for these scales were solid and impressive. With the exception of the Short Index for Self-Actualization (SISA), the observed Alpha ranged from .81 (the Death Anxiety Scale) to .90 (the Purpose-in-Life Scale). However, the SISA?s observed Alpha level was measured at .57, which is significantly lower than any other full scale employed in this study. Further examination of the reliability analysis revealed that removing items from the SISA would not improve reliability of the scale markedly. Table 5 also contains validity evidence for the existential scales through examination of the zero-order correlations of these scales. From a validity-confirming perspective, it was predicted that the existentially-oriented scales would correlate to a moderate extent, but would not correlate highly; as such findings would indicate that these measures are assessing different, but related, constructs. It was also predicted that the existential measures that purport to assess psychological health would correlate positively with one another (e.g., self- esteem, self-actualization, purpose-in-life, authenticity), while correlating negatively with the death anxiety measure. No prediction was made regarding the SONG correlations, as the search for purpose in life and its effect on psychological health is more difficult to forecast. The first prediction centered on assessing existential scale validity by examining the correlations among these instruments. The existential scale correlations ranged from 85 Table 5 Zero-Order Correlations of Existential Content Scales Scale RSE PIL SONG SISA DAS AI-3 Rosenberg Self-Esteem (RSE) (.88) .71 -.50 .53 -.31 .62 Purpose-in-Life (PIL) (.90) -.52 .49 -.36 .62 Seeking of Noetic Goals (SONG) (.85) -.35 .31 -.34 Short Index of Self-Actualization (SISA) (.57) -.44 .65 Death Anxiety Scale (DAS) (.82) -.28 Authenticity Inventory ? 3 (AI-3) (.85) Note: All correlations presented were statistically significant, p < .01. Alpha levels are presented on the main diagonal. 86 -.49 (between the SONG and the Rosenberg Self-Esteem scores) to .71 (between the Purpose-in-Life scores and the Rosenberg Self-Esteem scores). All correlations (as displayed in Table 5) were determined statistically significant (p < .0001). Though a correlation of .71 between two of the scales is somewhat higher than ideal, the prediction that these measures are truly assessing different, but related, constructs was supported. The second prediction regarding scale validity was that the existential measures that assessed positive aspects of functioning (self-esteem, authenticity, self-actualization, purpose-in-life) would correlate positively, while also correlating negatively with the measure of death anxiety. The findings again supported this prediction and the construct validity of these existential scales. The correlations among the scores derived from the Rosenberg Self-Esteem (RSE) scale, the Purpose-in-Life scale (PIL), the Short Index of Self-Actualization (SISA), and the Authenticity Inventory ? 3 (AI:3) were all in the positive direction, ranging from .49 (between the PIL scores and the SISA scores) to .71 (between the PIL scores and the RSE scores). The average of the RSE, PIL, SISA, and AI:3 scale correlations was .60. Again, all of these correlations were statistically significant (p < .0001). Regarding correlations of the self-esteem, purpose-in-life, self-actualization, and authenticity measures with the death anxiety measure (the Death Anxiety Scale, or DAS), all were in the predicted negative direction. These correlations ranged from -.28 to -.44, with an average correlation of -.35. All correlations were statistically significant (p < .0001). This finding further supported the construct validity of the existential scales, as one would conceptually expect those with greater death anxiety to have lower 87 self-esteem, a diminished sense of purpose-in-life, less capacity for self-actualization, and diminished authenticity. Finally, although no predictions were made regarding the Seeking of Noetic Goals scale (SONG) in relation to the other existential scales, findings emerged that are worthy of note. The SONG emerged as a possible indicator of psychological difficulties in this study, based on its correlations with the other existential scales employed. While the SONG scores correlated negatively with the scores from measures of psychological health (self-esteem, self-actualization, authenticity, purpose-in-life) with a range from - .34 to -.49, it also correlated positively with scores from the measure of death anxiety (r = .31). These findings suggested that the search for a sense of meaning and purpose in life was not experienced as a positive event by this particular sample, and that an active search for purpose and meaning in life correlated with enhanced death anxiety, diminished self-worth, reduced capacity for self-actualization, and reduced levels of authenticity. Preliminary Analyses of Demographic Variables in Relation to the Measures Employed No predictions were made regarding the supposed impact of the demographic variables assessed in this study. However, attempts were made to assess the impact of these variables nonetheless to rule out any effects that they might have on the outcome of the study. Unfortunately, the lack of variability within the sample for many of these variables rendered assessment of the impact a moot point. For example, the vast majority of the sample was White (86.3%), straight/heterosexual (88.4%), and of a Protestant form of Christian faith (78%). So few individuals responded in a manner inconsistent with this 88 trend that engaging in statistical analysis of group differences was viewed as of little import, because such analysis would provide very little information (or, worse yet, misleading information). Hence, the impact of race/ethnicity, religious faith, and sexual orientation was not assessed due to a lack of variability in the sample. This lack of variability on these particular demographic variables is a significant limitation of this dissertation study. However, enough variability did exist within the sample when considering the college participants attended, the age of participants, their sex, and their classification (freshman, sophomore, and so on). It was determined, through analysis using a series of independent-samples t-tests, that no significant score differences existed on any of the instruments used in the study (or on their subscales) between the Auburn University students and the Southern Union Community College students. Through use of a series of one-way Analyses of Variance (ANOVAs) comparing college classification groups to each of the scales used in the study, it was determined that no significant score differences existed based on college classification. College attended and student classification appeared to have no significant impact on the study and its overall findings. Interestingly, some significant differences in relation to particular scales and subscales did emerge when considering age and sex of participants. Table 6 contains results from a zero-order correlational analysis of age in relation to scores from the NPI (and all subscales) and all existentially-oriented instruments. Increasing age for this sample correlated positively (and significantly) with purpose-in-life, self-actualization, self-esteem, and authenticity scores, while correlating negatively (and significantly) with 89 total narcissism, exploitativeness/entitlement narcissism, superiority/arrogance narcissism, search for a sense of purpose in life, and death anxiety scores. While these correlations were statistically significant, they remained relatively small (the largest was measured at .26). However, age did appear to offer a relative advantage in relationship to psychological functioning for this particular sample based on this pattern of findings. Table 6 Correlations of Age with Scores from the Narcissistic Personality Inventory, NPI Subscales, and all Existential Scales Variable Correlation Coefficient p value (2-tailed) Narcissism -.23 < .01 Exploitativeness/Entitlement -.20 < .01 Leadership/Authority -.04 NS Superiority/Arrogance -.15 < .05 Self-Absorption/Self-Admiration -.14 NS Self-Esteem .19 < .01 Purpose-in-Life .19 < .01 Seeking of Noetic Goals -.18 < .05 Self-Actualization .26 < .01 Death Anxiety -.25 < .01 Authenticity .23 < .01 One way, independent-samples t-tests were conducted to examine the impact of participant gender on narcissism and existential scale scores. Though no significant mean 90 differences emerged for gender in relation to purpose-in-life, seeking of purpose, authenticity, self-actualization, or self-esteem scores, significant differences were noted for NPI total score, NPI Leadership/Authority and Superiority/Arrogance scores, and death anxiety scores. The mean score for males taking the total NPI in this sample was 22.43 (SD = 8.77), while the female mean was 19.46 (SD = 7.58). This mean score difference was statistically significant, F (1, 187) = 5.948, p = .016. In part, the total NPI mean score difference appeared to be driven by higher scores for males on both the Leadership/Authority and Superiority/Arrogance NPI subscales. Leadership/Authority means were as follows: males = 5.16 (SD = 2.32); females = 4.33 (SD = 2.56). This mean difference was determined statistically significant, F (1, 187) = 4.936, p = .028. Superiority/Arrogance means were 4.87 (SD = 2.49) for the male participants and 3.73 for the female participants (SD = 2.40). This mean difference was statistically significant as well, F (1, 187) = 9.46, p = .002. Furthermore, a mean difference was observed by gender in regard to death anxiety score. The mean score for males on the death anxiety measure was 39.47 (SD = 9.73), while the mean score for females was 45.48 (SD = 9.80). This mean score difference was determined statistically significant, F (1, 187) = 16.45, p < .0001. Therefore, females taking the Death Anxiety Scale displayed a greater level of death anxiety on average than did males taking the same instrument. Few significant and meaningful differences were found regarding demographic variables measured in relationship to the scales employed in the study. Analyses were not conducted for half of the demographic variables (race/ethnicity, sexual orientation, and 91 religious orientation) due to a lack of variability in the sample. Analyses were conducted for the remaining variables, which included participant age, gender, college attended, and college classification. No significant mean differences were found on any of the instruments for college attended or college classification. However, correlational analyses for participant age revealed a small (but statistically significant) positive correlation with purpose-in-life, self-actualization, self- esteem, and authenticity scores, while also revealing a small (but significant) negative correlation with total narcissism, exploitativeness/entitlement narcissism, superiority/arrogance narcissism, search for a sense of purpose in life, and death anxiety scores. Also, analyses in reference to participants? gender also displayed some significant mean differences in relation to NPI scores and death anxiety scores. Male participants scored higher on the total NPI and the Leadership/Authority and Superiority/Arrogance NPI subscales, while female participants scored higher on the death anxiety measure. Though the findings related to gender and age may have implications for future research and are interesting artifacts of the study, their effects did not appear to be significant in relation to the primary predictions of this dissertation study. Though statistically significant, the mean score differences and correlations obtained were relatively small. Also, examination of age and gender differences in relationship to narcissism (and to the existential scale scores) was not an intended primary focus of this study. Therefore, the effects of these variables were not taken into account when conducting the analyses which addressed the major hypotheses on which this study focused. 92 CHAPTER 5 DISCUSSION Review of the Empirical Findings In this study, an empirical examination was conducted in regard to the predictive abilities of the Narcissistic Personality Inventory and its four subscales (Exploitativeness/Entitlement, Leadership/Authority, Superiority/Arrogance, and Self- absorption/Self-Admiration) in relation to measures of important existential constructs that centered on psychological health or lack thereof (such as authenticity, self- actualization, and death anxiety). Hypotheses developed were based on a body of previous research using the NPI and on existential theory. The hypotheses developed and tested were as follows: 1. The total NPI score would predict greater self-esteem, a greater sense of purpose in life, greater capacity for self-actualization, greater authenticity, and reduced death anxiety. No prediction was made regarding the SONG score. 2. The Exploitativeness/Entitlement factor score would predict a lack of purpose in life, avoidance of searching for a sense of purpose, greater death anxiety, lowered capacity for self-actualization, less authenticity, and lowered self-esteem. This prediction would be augmented by partialing out the effects of the proposed healthier facets of narcissism (Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration). 93 3. The Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self- Admiration NPI subscale scores (to varying degrees) would predict a greater sense of purpose in life, less death anxiety, greater capacity for self-actualization, greater authenticity, and greater self-esteem. These predictions would strengthen when the effects of Exploitativeness/Entitlement were partialed out of the predictive equation. No prediction of the SONG score (indicative of a search for meaning and purpose) was made, as one may or may not continue searching if one already has a sense of meaning and purpose in life. The statistical findings from this dissertation study provided stronger support for hypotheses 2 and 3 than for hypothesis 1. The total Narcissistic Personality Inventory score was a less reliable predictor of existential psychological functioning than expected. The total NPI score served as a moderate predictor of increased self-esteem (r = .28) and as a statistically significant predictor of a greater sense of purpose-in-life (.20) and more intense searching for life purpose (.16). However, the total NPI score did not serve as a reliable predictor of self-actualization, death anxiety, or authenticity. These findings suggest that one may have more to gain from examining the component facets of narcissism and the functional capacity of those facets, rather than from examination of narcissism as a unidimensional construct. The findings related to the predictive abilities of the NPI subscales were closer to those hypothesized. The Exploitativeness/Entitlement (EE) subscale score served as a statistically significant predictor of a lack of purpose-in-life (-.18), a more intense search for life purpose (.31), less self-actualization (-.19), greater death anxiety (.18), and 94 reduced authenticity (-.24). However, this score did not significantly predict self-esteem (-.07). When the effects of the other, presumed healthier, NPI subscales were factored out, the EE subscale score predicted poor self-esteem (-.30), even less self-actualization (-.37) and authenticity (-.44), and even greater death anxiety (.27). The effects of partialing out the other subscales had no effect on predictive ability regarding the search for purpose in life (.32). The Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self- Admiration NPI subscale scores were shown to predict (to varying degrees) a greater sense of purpose in life, less death anxiety, greater capacity for self-actualization, greater authenticity, and greater self-esteem. The only hypothesized relationship that was not observed upon review of the data was that between Self-Absorption/Self-Admiration scores and death anxiety scores (.06). The removal of the effects of the presumed negative aspect of narcissism (Exploitativeness/Entitlement) increased the magnitude and level of statistical significance of each of these predictions. For example, the zero-order regression prediction for purpose-in-life in relation to Leadership/Authority was measured at .38, but this prediction increased to .49 when the effects of EE were partialed out. Initial analyses of the data suggested that the reliability and validity of most scales employed in the study was excellent. The internal consistency for the NPI and most of the existential measures was measured at above .80. The NPI subscales generally fell within an acceptable range using this statistical analysis as well. Furthermore, the NPI and its subscales intercorrelated in a pattern expected from a multidimensional, complex 95 instrument. Also, the existential scales related to one another as expected, providing evidence of construct validity for all of these important measures. Unexpected findings were observed when comparisons using the demographic variables of age and gender were conducted. As age increased in participants, so did a sense of purpose-in-life, self-actualization, authenticity, and self-esteem. Also, increased age correlated with lesser degrees of narcissism (including reductions in the Exploitativeness/Entitlement and Superiority/Arrogance subscale scores). Also, males scored higher on average on the Narcissistic Personality Inventory (especially the Leadership/Authority and Superiority/Arrogance subscales), while females scored higher on the death anxiety measure. These demographic findings were unexpected and were not part of the intended focus of this dissertation study. However, the findings did suggest avenues for future research. Meaning and Interpretation of Results This section of the manuscript will focus on the interpretation and meaning drawn from (1) analysis of the data related to each hypothesis, (2) from data analysis related to the reliability and validity of the measures used in the study, (3) the data regarding unexpected findings when gender and participant age were examined in relation to the NPI and existential scales, and (4) an overall summary of the meaning and interpretation of all statistical results. Interpretation and Meaning of Findings Related to Hypothesis 1. The zero-order regression analyses demonstrated that Hypothesis 1 was not as well supported as the Hypotheses 2 and 3. The total NPI score was shown to be a weak-to-moderate predictor 96 of self-esteem, a weak predictor of achieved sense of purpose-in-life, and an even weaker predictor of the ongoing search for meaning and purpose-in-life. Total NPI score did not, as hypothesized, predict levels of self-actualization, authenticity, or death anxiety. A surface level interpretation of these findings would be that narcissism, as measured by the Narcissistic Personality Inventory, exhibited no strong predictive relationships with existential measures of psychological health and/or dysfunction. Therefore, one could not claim (on the basis of data examining narcissism as a unidimensional construct) that higher or lower levels of narcissism offer any psychologically protective benefits from an existential perspective. The ability of the NPI total score to predict moderate gains in self-esteem could easily be explained away as a defensive maneuver on the part of those with narcissistic characteristics to protect themselves from exposure of their core sense of poor self-worth, which is a more traditional interpretation of narcissistic functioning (Kernberg, 1985). However, a more in-depth interpretation that takes previous NPI literature into account deserves exploration. Previous studies that employed the total NPI score as a predictor in relation to various measures of psychological health and dysfunction found that greater NPI scores predicted higher scores on measures of both psychological health and psychological dysfunction, that total NPI scores demonstrated a very inconsistent and unpredictable relationship to Rosenberg Self-Esteem scores (generally ranging from moderately positive to moderately negative), and that NPI total scores were generally unreliable predictors of psychological functioning (these findings are discussed in depth 97 in Chapter 2). In fact, the rationale for factor-analyzing the NPI was directly tied to these confusing and obfuscating findings. Given this literature-driven interpretation, the findings from this study as related to the total NPI score are not completely surprising, nor are they completely discouraging. The findings when assessing narcissism as a complex and multifaceted construct were much more useful and meaningful, as has been the case in a host of previous studies using the NPI as a narcissism measure. It is possible that, when the total NPI score was examined, that the ?healthy? NPI and ?unhealthy? NPI subscales cancelled one another out in relation to the existential measures, which would result in no predictable relationships among many of the constructs. Also, the pattern of findings from this study was similar in some ways to previous studies using the total NPI. Though the correlations were not strong, the total NPI score predicted both psychological health (greater self-esteem and sense of purpose-in-life), while also predicting greater levels on a measure that appeared to assess psychological dysfunction (the Seeking of Noetic Goals). Therefore, the total NPI findings were not inconsistent with previous research findings, despite the fact that the hypothesized predictions for this study did not emerge. More meaningful (and important) are the findings for the NPI subscales. Interpretation and Meaning of Findings Related to Hypothesis 2. Hypothesis 2 centered on the Exploitativeness/Entitlement facet of narcissism. As previous studies had demonstrated, this study sought to identify the Exploitativeness/Entitlement subscale of the NPI as a consistent predictor of psychological dysfunction and lack of psychologically healthy characteristics. The study also sought to demonstrate that this 98 facet of narcissism is most unhealthy when those who possess it do not also possess characteristics of narcissism that were assumed to be healthier and more psychologically protective (such as Leadership/Authority). Through examination of the zero-order and partial regression data, the findings strongly supported the hypothesized predictive relationships. Higher levels of Exploitativeness/Entitlement predicted an impaired sense of purpose-in-life, a more intense search for life purpose and meaning, reduced levels of self-actualization, increased death anxiety, and reduced levels of authenticity. Only the prediction that higher levels of this factor would predict reduced self-esteem did not hold. Also, as expected, each of these predictions increased in strength and statistical significance when the protective effects of the other NPI facets were statistically removed. In addition, higher levels of Exploitativeness/Entitlement did predict moderately reduced self-esteem when the protective narcissism facets were factored out. These findings supported the assertions that narcissism (1) must be studied as a multidimensional and complex construct (and not unidimensionally), (2) that facets of narcissism differentially predict psychological health and psychological dysfunction, and (3) that the least healthy facet of narcissism appears psychologically unhealthier when the protective effects of other narcissism facets are removed. Interpersonally exploitative behavior and a sense of entitlement seem, based on these results, to be the most problematic aspects of narcissism. Though possession of other facets of narcissism do seem to offer some protective effects, greater engagement in exploitation and a greater sense of personal entitlement consistently bodes poorly for an individual?s psychological health. 99 Also, the support of the hypotheses developed for the Exploitativeness/Entitlement facet suggests that existential theoretical ideas regarding narcissism have merit. While existential theorists have argued that narcissism is a natural result of the human condition (e.g., the need to create and maintain a sense of life purpose and meaning), they have also suggested that not all expressions and forms of narcissism are healthy (nor do they all promote psychological development) (Shaw, 2000; van Deurzen-Smith, 2000; Yalom, 1980). The adoption of interpersonal exploitation as a way of life and a distorted sense of personal entitlement as an attitude (one that does not take into account the needs/wants of others) are contrary to beliefs about what makes one psychologically healthy from an existential perspective. The findings from this study demonstrated that adoption and reliance on such a narcissistic motivational structure may lead to diminished psychological health. Interpretation and Meaning of Findings Related to Hypothesis 3. Hypothesis 3 centered on findings related to the Leadership/Authority, Superiority/Arrogance, and Self-Absorption/Self-Admiration narcissism facets. These facets were hypothesized to be the healthier aspects of narcissism. It was hypothesized that higher scores on all three facets would predict a greater sense of purpose-in-life, less death anxiety, greater self- actualization, greater authenticity, and greater self-esteem. It was also hypothesized that these predictions would strengthen when the already-established negative effects of Exploitativeness/Entitlement were factored out. The findings, as discussed previously, strongly supported hypothesized relationships. 100 These findings further supported the assertions that narcissism (1) must be studied as a multidimensional and complex construct (and not unidimensionally), (2) that facets of narcissism differentially predict psychological health and psychological dysfunction, and (3) that the healthier facets of narcissism appear psychologically more protective when the damaging effects of Exploitativeness/Entitlement facet are removed. It would appear that abilities to lead, abilities to possess and use authority, a strong sense of belief in one?s abilities, and the ability to take care of one?s own needs are key to a narcissistic form of expression that leads to enhanced psychological functioning. These abilities are even more psychologically protective when an individual does not possess a distorted sense of personal entitlement and/or engage in interpersonally exploitative behavior. These findings also lend even stronger support to the existential theories and beliefs in regard to narcissism. This study demonstrated that certain forms of narcissistic expression actually enhance one?s sense of purpose-in-life, self-esteem, self- actualization, and authenticity, while also diminishing one?s anxiety about one?s own death. These forms of expression are especially helpful when not accompanied by entitlement and exploitation of others. Shaw (2000) would likely not be surprised by these results, as he hypothesized that narcissism is a motivational structure that assists individuals in developing a sense of meaning and purpose. van Deurzen-Smith (2000) would also likely not seem shocked by findings that demonstrate certain narcissistic motivational structures to be beneficial to our health and psychological well-being (especially those structures that allow for the individual to take others? needs and wants into consideration). 101 Measure Reliability and Validity Data. The findings discussed in relation to the major hypotheses developed for this study were boosted by favorable findings regarding the reliability and validity of the measures employed. Not surprisingly, the Narcissistic Personality Inventory was again demonstrated to be a solid instrument, as its internal consistency and intercorrelational data were solid. As mentioned previously in this chapter, the data from this study also continued to support the assertion that narcissism is best studied through examination of its components or facets, as much data is distorted or lost with focus solely on the total scale score. The complex relationships among the subscales must be examined to find important existing relationship (that may be missed or overlooked without attention to these facets). The most important of the reliability and validity findings are those in relation to the existential scales used in this study. For far too long, existential perspectives have been derided for their questionable testability and apparent lack of scientific method. This empirical study included six existential measures with varying degrees of psychometric support derived through limited use (with the exception of the Rosenberg Self-Esteem Scale) in research studies. The reliability and validity data for these scales was generally impressive and comparable to data for scales developed from other theoretical perspectives. With the notable exception of the Short Index of Self-Actualization, the alpha level for all of the existential scales ranged from .82 to .90. In addition, the existential scales correlated in the patterns expected based on their content (e.g., Death Anxiety Scale scores correlated negatively with self-esteem, purpose-in-life, self- actualization, and authenticity). These impressive data support the continued use of 102 existentially-oriented research scales (and, hopefully, the development of new such scales). The data also support the existential perspective as a rich and rewarding avenue for future empirical research. Participant Gender and Age Findings. The surprising findings related to participant gender and age should not be overlooked, despite the fact that this study did not seek to further explore these findings. In regard to participant gender, it was found that males scored higher on average on the total NPI, the Leadership/Authority subscale, and the Superiority/Arrogance subscale, while females scored higher on average on the Death Anxiety Scale. Hence, it would appear that being male is related to greater levels on two of the presumed healthy narcissism facets, while being female is related to increased levels of death anxiety (which, in strict existential terms, is not psychologically healthy). The NPI findings are somewhat different than those reported by Watson, Taylor, and Morris (1987), which were that their sample of males scored higher on the total NPI, the Exploitativeness/Entitlement subscale, and on the Superiority/Arrogance subscale (but not on the Leadership/Authority subscale, where there was no gender difference). Without further examination of the effects of sex and/or gender roles, the present study?s results are difficult to explain and interpret. As Watson, et al., (1987) demonstrated, accounting for sex and gender roles is important in helping to explain and interpret findings such as these. It is possible that individual?s gender and/or sex roles play a mediating effect. The present study did not take such roles into account (only self- reported gender was requested demographically). The sample for this study was drawn 103 from a Southeastern student population, and this fact needs to be taken into account as well. It is highly likely that student samples from other parts of the country might display different patterns, as sex and gender roles may differ regionally. Future studies should seek to clarify these gender differences further. In regard to participant age, it was demonstrated that increasing age correlated with reduced levels of total narcissism, Exploitativeness/Entitlement, and Superiority/Arrogance, greater self-esteem, greater sense of achieved purpose-in-life, reduced search for meaning and purpose, enhanced self-actualization, reduced death anxiety, and greater authenticity. On the whole, it appeared that older individuals in this sample were psychologically healthier individuals as well. To date, NPI studies have not squarely focused on the process of how narcissistic motivational structures might change with age. Perhaps research in this area might better explain the findings from this dissertation study that suggested that older individuals were less narcissistic and more existentially healthy. Such studies would likely benefit from recruiting a broader sample than the one employed in this study, as the lifespan range of age would need to be expanded (as opposed to collecting data from college students, who generally fall into a restricted range of age). Summary of Empirical Findings, Interpretations, and Meanings. This dissertation study demonstrated that an existential conceptual view of narcissism is viable, and that such a view may be supported by empirical research. An existential view of narcissism proposes that one view narcissism as a personal motivational structure developed to cope with the human problem of personal insignificance (Shaw, 2000). Such a view also 104 proposes that development of such a motivational structure (especially in Western, individualistic cultures) is natural, normal, and common. Furthermore, such a theoretical view presupposes a balance between humans? natural tendency to be self-centered and selfish, while also balancing our needs with the community surrounding us (van Deurzen- Smith, 2000). The findings from this study support such a conceptualization by demonstrating that adoption of a narcissistic motivational structure (or personality/character structure) need not be damaging to self or others. Individuals in this study who possessed higher levels of characteristics consistent with the ability to lead, command authority, belief in their own abilities, and a certain amount of self-focus also possessed greater self-esteem, self-actualization, purpose-in-life, and achievement of authentic being. These characteristics are part and parcel of a narcissistic character structure. They also appear to be relatively healthy psychological characteristics. Furthermore, the effects of the Exploitativeness/Entitlement dimension of narcissism also supported an existential theoretical view. If van Deurzen-Smith (2000) was correct, individuals who make use of a narcissistic character structure may fare better leaving out an entitled, exploitative view of the world. Failing to do so might lead to too much focus on one?s own well-being, to the detriment of others (who then, in turn, will shun such an individual). Her theoretical assumptions were supported by these findings, as exploitative and entitled being-in-the-world was associated with decreases in self- esteem, self-actualization, authenticity, and purpose-in-life (while concurrently being associated with increases in the fear of death). Though the healthier facets of narcissism 105 offered some protection for individuals who operate in an exploitative and entitled manner, these findings still held even with the protective effects in place. This dissertation study also demonstrated that a more complex, multidimensional approach to personality characteristics bears fruit (both from a research and from a clinical perspective). A large body of NPI research now exists that clearly demonstrates that dimensional approaches to personality provide more accurate data and clarification of personality constructs. This study has also demonstrated that use of a dimensional approach assists clinicians and researchers in identifying healthier and more positive aspects of ?pathological? personality constructs. One might argue that other personality ?disorders? should also be examined as narcissism was in this study. If narcissism were still treated as a unidimensional personality construct (as it was early on), these data would never have been collected and researchers would have no empirical evidence of the positive aspects of narcissism detailed in this study. Perhaps other personality characteristics need to be examined in a similar way, as motivational constructs other than narcissism (e.g., paranoid, schizoid, borderline) may exhibit healthier forms and facets of being as well. The relatively unexpected findings in regard to participant gender and age suggest further opportunities in this area of research. The present study was not designed to fully examine the differences related to gender and relationships related to age. Future studies may need to expand the examination of sex/gender roles as a mediating variable in the gender differences found in regard to narcissism, narcissism facets, and death anxiety. It was also suggested that future studies further assess the development and path of 106 narcissistic functioning over time through use of more diverse and broad samples. The suggestion that some part of the aging process might prompt changes in narcissistic expression and enhance psychological well-being (from an existential viewpoint) is an exciting proposition that deserves further study. Implications for Research Future research in this area of study will likely need to employ more powerful and sophisticated statistical analyses that would assist in promoting greater understanding of the findings herein. For example, instead of using zero-order correlations to examine the relationships between the Narcissistic Personality Inventory total score and the existential measures, one might choose to use a multiple regression model or a path analysis to better elucidate the relative contributions of the existential constructs (i.e., determine which existential construct measures account for a greater proportion of the variance in NPI score). Use of such statistical techniques would provide a better opportunity to truly assess the full complexity of the narcissism construct from an existential perspective. Also, future studies may seek to assess whether certain variables (for example, self- esteem) serve as moderating variables in the prediction of narcissism scores. Based on the results of the present study, one cannot rule out the possibility that extraneous variables might be serving as moderators in the relationships observed between/among variables studied directly. Furthermore, concerns emerged regarding certain instruments employed in this study. For example, the observed alpha level for the NPI Exploitativeness/Entitlement subscale (.60) was troubling, in that one would expect a higher degree of internal 107 consistency for an 11-item scale. In fact, all of the NPI subscales produced a lower alpha level than expected, despite a solid alpha level for the total NPI. Future research in this area might be invested in reworking the NPI subscales such that they better measure the narcissism facets. Another suggested course of future research would include the development of more effective, reliable, and valid measures of the component narcissism facets. In addition, the Short Index of Self-Actualization should not be employed in future studies, as a .57 alpha level was not indicative of a coherent measure of this particular construct. Finally, the results of this study would have been easier to interpret and understand if normative data for each instrument employed had been available. These norms would provide important information regarding how the present sample would compare to a normative group in terms of narcissistic and existential functioning. Future research should include attempts to gather normative data on many of these instruments, while updating the data provided in Chapter 4. Also, researchers should adopt a uniform scoring system for each of these instruments, as normative data are difficult to obtain when versions of these instruments vary so widely. Review of the Goals of this Dissertation Study: Implications for Theory, Treatment, and the Field of Counseling Psychology Through interpretation of the results of the empirical study described above, this dissertation work sought to (1) examine a reinterpretation of the construct called narcissism from an existential perspective, (2) examine the construct of narcissism in relation to key existential constructs (such as self-actualization, purpose/meaning in life, 108 the search for meaning in life, authenticity, and death anxiety), and (3) explore narcissism as a complex and multidimensional construct that relates to a spectrum of functioning through use of a number of existential construct measures. These aims were accomplished herein through an empirical study examining the predictive ability of the Narcissistic Personality Inventory (NPI) and its subscales in relation to a host of scales that measured traditional existential constructs (such as purpose-in-life, authenticity, and self-actualization). However, this dissertation study also sought to demonstrate points that are larger than just finding statistical significance using regression analyses. For example, this study sought to provide a demonstration that existential theoretical constructs are important and should not be marginalized or overlooked by researchers nor by practitioners in the field. The existential/humanistic perspective has been criticized, possibly a bit excessively, as a theoretical perspective that does not lend itself to scientific study (and is, therefore, second-rate or of lesser import). This dissertation study attempted to provide evidence that existential theoretical constructs can be studied empirically, and that excellent measures of such constructs already exist for use in future research. It would seem that this study accomplished this goal. It is hoped that this study might serve as a source of hope for those who might have given up on convincing others of the importance of the existential perspective in psychology. It is also hoped that others will demonstrate rejuvenated interest in study of existential constructs as well. For clinicians, use of existential theory and research can lead to effective conceptualizations and treatment plans for clients. These conceptualizations and 109 treatment plans would be client-focused and would center on client strengths and abilities (rather than on pathology or character weaknesses). The Narcissistic Personality Inventory (Emmons, 1984, 1987) might well be added to traditional personality assessment instruments, as these instruments do not generally assess for narcissistic strengths. If existential clinicians hope to survive in a managed-care environment, it would be prudent to build a body of research that supports existentially-informed treatment planning and interventions. Therefore, these treatment plans and strength- focused interventions must be put to empirical test and demonstrated effective as compared to other treatment modalities. The instruments included in this study, in some cases, would make for excellent treatment outcome measures. For example, if one wished to test out the hypothesis that narcissistic treatment planning should focus on working with narcissistic strengths (e.g., building leadership skills, enhancing appropriate use of authority, engaging in healthy forms of self-focus and self-reflection) and avoiding engagement in damaging narcissistic motivational structures (e.g., acting in an interpersonally exploitative manner), then this treatment structure might be compared to other forms of treatment with use of the Purpose-in-Life Test (Crumbaugh & Maholick, 1969) or the Authenticity Inventory -3 (Goldman & Kernis, 2004) as an outcome measure. Related to the discussion above, this study also sought to support a more complex, dimensional view of personality than the categorical system proposed by the DSM-IV-TR (APA, 2000). Over-reliance on categorical systems of personality runs the risk of pathologizing clients who possess certain personality characteristics (or who exhibit 110 behavior consistent with certain ?pathological? personality styles). Systems such as these may prompt clinicians to focus more on diagnosis of a ?disorder? than on conceptualizing and developing a strength and wellness-based treatment plan for individual clients. The use of a more complex dimensional system of personality conceptualization was presented in this study that is more consistent with traditional counseling psychology values, as dimensional systems that allow for consideration of a continuum of personality functioning (rather than rely squarely on pathology), that focuses on strengths and weaknesses in a given personality configuration, and on developing treatment plans based on wellness and personal growth. Counseling psychology, as a field, has attempted to balance concerns regarding pathology and mental dysfunction with attempts to promote wellness, personal growth, and preventative interventions (Hoffman & Driscoll, 2000). Counseling psychologists are expected to be proficient in assisting individuals and groups that fall along a continuous dimension of psychological health and wellness (Hoffman & Driscoll, 2000). Even when working with individuals that a medical or categorical model of treatment may label pathological, counseling psychologists endeavor to consider such an individual?s strengths and abilities in their treatment plans and goals. In fact, many in the field have suggested that the promotion of wellness and psychological health is an emerging role for counselors and counseling psychologists (Gladding, 2004; Myers, Sweeney, & Witmer, 2001; Witmer & Sweeney, 1999). This focus on wellness and health, according to these authors, should apply to all individuals with whom we work (Myers, Sweeney, & Witmer, 2000). 111 Hence, the present study sought to support a view of narcissism (and psychological functioning in general) based on traditional counseling psychology values and beliefs. Instead of spending inordinate amounts of time clarifying personality disorder diagnoses (such that these problems can be treated and/or fixed), this study promotes greater clinician focus on assessing the adaptive strengths that all clients (no matter how pathological) possess. This study advocates for focus on using the adaptive facets of narcissism as strengths upon which to build in therapy/counseling, rather than maintaining sole focus on the less adaptive facets of such a motivational structure. Such a focus helps to reduce the stigma associated with a personality disorder diagnosis and shifts the focus to a consideration of how best to help the individual client achieve wellness and personal growth. Limitations of this Dissertation Study The most glaring weakness of this dissertation study resided in its sample. The generalizability of this study beyond young, White, heterosexual males and females is somewhat limited. The author was somewhat disheartened by the lack of variability in this sample in regard to many variables related to diversity. It is hoped that future studies in this area would seek to replicate these findings using a more diverse, representative sample. Also, given the interesting (and unexpected) findings related to the age and sex of participants, future studies would do well to seek out samples that go beyond selection from the college student pool. Surely, the findings related to males scoring higher on narcissism measures (especially the Leadership/Authority and Superiority/Arrogance subscales) and to the females scoring higher on death anxiety are related in large part to 112 cultural considerations not examined in this study. It would also be worth a researcher?s time to examine narcissism scores throughout the lifespan to truly examine the effects of age on expressions of narcissism. Also, this study should be replicated using a sample from a more collectivistic culture. It has been well-documented that Western, individualistic cultures tend to promote the development of narcissistic motivational structures (Lasch, 1979). However, one wonders if the ?healthier? aspects of narcissism would look as healthy to individuals raised in a collectivist culture. As personality and personality pathology are, to a large extent, culturally defined, it would be of great value to examine the construct of narcissism from a more diverse perspective. What seems pathological to individualistic Americans may not seem so in a collectivistic nation (and vice versa). These questions should be examined through further research. Finally, it must be mentioned that the examination of cause and effect relationships was beyond the scope of this study. This dissertation study was designed to examine possible predictive relationships. However, the statistical analyses chosen did not lend themselves to interpretation based on cause/effect relationships. One cannot argue, on the basis of these findings, that higher levels of exploitative behavior and a greater sense of entitlement caused increased death anxiety (as an example). One can only argue that such characteristics serve as predictors of greater death anxiety. These constructs may very well be related, but one cannot claim that any form of narcissistic motivational structure causes psychological health or dysfunction on the basis of the regression results discussed herein. 113 REFERENCES American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd edition). Washington, DC: Author. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th edition). Washington, DC: Author. American Psychiatric Assocation. (2000). Diagnostic and statistical manual of mental disorders. (4th edition, revised). Washington, DC: Author. Ashby, H. U., Lee, R. R., & Duke, E. H. (1979, September). A narcissistic personality disorder MMPI scale. Paper presented at the annual meeting of the American Psychological Association, New York. Barron, F. X. (1974). Basic research and aesthetic education. Washington, DC: U. S. Office of Education. Beck, A. T., Weissman, A., Lester, D., & Trexler, L. (1974). The measurement of pessimism: The Hopelessness Scale. Journal of Consulting and Clinical Psychology, 42, 861 ? 865. Block, J. (1961). The Q-sort method in personality assessment and psychiatric research. Springfield, IL: Charles C. Thomas. Bugental, J. F. T. (1965). The search for authenticity. New York: Holt, Rinehart, and Winston. 114 Cattell, R. B., Eber, H. W., & Tatsuoka, M. M. (1970). Handbook of the Sixteen Personality Factor Questionnaire (16PF). Champaign, IL: Institute for Personality and Ability Testing. Christie, R., & Geis, F. L. (Eds.). (1970). Studies in Machiavellianism. San Diego, CA: Academic Press. Cooper, A., & Ronningstam, E. (1992). Narcissistic personality disorder. American psychiatric press review of psychiatry (pp. 80-97). Washington, DC: American Psychiatric Press. Cooper, A. M. (1987). Histrionic, narcissistic, and compulsive personality disorders. In G. L. Tischler (Ed.), Diagnosis and classification in psychiatry: A critical appraisal of DSM-III (pp. 290-299). New York: Cambridge University Press. Costello, C. G., & Comrey, A. L. (1967). Scales for measuring depression and anxiety. Journal of Psychology, 66, 303-313. Crandall, R., McCown, D. A., & Robb, Z. (1988). The effects of assertiveness training on self-actualization. Small Group Behavior, 19, 134-145. Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 297 ? 334. Crumbaugh, J. C. (1968). Cross-validation of Purpose-in-Life Test based on Frankl?s concepts. Journal of Individual Psychology, 24, 74-81. Crumbaugh, J. C. (1977a). The Seeking of Noetic Goals Test manual. Huntsville, AL: Psychometric Associates. 115 Crumbaugh, J. C. (1977b). The Seeking of Noetic Goals Test (SONG): A complementary scale to the Purpose in Life Test (PIL). Journal of Clinical Psychology, 33, 900- 907. Crumbaugh, J. C., & Henrion, R. (1988). The PIL test: Administration, interpretation, uses, theory, and critique. The International Forum for Logotherapy, 11, 76 ? 88. Crumbaugh, J. C., & Maholick, L. T. (1964). An experimental study in existentialism: The psychometric approach to Frankl?s noogenic neurosis. Journal of Clinical Psychology, 20, 200-207. Crumbaugh, J. C., & Maholick, L. T. (1969). Purpose-in-Life Test Manual. Huntsville, AL: Psychometric Associates. Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85. Dawkins, R. (1976). The selfish gene. Oxford: Oxford University Press. Dougherty, K., Templer, D. I., & Brown, R. (1986). Psychological states in terminal cancer patients as measured over time. Journal of Counseling Psychology, 33, 357-359. Edwards, A. L. (1959). Manual for the Edwards Personal Preference Schedule. New York: Psychological Corporation. Emmons, R. A. (1981). Relationship between narcissism and sensation seeking. Psychological Reports, 48, 247-250. Emmons, R. A. (1984). Factor analysis and construct validity of the Narcissistic Personality Inventory. Journal of Personality Assessment, 48, 291-300. 116 Emmons, R. A. (1987). Narcissism: Theory and measurement. Journal of Personality and Social Psychology, 52, 11-17. Eysenck, H. J., & Eysenck, S. B. (1968). Manual for the Eysenck Personality Inventory. San Diego, CA: EDITS. Fenigstein, A., Scheier, M. F., & Buss, A. H. (1975). Public and private self- consciousness: Assessment and theory. Journal of Consulting and Clinical Psychology, 43, 522-527. Flett, G. L., Blankstein, K. R., & Hewitt, P. L. (1991). Factor structure of the Short Index of Self-Actualization. In A. Jones & R. Crandall (Eds.). Handbook of replication research. [Special Issue]. Journal of Social Behavior and Personality, 6, 321-329. Frankl, V. E. (1984). Man?s search for meaning. New York: Simon and Schuster. Freud, S. (1957). On narcissism: An introduction. In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 14, pp. 73-102). London: Hogarth Press. (Original work published in 1914). Frost, R. O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfectionsism. Cognitive Therapy and Research, 14, 449 ? 468. Gerson, M. J. (1984). Splitting: The development of a measure. Journal of Clinical Psychology, 40, 157-162. Gladding, S. T. (2004). Counseling: A comprehensive profession. (5th ed.). Upper Saddle River, NJ: Pearson. 117 Goldman, B. M., & Kernis, M. H. (2002). The role of authenticity in healthy psychological functioning and subjective well-being. Annals of the American Psychotherapy Association, 5, 18-20. Goldman, B. M., & Kernis, M. H. (2004). Manual for the Authenticity Inventory - 3. Unpublished document. Sent by author on 7/30/2004. Retrieved via e-mail on 7/30/2004. Gough, H. G. (1956). California Psychological Inventory. Palo Alto, CA: Consulting Psychologists Press. Gough, H. G., & Heilbrun, A. B., Jr. (1965). The Adjective Check List. Palo Alto, CA: Consulting Psychologists Press. Gray-Little, B., Williams, V. S. L., & Hancock, T. D. (1997). An item response theory analysis of the Rosenberg Self-Esteem Scale. Personality and Social Psychology Bulletin, 23, 443-451. Harder, D. W., & Zalma, A. (1990). Two promising shame and guilt scales: A construct validity comparison. Journal of Personality Assessment, 55, 729 ? 745. Hickman, S. E., Watson, P. J., & Morris, R. J. (1996). Optimism, pessimism, and the complexity of narcissism. Personality and Individual Differences, 20, 521-525. Hoffman, M. A., & Driscoll, J. M. (2000). Health promotion and disease prevention: A concentric biopsychosocial model of health status. In S. D. Brown & R. W. Lent (Eds.). Handbook of Counseling Psychology (3rd ed.) (pp. 532 ? 567). New York: John Wylie. 118 John, O. P., & Robins, R. W. (1994). Accuracy and bias in self-perception: Individual differences in self-enhancement and the role of narcissism. Journal of Personality and Social Psychology, 66, 206-219. Jones, A. (1980). The development of a short measure of self-actualization. Unpublished doctoral dissertation. Texas Christian University. Jones, A., & Crandall, R. (1986). Validation of a Short Index of Self-Actualization. Personality and Social Psychology Bulletin, 12, 63-73. Jones, R. G. (1969). A factored measure of Ellis? irrational belief system, with personality and maladjustment correlates. (Doctoral dissertation, Texas Technical College, 1968). Dissertation Abstracts, 69, 6443. Kernberg, O. (1985). Borderline conditions and pathological narcissism. New York: Jason Aronson. Kohut, H. (1971). The analysis of the self. New York: International Universities Press. LaForge, R., & Suczek, R. F. (1955). An interpersonal dimension of personality: An interpersonal checklist. Journal of Personality, 25, 94-112. Lasch, C. (1979). The culture of narcissism: American life in an age of diminishing expectations. New York: Norton. Leary, T. (1956). A manual for the use of the interpersonal system of personality. Berkeley: Psychological Consultation Service. Lonetto, R., & Templer, D. I. (1986). Death anxiety. Washington, DC: Hemisphere. Lowen, A. (1985). Narcissism: Denial of the true self. New York: Collier Books. 119 Mahler, M. S. (1972). Rapproachment subphase of the separation-individuation process. The Psychoanalytic Quarterly, 41, 487-506. Maslow, A. H. (1968). Toward a psychology of being. New York: Van Nostrand. Maslow, A. H. (1970). Motivation and personality. (2nd ed.). New York: Harper and Row. Maslow, A. H. (1982). Toward a psychology of being. New York: Van Nostrand. McHoskey, J. (1995). Narcissism and Machiavellianism. Psychological Reports, 77, 755 ? 759. McLeod, C. R., & Vodanovich, S. J. (1991). The relationship between self-actualization and boredom proneness. In A. Jones & R. Crandall (Eds.). Handbook of replication research. [Special issue]. Journal of Social Behavior and Personality, 6, 137-146. McMordie, W. R. (1979). Improving measurement of death anxiety. Psychological Reports, 44, 975-980. Miller, L. C., Murphy, M., & Buss, A. H. (1981). Consciousness of body: Private and public. Journal of Personality and Social Psychology, 41, 397-406. Millon, T. (1982). Millon Clinical Multiaxial Inventory manual. Minneapolis, MN: National Computer Systems. Millon, T. (1994). Manual for the MCMI-III. Minneapolis, MN: National Computer Systems. Millon, T. (1997). Millon Clinical Multiaxial Inventory-III manual (2nd ed.). Minneapolis, MN: National Computer Systems. 120 Myers, J. E., Sweeney, T. J., & Witmer, J. M. (2000). The Wheel of Wellness counseling for wellness: A holistic model for treatment planning. Journal of Counseling and Development, 78, 251-266. Myers, J. E., Sweeney, T. J., & Witmer, J. M. (2001). Optimization of behavior. In D. C. Locke, J. E. Myers, & E. L. Herr (Eds.), The handbook of counseling (pp. 641 ? 652). Thousand Oaks, CA: Sage. O?Brien, M. L. (1987). Examining the dimensionality of pathological narcissism: Factor analysis and construct validity of the O?Brien Multiphasic Narcissism Inventory. Psychological Reports, 61, 499 ? 510. Phares, E. J., & Erskine, N. (1984). The measurement of selfism. Educational and Psychological Measurement, 44, 597-608. Prifitera, A., & Ryan, J. J. (1984). Validity of the narcissistic personality inventory (NPI) in a psychiatric sample. Journal of Clinical Psychology, 40, 140-142. Raskin, R. N. (1980). Narcissism and creativity: Are they related? Psychological Reports, 46, 55-60. Raskin, R. N., & Hall, C. S. (1979). A narcissistic personality inventory. Psychological Reports, 45, 590. Raskin, R. N., & Hall, C. S. (1981). The Narcissistic Personality Inventory: Alternate form reliability and further evidence of construct validity. Journal of Personality Assessment, 45, 159-162. Raskin, R., Novacek, J., & Hogan, R. (1991). Narcissism, self-esteem, and defensive self- enhancement. Journal of Personality, 59, 19-38. 121 Raskin, R., & Terry, H. (1988). A principal-components analysis of the Narcissistic Personality Inventory and further evidence of its construct validity. Journal of Personality and Social Psychology, 54, 890-902. Richman, J. A., & Flaherty, J. A. (1990). Gender differences in narcissistic styles. In Plakun, E. M. (Ed.), New perspectives on narcissism (pp. 73-100). Washington, DC: American Psychiatric Press. Robbins, S. B., & Patton, M. J. (1985). Self-psychology and career development: Construction of the Superiority and Goal Instability scales. Journal of Counseling Psychology, 32, 221 ? 231. Robins, R. W., Hendin, H. M., & Trzesniewski, K. H. (2001). Measuring global self- esteem: Construct validation of a single-item measure and the Rosenberg self- esteem scale. Personality and Social Psychology Bulletin, 27, 151-161. Rogers, C. R. (1961). On becoming a person: A therapist?s view of psychotherapy. Boston: Houghton Mifflin. Rose, P. (2002). The happy and unhappy faces of narcissism. Personality and Individual Differences, 33, 379-391. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Scheier, M.F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment and implications of generalized outcome expectancies. Health Psychology, 4, 219 ? 247. 122 Shaw, J. A. (2000). Narcissism as a motivational structure: The problem of personal significance. Psychiatry, 63, 219-230. Shostrom, E. L. (1965). A test for the measurement of self-actualization. Educational and Psychological Measurement, 24, 207-218. Shostrom, E. L. (1966). Manual for the Personal Orientation Inventory. San Diego: Educational and Industrial Testing Service. Shostrom, E. L. (1975). Personal Orientations Dimensions. San Diego: Educational and Industrial Testing Service. Soares, A. T., & Soares, L. M. (1965). Self-perception inventory. Trumbull, CT: ALSO Corporation. Solomon, R. S. (1982). Validity of the MMPI narcissistic personality disorder scale. Psychological Reports, 50, 463-466. Soyer, R. B., Rovenpor, J. L., Kopelman, R. E., Mullins, L. S., & Watson, P. J. (2001). Further assessment of the construct validity of four measures of narcissism: Replication and extension. The Journal of Psychology, 135, 245-258. Staats, S. (1989). Hope: A comparison of self-report measures for adults. Journal of Personality Assessment, 53, 366 ? 375. Synder, C. R., & Fromkin, H. L. (1977). Abnormality as a positive characteristic: The development and validation of a scale measuring need for uniqueness. Journal of Abnormal Psychology, 86, 518-527. Synder, M. (1972). Self-monitoring of expressive behavior. Journal of Personality and Social Psychology, 30, 526-537. 123 Templer, D. I. (1970). The construction and validation of a death anxiety scale. The Journal of General Psychology, 82, 165-177. van Deurzen-Smith, E. (2000). Everyday mysteries: Existential dimensions of psychotherapy. New York: Routledge. Watson, P. J., & Biderman, M. D. (1993). Narcissistic personality inventory factors, splitting, and self-consciousness. Journal of Personality Assessment, 61, 41-57. Watson, P. J., Grisham, S. D., Trotter, M. V., & Biderman, M. D. (1984). Narcissism and empathy: Validity evidence for the narcissistic personality inventory. Journal of Personality Assessment, 48, 301-305. Watson, P., Hickman, S., & Morris, R. (1996). Self-reported narcissism and shame: Testing the defensive self-esteem and continuum hypotheses. Personality and Individual Differences, 21, 253-259. Watson, P. J., Hickman, S. E., Morris, R. J., Milliron, J. T., & Whiting, L. (2001). Narcissism, self-esteem, and parental nurturance. The Journal of Psychology, 129, 61-73. Watson, P. J., Hood, R. W. Jr., Morris, R. J., & Hall, J. R. (1987). The relationship between religiosity and narcissism. Counseling and Values, 31, 179-184. Watson, P., Little, T., Sawrie, S., & Biderman, M. (1992). Measures of the narcissistic personality: Complexity of relationships with self-esteem and empathy. Journal of Personality Disorders, 6, 433-448. Watson, P.J., & Morris, R. J. (1990). Irrational beliefs and the problem of narcissism. Personality and Individual Differences, 11, 1137-1140. 124 Watson, P. J., & Morris, R. J. (1991). Narcissism, empathy, and social desirability. Personality and Individual Differences, 6, 575-579. Watson, P. J., & Morris, R. J., & Miller, L. (1997). Narcissism and the self as continuum: Correlations with assertiveness and hypercompetitiveness. Imagination, Cognition, and Personality, 17, 249-259. Watson, P. J., Sawrie, S. M., Greene, R. L., & Arredondo, R. (2002). Narcissism and depression: MMPI-2 evidence for the continuum hypothesis in clinical samples. Journal of Personality Assessment, 79, 85-109. Watson, P. J., Taylor, D., & Morris, R. J. (1987). Narcissism, sex roles, and self- functioning. Sex Roles, 16, 335 ? 350. Watson, P., Varnell, S., & Morris, R. (1999). Self-reported narcissism and perfectionism: An ego-psychological perspective and the continuum hypothesis. Imagination, Cognition, and Personality, 19, 59 - 69. Whitmer, J. M., & Sweeney, T. J. (1999). Toward wellness: The goal of helping. In T. J. Sweeney (Ed.), Adlerian counseling: A practitioner?s approach. Philadelphia: Taylor & Francis. Widiger, T. A., & Frances, A. J. (2002). Toward a dimensional model for the personality disorders. In P. T. Costa , & T. A. Widiger, (Eds.). (2002). Personality disorders and the Five-Factor Model of personality (2nd ed.). Washington, DC: American Psychological Association. Wink, P. (1991). Two faces of narcissism. Journal of Personality and Social Psychology, 61, 590-597. 125 Yalom, I. D. (1980). Existential Psychotherapy. New York: Basic Books. Zuckerman, M. (1971). Dimensions of sensation seeking. Journal of Consulting and Clinical Psychology, 36, 45-52. 126 APPENDIX A Demographic Sheet Age: _______ Sex: _______ Race or Ethnicity: _______________ Religious Preference: _______________ Sexual Orientation: ______________ Classification (Freshman, Sophomore, Junior, Senior, Grad Student): Estimated GPA: __________ Major/Area of Study: _________________