Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Narcissistic personality and academic underachievement in school age children Mah, Terry 1988

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1988_A8 M33.pdf [ 6.95MB ]
[if-you-see-this-DO-NOT-CLICK]
Metadata
JSON: 1.0097751.json
JSON-LD: 1.0097751+ld.json
RDF/XML (Pretty): 1.0097751.xml
RDF/JSON: 1.0097751+rdf.json
Turtle: 1.0097751+rdf-turtle.txt
N-Triples: 1.0097751+rdf-ntriples.txt
Original Record: 1.0097751 +original-record.json
Full Text
1.0097751.txt
Citation
1.0097751.ris

Full Text

NARCISSISTIC PERSONALITY AND ACADEMIC UNDERACHIEVEMENT IN B.Ed., The University of British Columbia, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF THE FACULTY OF GRADUATE STUDIES (Department of Educational Psychology) (Graduate Programme in School Psychology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 1988 (c) Terry Man, 1988 SCHOOL AGE CHILDREN By TERRY MAH MASTER OF ARTS in In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Educational Psychology The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date March 30, 1988  DE-6(3/81) i i ABSTRACT The achievement (as indexed by standardized test scores) of 56 (54 females and 2 males) private school children was studied in relation to demographic (social status and gender), behavioral (three indicators of persistence), dispositional (clinical and psychometric measures of narcissism), and ability (Otis-Lennon) factors. A clinical procedure and device were developed to augment the information yielded by those procedures whose purpose was primarily the generation of quantitative data. The qualitative and quantitative material was studied together to explore Freud's distinction between libidinal types, which might be implicated in differences in cultural (e.g., school) achievement. Results are discussed in relation to research, assessment, and educational issues. Thesis Supervisor iii TABLE OF CONTENTS CHAPTER PAGE List of tables ..v List of figures vii Acknowledgement viiI Derivation and statement of the problem 1 Kohut' s conception and assessment 6 Narcissism and school achievement: recent work 10 Statement of the problem 14 II Review of literature 16 Scope and delimitationThe early work of Freud and the recent contributions of Kohut and Kernberg 17 Psychometric assessment of the narcissistic personality 23 Recent studies on behavioral persistence 29 Recent studies on academic achievement 30 Direct lineage of the present study 33 III Design and method...-. 35 Hypotheses 36 Subjects ....3Procedures 7 Narciss ism 8 Measures one to four: assessment through pro ject ives 3Measure five: personality inventory 48 Persistence 4Measure one: "naturalistic": In Class Observation 48 Measure two: "contrived": Picture Reproduction 9 Measure three: "contrived": Letter Deletion .....50 Ability 5Achievement 1 Evaluat ionIV Results 3 Section A: description of the sample 55 (1) The homogeneity of the sample(2) Comparison of the two subsets of the sample 57 Section B: measurement of narcissism with projectives 6(1) Evaluation of the projective instrument 67 (2) Assessment of narcissism with the projective and NPI 70 Section C: the obsessional and non-obsessional types 78 (1) Defining the obsessional type 79 (2) Defining obsessionals by clinical judgment ....80 (3) Defining obsessionals as overachievers . 85 Section D: examining relationships between the measures.... 91 (1) Narcissism and persistence: rank order consistency 91 (2) The relationship of the measures to achievement 94 (3) Explaining variance in achievement: regression analysis 96 D iscuss ion .98 Special nature of the study..." 9Narcissism: construct and assessment 100 The merits of the projective instrument and procedures 102 Issues emerging from use of the projective instrument 105 (1) Measurement and assessment issues 105 (2) Psychological and educational issues..106 (3) Conceptual issues 107 Narcissism, persistence, and achievement 108 Narcissism and cross-situational persistence..110 Evaluation of hypotheses 114 Future directions 117 Glossary 12References 4 Appendix A: projective device 131 Appendix B: record form for projective device...147 Appendix C : NPI 152 LIST OF TABLES TABLE PAGE I Demographic data for the aggregate sample 56 II Summary statistics on the measures of ability, reading achievement, persistence, and narcissism for the aggregate sample 58 IIIA Summary of oneway analysis of variance on each independent measure for School 1 versus School 2 60 IIIB Summary of Chi-square analysis of the distributions of subjects on the social status measure for School 1 versus School 2....62 IIIC Summary of Chi-square analysis of the distributions of subjects on judged degree of narcissism for School 1 versus School 2 63 IV Summary of totals for the 14 stimulus cards and for the representational (R) and presentational (P) indicators on the projective measure of narcissism for the aggregate sample 68 VA Summary of oneway analysis of variance on each measure for subjects in the top quartile (T) versus subjects in the bottom quartile (B) of the projective measure of narcissism 71 VB Summary of Chi-square analysis of the distributions of subjects on overall judgment of narcissism for the top and bottom quartiles of the projective 73 VC Summary of Chi-square analysis of the distributions of subjects on judged degree of narcissism for the top and bottom quartiles of the projective 74 VD Summary of Chi-square analysis of the distributions of subjects on judged type of narcissism for the top and bottom quartiles of the projective 75 VI Summary of oneway analysis of variance on each measure for subjects in the top quartile (T) versus subjects in the bottom quartile (B) of the NPI 76 V 1 VIIA Correlation coefficients (Pearson r) of the narcissism measures with the reading achievement and persistence measures for obsessionals (defined by,clinical judgment)....81 VIIB Correlation coefficients (Pearson r) of the narcissism measures with the reading achievement and persistence measures for non-obsessionals (defined as all those not judged as obsessionals) 81 VIIIA Summary of oneway analysis of variance on each measure for obsessionals (OB) versus non-obsessionals (N-OB) (defined by clinical judgment) 84 VIIIB Summary of Chi-square analysis of the distributions of subjects on judged degree of narcissism for obsessionals and non-obsessionals (defined by clinical judgment) 86 IXA Correlation coefficients (Pearson r) of the narcissism measures with the reading achievement and persistence measures for obsessionals (defined as overachievers) 87 IXB Correlation coefficients (Pearson r) of the narcissism measures with the reading achievement and persistence measures for non-obsessionals (defined as other than overachi evers ) 87 X Summary of oneway analysis of variance on each measure for obsessionals (OB) versus non-obsessionals (N-OB) (defined by achievement) 90 XIA Cronbach's alpha coefficients (CA) showing the rank order consistency across the three persistence measures for subjects high in narcissism versus subjects low in narcissism. 92 XIB Cronbach's alpha coefficients (CA) showing the rank order consistency across the three persistence measures for obsessionals versus non-obsessionals 92 XII Correlation coefficients (Pearson r) of the independent measures and the reading achievement measures (STEP from School 1 and CTBS from School 2) 95 vii LIST OF FIGURES FIGURE PAGE 1 Plot of the ability measure (Otis-Lennon) and the reading achievement measure (STEP) from School 1 64 2 Plot of the ability measure (Otis-Lennon) and the reading achievement measure (CTBS) from School 2.. 65 3 Plot of the ability measure (Otis-Lennon) and the reading achievement measures from both schools 66 ACKNOWLEDGEMENTS My deepest gratitude goes to Dr. LeRoy D. Travis, whose intellectual sophistication inspired me to undertake and complete this work of psychological, educational, and cultural significance. Rarely have I met an individual as devoted to the pursuit of knowledge and scholarly research. I am grateful to have had the opportunity to grow and develop both as a clinician and as a researcher under his tutelage. I wish to extend my sincere appreciation to Dr. John Allan and Dr. David Whittaker for their thoughtful contributions. I would like to express a special note of thanks to Dr. William B. White for giving me his experience and friendly advice. I also want to thank Miss Rosalind Addison, Miss Ursula Bell, and Mrs. Christine Addison for their hospitality and helpfulness. And finally, I should thank my family for their understanding, encouragement, and continuing support. 1 CHAPTER ONE DERIVATION AND STATEMENT OF THE PROBLEM The problem of underachievement has been one of the most discussed topics in education and psychology. The term "academic underachievement" is defined as a level of scholastic performance that is not commensurate with one's intellectual abilities, as measured by the discrepancy between a subject's respective scores on standardized achievement and intelligence tests. The condition of underachievement can be produced by different concatenations or combinations of factors. That is, just as underachievement varies in content or domain, in degree of severity or seriousness, in generality or consistency, and in intractability or insensitivity to remediation efforts, so it varies with respect to the factors or conditions which are responsible for it (i.e., in etiology). While certain matters are commonly associated with underachievement (e.g., unfavorable living conditions, inadequate instruction, sensory impairments), conceptualizations and investigations of what personal attributes might be implicated in underachievement have not been very illuminating. For example, work on need for achievement has been disappointing (Anderson & Travis, 1983). Similarly, self studies which focus on constructs like self-concept (Allport, 1965; Epstein, 1973; Gergen, 1961; Mead, 1934) have also contributed little insight into 2 the problem of underachievement (e.g., Scheirer & Kraut, 1979). Even so, the importance of affect and personality in an individual's endeavors is difficult to abandon. We intuitively sense that it is the person that is moved to seek coherence, meaning, or correspondences which help him make his way in the world. Over the years, clinical psychologists have offered ideas about what impairs people in their endeavors. But not all these notions, in fact few of them, when they are brought under empirical scrutiny, yield or add much to our understanding of what produces underachievement. One avenue that begs for exploration is the realm known as affective disorders and so-called borderline conditions. Those of less than acute character are normal in the sense that they are so widespread that people generally do not register "official complaints" by taking themselves to a therapist for relief. One category of such conditions is narcissism. The term "narcissistic personality" (pathological narcissism), as described in the Diagnostic and Statistical Manual of Mental Disorders (Third Edition) (DSM III) (1980, pp. 315-317), refers to a personality disorder in which there are a grandiose sense of self-importance or uniqueness; preoccupation with fantasies of unlimited success, ability, power, wealth, beauty, ideal love, brilliance, and omniscience; exhibitionistic need 3 for constant attention and admiration; characteristic responses to threats to self-esteem (e.g., to criticism, indifference, or defeat), such as cool indifference or marked feelings of rage, inferiority, shame, humiliation, or emptiness; and characteristic disturbances in interpersonal relationships, such as feelings of entitlement, interpersonal exploitativeness, relationships that alternate between the extremes of overidealization and devaluation, and lack of empathy. The exaggerated sense of self-importance may be manifested as extreme self-centeredness and self-absorption. The discussions of Kohut (1977) and Kernberg (1975), two of the most highly regarded authorities on this subject, reflect conceptions that are consistent with this description; and it is primarily from their works and those of Sigmund Freud that the present work draws its sense of what narcissism entails. Freud (1914), with whom this psychological construct originated, explains that narcissism results from the withdrawl of libidinal interests from external objects and individuals, and the subsequent reinvestment of libido back into the ego. Pathological narcissism results from the inability to redirect libidinal interests into external objects and the undue concentration of libido in the ego. 4 This condition of being in love with one's own image is often referred to as "self-love" (Fromm, 1956; Lowen, 1983). The critical distinctions and definitions associated with the concept of narcissism are elaborated in the context of a discussion of pertinent literature (in Chapter Two). According to Baker (1979) and the DSM III (1980), narcissistic personalities tend to overestimate their abilities and achievements. Individuals with this disorder seek external admiration and attention with unusual frequency. Because of their fragile self-esteem, they may respond to criticism, disappointment, or defeat with marked feelings of rage, followed by feelings of inferiority, shame, and emptiness in light of their "acting-out" behavior (direct expressions of conflictual tensions in annoying or anti-social behavior or in fantasies). Accordingly, narcissistic personalities are also intensely envious of others, who they perceive as being sources of narcissistic supply (Kernberg, 1974). Of course, such qualities can create difficulties for persons in social situations such as those of the school. Another quality which clinicians attribute to narcissistic people might be regarded as potentially inimical to optimal functioning in an educational context. Usually present in these individuals is the sense of "narcissistic entitlement," or the expectation of special favors without assuming reciprocal responsibilities. The grandiose self demands absolute control of the environment. 5 Such individuals feel as if they have the right to control and possess others and to exploit them without guilt feelings (Kernberg, 1975). Many teachers might lack sympathy for those who manifest such qualities. Indeed, they might express negative reactions toward such individuals. These are further reasons for studying personal dimensions like narcissism and their relationships with achievement. Of course, such investigation requires both reliable and valid identification and observation of these variables. This fact may account for the paucity of studies. Just as assessment of other personality variables has been problematical (Mischel, 1968; Travis, Violato, & White, 1982) assessment of narcissism has been a problem. Even so, some attempts have been made to develop appropriate instrumentation and assessment techniques to identify and measure the narcissistic personality. Progress in these matters has been notably modest. Several attempts at constructing a measure of narcissism have yielded mixed results (Grayden, 1958; Harder, 1979; Young, 1959; Ashby, Lee, & Duke, 1979; Raskin & Hall, 1979, 1981). Such efforts have not been especially assiduous in making strong links between the constructs and clinical manifestations on the one side and the measures on the other. All should notice Kohut's (1971) claim that the narcissistic personality is assessed necessarily by means which recognize that a mirroring or idealizing transference provides the most 6 theoretically compelling indicators. The present study attempted to resolve some of the difficulties encountered by previous researchers. Accordingly, Kohut's views on this matter warrant attention. Kohut's Conception and Assessment An appreciation, of Kohut's ideas and thoughts requires a grasp of a central idea or concept from the psychoanalytic literature and its referents. The concept is "transference." Transference refers to the mobilization or revival of repressed, infantile, object-directed, libidinal urges that are related to objects in the present. In psychoanalysis, the patient re-enacts earlier experiences and emotions in relation to the analyst. But the transference phenomena are not confined only to the psychoanalytic situation; they also appear in social life generally. Freud (1962, p.83) states: Transference arises spontaneously in all human relationships just as it does between the patient and the physician.... The less its presence is suspected, the more powerfully it operates.... Psychoanalysis does not create it, but merely reveals it... . With these ideas in mind, we can consider some of Kohut' s thoughts about narcissism and its appraisal. In an "idealizing transference," says Kohut (1971), the patient perceives the therapist as the idealized parent figure during the analysis. The narcissistic personality 7 will feel vulnerable to and envious of the idealized parent figure, as it represents his source of narcissistic supply. In the course of the analysis, the primary defence is erected against the possibility of depending on the analyst, since the development of a situation in which the patient feels dependent immediately revives the threatening and ego-weakening experiences of oral frustration, anxiety, depression, anger, and envy from early childhood. The patient exhibits acting-out behavior, which appears to stem from his unconscious fantasies of omnipotence, as a defence against feeling vulnerable to the environment and as a denial of his helplessness or dependency on others. Acting-out behavior is further manifested in the "mirroring transference" (Kohut, 1971), wherein the patient sees the therapist as merely an extension of the patient's personality. Here, the therapist senses that he is not being perceived as an "individuated person." (Essentially, Kohut states that an individuated person is an autonomous individual, whose independence of the environment implies a greater stability in facing deprivations, frustrations, and other problems.) The patient looks to the therapist for admiration and acclaim. However, the patient is unfortunately unable to acknowledge receipt of the help and attention from the therapist, for recognition of such would arouse intense envy in the patient. To defend against feeling dependent upon and envious of the therapist, and to keep his perfect image of the grandiose self intact, the 8 patient devalues the therapist's help. If confronted or criticized by the therapist for appearing distant and unresponsive to treatment, the patient usually becomes enraged with bitterness and sarcasm in analysis (Kohut, 1971). Subsequently, signs of a sense of "omnipotence" and "narcissistic rage" become evident. Omnipotence (fantasies of unlimited success, power, brilliance, and beauty or a grandiose sense of self-importance, i.e., exaggeration of achievements and talents) is asserted as a denial of feelings of vulnerability. Narcissistic rage (a manifestation of the vulnerable patient's aggressive responses to actual or anticipated injury to the aggrandized sense of self, i.e., the need for revenge and the unrelenting compulsion in the pursuit of undoing a hurt) is a reaction to an experience of injury to the vulnerable grandiose self. Patients who are difficult to treat may prompt serious "countertransference" problems in the therapist. The countertransference refers to the therapist's overt and covert reactions to the patient's acting-out behaviors in analysis. Because of the patient's recalcitrant disposition toward treatment, the therapist is often left feeling helpless and impotent. In effect, the patient's devaluative remarks and rage reactions toward the therapist can, quite frequently, leave the therapist with marked negative feelings toward the patient. 9 Since transference phenomena are not restricted to the therapist-patient context, but are found, as we have seen, in social situations of the most common sort, we might have been blind to the significance of these phenomena in schools. The ability of teachers to teach narcissistic pupils can be greatly impaired by said narcissism. Presumably, narcissism in teachers would compound such problems. In any case, Kohut's thoughts are provocative when we bring them to bear on the problems of teaching and learning. Perhaps our grasp of the underachievement problem will be made more sure and firm if we pursue this matter through research like the present study. Despite the current interest in narcissism both as a societal and a clinical phenomenon (Berkowitz, 1977; Bleiberg, 1984; Bromberg, 1983; Coles, 1977; Goldstein, 1985; Lasch, 1978) and despite the multitude of studies conducted on academic underachievement (Haertel, Walberg, & Weinstein, 1983; Parkerson, Lomax, Schiller, & Walberg, 1984; Thomas, 1980; Thompson, 1985; Violato & Travis, 1985) little work has been done to clarify the nature of the relationship between narcissism and underachievement in school. Although evidence which clarifies the nature of the relationship of narcissism to educational underachievement is very scarce, there are reasons to suppose that such understanding may be important to resolve some of the more intractable problems pertaining to the teachability of students with narcissistic personalities. To repeat: 10 narcissism "may be a critical feature of some underachievement. The work of a small number of researchers suggests that some others are thinking along these lines. Narcissism and School Achievement: Recent Work Baker (1979) is one of the few who proposed that the development of narcissistic disturbances impedes scholastic performance. Baker's reasoning, which links narcissism and underachievement, is quite clear. Narcissistic individuals have an aggrandized image of themselves and are thus unable to accept events which do not confirm the perfection of the "grandiose self." Since academic studies require an implicit admission of ignorance as well as regular struggles with difficult intellectual material, one might expect students with high degrees of narcissism to have difficulty. Because these struggles imply challenge to the perfection of the grandiose self and thereby threaten self-esteem, narcissistic students might be expected to react with notable anxiety, depression, or narcissistic rage. These reactions would hinder the capacity of such people to think clearly and study effectively. In order to escape from the source of the dysphoria, one would expect these students to withdraw into a state of apathy and passivity regarding challenges in school. Rather than study harder, they may avoid studying when presented with difficult material. Avoidance of study could maintain the delicate cohesion of the grandiose self in two ways: failure on an exam could be attributed to lack of study rather than to lack of ability; 11 however, passing an exam without studying verifies the sense of brilliance and omniscience. In both cases, work is minimized and so is achievement. If students do not prepare for an examination, a crisis usually results. These students feel hopelessly inadequate if they are unable to solve an early question. The crisis cripples their ability to concentrate; a poor grade is received; and further damage to self-esteem occurs. Some students withdraw further by refraining from attending classes and even miss important examinations, while others undertake time-consuming activities which will reaffirm their grandiose sense of self. These tactics increase the potential for academic failure and diminish self-esteem. At this time, reports on empirical investigations of these matters are extremely hard to find. The only one found (besides that of Baker) is that of Vigilante (1983), who studied students' narcissism and academic performance. She distinguished between healthy narcissism, which she called "situational self-preoccupation" and pathological narcissism, which she named "characterological self-preoccupation" (pp. 603-604). In the former state, the individual does not require the exploitation of others to maintain his self-esteem. Conversely, in the latter condition, the individual needs to exploit others for self-gratification. Kohut (1971) explains this tendency to exploit others in notably narcissistic people as follows. He says that 12 narcissistic personalities need other individuals as external objects for maintaining self-esteem. If the grandiose self persists into adulthood without modification, the individual is burdened with an array of internalized and unconscious needs, which cannot be satisfied. Without realistic personal standards, healthy narcissism is impossible, and so the individual must constantly look to the environment for admiration and acclaim. A crushing loss of self-esteem results whenever the craved perfectionistic reflections of the grandiose self are not "mirrored" back by the outer world. Thus, it is probable that narcissistic individuals engage in shallow, exploitative relationships with others. Vigilante's (1983) study reports data taken from responses to a questionnaire she developed to identify and differentiate subjects with healthy narcissism from those with pathological narcissism. Their respective relationships with learning processes were also examined. Her findings suggested that both types of self-preoccupation were associated with learning problems. Specifically, students identified as having situational self-preoccupation had only transient learning problems, whereas those identified as having characterological self-preoccupation appeared to have continuing learning difficulties. However, the subjects in the study consisted of a select group of university students, who qualified for admission into a program of studies in social work. Accordingly, the range 13 of her generalizations is unknown. Moreover, other considerations must be weighed too. Although Baker (1979) and Vigilante (1983) both argue that a negative relationship between the narcissistic personality and academic achievement is probable, Freudian theory suggests that in some cases the reverse may be true (Freud, 1931). From this perspective, one might argue that one of the varieties of the narcissistic personality (the so-called mixed type known as the narcissistic-obsessional) should be positively associated with academic achievement. In his discussion of "libidinal types," Freud (1931) says that a particular libidinal type is the most valuable variant from a cultural standpoint. This variant, which Freud called the "narcissistic-obsessional type," is culturally inclined. Freud says he is able to persist through adversity to meet his high standards for achievement so that he may be recognized as a stalwart from a cultural perspective. Kernberg (1976) adds that some narcissistic personalities have the capacity for active and consistent work in some areas, which permit them to partially fulfill their ambitions of greatness and of obtaining admiration and acclaim from others. Naturally, their capacity for earnest endeavor may be exercised in the educational realm. This schizm between original theory and contemporary investigation called for inquiry. Therefore, the focus of the present study entailed exploration of the narcissistic construct and the validity of claims that two configurations 14 of narcissism — obsessional and non-obsessional — exist. It also entailed exploration of the relationships of manifestations of narcissism to academic achievement. Specific predictions about these two putatively distinguishable narcissistic configurations and achievement were investigated within a sample of private school children. Statement of the Problem This study was designed to explore the relationship between manifestations of narcissism and academic achievement in school-age children. A study of this nature has both a theoretical and a pedagogical purpose. Of theoretical importance is the identification of the variations of the narcissistic personality and the delineation of the narcissistic factors which influence children's school performance. Of pedagogical significance is the interpretation of the psychodynamic events that occur between a teacher and a student in the classroom setting: we have reason to ask if the interactions resemble those that arise between a therapist and a patient in the psychoanalytic situation. As we have seen, Freud (1962, p.83) explicitly claimed that the transference phenomena were not confined to the analytic situation. Transference arises spontaneously in all human relationships just as it does between the patient and the physician.... The less its presence is suspected, the more powerfully it operates.... 15 Psychoanalysis does not create it, but merely reveals it... . The present study attempted to draw parallels between the psychodynamic events that occur between a therapist and a patient in the psychoanalytic context and those that arise between a teacher and a student in the classroom situation. Since narcissistic factors may account for the unteachability of capable students, a clearer understanding of the psychodynamic events that occur in the classroom can help us deal with our feelings of impotence, frustration, and anger at them. Moreover, narcissism may mediate or moderate persistence. This possibility emerges from the literature. 16 CHAPTER TWO REVIEW OF LITERATURE Scope and Delimitation Since Freud's introductory work on narcissism in the first years of this century but prior to World War One, the analytical writings of Kohut and Kernberg, despite their contradictions, have thrown more light on this phenomenon of personality. In order to achieve a greater understanding of narcissism, the contrasting theories of these two most prominent authorities on the subject are worthy of examination. Although both Kohut and Kernberg rely on transference manifestations in psychoanalysis as diagnostic indicators of the narcissistic personality, others have attempted to measure the construct using questionnaires and inventories. Accordingly, the adequacy of these attempts is also assessed, and the merits of projective techniques is explored as an alternative assessment procedure. Finally, since tenacity in the face of difficulty is necessary for significant educational attainments, persistence is studied as an observable behavioral pattern that mediates the personal psychodynamics and educational performance and attainment for persons with whatever degree of narcissism. Accordingly, a review of the recent research on both persistence and achievement, followed by a discussion of the relationship of the narcissistic personality and behavioral persistence to academic 17 achievement concludes a discussion of literature and concepts pertinent to the present study. The Early Work of Freud and the Recent Contributions of Kohut and Kernberg Freud (1914) introduced narcissism as a construct to clarify the course of human sexual development in the context of his libido theory. Freud's central tenet was that the libidinal cathexis in another subject is object love whereas the libidinal investment in the self is narcissism. He assumed that the individual has a fixed quantity of libido. Despite subsequent transformations in usage, the term narcissism has nonetheless retained the notion of a positive libidinal feeling toward the self (Drever, 1952; Lasch, 1978). Freud (1914) differentiated between two developmental stages of narcissism: primary and secondary narcissism. In primary narcissism, all libido is invested in the self. The construction of ego boundaries enables libido to be invested in external objects, while some libido remains attached to the self as residual primary narcissism. In secondary narcissism, libido is withdrawn from external objects and reinvested in the self, augmenting residual primary narcissism. This withdrawl of love from the object onto the self is regarded as a pathological condition. Freud therefore perceived narcissism to be an immature self-centered state in which an individual indulges at the expense of object love. 18 Building on Freud's early work on narcissism are the two prominent psychoanalysts, Heinz Kohut and Otto Kernberg. Their contrasting viewpoints have ordered what are perhaps the two main schools of thought on the narcissistic personality disorder. Kohut's view on narcissism deviates from Freud's definition of the condition as the investment of libido in the self. Kohut (1971, 1972) defines narcissism not by the target of libido as between either self or object, but by the nature of the libidinal cathexis; the attachment of libido to self or object is considered to be narcissistic if it is "idealizing" (i.e., the analyst serves as an idealized parent image) or "self-aggrandizing" (i.e., the analyst serves as a mirror for the patient's grandiose self). These two qualities or states pertain to the two narcissistic transferences which Kohut regards as being crucial for an accurate diagnosis of the narcissistic personality: the "idealizing transference" and the "mirror transference." In the idealizing transference, the patient's early need for merging with an idealized parent image or omnipotent object is revived in the psychoanalytic situation. Perfection is attributed to an admired "self-object" (i.e., the "you are perfect, but I am part of you" view of the parent figure). In the mirror transference, the patient's early need for parental acceptance or "mirroring" is revitalized in the assessment situation. Perfection is ascribed to the self (i.e., the "grandiose self" or the "I 19 am perfect" image of the self). Kohut emphasizes the importance of allowing these transferences to arise spontaneously, uninhibited by premature interpretation. The emergence of the idealizing and the mirror transferences respectively reflects the mobilization of the two narcissistic configurations in the patient: the ideal parent image and the grandiose self. Kohut believes that these two narcissistic structures originate from developmental deficiencies that can arise between birth and latency. Although these narcissistic configurations are archaic, they are nevertheless stable, cohesive, and resistant to fragmentation. This point is critical to Kohut, since it is the stability and cohesiveness of the narcissistic personalities that distinguishes them from the more pathological, less stable borderline personalities, who are more susceptible to fragmentation. It is this stability and cohesiveness that enables the narcissistic personality to be distinguished from the borderline. Hence in the present study, it was expected that the stability and cohesiveness of narcissistic subjects would be evidenced by responses that are consistent across different situations (as opposed to responses that are situation-specific). According to Kohut (1977), both narcissistic structures are natural occurrences in the course of a child's normal development. Under favorable conditions, the child matures, and these narcissistic configurations become transmuted and integrated into the rest of the adult personality. The 20 idealized parent image becomes internalized as the idealized superego and provides one's ideals and guidelines, while the grandiose self is transformed into normal self-esteem, ambition, and self-confidence. Essentially, Kohut postulates that the idealized parent image and the grandiose self persist into adulthood as a result of a "developmental arrest." When the transformation of the primitive structures stops, they remain unaltered and detached from the rest of the personality. The result is pathological narcissism. (In pathological narcissism, the DSM III descriptive features of the disorder are manifested to a notably greater degree than in normal narcissism.) The patient's fixation at infantile narcissistic stages of development is directly related to inevitable shortfalls in maternal care. If the child experiences the mothering figure or another significant caretaker figure as cold, rejecting, or destructive, it becomes painfully aware of its relative insignificance and vulnerability. If the child does not experience maternal empathy, the narcissistic configurations do not become integrated into the rest of the personality. Instead, they remain as archaic and primitive structures, which comprise the core of the narcissistic personality. Unlike Kohut, Kernberg views the narcissistic personality as the outcome of the early development of specific psychopathological structures rather than the result of the arrested development of normal narcissistic 21 configurations. Kernberg (1974, p. 216; 1975, pp. 265, 271, 325) defines pathological narcissism not merely as the libidinal investment in one's self, but as the libidinal investment in a pathological, although integrated, self-structure called the grandiose self. The grandiose self is characterized by a fusion of three pathological structures: the "actual self" (i.e., the uniqueness of the child that was reinforced by early experiences); the "ideal self" (i.e., the fantasies of unlimited power, wealth, and beauty that compensated the child for the ego-weakening experiences of intense oral frustration, rage, and envy); and the "ideal object" (i.e., the fantasy of an ever-giving,, ever-loving, and accepting caretaker, in contrast to the experience in reality). Essentially, Kernberg believes that these structures originate from a "pathological development" that can arise during the late oral stage of development. Like Kohut, Kernberg (1980) claims that these pathological structures develop in the early environment of a chronically cold parent figure. The grandiose self serves as a defense against the child's traumatic experiences of severe oral frustration and of a world without food and love in the mother-child relationship. These injurious experiences result in the child's envy, hatred, and rage toward the caretaker. Enraged at the mothering figure, the child withdraws affection from her and invests it in the grandiose self. The result is pathological narcissism. 22 Both Kohut and Kernberg rely on transference manifestations in psychoanalysis as diagnostic indicators of the narcissistic personality. Kernberg regards the patient's denial of the analyst as an independent person as the key diagnostic sign. The patient's perception of the analyst as merely an extension of himself reveals the patient's grandiose denial of his underlying feelings of dependency, vulnerability, and inadequacy. Kernberg believes that idealized others represent projections of the aggrandized sense of self, and unacceptable features of the sense of self are projected onto others who are devalued and regarded as unreliable. Accordingly, the combination of an inflated sense of self and a devaluation of others provides a defense against feeling dependent upon anyone who is experienced as frustrating. Both Kernberg and the DSM III devote considerable space to the descriptive features of the narcissistic personality. The DSM III describes narcissistic patients as being expressively absorbed in themselves and as having serious "distortions" in their interpersonal relationships. (For example, these distortions may take the form of any of the following: (1) entitlement: expectation of special favors without assuming reciprocal responsibilities; (2) interpersonal exploitativeness: taking advantage of others to indulge own desires or for self-aggrandizement; disregard for the personal integrity and rights of others; (3) relationships that alternate between the extremes of 23 overidealization and devaluation; (4) lack of empathy: inability to recognize how others feel (DSM III, 1980, p. 317)). Kernberg (1974) delineates the narcissistic patients' tendency to envy others and erect defenses against envy (e.g., devaluation, omnipotent control, and narcissistic withdrawl); their tendency to idealize those from whom they expect narcissistic supplies, while devaluing those from whom they expect nothing; their exploitative and parasitic relationships; their sense of entitlement; and their inability to empathize with others. Kernberg notices the curious contradiction that although narcissistic patients have inflated self-concepts, they manifest an inordinate desire for acceptance and acclaim from others to keep their grandiose sense of self intact. Psychometric Assessment of the Narcissistic Personality After the ideas and work of Kohut (1971, 1972, 1977) and Kernberg (1974, 1975, 1980) were published, the diagnosis of the narcissistic personality became a topic of growing interest. However, attempts to develop instruments to measure narcissism have not been singularly successful. Ashby, Lee, and Duke (1979) constructed the MMPI Narcissistic Personality Disorder Scale (NPDS), which consists of 19 selected items from the MMPI. Solomon (1982) discovered that the NPDS differentiated between individuals with healthy and pathological self-esteem. Although the NPDS certainly represents an expedient means of measuring narcissism, it may not be an acceptable assessment device. 24 For it does not have the capability of registering the transference phenomena, which are the critical diagnostic indicators of the construct in clinical practice. The need for a procedure that can generate valid and reliable indicators and yield sufficient variance in the manifestation of such indicators is crucial to the present study. Accordingly, the search for same yielded another questionnaire/inventory, which is more likely to yield a greater dispersion in test scores than would the NPDS. With the exception of the NPDS, the lack of a measuring instrument impeded the empirical study of narcissism until Raskin and Hall (1979) developed the Narcissistic Personality Inventory (NPI) (see Appendix C). The authors believe that the NPI possesses desirable psychometric properties that will facilitate the identification of pathological narcissism. The NPI is a 54 item, forced-choice questionnaire designed to measure individual differences in narcissism as a personality trait. The construction of the device was predicated on the descriptive features of the narcissistic personality disorder as specified in the DSM-III. Raskin and Hall emphasize that only extreme manifestations of those specified behaviors constitute pathological narcissism. The NPI appears to be an objective self-report inventory of narcissism as a personality trait. Because of its greater number of items and hence its greater reliability as a measuring instrument and its capacity to 25 yield greater variance, the NPI was preferred over the NPDS as a questionnaire/inventory measure of narcissism for the current investigation. The development of the questionnaire has been accompanied by several attempts to assess its psychometric qualities including reliability and validity (Emmons, 1981; Raskin, 1980; Raskin & Hall, 1979, 1981). Emmons (1984) factor analyzed the NPI into four disparate factors: "Exploitativeness / Entitlement", "Leadership / Authority", "Superiority / Arrogance", and "Self-Absorption / Self-Admiration" and discovered that all of the factors except Exploitativeness / Entitlement showed a significant positive correlation with self-esteem (the Self-Perception Inventory). (Pearson Product-Moment correlations between self-esteem and each of the four NPI factors are as follows: r=-.02; r=.53; r=.38; r=.41 respectively.) Emmons also found that the total NPI score was positively related to the need for uniqueness (the Body Self-Consciousness Scale) (r=.54), extraversion, and acting (exhibitionism) (the Self-Monitoring Scale) (r=.52 and r=.55 respectively). The four variables: self-esteem, uniqueness, extraversion, and acting accounted for 67% of the variance in the NPI measure of narcissism. Finally, Watson, Grisham, Trotter, and Biderman (1984) found that the Exploitativeness / Entitlement subscale of the NPI was positively related to the NPDS (r=.25) and negatively related to three questionnaire measures of empathy: the Mehrabian and Epstein Scale (MEES) (r=-.24); the Hogan Empathy Scale (HES) 26 (r=-.20); and the Smith Empathic Personality Questionnaire (SEPQ) (r=-.35). But despite the recent attempts to assess the technical aspects of the instrument, and despite the apparently favorable evidence of its reliablility, there remains some serious drawbacks to using the NPI as the sole measure of narcissism. Although the NPDS and the NPI both represent attempts to assess pathological narcissism, the studies designed to validate these instruments usually employ college students as subjects. The degree of generalizabi1 ity of these results is unknown. Certainly, little is known about how children perform on these instruments. Furthermore, and most importantly, the use of questionnaires/inventories to identify pathological narcissism is criticized by Kohut. The essential distinction is diagnosis by transference as opposed to diagnosis by symptoms. Kohut (1971, 1977) asserts that narcissism cannot be assessed by symptoms because narcissistic patients tend to conceal rather than reveal their condition to avoid feeling shame. Questionnaires represent an overt content-oriented approach, which is likely to produce defenses (like denial or disguise) against shame and promote closure in narcissistic patients. Kohut (1971, 1977) maintains that the narcissistic personality is tapped through a transference either in the form of the patient's desire to be in the presence of an idealized parent image or the patient's longing to be mirrored by the analyst. 27 More in keeping with Kohut's psychoanalytic approach to the identification of pathological narcissism is the use of projective procedures. Projectives represent a procedural recognition of human subtleties which subvert obtrusive and impersonal static procedures. They are devised to help clinicians gain insight into the patient's psychodynamics and personality in a non-obtrusive manner which minimizes defenses against direct inspection of sensitive content. The patient's interpretation of the typically ambiguous visual stimuli may indicate the presence of any narcissistic configurations within the subject through the emergence of an idealizing or mirroring transference. Because subjects are asked to respond to generally vague images with their own constructions, the presentation of a series of stimulus cards is less likely to promote defenses against shame and closure in narcissistic patients since there is no apparent reason or signal for defensiveness in these matters. Moreover, the projection of qualities which are threatening onto content external to the person allows for an analysis of such projections for signs of narcissism. In addition, Kernberg's (1975) claim that narcissistic subjects tend to emit a constant projection of self and object images may be further justification of the use of projectives to assess narcissism. Kernberg (p.36) states: Constant projection of "all bad" self and object images perpetuates a world of dangerous, threatening objects, against which the "all good" 28 self images are used defensively, and megalomanic ideal self images are built up. According to Kernberg, narcissistic subjects' distorted perception of significant parental images together with their unrealistic self images are easily projected onto content external to the subjects. Hence, Kernberg's assertion supports the assessment of narcissism through projectives, as do the considerations of Kohut. Kohut's criticism of questionnaire/inventory measures of narcissism has face validity and must be taken seriously. Since narcissistic subjects can defend themselves through impression management, the construct validity of personality inventories such as the NPI remains questionable. Nevertheless, the NPI was included in the current study to discover the relationship of the performances on the NPI to the transference indicators (and to the achievement and persistence phenomena). It was expected that the relationship between performances on the NPI and performances on the projective device (i.e., transference indicators) would be negative due to defensive distortion associated with performances on the NPI. Accordingly, the present investigation adopted Kohut's position that the narcissistic personality is probably best verified diagnostically through the emergence of an idealizing or mirroring transference. The current study included a projective procedure as a means to observe the subjects' orientation to social situations wherein 29 transference phenomena are incipient or usual. Through analysis of the meanings subjects project onto a series of ambiguous social situations in response to invitations to tell what they imagine is happening, there is reason to suppose we may thereby tap any narcissistic configurations within the subject. The graphic depiction of adult-child relationships was used as the stimulus material to which subjects were to orient and respond. The individuals' interpretations of the visual representations was expected to reflect the mobilization of such narcissistic configurations as were present. Proper diagnosis of the narcissistic personality will facilitate a clearer understanding of the relationship of narcissism to academic achievement. The specific procedures for generating responses and for analyzing and interpreting such responses are described later (in Chapter Three). Recent Studies on Behavioral Persistence To clarify the relationship between the narcissistic personality and school achievement, the direct measurement of behavioral persistence was included in the present study. Although behavioral persistence is of interest in its own right, this personal attribute also appears to be related to both narcissism and achievement. Baker (1979) suggested that some narcissistic individuals lack the willingness to persist in the face of difficult intellectual material because they are unable to tolerate such personal struggles that challenge the perfection of the grandiose self. Travis 30 and Violato (1982) argue that the disposition to persist in efforts to overcome difficulties is highly related to achievement. Shrauger and Sorman (1977) found an overall positive correlation between persistence and performance on an anagram task. Travis and Violato's (1982) contention that the individuals who excel in achievement tend to be those who are disposed to persist in the face of difficulty is of particular interest. It raises Freud's (1931) description of the narcissistic-obsessional personality. Freud essentially claims that this libidinal type is able to persist through adversity to meet high standards for achievement so that he may be highly regarded from a cultural standpoint. Thus, the degree to which narcissistic individuals persist in relation to a significant cultural expectation such as achievement in school may provide valuable insight into the current problem under investigation. Recent Studies on Academic Achievement Clearly, academic achievement is of cultural and educational significance. The research literature documenting the relationship of student achievement with other variables is voluminous. Parkerson, Lomax, Schiller, and Walberg (1984) studied the effect of eight important factors on academic achievement: ability," classroom social environment, home learning environment, use of media, motivation, peer groups, quality of instruction, and 31 quantity of instruction. These researchers found, as have many others over the years, that of the eight predictor variables, ability was the most powerful predictor of student achievement. Similarly, Walberg (1984) conducted ten disparate studies of the relationship between achievement and abilities and found that correlations ranged from r-.55 to r=.80. But despite the well established status of abilities in explaining variance in achievement, there still remains some unexplained variance. The current study explored how affective factors like narcissism mediate or moderate achievement through observable patterns of behavior (i.e., persistence) upon which achievement depends. The present study investigated how variance that is unexplained by references to abilities and social privilege is affected by narcissism and persistence. Much of the previous research on achievement appears to acknowledge the large constellation of variables intended to capture the complexities of school performance. Both theory and research suggest that academic achievement is influenced by numerous factors including learner characteristics (e.g., motivation, study habits, abilities), socioeconomic conditions, home and school environment, cultural context, exposure to mass media, peer group, teacher attributes, quality and quantity of instruction, curriculum materials, performance contingencies, and persistence (Haertel, Walberg, & Weinstein, 1983; Parkerson, Lomax, Schiller, & Walberg, 1984; Thomas, 1980; Thompson, 1985; Violato & 32 Travis, 1985). In an effort to impose order on apparent chaos, Haertel et al. (1983) presented eight holistic models of student learning in classroom settings. Still, despite much work in recent years, few factors which conclusively influence student performance have been identified and explained. Further study of the variables which affect academic achievement decisively seems warranted. The current investigation examined both psychodynamic and sociodynamic factors in relation to academic achievement. The present study was designed to explore how narcissistic factors and behavioral persistence influence private school children's performances. Of theoretical interest is the identification of the culturally inclined, narcissistic-obsessional variant of the narcissistic personality. The narcissistic-obsessional libidinal type was identified by clinical judgment and by achievement (as described in Chapter Three). Of pedagogical importance is the interpretation and understanding of the teacher-pupil interactions as they are influenced by phenomena not unlike those which affect therapist-patient exchanges in the psychoanalytic context. Proper assessment of the narcissistic personality may be crucial for gaining a greater understanding of the relationship of narcissism to academic achievement. This study was an attempt to make a contribution to a more complete understanding of these matters. 33 Direct Lineage of the Present Study The studies by Travis and his associates (e.g., Travis & Violato, 1982; Violato & Travis, 1985) provide the lineage from which the present study comes. The number and type of measures and amount of time which each subject contributes necessitates a reliance on relatively small samples. As a consequence, the number of subjects from the less populous (more privileged) strata of society has been small in the Travis studies. Accordingly, the time was propitious for securing a sample of more privileged children so that their characteristics can become better known. The need for this information has been summarized by Coles (1977) as well as Travis (Violato & Travis, 1985) and Feshbach (1978). In effect, personality is differentially developed and expressed in environments which vary. Hence, there is need to study children of the privileged. Private school children were of interest in this study because they possess above average levels of "cultural capital." The expression denotes the cultural resources (intellectual possessions such as knowledge, social savvy, tact, linguistic refinement, aesthetic and moral sensibilities), which are differentially distributed among individuals and groups according to their positions in the political economy, cultural ecology, and social structure. Since these are cultural resources, they are acquired. Hoffman (1986) asserted that cultural capital is needed for educational success. Similarly, Bourdieu (1984) contended 34 that the children of the most privileged elements in society have greater chances for educational success because they (by virtue of their life circumstances) acquire the knowledge, social graces, expectations, language, and social savvy naturally. In contrast, the less privileged must acquire such cultural resources through the investment of effort and the expenditure of time in order to succeed. Furthermore, Coles (1977) claims that children from advantaged backgrounds are likely to have developed a constellation of expectations or a sense of entitlement from learning to control and organize their environment to their liking and the habit of manipulating that environment and those in it. Since so few children from more privileged families are readily available from public schools, it would not be possible to undertake quantitative descriptions and analyses of the psychological factors of significance which are associated with privilege. Accordingly, a sample of private school children was studied to explore the nature of social psychological life at the privileged end of the socio economic spectrum. The present sample of private school children was of interest because of the need to study how a variety of personal attributes (i.e., narcissism and persistence) vary in children who come from families which afford them possession of socio-cultural privilege, and because of the need to understand how privileged position affects educational development. 35 CHAPTER THREE DESIGN AND METHOD Studies which examine the relationship of socioeconomic status and ability to academic achievement have generally found high correlations with some unexplained variance (Thompson, 1985; Violato & Travis, 1985). Although some of this variance may be attributable to measurement error, it is also likely to include variance due to the impact of affective factors. Narcissism may well represent a not inconsiderable contribution to this unexplained variance. In the current investigation, an attempt was made to identify two types of narcissism (obsessionals and non-obsessionals) two ways: by clinical judgment and by achievement. In the former procedure, the differentiation between an obsessional and a non-obsessional was made on the basis of clinical judgment before the actual scoring of the protocols for the projective device. The judgment, then, could not be influenced by the data that was subsequently scored. In the latter procedure, obsessionals were defined as overachievers (those who show greater achievement than one would expect from given levels of ability and social status). The criterion for overachievement was a minimum of a seven percentile rank discrepancy between ability and achievement. All others who were not overachieving were classified as non-obsessionals. 36 (It was expected that narcissistic-obsessionals and narcissistic non-obsessionals could both exhibit high degrees of narcissism. Hence, the basis for differentiating between the two libidinal types was not the degree of narcissism, but rather the degree of persistence and subsequently the level of achievement. Narcissistic-obsessionals were expected to show greater persistence and achievement than narcissistic non-obsessionals.) Hypotheses (1) Performances on the NPI should be negatively related to scores from the projectives (transference indicators) and to the achievement and persistence phenomena. (2) For subjects identified as narcissistic-obsessional, narcissism should be positively related to persistence. (3) For all other subjects (non-obsessionals), narcissism should be negatively related to persistence. (4) For all subjects, persistence should be positively related to achievement. (5) For all subjects, ability should be positively related to achievement. Subjects Fifty-six children enrolled in the sixth and seventh grades of two private elementary schools situated in a large metropolitan center in Western Canada comprised the subject pool for the current investigation. Participation in the study was voluntary. To ensure that the subjects were able to read and understand the questions on the personality 37 inventory without encountering unfamiliar vocabulary and concepts, only subjects in the upper intermediate age range were included in the study. To ensure that children from the upper socioeconomic bracket were sufficiently represented in the study, a sample from two private schools was chosen, and all subjects were asked by their teacher to give a brief written description of their parents' occupations. These descriptions were analyzed to assign each subject to one of five categories according to the top occupation in the family. The categories for social status were entrepreneur, executive/professional, managerial/small business, skilled labor, and clerical/homemaker. Procedures Regard for reliability and ecological validity requires that narcissism and persistence should not be assessed in one situation only (Borg & Gall, 1983; Bronfenbrenner, 1976). For these reasons, multiple assessments were incorporated in this study. Because the correlation between abilities and achievement is significant and substantial (Walberg, 1984), academic ability was added to the present study as a third predictor variable. Accordingly, the current investigation included for each subject five measures of narcissism (the total index from the projective device, overall clinical judgment, judged degree, judged type, and the NPI scale), three measures of persistence (In Class Observation, Picture Reproduction, and Letter 38 Deletion),, percentile ranks of ability (Otis-Lennon School Ability Test), as well as percentile ranks of reading achievement - the criterion variable - (on either the reading subtest of the Sequential Tests of Educational Progress (STEP) and the Canadian Tests of Basic Skills (CTBS)). Narcissism (Measures One To Four: Assessment Through Projectives) A projective measure of narcissism was obtained for all students by presenting a series of stimulus cards each containing a graphic representation of an adult-child encounter (see Appendix A). The series of cards, with which narcissism was assessed, enabled one to obtain an indication of the internal consistency and inferred reliability of the projective measure; an adequate dispersion of scores, which facilitates calculation of correlations between the variables of interest; and an appraisal of the construct validity of the NPI. The use of a series of stimulus cards offered still another important advantage. It was necessary to control for the considerable possibility that subjects may have specific (as opposed to more general, structural, or prototypic) reactions to a picture of a teacher and a student engaged in a classroom activity. Since the transference construct is predicated on the assumption that a structural or prototypic adaptation has been introjected, it was necessary to know whether in fact given reactions are 39 manifestations of this introjection or whether they are more situation specific. The multiplicity and variety of stimulus cards, each representing a different adult-child situation, allows for an exploration and potential control of the situation specificity of subjects' responses. Since the gender of the figures in the depicted situations may be a significant factor in subjects' reactions to them, care was taken to ensure that both males and females perceive equivalent social situations and models. Accordingly, the projective device consisted of 14 stimulus cards which depict 8 different situations. The adult figures in the 8 situations consisted of 5 males and 5 females, and the gender of the subordinate figure (child) in each situation was alternated as follows: (1) a male dentist with a boy; (8) a male dentist with a girl; (2) a mother with a son; (9) a mother with a daughter; (3) a female recreation leader with a boy; (10) a male recreation leader with a girl; (4) an adult with a child (both of ambiguous sex); (11) a man and a woman with a child (sex ambiguous); (5) a female nurse with a boy; (12) a female nurse with a girl; (6) a male teacher with a boy; (13) a male teacher with a girl; (7) a boy scout leader with a boy scout; (14) a girl guide leader with a girl guide. (See Appendix A for more elaboration on and plates of the 14 cards.) Each subject was asked to look at each picture, imagine what is happening, and describe what he or she imagined is 40 taking place between the two figures in the picture. Each child was given a specific set of verbal instructions (i.e., "Here is a picture. I want you to look at the picture and imagine what is happening. Then tell me what is happening."). To ensure that the subject's responses were sufficiently elaborate for scoring, the child's account of what is imagined was supplemented with responses to a series of specific supplementary questions (i.e., "What is this person thinking?" (The child was indicated by pointing.) "How is he feeling?" "What is this person thinking?" (The adult was indicated by pointing.) "How is he feeling?" "How do they get along?" "Is this person (adult) helping this person (child)?" "How does this person (child) feel about that?" "What does this person (adult) think of this person's (child's) work/behavior?" "How does this person (child) react to this person's (adult's) remarks/actions?" "Does this person (adult) understand what this person (child) is thinking?" "Does this person (adult) understand how this person (child) is feeling?"). (Rather than repeatedly saying "child" and "adult" in each of the supplementary questions, the words "this person" was used while pointing to the appropriate figure on the stimulus card. In addition, the order of the questions was varied to make the presentation appear less mechanical.) All of the responses were audiotaped to permit a subsequent measure of interrater reliability and analysis of presentational (expression) as well as representational content (signed 41 meaning) (Edelson, 1984). The subject's responses for each stimulus card were then translated into scores on a protocol (see Appendix B or Chapter Four, Table IV (the composite table)). Notice that each protocol contained the scores to all 14 cards for a particular subject. Once the responses were audiotaped, the scoring and interpretation of the responses entailed two approaches. In the first method, each subject is classified by an overall clinical judgment of his performance on the entire series of stimulus cards taken as a whole. This method is quicker, but it does not reveal as much fine detail as does the second approach. The second procedure requires the classification of reactions to each test item (stimulus card) in order to generate a total score (or classification) for a particular subject. A scoring matrix was used to record and designate how the reactions to the test items were classified for each subject. Because this study incorporated what is perhaps the first attempt at assessing narcissism through projectives, the more appropriate of the two approaches to scoring and interpreting the responses was not known. Accordingly, a procedure which incorporated both approaches was adopted to help shed light on assessment questions. These considerations suggested that a combination of clinical judgment and dimension scale scores be incorporated in the present study. 42 In the former approach, subjects were classified on the basis of their responses across the entire set of stimulus cards. Initially, each subject was classified as either manifesting notable signs of narcissism or not showing notable signs of narcissism by an overall clinical judgment. Secondly, those subjects who were classified as manifesting notable signs of narcissism were given a categorical assignment of high, medium, or low based on judgment of the degree of narcissism manifested. Finally, those subjects who were classified as manifesting notable signs of narcissism were identified as either obsessionals or non-obsessionals based on judgment of type. (Typically, a subject was judged as obsessional if, for example, he said that the child on a given stimulus card was unhappy about his performance because he failed to meet expectations. Thus, one child said, "The girl is upset about getting a poor grade on her science project because she knows she could do better, and now she's afraid that her parents will be disappointed." Another said, "The boy is being scolded by his mother for not cleaning up his room, so now he's mad at himself because his mother expects him to have a clean room." All others who were not identified as obsessionals (including those seven who initially did not show notable signs of narcissism) were classified as non-obsessionals. These classifications were recorded as per page one on Appendix B. This procedure immediately followed the testing session and preceded the actual scoring of the responses so 43 that the judgment could not be influenced by the data that was subsequently scored. Subsequently, the subject's responses to each of the 14 stimulus cards were analyzed according to a list of the indicators of narcissism. Both presentation (expressive) indications and representation (symbolic or verbal) indications were recorded for purposes of assessment. A response was scored either "1" or "0" depending on whether evidence of narcissism was present in the subject's response to that card. A score of "1" indicated evidence of narcissism, and a score of "0" denoted a lack of evidence of narcissism. The score (nominal classification) for each item was entered on a scoring matrix with the 14 stimulus cards forming the columns and the indicators of narcissism comprising the rows. (i.e., for each card, a score of "1" was given for each indicator that appeared in the subject's response, and a score of "0" was given for each indicator that was not present in the response.) Sub-totals for each indicator across each row were entered at the right of the protocol, and sums were generated for a total score. A check procedure consisted of adding column sums, which were identical to the sum of row sums. The subject's total score (i.e., the total of both the representational and presentational indices) was entered at the bottom right corner of the protocol and was taken as a measure of his degree of narcissism. Hence, a projective measure of narcissism was obtained by measuring the frequency with 44 which a subject manifested evidence of narcissism (i.e., the frequency of "1" responses) across the series of stimulus cards. Indications of "idealizing" or "mirroring" transference were taken as evidence of narcissism. Observations of manifestations in the testing situation were scored or tallied by recording "1" in the appropriate classification table space. Concomitant to an idealizing transference was the revival of repressed infantile emotions or experiences of inadequacy, vulnerability, and dependence in relation to the ideal parent image. Acting-out behavior was expected to be erected as a primary defence against the possibility of dependence on the ideal parent figure (helper) and against narcissistic envy of the helper as a source of narcissistic supply. Since the subjects were expected to project acting-out behaviors as a result of the transference(s) , the presence of any of the following indicators in a subject's response was taken as evidence of narcissism. Two types of evidence were recorded: representational (verbal-symbolic) indicators and presentational (expressive) (e.g., tone, inflection, or other non-verbal indications of affective orientation or response) indicators of: narcissistic envy; a devaluation of help or support; depreciatory attacks on the helper; and aggrandizing of the self (mirror transference) (i.e., overestimation of one's abilities and achievements). 45 The denial or absence of any acknowledgement of help or support and the denial or absence of any sign of dependence were expected to be critical indicators of narcissism in the interpretation of the protocols. Other indicators (acting-out behaviors) included the following: subjects' expressed sense of omnipotence (i.e., preoccupation with fantasies of unlimited success, power, brilliance, beauty, or ideal love); exhibitionistic need for intense or concerted attention and admiration; fragility of self-esteem (i.e., cool indifference or feelings or expressions of rage, revenge, hostility, inferiority, shame, humiliation, or emptiness in response to criticism, indifference of others, or defeat); sense of entitlement; tendencies to exploit others in interpersonal relationships; over idealization and devaluation of others in interpersonal relationships; and lack of capacity for empathy. Presentational (expressive) indications included dynamics of inflection; amplitude; pitch; cadence; lability of emotion; tone; sense. Illustrative indicators for each of the manifestations are provided in the scoring guide (Appendix B). Manifest evidence of narcissism should bear at least a "family resemblance" (Wittgenstein, 1953; Bruner, Olv er, Greenfield, et al., 1966, p.79) to the example provided for a particular indicator. While all of the above indicators are dependent upon the projection of qualities onto content (i.e., objects, persons) external to the subject, resistance to engage in 46 imaginative interpretation of the portrayed content can also be interpreted as indicative. For a subject's resistance to imagine or project is taken as a clinical indicator of narcissism (Brodey, 1965; Kernberg, 1975; Kohut, 1971; Lasch, 1978). The reluctance to engage in imaginative interpretation (as opposed to mere description of depicted detail or features) can be regarded as indicative of an inability to decathect from the immediate situation, from the objective detail of the surround. Lasch (1978) adds that some narcissistic persons manifest an "inability to tolerate the decathexis of current reality and to accept the ambiguity of the analytic situation" (p.177). Refusal to make imaginative interpretations of line drawings was taken as a test of this ability to "decathect" and accept ambiguity. Lasch maintains that these subjects' resistance to suspending the secondary process (i.e., to distance themselves from everyday life) is associated with the fear of fantasy. Accordingly, provision was also made for this indicator on the subjects' protocols. No projection (i.e., an objective account of the stimulus card) was recorded as "NP" on the scoring matrix. A frequency count of a subject's "NP" responses was recorded on his protocol and was included in the comparison of his performances on the NPI and projective measures of narcissism. (No "NP" responses were observed in the administration of the stimulus cards.) 47 Despite its potential clinical value, assessment of narcissism through projectives also has its limitations. Because the subjects were unable to interact with the helper/therapist (adult figure in each of the stimulus cards), some indicators of narcissism might not appear in the testing situation. For example, in the absence of criticism from the helper/therapist, the chances of seeing signs of narcissistic rage or signs of revenge in the subjects were greatly reduced. Similarly, it was not possible to see the helper/therapist's reactions to the subject's acting-out behavior, as indications of countertransference were precluded. This, in turn, reduced the probability that the subjects manifested narcissism indicators which are typically evoked by countertransference signs in the clinical situation. In addition, one other potential problem was anticipated: since the use of projectives requires that subjects- respond verbally (and/or gesturally) to the stimulus material, a wide range of expressive competence and confidence may be expected. This includes the possibility of non-response. At this time, it is not known whether no response to a stimulus card is due to a lack of ability to articulate, or indicative of the ultimate defensive posture. While this question can form the basis of a future study, in the present study, the tester attempted to elicit compliance to request for imaginative interpretation; and refusal to do so was taken as evidence of narcissism (for the reasons set forth above). 48 Accordingly, for present purposes, no response was recorded as "NR" on the scoring matrix. Failure to respond initially was followed by encouragement via a second attempt to solicit a response. The total scores from this projective measure were correlated with the total NPI scores from the questionnaire/inventory measure of narcissism. (Measure Five: Personality Inventory) A questionnaire/inventory measure of narcissism (see Appendix C) was obtained for each subject by administering the Narcissistic Personality Inventory (NPI) (Raskin & Hall, 1979). The total NPI scores were correlated with the total scores from the projective measure of narcissism (as well as with the other variables). Other exploratory analyses also entailed use of the scale scores (see Chapter Four). Persistence (Measure One - "Naturalistic": In Class Observation) A measure of behavioral persistence was obtained by observing the frequency of a subject's off-task behavior under naturalistic classroom conditions. The observation of each child's behavior in the regular classroom was conducted in two separate 10-minute sessions. Observations were recorded only after the students were assigned individual seatwork. Assignments consisting of regular classwork (arithmetic problems, language tasks, and reading) were attempted by each subject without assistance. Students were considered to be on-task if they appeared to be looking at the appropriate task materials. Pupils were assumed to be 49 off-task if they looked away from the relevant task materials for a minimum of three consecutive seconds. The 10-minute observation periods were divided into 10-second intervals, and each subject's behavior was recorded as either on-task or off-task for each 10-second segment. Off-task behavior was scored "1", whereas on-task behavior was not scored. Because two 10-minute recording sessions were used for each child, the frequency of off-task behavior in 20 minutes was tallied for each subject. This frequency tally was taken as an index of behavioral persistence. One must realize that this index of persistence is a negative index (i.e., a high score indicates low persistence, while a low score denotes high persistence.) Accordingly, for purposes of correlational analyses, these scores were transformed so that the index is positive. The transformation equation was Pt=100-X, where Pt is the transformed score for the in class measure of persistence, and X is the frequency tally of off-task behavior in 20 minutes. (Measure Two - "Contrived": Picture Reproduction) A second measure of behavioral persistence was acquired by giving each subject paper and crayons and asking the child to reproduce a printed color picture to the best of his abilities (Violato & Travis, 1985). Time on task was used as the index of behavioral persistence. Time on task was operationally defined as beginning when the subject put the crayon on the paper and terminating when he declared 50 completion. (The maximum time allowed for this activity was 65 minutes.) Previous research which incorporated this procedure provides evidence that such performances are correlated with standardized achievement measures at a respectable level (r=.60) (Violato & Travis, 1985). The picture used was published in The Sciences (January/February 1987, volume 27, number 1, page 24). (Measure Three - "Contrived": Letter Deletion) A third measure of behavioral persistence was secured by giving each subject a six-page history paper in small print and a pencil. Each child was asked to read through the paper and cross out every letter "e". The subject was also informed that a pencil maze was printed on the back of the history paper and that he may attempt to solve the maze only after he deleted as many "e's" as he could. Time on task was used as the index of behavioral persistence. Time on task was operationally defined as beginning when the subject put the pencil on the paper and terminating when he turned the paper over to do the maze. (The maximum time allowed for this activity was 35 minutes.) Ability Percentile ranks were derived for each subject by referring to school records of standardized intelligence test scores (Otis-Lennon School Ability Test). Since this test provided an adequate measure of abilities for present purposes and was the common metric for both schools, the 51 percentile ranks from the Otis-Lennon was used as the measure of academic ability in this study. Achievement Percentile ranks for each subject were derived by referring to school records of standardized achievement test scores (the Sequential Tests of Educational Progress (STEP) (Educational Testing Service, 1971) from School 1 and the Canadian Tests of Basic Skills (CTBS) (King, 1971) from School 2). Since proficiency in reading is necessary for academic success, and because reading is generally regarded as a powerful indicator of educational achievement, the percentile ranks from the reading scales of the STEP and the CTBS were used as the measures of achievement in this study. Evaluation For each subject, comprehensive measures of narcissism (five), persistence (three), ability (one), and reading achievement (one) were obtained. Pearson correlations were computed to examine the associations among the variables for the obsessional and non-obsessional subgroups and for the aggregate sample. In addition, oneway univariate analyses of variance (ANOVA) and Chi-squares were performed to compare obsessionals and non-obsessionals on the measures. Finally, multiple regression analyses were executed to estimate the proportions of variance in achievement which are associated with each of the predictor variables: ability, social status, persistence, and narcissism. 52 Since narcissism, persistence, and ability are likely to operate together to influence achievement, any interaction among the predictor variables was discussed in relation to the five hypotheses. 53 CHAPTER FOUR RESULTS The present study was designed to explore school achievement in a sample of private school children and the influence of some demographic and personal factors on it. Of particular interest was the impact of persistence and narcissism on achievement. Accordingly, along with indices of social class and mental abilities (well established influences on scholastic achievement) several measures of persistence and narcissism were secured for the sample. The latter was posited as a construct that represents a personality or dispositional factor that moderates striving or persistence. The rationale for positing narcissism as a moderator variable turns on the theoretical distinction between varieties of narcissism which are associated in Freud's work with differences in cultural absorption and attainment. This study incorporated an attempt to identify two variations of narcissistic personality - "narcissistic-obsessionals" and "non-obsessionals" - and factors which influence children's school performance. Hypotheses about the relationship of various indicators of narcissism to achievement and about the anticipated differences between the two narcissistic types (as stated in Chapter Three) were explored through study of 56 private school children. The present chapter is devoted to a description of the results 5^ of data analyses which were undertaken to compare the children designated as obsessionals with the remainder of the sample (the non-obsessionals). The empirical results of this investigation are organized to inform the reader about (a) the nature of the sample, (b) the evidence gathered from the projective procedures, (c) the data from comparisons made between obsessionals and non-obsessionals, and (d) the data from the other measures of narcissism, persistence, social status, ability, and achievement with respect to the aggregate sample. Section A (description of the sample) highlights data on gender composition, social class composition, intellectual ability, achievement, and persistence. The description of the sample should be kept clearly in mind by the reader who wishes to understand the remainder of the data, since this sample is not (and was not intended to be) a representative sample. Indeed, the special character of the sample must be emphasized at the outset. Section B (measurement of narcissism with projectives) presents data on the projective instrument. These data include an examination of the efficacy of the 14 stimulus cards as measured by the representational and presentational indicators of narcissism, as well as a comparison of the projective instrument with the NPI. Section C (comparing the obsessional and non-obsessional types) highlights the data on obsessionals and 55 non-obsessionals defined by judgment and by achievement. These data include comparisons of the two variants of narcissistic personality on all the measures. Section D (examination of relationships between the measures) presents the data for the measures of narcissism, persistence, social status, ability, and achievement. These data include a description of the transsituational consistency for subjects high in narcissism versus subjects low in narcissism and for obsessionals versus non-obsessionals. The data also includes an examination of the relationships between the measures within the aggregate sample. Section A: Description of the Sample It is of primary importance to realize that a particular segment of the school population - private school children - was purposely selected in order that more could be learned about these children vis-a-vis some variables of educational and psychological significance. This, in effect, resulted in a sample that was drawn from a highly atypical (unrepresentative) population of children. (1) The Homogeneity of the Sample Evidence of the homogeneity of the sample is presented in Table I. Firstly, virtually all of the subjects in the sample were girls (54 of 56). Secondly, the majority of the children in the sample were from families where the head of the household was either an executive or a professional (43 of 56). These occupations are traditionally associated with 56 Table I Demographic data for the aggregate sample (n=56) School Grade Female/Male Ratio 1 n = 36 7 n = 36 36/0 2 n = 20 6 n = 20 18/2 Birthorder n Siblings n Social Status (Head) n Only Child 14 None 14 Entrepreneur 5 First Born 10 One 23 Executive/Professional 43 Middle 12 Two 13 Managerial/Small Business 7 Youngest 20 Three 5 Skilled Labor 1 Four 1 higher than average levels of cultural as well as economic capital. Further evidence of the homogeneity of the sample is presented in Table II. The summary statistics on the measures for the sample indicate that these children are clearly different from those one might expect to find in the public school population. The mean percentile ranks on ability (82.14) and reading achievement (STEP, 81.89 and CTBS, 81.95) are nearly one standard deviation above the mean for a normal population. Moreover, the mean scores on the In Class (96.46), Picture Reproduction (61.58), and Letter Deletion (33.23) measures of persistence suggest that these children are also remarkably persistent in tasks that require striving or willful effort with little return on investment. It was anticipated in advance that selecting a restricted and homogeneous sample would likely result in low variance on some measures of the constructs and hence low correlations between measures when values from this sample are considered by themselves (i.e., when a highly homogeneous or unrepresentative sample is considered). (2) Comparison of the Two Subsets of the Sample The subjects for this study were drawn from two private elementary schools to ensure that the sample size was adequate. However, this expediency created a problem because two different measures of the criterion variable (achievement) were provided by the two schools. Certainly, 58 Table II Summary statistics on the measures of ability, reading achievement, persistence, and narcissism for the aggregate sample Measure n Mean+ S . D • Range Ability (Otis-Lennon) 50 82. 14 17 . 61 25- 99 Reading Achievement (STEP) 35 81. 89 15. 94 47- 99 Reading Achievement (CTBS) 19 81. 95 17 . 83 46- 99 Persistence-In Class 56 96. 46 5 . 79 71-100 Persistence-Picture Reproduction 36 61 . 58 9. 22 26- 65 Persistence-Letter Deletion 56 33 . 23 5 . 21 12- 35 Narcissism-NPI 56 16. 07 8. 15 1- 39 Narcissism-Projective Total 56 9. 59 5 . 94 1- 32 + The units of measure for the means are as foilows: Otis-Lennon - percentile ranks STEP - percentile ranks CTBS - percentile ranks In Class - transformed score (maximum score: 100) Picture Reproduction - minutes (maximum time: 65 minutes) Letter Deletion - minutes (maximum time: 35 minutes) NPI - raw score (maximum score: 54) Projective Total - raw score (maximum score: 308) 59 obtaining a common metric for achievement would have been more desirable, but doing so was not feasible. However, since standardized reading achievement scores were available from both schools, reading achievement was used as the dependent variable in this study. The sample size enhancement was necessary even at the cost of having two instead of one common measure of achievement. As a consequence of the fact that the two schools used different standardized achievement tests, it was necessary to determine whether the two subsets of the sample differed. Two operational definitions of achievement required that recognition be given to two subsets of the sample. Analyses were necessarily more elaborate than was planned because of this complication. Oneway univariate analysis of variance (ANOVA) and Chi-square were used to test the two subsets (as comparison groups) for significant differences, and the criterion for judgment of statistical significance was p<.05. The means and standard deviations of the measured performances for the two schools are presented in Table IIIA. The results of the ANOVA did not indicate that there were any significant differences between the two groups on intellectual ability, narcissism (NPI and projective total), and persistence (In Class and Letter Deletion). (The Picture Reproduction measure of persistence was not administered in School 2. Hence, no comparison was made on this measure.) 60 Table IIIA Summary of oneway analysis of variance on each independent measure for School 1 versus School 2 -Measure School n Mean + S . D • Ability (Otis-Lennon) 1 31 85. 77 13. 03 (F(l,48)=3.67, p=.062) 2 19 76. 21 22. 38 Persistence-In Class 1 36 95. 56 6. 47 (F(l,54)=2.56, p=.116) 2 20 98. 10 3. 95 Persistence-Letter Deletion 1 36 34. 08 2. 74 (F(l,54)=2.78, p=.102) 2 20 31. 70 7. 81 Narciss ism-NPI 1 36 17. 00 8. 86 (F(l,54)=1.32, p=.256) n 20 14. 40 6. 56 Narcissism-Projective Total 1 36 9. 75 6. 12 (F(l,54)=0.07, p=.789) 2 20 9. 30 5. 75 + The units of measure for the means are as follows: Otis-Lennon - percentile ranks In Class - transformed score (maximum score: 100) Letter Deletion - minutes (maximum time: 35 minutes) NPI - raw score (maximum score: 54) Projective Total - raw score (maximum score: 308) 61 A summary of the Chi-square analysis on the two schools is presented in Tables IIIB and IIIC. The results of the Chi-square revealed that the only measures where the groups differed significantly were social status, wherein consideration was given to the top occupation in each family (p=.010); and judged degree of narcissism (p=.045). There were no significant differences between the two groups on the other two judgment categories of narcissism: overall and type. Moreover, scatterplots of reading achievement with ability are illustrated in Figures 1-3. Upon examination of the three plots, the two subsets do not appear to differ significantly from each other, and neither of the two subsets appear to differ significantly from the aggregate sample. However, one should still have reservations about collapsing the two separate sets of scores for reading achievement into one common scale. The Sequential Tests of Educational Progress (STEP) and the Canadian Tests of Basic Skills (CTBS) are different tests, and the norms were derived from performances from different populations. Accordingly, the two subsets will be considered separately when determining the relationship of reading achievement to the independent measures (as stated in Chapter Three), and considered together for the purpose of identifying the obsessional and non-obsessional variants within the aggregate sample. 62 Table IIIB Summary of Chi-square analysis of the distributions of subjects on the social status measure for School 1 versus School 2 occupation of head of household COUNT EXP VAL ROW PCT COL PCT TOT PCT RESIDUAL STD RES ADJ RES school 1| 1 entrepreneur execut i ve/profes manager i a 1/sma11 sk i11ed 1abor COLUMN TOTAL 2| 3 2 3.2 1 .8 60.0% 40.0% 8.3% 10.0% 5.4% 3.6% -.2 . 2 - . 1 .2 -.2 . 2 32 1 1 27.6 15.4 74 . 4% 25 .6% 88.9% 55.0% 57 . 1% 19 .6% 4 . 4 -4.4 .8 -1.1 2.9 -2.9 1 6 4 . 5 2.5 14 . 3% 85. 7% 2.8% 30.0% 1 .8% 10.7% -3.5 3.5 -1.6 2.2 -3.0 3.0 0 1 .6 . 4 .0% 100.0% .0% 5 .0% .0% 1 .8% - .6 .6 - .8 1 . 1 -1.4 1 .4 36 20 64 . 3% 35 . 7% ROW TOTAL 5 8.9% 43 76.8% 7 1 2 .5% 1 .8% 56 100.0% CHI - SQUARE 0 . F . SIGNIFICANCE 1 1 .38522 O.0098 63. Table IIIC Summary of Chi-square analysis of the distributions of subjects on judged degree of narcissism for School 1 versus School 2 narcissism -judgment of degree high COUNT EXP VAL ROW PCT COL PCT TOT PCT RESIDUAL STD RES ADJ RES school 1| 1 medium 1 ow COLUMN TOTAL 2| 1 - 1 1 .3 .7 50.0% 50.0% 2. ay. 5.0% 1 .8% 1 .8% - .3 . 3 - .3 .3 • -.4 . 4 14 14 18.0 10.0 50.0% 50.0% 38 .9% 70.0% 25 .0% 25.0% -4.0 4.0 - .9 1 .3 -2.2 2.2 17 2 12.2 6.8 89.5% 10.5% 47 . 2% 10.0% 30. 4% 3.6% 4.8 -4.8 1 . 4 -1.8 2.8 -2.8 4 3 4 . 5 2 . 5 57 . 1% 42 .9% 1 1 .1% 15 .0% 7 . 1% 5 . 4% - . 5 . 5 - .2 . 3 - . 4 . 4 36 20 64 . 3% 35.7% ROW TOTAL 2 3.6% 28 50.0% 19 33 .9% 7 12 . 5% 56 100.0% CHI-SQUARE D.F. SIGNIFICANCE 8.07251 3 0.0445 64 Figure 1 Plot of the ability measure (Otis-Lennon) and the reading achievement measure (STEP) from School 1 CN *— -~ 0) 05 in 0) + in O 0) m co in 6 CO HI 10 01 15 >. a in CN a -— in £ 10 c ip c (1) in — r- ! if) *~ — 10 r~ 0 in , ' if) in Qi CN CN LI •— LO -— r-T TJ 0) 0 Q. in C i/i (3 O ID C7J o CO + ID in e u ; - D > a; = i> c ; in a o. CN -n S~ 65 Figure 2 Plot of the ability measure (Otis-Lennon) and the reading achievement measure (CTBS) from School 2 o a. — i- m <c T5 — C DI o c s a • — 66 Figure 3 Plot of the ability measure (Otis-Lennon) and the reading achievement measures from both schools _UI<QHZ(J 67 Section B: Measurement of Narcissism With Projectives As was stated in Chapters Two and Three, the current project incorporated an attempt to assess the construct of narcissism through projectives. The projective instrument and the data generated from its use are central features of this study, and warrant emphasis. (1) Evaluation of the Projective Instrument In view of the known difficulty in assessing narcissism with questionnaires (because they typically invite defensive distortion and impression management), the said projective instrument was devised. It was designed to allow for an inspection of the psychodynamics of narcissistic subjects in a non-obtrusive manner which minimizes defenses against direct examination of sensitive content. Because subjects were asked to respond to generally vague pictures with their own constructions, there was no apparent reason or signal for defensiveness on these matters. Consequently, the presentation of the series of 14 stimulus cards allowed for an analysis of such projections for indications of narcissism. A summary of the aggregate totals for the 14 stimulus cards and for the distributions of these totals across the representational and presentational indicators is presented in Table IV. The relative yield of each of the 14 stimulus cards was determined by adding the frequency of indications for the cards and the indicators. It is noteworthy that the cards do not appear to be equally effective in eliciting 68 Table IV Summary of totals for the 14 stimulus cards and for the representational (R) and presentational (P) indicators on the projective measure of narcissism for the aggregate sample Stimulus Cards** T Indicators 0 of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 T Narciss ism A L envy R 0 0 0 0 1 0 3 0 0 2 0 1 0 2 9 P 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 devaluate R 3 9 3 0 4 10 5 7 5 2 0 2 6 4 60 P 0 0 0 0 0 2 0 0 1 0 0 0 0 0 3 depreciate R 2 1 0 0 0 4 2 1 0 0 0 1 3 1 15 P 0 0 0 0 1 2 1 0 0 0 0 0 0 1 5 aggrandi ze R 0 0 6 34 0 5 2 1 0 1 39 1 3 1 93 P 0 0 2 0 0 1 0 0 0 0 0 0 0 0 3 omnipotence R 0 0 1 0 0 3 0 1 0 0 3 0 2 1 11 P 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 attent ion R 0 0 6 1 1 1 1 3 1 3 1 0 2 4 24 P 0 0 1 0 0 0 0 1 0 0 0 0 0 1 3 self-esteem R 4 25 7 3 6 20 16 10 18 4 2 2 21 7 145 P 1 3 1 0 0 3 0 2 0 2 0 1 3 1 17 entitlement R 0 4 0 0 0 0 3 0 3 0 0 0 1 0 11 P 0 1 0 0 0 0 0 0 2 0 0 0 0 1 4 exploit R 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 P 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 idealize R 5 12 4 2 6 17 10 7 9 5 0 2 20 12 111 P 2 1 2 0 1 2 1 2 1 1 0 0 0 2 15 empathy R 0 1 0 2 0 0 0 0 0 0 2 0 0 0 5 P 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 TOTAL 17 57 33 42 20 70 44 35 40 20 48 io 62 39 537 ** The 14 cards depicted the following subject matter: (1) male dentist with boy (2) mother with son (3) female coach with boy (4) adult with child (5) female nurse with boy (6) male teacher with boy (7) male leader with boy , (8) male dentist with girl (9) mother with daughter (10) male coach with girl (11) man and woman with child (12) female nurse with girl (13) male teacher with girl (14) female leader with girl (See Appendix A for elaboration of the stimulus cards.) (See Appendix B for elaboration of the indicators.) 69 signs of narcissism, as evidenced by the totals for the cards. Interestingly, the two most productive cards were Cards 6 and 13 (both depict a classroom scene). The two indicators which appear most frequently throughout the cards are "fragility of self-esteem" and "idealization." Moreover, "aggrandization" appears most frequently in Cards 4 and 11 (the only two cards wherein the child is depicted as the person in a position of power and control). "Devaluation of help" (considered to be a crucial diagnostic indicator of narcissism) was observed regularly and frequently across the cards, while "envy" and "lack of empathy" were witnessed with lesser frequency. Furthermore, only one indicator ("exploitativeness") was not apparent in any of the cards. From examining the scores for each of the indicators, it is clear that the representational (verbal-symbolic) indicators were observed with far greater frequency than the-presentational (expressive) indicators. Accordingly, an inspection of how well the representational and presentational indicators correlate was necessary. To determine the direction and magnitude of the relationship between the representational and presentational indices of the projective instrument, a Pearson correlation coefficient and two rank order correlation coefficients (Kendall's Tau and Spearman's Rho) were computed. The Pearson correlation revealed that the representation subtotal showed a respectable positive 70 correlation with the presentation subtotal of the projective (r=.46). Similarly, the two rank order correlations between the two indices uncovered the same finding (Kendall's Tau, rs=.34; Spearman's Rho, rs=.41). Moreover, the reliability of the projective instrument was assessed using the aggregate sample. A split-half (internal consistency) test was performed using the Spearman-Brown prophecy formula, which corrects for attenuation. The results from the computation indicate that the reliability of the projective is at a respectable level (equal length Spearman-Brown=.74) (Cronbach's Alpha for first 7 cards=.44; Cronbach's Alpha for second 7 cards=.49). (2) Assessment of Narcissism With the Projective and NPI With narcissism defined alternatively by the total from the projective instrument and by the NPI, comparisons were made between subjects high in narcissism and those low in narcissism as measured by the two tests. The means and standard deviations of the measured performances for the top and bottom quartiles of the projective are presented in Table VA. When the projective instrument was used as the operational definition of narcissism, the results of oneway ANOVA applied to subjects in the top versus those in the bottom quartiles of the projective reveal that subjects in the top quartile have a significantly higher mean score on the NPI measure of narcissism than those falling in the bottom quartile (F(1,24)=5.12, p=.033). Moreover, subjects in the top 71 Table VA Summary of oneway analysis of variance on each measure for subjects in the top quartile (T) versus subjects in the bottom quartile (B) of the projective measure of narcissism Measure Group n Mean* S.D. Ability (Otis-Lennon) T 9 80 . 00 14. 24 (F(l,19)=0.02, p=.898) B 12 78. 92 21. 54 Reading Achievement (STEP) T 8 81. 38 17. 32 (F(i,14)=1.06, P-.320) B 8 73. 25 14. 02 Reading Achievement (CTBS) T 3 77. 33 20 . 21 (F(l,7)=0.10, p=.766) B 6 81. 83 20 . 70 Persistence-In Class* T 12 93. 33 8. 65 (F(l,24)=4.34, p=.048) B 14 98. 36 2. 47 Persistence-Picture Reproduction T 8 59. 13 11. 17 (F(l,14)=0.03, p=.876) B 8 60. 13 13. 79 Persistence-Letter Deletion T 12 31. 58 6. 99 (F(l.,24)=0.17, p=.688) B 14 32. 57 5 . 40 Narcissism-NPI* T 12 17. 67 6. 85 (F(l,24)=5.12, p=.033) B 14 12. 50 4. 74 * Significant differences at p< + The units of measure for the . 05 means are as foilows: Otis-Lennon - percentile ranks STEP - percentile ranks CTBS - percentile ranks In Class - transformed score (maximum score: 100) Picture Reproduction - minutes (maximum time: 65 minutes) Letter Deletion - minutes (maximum time: 35 minutes) NPI - raw score (maximum score: 54) Projective Total - raw score (maximum score: 308) 72 quartile have a significantly lower mean score on the In Class measure of persistence than those in the bottom quartile (F(1,24)=4.34, p=.048). No other significant differences were observed between the two quartiles on the measures of ability, reading achievement, and the other two measures of persistence. In addition, a summary of the Chi-square analysis on the top and bottom quartiles of the projective is presented in Tables VB, VC , and VD. The results from Chi-square applied to the two quartiles indicate that the subjects in the top quartile differ significantly from those in the bottom quartile of the projective on all three judgment categories of narcissism (overall judgment, p=.015; judged degree, p=.002; and judged type, p=.008). However, no significant difference was found between the two quartiles with repect to social status. The means and standard deviations of the measured performances for the top and bottom quartiles of the NPI are presented in Table VI. When the NPI was used as the operational definition of narcissism, the results from oneway ANOVA applied to subjects in the top versus those in the bottom quartiles of the NPI indicate that subjects in the top quartile of the NPI have a significantly higher mean score on the projective measure of narcissism than those falling in the bottom quartile (F(1,27)=4.75, p=.038). No other significant differences were found between the two 73 Table VB Summary of Chi-square analysis of the distributions of subjects on overall judgment of narcissism for the top and bottom quartiles of the projective narc1ss i sm COUNT EXP VAL ROW PCT COL PCT TOT PCT RESIDUAL STD RES AOd RES overall judgment notable signs top 12s bottom 1 cores on 4 scores 1.00| 2.00| 1 not apparent COLUMN TOTAL 12 7 8.8 10 .2 63 . 2% 36 .8% 100.0% 50 .0% 46.2% 26 .9% 3 . 2 -3 .2 1 . 1 - 1 .0 2.9 -2 .9 0 7 3.2 3 .8 . 0% 100 .0% .0% 50 0% .0% 26 9% -3.2 3 2 -1.8 1 7 -2.9 2 9 12 14 46 . 2% 53 8% ROW TOTAL 19 73 . 1% 7 26 . 9% 26 100.0% CMI-SQUARE D . F . SIGNIFICANCE 5.86583 0.0154 74 Table VC Summary of Chi-square analysis of the distributions of subjects on judged degree of narcissism for the top and bottom quartiles of the projective narc i ss i sm judgment of degree high COUNT EXP VAL ROW PCT COL PCT TOT PCT RESIDUAL STD RES ADO RES top 12 s bottom 1 cores on 4 scores 1.00| 2.00| + + 1 med i um 1 ow COLUMN TOTAL 2 0 .9 1 . 1 100 .0% .0% 16 .7% .0% 7 .7% .0% 1 . 1 - 1 . 1 1 . 1 -1 .0 1 .6 - 1 .6 10 4 6 .5 7 . 5 71 .4% 28 .6% 83 .3% 28 .6% 38 .5% 15 .4% 3 .5 -3 .5 1 .4 - 1 . 3 2 . 8 -2 .8 0 3 1 . 4 1 .6 .0% 100 .0% .0% 2 1 .4% .0% 1 1 .5% -1 .4 1 .4 - 1 . 2 1 . 1 - 1 . 7 1 . 7 0 7 3 . 2 3 8 .0% 100 0% .0% 50 0% .0% 26 9% -3 . 2 3 2 - 1 .8 1 7 -2 .9 2 9 12 14 46 .2% 53 8% ROW TOTAL 2 .7% 14 53 . 8% 1 1 3 .5% 7 26 . 9% 26 100.0% CHI-SOUARE D.F. SIGNIFICANCE 14.50340 3 0.0023 Table VD 75 Summary of Chi-square analysis of the distributions of subjects on judged type of narcissism for the top and bottom quartiles of the projective narcissism -judgment of type COUNT EXP VAL ROW PCT COL PCT TOT PCT RESIDUAL STD RES ADJ RES top 12 s bottom 1 cores on 4 scores 1 .001 2 .001 1 obsess i ona1 non-obsess1ona1 non-obsess ionals (judged as showing no apparent signs of narcissism) COLUMN TOTAL 5 1 2 8 3 2 83 3% 16 7% 41 7% 7 1% 19 2% 3 8% 2 2 -2 2 1 3 -1 2 2 1 -2 1 7 6 6 .0 7 0 53 .8% 46 2% 58 .3% 42 9% 26 .9% 23 . 1% 1 .0 - 1 0 .4 - .4 . 8 - .8 0 7 3 .2 3 .8 .0% 100 .0% .0% 50 .0% .0% 26 .9% -3 . 2 3 . 2 - 1 .8 1 . 7 -2 .9 2 .9 12 14 46 .2% 53 .8% ROW TOTAL 6 23. 1% 13 50.0% 7 26.9% 26 100.0% CHI-SQUARE 0 . F . SIGNIFICANCE 9 .64683 0.0080 76 Table VI Summary of oneway analysis of variance on each measure for subjects in the top quartile (T) versus subjects in the bottom quartile (B) of the NPI Measure Group n Mean + S . D 1 Ability (Otis-Lennon) T 12 81. 83 19. 62 (F(l,23)=0.17, p=.687) B 13 78. 46 21. 53 Reading Achievement (STEP) T 11 92 . 36 7. 17 (F(l,18)=4.12, p=.058) B 9 80 . 33 18. 09 Reading Achievement (CTBS) T 3 63. 67 17. 04 (F(l,6)=0.36, p=.573) B 5 72. 60 22. 03 Persistence-In Class T 14 97. 50 3. 11 (F(l,27)=0.00,. p=.946) B 15 97. 40 4. 61 Persistence-Picture Reproduction T 11 64. 09 3. 02 (F(1,I9)=0..63, p=.438) B 10 62. 20 7. 27 Persistence-Letter Deletion T 14 33. 57 •3 . 57 (F(l,27)=0.00, p=.953) B 15 33. 47 5 . 67 Narcissism-Projective Total* T 14 12. 57 7 . 27 (F(l,27)=4.75, p=.038) B 15 7. 80 4 . 23 * Significant differences at p<.05 + The units of measure for the means are as follows: Otis-Lennon - percentile ranks STEP - percentile ranks CTBS - percentile ranks In Class - transformed score (maximum score: 100) Picture Reproduction - minutes (maximum time: 65 minutes) Letter Deletion - minutes (maximum time: 35 minutes) NPI - raw score (maximum score: 54) Projective Total - raw score (maximum score: 308) 77 quartiles on the measures of ability, reading achievement, and persistence. In addition, the results from Chi-square applied to the two quartiles of the NPI reveal that there are no significant differences between the two quartiles with respect to the judgment categories of narcissism and social s t atus. The relationship between the NPI and the projective total is yet another topic of interest in this study. As stated in Chapters Two and Three, it was anticipated that the relationship between performances on the NPI (i.e., personality inventory) and performances on the projective device (i.e., transference indicators - total of the representational and presentational indices) would be negative due to defensive distortion associated with performances on the NPI (Hypothesis 1). To determine the direction and magnitude of the relationship between the NPI and the projective total, a Pearson correlation coefficient and two rank order correlation coefficients (Kendall's Tau and Spearman's Rho) were computed. The Pearson correlation between scores on the NPI and scores on the projective (for the aggregate sample) revealed a low positive relationship (r=.19). Similarly, the two rank order correlations between these two measures of narcissism (for the aggregate sample) uncovered the same result (Kendall's Tau, rs=.17; Spearman's Rho, rs = .23). 78 Although the correlations are not negative, they do indicate nonetheless that the two measures are not strongly related. The findings might also suggest that the two instruments measure different dimensions of narcissism. Section C: The Obsessional and Non-Obsessional Types Though examination of aggregates is necessary in order to acquire an understanding of the nature and test performances of the sample in general, aggregates tend to conceal the nature of their subgroups. An investigation of the subgroups within an aggregate sample usually reveals valuable information about such subgroups and their relation to the aggregate. In the present study, two subgroups are of particular interest: the narcissistic-obsessional and the non-obsessional variants of narcissistic personality. It was expected, as stated in Chapter Three, that for obsessionals, narcissism would be positively related to persistence (Hypothesis 2). It was also anticipated that for non-obsessionals, narcissism would be negatively related to persistence (Hypothesis 3). Because the theoretical distinction between the two variants of narcissistic personality was a primary interest of this study, differences between the obsessional and non-obsessional types in addition to those postulated in Hypotheses 2 and 3 were observed. These differences are also worthy of examination. 79 To determine the direction and magnitude of the relationship between narcissism and persistence for obsessionals and non-obsessionals, Pearson correlation coefficients were computed. In addition, oneway univariate analyses of variance (ANOVA) and Chi-squares were also computed to explore differences between the two subgroups. (1) Defining the Obsessional Type To facilitate comparisons, obsessionals and non-obsessionals were each defined two ways: by clinical judgment and by achievement. In the former procedure, the discrimination between an obsessional and a non-obsessional was made on the basis of clinical judgment prior to the actual scoring of the protocols for the projective (see Chapter Three). In the latter procedure, the marker for an obsessional was overachievement (defined as greater achievement than one would expect from given levels of ability and social status), and all others who were not overachieving were classified as non-obsessionals. A minimum of a seven percentile rank discrepancy between ability and achievement was used as the criterion for overachievement. (The seven percentile rank discrepancy is as arbitrary as any chosen demarcation. However, because it allowed for a minimum of 12 subjects to comprise the overachievement group, this criterion was adopted as adequate. While a higher discrepancy value (e.g., ten percentile ranks) might have been more desirable, the size of the overachievement group would have been insufficient 80 for purposes of comparing subgroups using oneway ANOVA. In fact, a ten percentile rank discrepancy would have reduced the number of subjects in the overachievement group to only 9. ) (2) Defining Obsessionals by Clinical Judgment The correlation coefficients showing the relationship between the measures of narcissism and persistence for obsessionals (defined by judgment) are presented in Table VIIA. When clinical judgment was used as the procedure for discriminating between obsessionals and non-obsessionals, the intercorrelations among the variables indicate that for obsessionals, the NPI measure of narcissism shows low and modest positive correlations with the Picture Reproduction and Letter Deletion measures of persistence (r=.21 and r=.42 respectively) (n=5 for Picture Reproduction) but does not appear to be related to the In Class measure of persistence (r=.00). In addition, the NPI displays moderate positive and moderate negative correlations with the STEP and CTBS measures of reading achievement (r=.69 and r=-.73 respectively) (n=4 and n=6 respectively). The projective measure of narcissism manifests a modest positive relationship with the Letter Deletion measure of persistence (r=.42) but shows moderate and low negative correlations with the In Class and Picture Reproduction measures of persistence (r=-.50 and r=-.22 respectively) (n=5 for Picture Reproduction). Furthermore, the projective total displays modest and low negative correlations with the STEP 81 Table VIIA Correlation coefficients (Pearson r) of the narcissism measures with the reading achievement and persistence measures for obsessionals (defined by clinical judgment) Measure n NPI Projective Total Reading Achievement (STEP) 4 . 69 -.44 Reading Achievement (CTBS) 6 -.73 -.08 Persistence-In Class 12 . 00 - . 50 Persistence-Picture Reproduction 5 .21 -.22 Persistence-Letter Deletion 12 . 42 . 42 Table VIIB Correlation coefficients (Pearson r) of the narcissism measures with the reading achievement and persistence measures for non-obsessionals (defined as all those not judged as obsessionals) Measure n NPI Projective Total Reading Achievement (STEP) 31 . 19 . 20 Reading Achievement (CTBS) 13 -.22 -.01 Persistence-In Class 44 . 00 -.07 Persistence-Picture Reproduction 31 . 14 . 07 Persistence-Letter Deletion 44 . 07 -. 15 82 and CTBS measures of reading achievement (r=-.44 and r=-.08 respectively) (n=4 and n=6 respectively). It is noteworthy that with the definition of narcissism operationalized alternatively by the NPI and by the total of the representational and presentational indicators from the projective instrument, the resulting correlations with the In Class and Picture Reproduction measures of persistence differ (r=.00 and r=-.50 for the former; r=.21 and r=-.22 for the latter) (n=5 for the latter). The same is true for the STEP and CTBS measures of reading achievement (r=.69 and r=-.44 for the former; r=-.73 and r=-.08 for the latter) (n=4 for the former; n=6 for the latter). The correlation coefficients showing the relationship between the measures of narcissism and persistence for non-obsessionals (defined as all those not judged as obsessionals) are presented in Table VIIB. For non-obsessionals, the NPI measure of narcissism shows low positive correlations with the Picture Reproduction and Letter Deletion measures of persistence (r=.14 and r=.07 respectively) and does not appear to be related to the In Class measure of persistence (r=.00). In addition, the NPI displays low positive and low negative correlations with the STEP and CTBS measures of reading achievement (r=.19 and r=-.22 respectively). The projective measure of narcissism shows low negative correlations with the In Class and Letter Deletion measures of persistence (r=-.07 and r=-.15 respectively) but manifests a low positive relationship with 83 the Picture Reproduction measure of persistence (r=.07). Moreover, the projective total displays low positive and low negative correlations with the STEP and CTBS measures of reading achievement (r=.20 and r=-.01 respectively). One should notice that when judgment was used to differentiate obsessionals from non-obsessionals (see Tables VIIA and VIIB), the two sets of correlations of the measures of persistence and reading achievement with the projective measure of narcissism appear to differ. This is especially noteworthy given the0 homogeneity of the sample and the corresponding small variance on the measures. Though the size of the two subgroups differ, the intercorrelations nonetheless suggest that scores on the projective can be used to discriminate obsessionals from non-obsessionals. The means and standard deviations of the measured performances for obsessionals and non-obsessionals (defined by judgment) are presented in Table VIIIA. When clinical judgment was used as the procedure for discriminating between obsessionals and non-obsessionals, the results from oneway ANOVA applied to obsessionals versus non-obsessionals indicate that obsessionals have a significantly higher mean score on narcissism than non-obsessionals as measured by the total of the representational and presentational indices of the projective instrument (F(1,54)=16.16, p=.000). (When measured by each of the two indices individually, the representational subtotal (F(1,54)=16.05, p=.000) and the presentational subtotal (F(1,54)=4.24 , p=.044) also reveal 84 Table VIIIA Summary of oneway analysis of variance on each measure for obsessionals (OB) versus non-obsessionals (N-OB) (defined by clinical judgment) Measure Type n Mean + S . D • Ability (Otis-Lennon) OB 11 79. 27 17. 52 (F(l,48)=0.37, p=.546) N--OB 39 82. 95 17 . 77 Reading Achievement (STEP) OB 4 87. 75 14. 41 (F(l,33)=0.60, p=.443) N--OB 31 81. 13 16. 19 Reading Achievement (CTBS) OB 6 85 . 17 14. 22 (F(l,17)=0.27, p=.607) N--OB 13 80. 4 6 19. 62 Persistence-In Class OB 12 94. 75 8. 45 (F(l,54)=1.35, p=.251) N--OB 44 96. 93 4. 85 Persistence-Picture Reproduction OB 5 58. 60 14. 31 (F(l,34)=0.60, p=.444) N--OB 31 62. 06 8. 37 Persistence-Letter Deletion OB 12 34. 92 0. 29 (F(l,54)=1.61, p=.210) N--OB 44 32. 77 5 . 81 Narciss ism—NPI OB 12 18. 00 9. 69 (F(l,54)=0.85, p=.360) N--OB 44 15. 55 7 . 72 Narcissism-Projective Total* OB 12 15. 00 7 . 58 (F(l,54)=16.16, p=.000) N--OB 44 8. 11 4. 48 * Significant differences at p<.05 + The units of measure for the means are as follows: Otis-Lennon - percentile ranks STEP - percentile ranks CTBS - percentile ranks In Class - transformed score (maximum score: 100) Picture Reproduction - minutes (maximum time: 65 minutes) Letter Deletion - minutes (maximum time: 35 minutes) NPI - raw score (maximum score: 54) Projective Total - raw score (maximum score: 308) 35 significant differences between the two subgroups.) No other significant differences were found between obsessionals and non-obsessionals on the measures of ability (Otis-Lennon), reading achievement (STEP and CTBS), persistence (In Class, Picture Reproduction, and Letter Deletion), and narcissism (NPI). In addition, a summary of the Chi-square analysis on obsessionals and non-obsessionals (defined by judgment) is presented in Table VIIIB. The results from Chi-square applied to obsessionals versus non-obsessionals reveal that obsessionals have a significantly higher index of narcissism than non-obsessionals as measured by judged degree of narcissism (p=.003). No other significant differences were found between the two subgroups on overall judgment (as to whether signs of narcissism are notable or not apparent) and social status (top occupation in the family). (3)~ Defining Obsessionals as Overachievers The correlation coefficients showing the relationship between the measures of narcissism and persistence for obsessionals (defined as overachievers) are presented in Table IXA. When achievement was used as the procedure for distinguishing obsessionals from non-obsessionals, the intercorrelations among the variables indicate that for obsessionals (overachievers), the NPI measure of narcissism shows a modest positive relationship with the Picture Reproduction measure of persistence (r=.31) (n=5) but manifests low negative correlations with the In Class and 36 Table VIIIB Summary of Chi-square analysis of the distributions of subjects on judged degree of narcissism for obsessionals and non-obsessionals (defined by clinical judgment) narcissism -judgment of degree COUNT EXP VAL ROW PCT COL PCT TOT PCT RESIDUAL STD RES ADO RES obsessio the rest ROW nals TOTAL 1 .001 1 high medium 1 ow COLUMN TOTAL 21 2.001 + 2 0 4 1 .6 100 0% .0% 16 7% .0% 3 6% .0% 1 6 -1 .6 2 4 -1 .3 2 8 -2 .8 9 19 6 0 22 .0 32 1% 67 .9% 75 0% 43 .2% 16 1% 33 .9% 3 0 -3 .0 1 2 - .6 2 0 -2 .0 1 18 4 1 14 .9 5 3% 94 .7% 8 3% 40 .9% 1 8% 32 . 1% -3 - 1 1 5 3 . 1 .8 -2 1 2 . 1 2 3.6% 28 50.0% 19 33.9% 0 7 7 . 5 5.5 1 2 . 5% .o% 100.0% .0% 1 5 . 9% .0% 1 2 . 5% . 5 1 . 5 . 2 . 6 . 5 1 . 5 + + 12 44 56 .4% 78 .6% 100.0% CHI-SOUARE O.F. SIGNIFICANCE 14.10047 3 0.0028 87 Table IXA Correlation coefficients (Pearson r) of the narcissism measures with the reading achievement and persistence measures for obsessionals (defined as overachievers**) Measure n NPI Projective Total Reading Achievement (STEP) 5 . 80 .39 Reading Achievement (CTBS) 7 -.26 -. 12 Persistence-In Class 12 -. 17 -.05 Persistence-Picture Reproduction 5 .31 . 23 Persistence-Letter Deletion 12 -. 12 -.44 ** The criterion for overachievement was a minimum of a seven percentile rank discrepancy between ability and achievement. Table IXB Correlation coefficients (Pearson r) of the narcissism measures with the reading achievement and persistence measures for non-obsessionals (defined as other than overachievers) Measure n NPI Projective Total Reading Achievement (STEP) 30 . 19 . 11 Reading Achievement (CTBS) 12 -.37 . 10 Persistence-In Class 44 -.02 -.32 Persistence-Picture Reproduction 31 . 13 -. 10 Persistence-Letter Deletion 44 . 11 . 03 88 Letter Deletion measures of persistence (r=-.17 and r=-.12 respectively). In addition, the NPI displays high positive and low negative correlations with the STEP and CTBS measures of reading achievement (r=.80 and r=-.26 respectively) (n=5 and n=7 respectively). The projective measure of narcissism shows a low positive relationship with the Picture Reproduction measure of persistence (r=.23) (n=5) but manifests low and modest negative correlations with the In Class and Letter Deletion measures of persistence (r=-.05 and r=-.44 respectively). Furthermore, the projective total displays modest positive and low negative correlations with the STEP and CTBS measures of reading achievement (r=.39 and r=-.12 respectively) (n=5 and n=7 respectively). The correlation coefficients showing the relationship between the measures of narcissism and persistence for non-obsessionals (defined as other than overachievers) are presented in Table IXB. For non-obsessionals, the NPI measure of narcissism shows a low negative relationship with the In Class measure of persistence (r=-.02) but manifests low positive correlations with the Picture Reproduction and Letter Deletion measures of persistence (r=.13 and r=.ll respectively). In addition, the NPI displays low positive and modest negative correlations with the STEP and CTBS measures of reading achievement (r=.19 and r=-.37 respectively). The projective measure of narcissism shows modest and low negative correlations with the In Class and 89 Picture Reproduction measures of persistence (r=-.32 and r = -.10 respectively) but manifests a low positive relationship with the Letter Deletion measure of persistence (r=.03). Moreover, the projective total displays low positive correlations with the STEP and CTBS measures of reading achievement (r=.ll and r=.10 respectively). Here too, it is noteworthy that with the definition of narcissism oper at i onal i zed alternatively by the NPI and by the projective instrument, the resulting correlations with the In Class and Picture Reproduction measures of persistence differ (r=-.02 and r=-.32 for the former; r=.13 and r=-.10 for the latter). The same is true for the CTBS measure of reading achievement (r=-.37 and r=.10). As was the case with clinical judgment, one should notice here too that when achievement was used to discriminate between obsessionals and non-obsessionals .(see Tables IXA and IXB), the two sets of correlations of the measures of persistence and reading achievement with the projective measure of narcissism appear to differ. Again, this is especially noteworthy given the homogeneity of the sample and the concomitant low variance on the measures. And again, though the size of the two subgroups differ, the intercorrelations nonetheless suggest that scores on the projective can be used to differentiate obsessionals from non-obsessionals. The means and standard deviations of the measured performances for obsessionals and non-obsessionals (defined by achievement) are presented in Table X. No significant 90 Table X Summary of oneway analysis of variance on each measure for obsessionals (OB) versus non-obsessionals (N-OB) (defined by achievement) Measure Type n Mean + S . D . Ability (Otis-Lennon)* OB 12 64. 83 18. 93 (F(l,48)=21.71, p=.000) N-OB 38 87. 61 13 . 27 Reading Achievement (STEP) OB 5 84. 80 13. 01 (F(l,33)=0.19, p=.666) N-OB 30 81. 40 16. 52 Reading Achievement (CTBS) OB 7 80. 00 18. 66 (F(l,17)=0.13, p=.727) N-OB 12 83. 08 18. 07 Persistence-In Class OB 12 97. 75 3 . 19 (F(l,54)=0.75, p=.390) N-OB 44 96. 11 6. 30 Persistence-Picture Reproduction OB 5 60. 20 6. 72 (F(l,34)=0.13, p=.723) N-OB 31 61. 81 9. 63 Persistence-Letter Deletion OB 12 33. 92 3. 45 (F(l,54)=0.26, p=.612) N-OB 44 33 . 05 5 . 62 Narcissism-NPI OB 12 18. 42 8. 82 (F(l,54)=1.27, p=.265) N-OB 44 15. 43 7. 95 Narcissism-Projective Total OB 12 10. 08 4. 62 (F(l,54)=0.10, p=.749) N-OB .44 9. 45 6. 29 * Significant differences at p< + The units of measure for the . 05 means are as foilows: Otis-Lennon - percentile ranks STEP - percentile ranks CTBS - percentile ranks In Class - transformed score (maximum score: 100) Picture Reproduction - minutes (maximum time: 65 minutes) Letter Deletion - minutes (maximum time: 35 minutes) NPI - raw score (maximum score: 54) Projective Total - raw score (maximum score: 308) 91 differences were observed between the two subgroups on the measures of reading achievement, persistence, and narcissism. In addition, the results from Chi-square applied to obsessionals and non-obsessionals (defined by achievement) indicate that there are no significant differences between the two subgroups with respect to the judgment categories of narcissism and social status. Section D: Examining Relationships Between the Measures The transsituational consistency of subjects high in narcissism versus subjects low in narcissism and of obsessionals versus non-obsessionals was explored across the persistence measures. In addition, an examination of the relationships between the measures for the aggregate sample was undertaken. (1) Narcissism and Persistence: Rank Order Consistency The rank order consistency of subjects high in narcissism versus subjects low in narcissism and of obsessionals versus non-obsessionals was examined across the three measures of persistence (In Class, Picture Reproduction, and Letter Deletion). To compare the cross-situational stability of the subgroups, Cronbach's Alpha coefficients (CA) were computed. The alpha coefficients showing the rank order consistency for subjects high in narcissism versus subjects low in narcissism are presented in Table XIA. The alpha coefficients indicate that the subjects in the top quartile 92 Table XIA Cronbach's Alpha coefficients (CA) showing the rank order consistency across the three persistence measures for subjects high in narcissism versus subjects low in narcissism Subgroup CA n Top quartile (projective) .55 8 Bottom quartile (projective) .15Top quartile (NPI) .50 11 Bottom quartile (NPI) .07 0 Table XIB Cronbach's Alpha coefficients (CA) showing the rank order consistency across the three persistence measures for obsessionals versus non-obsessionals Subgroup CA n Obsessionals (judged) .70 5 Non-obsessionals (the rest) .13 31 Obsessionals (overachievers) .55 5 Non-obsessionals (the rest) .38 31 93 of the projective show far higher stability (CA=.55) than those in the bottom quartile ' (CA=. 15) . One should notice that because subjects from School 2 do not have the Picture Reproduction measure of persistence, the number of subjects in both cases is consequently smaller than one quarter of the 56 subjects in the total sample (n = 8 in both cases). This finding suggests that subjects high in narcissism (as defined by scores on the projective device) are more consistent in their persistence than subjects low in narcissism. (A comparison of the rank order consistency of subjects judged as high in narcissism and those judged as low in narcissism could not be made due to the small number of subjects in the former group. Since only 2 subjects were classified (judged) as high in narcissism, an alpha coefficient could not be computed for this group.) Similarly, the computations reveal that the subjects in the top quartile of the NPI manifest far greater response consistency (CA=.50) (n=ll) than those in the bottom quartile (CA=.07) (n=10). This finding suggests that subjects high in narcissism (as defined by performances on the NPI) are more stable or consistent in their persistence across situations than are the low narcissism cohort. The alpha coefficients showing the rank order consistency for obsessionals versus non-obsessionals are presented in Table XIB. The alpha coefficients reveal that obsessionals (defined by judgment) show far higher stability (CA=.70) (n=5) than non-obsessionals (CA=.13) (n=31). This 94 finding is provocative; but we must stress the necessity of caution given the small number of judged obsessionals involved. While the computations indicate that obsessionals (defined as overachievers) manifest greater response consistency (CA=.55) (n=5) than non-obsessionals (CA=.38) (n = 31),. the small number of overachievers signals need for more study with larger numbers. (2) The Relationship of the Measures to Achievement The two sets of reading achievement scores (STEP and CTBS) were examined separately to determine the relationship between the measures of reading achievement and the independent measures. It was expected, as stated in Chapter Three, that persistence would be positively related to achievement for all subjects (Hypothesis 4). It was further anticipated that ability would also correlate positively with achievement for all subjects (Hypothesis 5). To determine the direction and magnitude of the relationship between the reading achievement measures and the independent measures, Pearson correlation coefficients were computed. The correlation coefficients showing the relationship between the independent measures and the two reading achievement measures are presented in Table XII. The intercorrelations among the variables indicate that the reading achievement subtest of the STEP show's low positive correlations with the Picture Reproduction and Letter 95 Table XII Correlation coefficients (Pearson r) of the independent measures and the reading achievement measures (STEP from School 1 and CTBS from School 2) School 1 School 2 Measure STEP** CTBS*** Ability (Otis-Lennon) .63 .83 Persistence-In Class .30 -.22 Persistence-Picture Reproduction .22 Persistence-Letter Deletion .11 .00 Narcissism-NPI .26 -.33 Narcissism-Projective Total .14 .04 ** n=35 for all measures except ability (n=30) *** n=19 for all measures 96 Deletion measures of persistence (r=.22 and r=.ll respectively) and a modest positive relationship with the In Class measure of persistence (r=.30). The STEP reading scale also manifests a moderate positive correlation with ability (Otis-Lennon) (r=.63). The relationship between STEP reading achievement and ability is illustrated in Figure 1. In addition, the STEP reading scale shows low positive correlations with the NPI and projective total measures of narcissism (r=.26 and r=.14 respectively). Alternatively, the reading achievement subtest of the CTBS does not appear to be related to the Letter Deletion measure of persistence (r=.00) and shows a low negative correlation with the In Class measure of persistence (r=-.22). (Curiously, this result differs from that found earlier with the STEP.) The CTBS reading scale also shows a strong positive correlation with ability (Otis-Lennon) (r=.83).-The relationship between CTBS reading achievement and ability is illustrated in Figure 2. In addition, the CTBS reading scale manifests .modest negative and low positive correlations with the NPI and projective total measures of narcissism (r=-.33 and r=.04 respectively). (3) Explaining Variance in Achievement: Regression Analysis Finally, the present investigation also incorporated an attempt to determine how variance that is unexplained by references to abilities and social status is affected by narcissism and persistence. Multiple regression analyses were executed for each of the two samples to estimate the 97 proportions of variance in achievement which are associated with ability, social status, persistence, and narcissism. For the sample which used the reading subtest of the STEP as its measure of reading achievement, the regression analysis revealed that only the ability measure was significant in explaining variance in achievement (p=.003). Aside from ability, the measures of social status, persistence, and narcissism did not significantly reduce the unexplained variance in achievement. For the group which took the CTBS, the regression analysis revealed that three measures were significant in explaining variance in achievement: ability (p=.000); the Letter Deletion measure of persistence (p=.019); and the NPI measure of narcissism (p=.040). However, the measure of social status and the remaining measures of persistence and narcissism did not significantly reduce the unexplained variance in achievement. Although the regression analyses showed that the measures of narcissism and persistence generally did not significantly reduce the unexplained variance in achievement, one should not conclude that narcissism and persistence are not salient factors which affect achievement. Rather, these results were anticipated prior to the regression analyses in light of the homogeneity of the sample and the resulting small variance on the measures. 98 CHAPTER FIVE DISCUSSION Special Nature of the Study Because of the exploratory nature of the current investigation, a sample consisting solely of private school children was purposely chosen in order that more could be learned about these children with respect to some variables of educational and psychological importance. However, this meant that the sample was drawn from a highly atypical (unrepresentative) population of children. The sample is atypical for two reasons. Firstly, nearly all of the subjects are girls. Secondly, children who attend private schools must have parents who are sufficiently privileged so that they can afford to pay the fees and in other ways provide an unusually well buttressed support system for these children. Indeed, the majority of the children in the sample are from families where the head of the household is either an executive or a professional. In effect, these children are clearly different from those one would expect to find in public schools, as shown by not only the occupational data, but by the high mean scores on ability, reading achievement, and persistence for the sample as well. It follows, then, that because the subjects in this study were purposely selected from private elementary schools, it is highly probable that the homogeneity of the sample precluded the generation of much variance on the 99 measures of the constructs in this work. It was anticipated in advance of the data analyses that there would not be sufficient variability on the measures to facilitate the settling of hypotheses or the formulation of conclusions, which could be generalized to a more heterogeneous population. So while analyses of the sort one would undertake with a representative sample were carried out with this sample, they were executed in the spirit of exploration rather than with the intention of settling hypotheses by rigorous testing. The testing of hypotheses about relationships that are generally characteristic can only be done when a representative sample is studied. Accordingly, one must understand the reason for studying this sample. Children from advantaged backgrounds appear with low frequency in the general public school population. The raison d'etre for this project was to observe the factors of psychological and educational interest (narcissism and persistence) in children of the sort drawn from private schools. The interpretation of the data and the significance of the findings are discussed in this chapter. The chapter includes a discussion of: the narcissism construct and the adequacy of two contrasting measurement approaches (with respect to the data from the NPI and the projective instrument); the merits of the projective instrument and procedures; the issues that arose from use of the projective device; narcissism as a moderator of persistence (with 100 respect to subjects high in narcissism versus subjects low in narcissism and obsessionals versus non-obsessionals); the hypotheses set forth in Chapter Three; and the directions for future research. Narcissism: Construct and Assessment The measurement of narcissism with questionnaires was attempted by Raskin and Hall (1979) with the NPI. Although Raskin and Hall believe that the NPI possesses desirable psychometric properties that would facilitate the assessment of narcissism, it does not have the capability of registering the transference phenomena, which both Kohut and Kernberg consider to be critical diagnostic indicators of the construct in clinical practice. This is concordant with Kohut's criticism of questionnaire and inventory measures of narcissism, for they typically invite defensive distortion and impression management in narcissistic subjects who try to conceal rather than reveal their condition to avoid feeling shame. While there has been some work that has found some psychometric merit in the NPI (Prifitera & Ryan, 1984), a finding in the current literature provides further evidence that the NPI promotes defensiveness in respondents. Auerbach (1984) has recently shown that undergraduates who took the NPI were inclined to "manage" the impressions they convey so that others see them in a favorable light. In addition, Auerbach has found that the NPI correlates respectably with a recently developed Narcissistic Scale of the Millon Clinical Multiaxial Inventory (MCMI), and reports 101 that the MCMI also elicits defensive* camouflage, which material whether scaled or otherwise, raises doubts about content and construct validities. These findings underscore the contention (e.g., Travis & Mah, 1986) that inventory measures of narcissism like the NPI yield defensive output, which is worthy of study in its own right, but which misleads the unwary observer. Accordingly, the merits of procedures (such as those which entail projectives) that circumvent disguise, that probe the depths covered by surface detail, warrant serious attention. The data generated in the present study can be studied for clues to answer the question of whether the NPI and the projective instrument measure different dimensions of narcissism. One notices that the two measures typically produced different results and correlated poorly with one another. In addition, the total score from the projectives was found to discriminate between obsessionals and non-obsessionals when clinical judgment was used to identify the two variants. (The NPI did not discriminate between the two types.) Given the homogeneity of the sample, this finding is especially noteworthy. Hence, the construct validity of personality inventories like the NPI remains questionable if they fail to produce empirical evidence of Freud's distinction between the libidinal types and the multiple dimensions of narcissism upon which the distinctions depend (Travis & Mah, 1986). 102 The Merits of the Projective Instrument and Procedures Despite the attempts that have been made to develop appropriate instrumentation and assessment techniques to measure narcissism, assessment of this psychological construct has been problematical. Accordingly, the current investigation attempted to resolve some of the difficulties encountered by previous researchers by exploring the assessment of narcissism with projectives. The projective procedures were designed to make strong conceptual links between the construct of narcissism and its clinical manifestations on the one side and the measures on the other. Essentially, the development and use of the projective instrument in this study was inspired by Kohut's (1971) claim that narcissism is assessed necessarily by means which recognize the emergence of transference indicators. The use of projectives was also supported by Kernberg's (1975) assertion that narcissistic subjects tend to emit a constant projection of self and object images which are easily projected onto content external to the subjects. The projectives, through the presentation of a series of cards depicting social situations wherein the transference phenomena is usual, allowed for an analysis of the subjects' projections for signs of narcissism. More importantly, the development and use of the projective device and procedures represents what is perhaps the first attempt to operationalize the construct of 103 narcissism so that it reflects the multidimensional basis for empirical distinctions which correspond with the theoretical distinctions set out in theory (i.e., in Freud's libidinal types). Nowhere else in the literature is there evidence of such conceptual and clinical innovation. The complexity and multiplicity of this invention offers several salient features. The projective instrument and procedures are in principle capable of registering the transference phenomena which are essential for proper diagnosis of narcissistic personality. The projective procedures also offer provision for classification of narcissistic types by clinical judgment as well as by inventory scale scores. The projective instrument incorporates multiple indicators of narcissism, each of which consists of two further dimensions: the representational and presentational indices. (The former are verbal utterances which signal denotative meaning; and the latter are expressive nuances which signal affective states, commitments, and conation.) The invention of the scoring scheme and the accompanying scoring matrix' (see Appendix B) for studying narcissism was not only central to the present study, but it also has potential value in future research. The incorporation of the representational and expressive dimensions in addition to the analytic categories ~ for observing signs of narcissism is methodologically as well as conceptually innovative. As anticipated by Travis and Mah 104 (1986), this scoring scheme holds promise for an analytic framework - one which can be used to quantify qualitative material. Furthermore, the psychometric qualities of the projective device are very respectable. The split-half reliability (internal consistency) of the projective instrument is substantial (Spearman-Brown=.74). In addition, the correlations between the representational and presentational subtotals of the projective (Pearson r=.46; Kendall's Tau, rs=.34; Spearman's Rho, rs=.41) also provide evidence df the technical merits of the projective device. Although the representational and presentational indices correlate at a respectable level, an examination of the composite scoring matrix (as was shown in Chapter Four, Table IV) revealed that the representational (verbal-symbolic) indicators were observed with far greater frequency than the presentational (expressive) indicators. Though it was anticipated that the presentational indicators would yield the richest data, the outcome should not be surprising in view of the nature of the sample, the very limited range of discussion topics in and the brevity of the encounters with each of the subjects. (Diagnosis of narcissism through transference indicators in psychoanalysis frequently requires months and entails the generation of wide ranging talk and much affect.) A characteristic observed in most of the subjects examined was an underlying tone of politeness and social sensitivity. This would 105 suggest that the establishment of trust in the examiner— subject relationship takes more time than the brief periods afforded in the present study. In effect, what was noticed during the administration of the cards were signs of deference, cooperation, acceptable demeanor, and social grace. The subjects in the study were, after all, children from culturally and economically advantaged backgrounds. Issues Emerging From Use of the Projective Instrument Some measurement and assessment, psychological and educational, and conceptual issues which arose from use of the projective device and procedures are also worthy of discussion. (1) Measurement and Assessment Issues First, it is probable that the encounters during the sessions devoted to administration of the projective instrument and procedures were too brief to allow the transference phenomena to develop fully. This would account for the relatively low yields of the presentational indicators of narcissism (as compared to the representational indicators). Modification of the projective instrument and procedures to produce greater frequency of expressive responses should allow for fuller observation of the transference indicators. Second, the need to generate a higher degree and frequency of affective responses from the subjects raises a question of delicacy: How can sufficient affect be generated in order to observe the expressive indications of 106 narcissism without creating unnecessary (and certainly undesirable) friction between the subject and the examiner? This is a complex issue which entails social as well as ethical considerations. It is a serious issue which faces the clinician who wishes to assess narcissism through transference in a school setting. Third, two procedures were incorporated in the scoring of the responses for the projectives: classification by clinical judgment and classification by dimension scale scores. While the former approach was a more expedient means of classifying responses, it was less revealing than the latter, more systematic procedure. However, the use of both approaches to classify responses offered an indication of the accuracy of the clinician's judgment. Obviously, further exploration is warranted - especially with other population samples. Fourth, because the subjects in this study were generally articulate, there was little evidence of resistance to engage in imaginative interpretation of the portrayed content of the cards. Hence, this resistance to project, which Brodey (1965) and Lasch (1978) believed to be a clinical indicator of narcissism, was not observed. This state of affairs may one day be understood better than it is presently. Its meaning is unclear. (2) Psychological and Educational Issues Because it was not possible to observe the adult's (helper's) reactions to the subject's acting-out behavior 107 with respect to the stimulus cards, indications of countertransference were precluded. This, in turn, precluded the possibility of observing signs of rage, revenge, or hostility, which are typically evoked by signs of countertransference in the clinical situation. However, (as Freud, 1962 asserted) the countertransference phenomena are not restricted to the clinical situation, but are also found in social situations of the most common sort, including those of the school. Hence, one must be aware of such indications of narcissism in the educational context. Unfortunately, the restricted range of content and transitory nature of the encounters during the sessions devoted to use of the projective procedure did not allow much observation or insight into such matters. Modification of this procedure and instrument (to include more varied and evocative material) should improve such opportunities for generating more observations and insight. (3) Conceptual Issues In the current "study, an attempt was made to identify obsessionals and non-obsessionals two ways: by clinical judgment and by achievement. Although the use of clinical judgment was an expedient means of identifying the culturally inclined variant of narcissistic personality, this procedure yielded valuable insight into the identification and assessment of the types and multiple dimensions of narcissism with projectives. And although the second method of defining obsessionals as overachievers may 108 not have been a fine enough definition to allow for the identification of some overachievers in the present sample (i.e., it is difficult to identify overachievement in a homogeneous sample of children whose scores are mostly very high in both ability and achievement), the defintion is nonetheless conceptually correct and in agreement with Freud's definition of obsessionals as strivers. Since this study incorporated what may be the first attempt at identifying the two variants of narcissistic personality using projective procedures, it is not known at this time which is the more appropriate approach. Accordingly, it is sensible that both procedures be adopted in subsequent investigations. Narcissism, Persistence, and Achievement The current study also attempted to clarify the nature of the relationship between narcissism and underachievement in school. Drawing from the work of Baker (1979), who proposed that the development of narcissistic disturbances impedes scholastic performance, and that of Vigilante (1983), who found that self-preoccupation was associated with learning problems, it was supposed that narcissism may be a critical feature of some underachievement. According to Baker, narcissistic subjects are unable to tolerate events which do not confirm their sense of perfection and grandiosity. Specifically, Baker suggested that some narcissistic students lack the willingness to persist in the face of difficult academic material because they cannot 109 accept personal struggles that challenge the perfection of the grandiose self. Because academic studies require regular struggles with challenging intellectual material, one might expect students with high degrees of narcissism to have difficulty in school. However, Freud's (1931) contention that narcissism can also be a source of cultural accomplishment seems to conflict with Baker's (1983) reasoning. Freud essentially stated that a variant of narcissistic personality (the obsessional type) is able to persist through adversity to meet high standards for significant cultural attainment so that he may be highly regarded from a cultural perspective. Similarly, in the current psychiatric literature, there is recognition that many highly successful people meet the criteria for a narcissistic disorder (Grindlinger, 1986). The central issue of this thesis was to contrast obsessional and non-obsessional narcissism as dispositional factors which differentially moderate striving. The moderation was expected to occur in two ways: it was posited that (given rough equivalence in all else) obsessionals would show greater persistence and hence higher achievement than non-obsessionals, and that non-obsessionals would manifest a lower consistency in their persistence and thus also lower achievement than obsessionals. Accordingly, it was believed that overachievement would be an identifiable characteristic of obsessionals and that underachievement would be a salient feature of those non-110 obsessionals who manifested high degrees of narcissism and low degrees of striving. As such contrasts in larger and more heterogeneous samples reveal merit in this study strategy, the full significance of the present work will be fully evident. Narcissism and Cross-situational Persistence Because human beings are situation sensitive, trait evidence cannot easily be established. Any given instance of observed behavior may be situation-specific (i.e., be a state expression rather than a trait expression). Hence, it was necessary to study the patterns across different situations for evidence of commonality and consistency of response behavior, which can be regarded as a personal trait. Accordingly, the degree of rank order consistency across the three persistence situations for the obsessionals and the non-obsessionals was assessed to determine whether the striving patterns for these libidinal types vary significantly from one another. The persistence patterns suggested that obsessionals (defined either by judgment or as overachievers) were more consistent than non-obsessionals in their striving behavior. This finding supports the operational hypothesis of this thesis - that obsessional and non-obsessional narcissism differentially moderate striving. In addition, the striving patterns also suggested that subjects high in narcissism (on either the projective or the NPI) were more stable or consistent in their persistence than subjects low in Ill narcissism. This latter finding indicates that degree of narcissism also differentially moderates striving. One must remember that these differences between the subgroups were observed in a homogeneous sample with low variance on the persistence measures. It is probable that even greater differences would be observed in a larger and more representative sample of children. Conceivably, the present sample consisted primarily of the obsessional type. This possibility is suggested by the data on striving and achievement as well as demographic data. A comparison with public school samples would yield suitable and illuminating contrasts on these variables. The present research was consciously undertaken with an atypical sample to explore the character of narcissistic factors and behavioral persistence associated with the school performance of private school children. Earlier studies by Travis and his associates (Travis & Violato, 1982; Travis, Violato, & White, 1982; Violato & Travis, 1985) had documented the difficulty of finding privileged children in samples of public school children in numbers sufficiently large to permit adequate study of social class related dispositional differences such as those entailed herein. This present work therefore constituted an attempt to identify and study the culturally inclined, narcissistic-obsessional variant of narcissistic personality, which the theorizing of Travis and others suggest is most readily found in the institutions of the privileged (Travis, 1975; 112 Travis, Violato, & White, 1982; Coles, 1977; Feshbach, 1978). Further, predictions about the relationship of various indicators of narcissism to achievement, and conjectures about the anticipated differences between the two narcissistic types were incorporated in the following five hypotheses: (1) Performances on the NPI should be negatively related to scores from the projectives (transference indicators) and to the achievement and persistence phenomena. (2) For subjects identified as narcissistic-obsessional, narcissism should be positively related to persistence. (3) For all other subjects (non-obsessionals), narcissism should be negatively related to persistence. (4) For all subjects, persistence should be positively related to achievement. (5) For all subjects, ability should be positively related to achievement. While these hypotheses were formulated in accordance with the reasoning set forth in the early sections of this thesis, their primary function was that of defining and refining focal questions. As Cronbach (1982) and others (Cook & Shadish, 1986; McCutcheon, 1981; Smith, 1978) say, this is to be contrasted with the formal and definitive testing of posited relationships, since such formal tests are more suitable when nomothetic-theoretical objectives are pursued through investigations with representative samples. The present sample, which was drawn from the sixth and 113 seventh grades of two private schools, was very unrepresentative of all' children of this age. Even so, considerations of reliability and ecological validity (Borg & Gall, 1983; Bronfenbrenner, 1976) were incorporated in that multiple assessments were incorporated in this study. Accordingly, the study included five indices of narcissism (projective; overall judgment; judged degree; judged type; and NPI); three measures of persistence (In Class Observation; Picture Reproduction; letter Deletion); a classification for social status (in which subjects were assigned to one of five categories based on the economic involvement of the parents); percentile ranks of ability (Otis-Lennon School Ability Test); and percentile ranks of reading achievement - the criterion variable - (on STEP and CTBS reading subtests). Reliance on standardized measures (e.g., of aptitude and achievement) and other factors which are regularly studied in connection with achievement (e.g., persistence, social class, and gender) furthers exploration, extrapolation, and speculation - all of which can extend our understanding. More than this, and above all, the design allowed for consideration of all of the factors just mentioned in conjunction with an essay in clinical exploration. To determine the associations among the variables, Pearson correlations and multiple regression analyses were executed. The regression analyses were also performed to estimate the proportions of variance in achievement which 114 are associated with each of the predictor variables: ability (one), social status (one), persistence (three), and narcissism (five). In addition, oneway univariate analyses of variance (ANOVA) and Chi-squares were computed to compare the narcissistic-obsessional and the non-obsessional libidinal types. The two narcissistic types were classified on the basis of clinical judgment and achievement. Firstly, judgment of type was made via the clinician's formulation of an overall impression of the subject's responses to the stimulus cards ) (projective instrument) before the actual scoring of the protocol (so that the judgment is not influenced by the data that is scored). Secondly, subjects were classified as obsessional on the basis of overachievement (higher achievement than expected from knowledge of ability and social background). All other subjects who were not overachieving were classified as non-obsessional. (The criterion for overachievement was a minimum of a seven percentile rank discrepancy between ability and achievement. ) Evaluation of Hypotheses (1) It was hypothesized that performances on the NPI would be negatively related to scores on the projective due to defensive distortion associated with performances on the NPI. The NPI scores were also expected to be negatively related to the achievement and persistence phenomena. The correlations between the scores on the NPI and the 115 projective were not negative but insignificant nonetheless. Such signs suggest that the two instruments may measure different dimensions of the construct - or may not measure the same construct. In addition, the NPI also correlated either negatively or insignificantly with both the achievement and persistence measures again as was expected. These are reasons to suspect the construct validity of the NPI. (2) It was anticipated that narcissism is positively related to persistence for narcissistic-obsessionals (defined either by judgment or by achievement). However, given the nature of the present sample, this prediction could not be fully validated due to the extremely small variances on the persistence measures.. As might be expected, the correlations were low to modest because of the homogeneity of striving data in the sample: with respect to the variables of interest (achievement, ability, social status, persistence, and narcissism), all ranges are very much abbreviated. Moreover, the small subset size of both judged obsessionals and overachievers (n=12 for both) as well as the small variance on all three measures of persistence must also be considered. The low variability on the measures in particular would account for the low correlations between narcissism and persistence. These data (or data from an analogous sample) must be considered and analyzed together with data from subjects who would comprise a representative sample to test the conjectures properly. 116 (Though narcissism and persistence were found to correlate only modestly for obsessionals (defined either by judgment or by achievement), obsessionals were also found to be consistent in their persistence (although the number of obsessionals was small). This second finding may lend credence to the supposition that narcissism and persistence are positively related for obsessionals.) (3) Narcissism and persistence were expected to be negatively related for non-obsessionals. Whether defined by judgment or by achievement, the correlations for this subgroup were also low to modest. Similarly, the low variability on the measures of the constructs accounts for this observation. (Though only modest correlations were found between narcissism and persistence for non-obsessionals (defined either way), non-obsessionals were also found to be considerably less stable or consistent in their persistence than obsessionals. The supposition that narcissism and persistence is negatively related for non-obsessionals may be supported by the latter finding.) (4) For the aggregate, it was expected that persistence would be positively related to achievement. Correlations between each of the three measures of persistence with each of the two measures for reading achievement revealed a low positive relationship between the two constructs. Since previous researchers have found a strong positive relationship between persistence and achievement (Travis & Violato, 1982; Shrauger & Sorman, 1977), the low variance on 117 all three measures of persistence in the current investigation is again a likely explanation for the reported findings. (5) Ability and achievement were predicted to be positively related for the aggregate. The reading achievement subtest of the STEP correlated moderately with the Otis-Lennon School Ability Test (r=.63), while that of. the CTBS correlated strongly with the said measure of ability (r=.83). This is concordant with the usual findings such as those of Walberg (1984), who found that correlations between abilities and achievement is significant and substantial. However, the correlation between the STEP reading measure and the Otis-Lennon deserves some discussion. While it appears that this correlation is the more modest of the two said correlations, one must realize that first, the STEP and the CTBS are different tests and the norms were derived from performances from different populations (with the CTBS norms being the more suitable of the two), and second, the two correlations were based on different sample sizes (n=35 for STEP; n=19 for CTBS). Future Directions The current research provoked the formulation of several questions which should inform future work: (1) What are the characteristics of those subjects who are both judged obsessionals and overachievers? Although obsessionals were defined alternatively (by judgment and by achievement) in the present study, subjects who fall in both 118 categories need to be examined in connection with judged obsessionals and overachievers to explore whether any differences exist. There is also reason to believe that judged obsessionals are overrepresented in the overachievement subgroup. A close examination of the profiles for the high achievers in the present sample may provide clues which can direct future work. (2) What are the characteristics of those subjects who are both non-obsessionals and underachievers? Non-obsessionals need to be examined in relation to underachievers to determine how they may be identified vis-a-vis their performance on the indicators of the projective instrument. Unfortunately, the present sample included too few underachievers to allow for formal correlational analyses. However, those present have profiles in the projective data that may be fruitfully studied for clues which can guide future research. (3) Would the results obtained in the present study have been similar for a sample of males? A probe similar to the present study needs to be conducted using a sample with an equivalent number of males and females to estimate the distribution of males and females in obsessionals and non-obsessionals. Again, we see reason to extend this exploration with samples of other contrasting populations. (4) How can the series of stimulus cards be modified to generate greater affect in the subjects? The existing series of stimulus cards can be augmented to include more 119 cards which, depict the child in a 'position of power. Perhaps this modification to the projective instrument will produce higher yields in the presentational index of narcissism. (5) What clusters may be formed with the indicators on the projective measure of narcissism? A factor analysis needs to be computed to determine the possible groupings among the indicators on the projective device. Subsequently, particular factors may be identified as a basis for differentiating between obsessionals and non-obsessionals. (6) How valid is the projective device as a measure of narcissism? Further assessment of the technical qualities of the projective instrument is necessary to determine its validity as a measure of narcissism. The main difficulty with this project is shared with the present study: in the absence of a solid bench mark, concurrent validity is elusive. Clearly, the advantage of behavioral correlates like persistence, which can be linked theoretically as well as empirically lies in their promise for providing an independent and stable basis for establishing better understanding of personality or dispositional factors that are implicated in achievement. 120 GLOSSARY Acting out. Direct expressions of conflictual tensions in annoying or anti-social behavior or in fantasies. Countertransf erence. The arousal of the analyst's repressed feelings by the analytic situation, especially the transference by the analyst of his repressed feelings upon the patient. Cultural capital. The cultural resources (intellectual possessions such as knowledge, social savvy, tact, linguistic refinement, and aesthetic and moral sensibilities) which are differentially distributed among individuals and groups according to their places in the political economy, cultural ecology, and social structure. Since these are cultural resources, they are acquired naturally . from living in privileged circumstances. Extension of the self. The investment of new objects or individuals with the same meaning and values as were engendered by experience with objects or individuals in the past, to which the new ones bear no actual resemblance. In psychoanalysis, this extension is used as a defence against the patient's feelings of dependence, vulnerability, or inadequacy in relation to the analyst. The property of a body of occupying space. Grandiose self. A stable and cohesive narcissistic structure, which is mobilized within the patient during a mirror transference. This configuration serves as a defense against the patient's traumatic childhood experiences of a chronically cold parent figure. It is characterized by an exaggerated belief in or feelings of one's importance or identity, often manifested by delusions of great wealth, power, beauty, fame, or ability. Ideal parent image. A stable and cohesive narcissistic structure, which is mobilized within the patient during an idealizing transference. This structure represents the patient's fantasy of an ever-giving, ever-loving, and accepting caretaker, in contrast to the experience in reality. The structure is conceived as being perfect, or better than it is with an exaggeration of its virtues and a minimizing of its faults. Idealizing transference. The mobilization of the ideal parent image during analysis, wherein the patient perceives the analyst as the idealized parent figure. 121 In effect, the patient's early need for merging with an omnipotent object is revived. Individuated person. One who has achieved the most differentiated and integrated state of personality development. Under positive circumstances, the ultimate outcome of the path of psychological progression is a well-articulated, smooth-functioning physical system within the self. Libidinal type. The person, or thing with which libido is concerned, which provokes or excites instinctual activity related to it. Mirror transference. The mobilization of the grandiose self during analysis, wherein the patient sees the analyst as merely an extension of the patient's personality. In effect, the patient's early need for parental acceptance is revitalized. Narcissism. The condition of being in love with one's own image. Narcissism results from the withdrawl of libidinal interests from external objects and individuals, and the subsequent reinvestment of libido back into the ego. Pathological narcissism results from the inability to redirect libidinal interests into external objects and the undue concentration of libido in the ego. The term is thus applied to high valuation of one's own bodily qualities and, by extension, of one's deeds and personal qualities. Narcissistic entitlement. The expectation of special favors without assuming reciprocal responsibilities. Narcissistic-obsessionaJ type. The culturally inclined variant of the narcissistic personality. This particular libidinal type is able to persist through adversity to meet high standards for achievement so that he may be highly regarded from a cultural standpoint. Narcissistic personali ty. A personality disorder in which there are a grandiose sense of self-importance or uniqueness; preoccupation with fantasies of unlimited success, ability, power, wealth, beauty, ideal love, brilliance, and omniscience; exhibitionistic need for constant attention and admiration; characteristic responses to threats to self-esteem, such as cool indifference or marked feelings of rage, inferiority, shame, humiliation, or emptiness; and characteristic disturbances in interpersonal relationships, such as feelings of entitlement, interpersonal exploitativeness, relationships that alternate between the extremes of overidealization and devaluation, and lack of empathy. 122 Narcissistic rage. A manifestation of the vulnerable patient's aggressive responses to actual or anticipated injury to the aggrandized sense of self (i.e., the need for revenge and the unrelenting compulsion in the pursuit of undoing a hurt). It is a violent or intense anger, marked by threatened or actual attack. Narcissistic supply. Any source of admiration, acclaim, attention, or acceptance, which will reaffirm the patient's sense of omnipotence. The overidealization of other individuals (ideal parent figures) from whom the patient expects said supplies exemplifies the typically exploitative interpersonal relationships in which narcissistic persons engage. Negative transference. The development of a hostile attitude toward the analyst. The term refers to the transference of a negative attitude, not to a transference in a negative direction. Omnipotence. Fantasies of unlimited success, power, brilliance, and beauty or a grandiose sense of self-importance (i.e., exaggeration of achievements and talents). Omnipotence is asserted as a denial of feelings of vulnerability. The individual literally regards himself as able to control all the persons around him. Positive transference. The development of an affectionate attitude toward the analyst wherein the patient overvalues the analyst unrealistically. Primary narcissism. The early stage when the infant's libido is turned toward his own body. The partial persistence of this stage, or return thereto, is a major factor in some neuroses. Secondary narcissism. The withdrawing of libido from objects and investing it in the ego, especially investing it in the image of one's ego built up in childhood by identification with parents. This withdrawl of love from objects onto the self is regarded as a pathological condition. Transference. Displacement of affect from one object to another. The process whereby the patient shifts affects applicable to another person onto the analyst. (e.g., The patient directs upon the analyst the hatred he feels toward his father.) 123 Preparation of this glossary is based upon significant contributions from the following source: English, H. & English, A. (1958). A comprehensive dictionary of psychological and psychoanalytical terms. New York: David McKay. 124 REFERENCES Allport, G. (1965). Pattern and growth in personality. New York: Holt, Rinehart & Winston. American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: American Psychiatric Association. Ammons, R. & Ammons, C. (1962). The Quick Test (QT): Provisional manual. Psychological Reports, Monograph Supplement, 1(7). Ammons, R. & Ammons, C. (1979). Use and evaluation of the Quick Test (QT) — Partial summary through October, 1979: I. Published papers. Psychological Reports, 45, 943-946. Anderson, C. & Travis, L. (1983). Psychology and the liberal consensus. Waterloo, Ontario: Wilfred Laurier University Press. Ashby, H.U., Lee, R.R., & Duke, E.H. (1979). A narcissistic personality disorder MMPI scale. Paper presented at the meeting of the American Psychological Association, New York. Auerbach, J.S. (1984). Validation of two scales for narcissistic personality disorder. Journal of Personality Assessment, 48(6), 649-653. Baker, H.S. (1979). The conquering hero quits: Narcissistic factors in underachievement and failure. American Journal of Psychotherapy, 33(3), 418-42 7. Berkowitz, D.A. (1977). The vulnerability of the grandiose self and the psychotherapy of acting-out patients. International Review of Psychoanalysis, 4(1), 13-21. Bleiberg, E. (1984). Narcissistic disorders in children: A developmental approach to diagnosis. Bulletin of the Menninger Clinic, 48(6), 501-517. Borg, W.R. & Gall, M.D. (1983). Educational research (4th ed.). New York: Longman. Bourdieu, P. (1984). Distinction: A social critique of the judgment of taste. (Translated by R. Nice). Cambridge, MA: Harvard University Press. Brodey, W. (1961). Image, object and narcissistic relationships. American Journal of Orthopsychiatry, 31, 69-73. 125 Brodey, W. (1965). On the dynamics of narcissism. Psychoanaly tic Study of the Child, 20, 165-193. Bromberg, P.M. (1983). The mirror and the mask: On narcissism and psychoanalytic growth. Contemporary Psychoana1y sis, 19(2), 359-387. Bronfenbrenner, U. (1976). The experimental ecology of education. Educational Researcher, 5(9), 5-15. Bruner, J., Olver, R., Greenfield, P., et al. (1966). Studies in cognitive growth. New York: Wiley. Chaplin, J.P. (1968). A dictionary of psychology. New York: Dell. Coles, R. (1977). The children of affluence. The Atlantic, 240(3), 52-66. Collins, J.K. & LaGanza, S. (1982). Se1f-recognition of the face: A study of adolescent narcissism. Journal of Youth and Adolescence, 11(4), 317-328. Cook, T. & Shadish, W. (1986). Program evaluation: The worldly science. Annual Review of Psychology, 37, 193-232. Cronbach, L. (1982). Designing evaluations of educational and social programs. San Francisco: Jossey-Bass. Davies, M.V. (1985). The relationship of narcissistic vulnerability to academic achievement. Dissertation Abstracts International, 46, 2852. Drever, J. (1952). A dictionary of psychology. Harmondsworth, Middlesex, U.K.: Penguin. Edelson, M. (1984). Language and interpretation in psychoanalysis. Chicago: University of Chicago Press. Educational Testing Service. (1971). Sequential tests of educational progress (Series II): Teacher's manual for administering, scoring, and interpreting. Princeton: Educational Testing Service. 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(3), 291-300. Emmons, R.A. (in press). Narcissism: Journal of Personality and Social Theory and measurement. Psychology, in press. 126 English, H. & English, A. (1958). A comprehensive dictionary of psychological and psychoanalytical terms. New York: David McKay. Epstein, S. (1973). The self-concept revisited: Or a theory of a theory. American Psychologist. Feather, N.T. (1962). The study of persistence. Psychological Bulletin, 59, 94-115. Feshbach, S. (1978). The environment of personality. American Psychologist, 33, 447-455. Freud, S. (1914). On narcissism: An introduction. The Standard Edition, 14, 69-102. Freud, S. (1931). Libidinal types. The Standard Edition, 25, 216-220. Freud, S. (1962). Two short accounts of psychoanalysis. Harmondsworth, Middlesex, U.K.: Penguin. Fromm, E. (1956). The art of loving. New York: Harper & Row. Gergen, K. (1961). Theconceptofself. New York: Holt, Rinehart & Winston. Goldstein, W.N. (1985). DSM III and the narcissistic personality. American Journal of Psychotherapy, 39(1), 4-16. Grayderi, C. (1958). The relationship between neurotic hypochondriasis and three personality variables: Feelings of being unloved, narcissism, and guilt feelings. Dissertation Abstracts International, 18, 2209-2210. Grindlinger, H.M. (1986). Narcissistic disorder in the highly successful. American Journal of Psychiatry, 143(2), 269. Haaken, J. (1983). Sex differences and narcissistic disorders. The American Journal of Psychoanalysis, 43(4), 315-324. Haertel, G., Walberg, H., & Weinstein, T. (1983). Psychological models of educational performance: A theoretical synthesis of constructs. Review of Educational Research, 53, 75-91. Hoffman, S. (1986). Monsieur taste. The New York Review of Books, 33(6), 45-48. 127 Jung, C. (1954). The development of personality. New York: Princeton University Press. Kernberg, 0. (1970). Factors in the treatment of narcissistic personality disorder. Journal of the American Psychoanalytic Association, 18, 51-85. Kernberg, 0. (1974). Further contributions to the treatment of narcissistic personalities. International Journal of Psychoanalys is, 55, 215-240. Kernberg, 0. (1975). Borderline conditions and pathological narcissism. New York: Jason Aronson. Kernberg, 0. (1980). Internal world and external reality. New York: Jason Aronson. King, M. (Ed.). (1971). Canadian tests of basic skills: Teacher's guide. Toronto: Houghton-Mifflin. Kohut, H. (1968). The psychoanalytic treatment of narcissistic personality disorders. Psychoanalytic Study of theChild, 23, 86-113. Kohut, H. (1971). The analysis of the self: A systematic approach to the psychoanalytic treatment of narcissistic personality disorders. New York: International Universities Press. Kohut, H. (1972). Thoughts on narcissism and narcissistic rage. Psychoanalytic Study of the Child, 27, 360-400. Kohut, H. (1977). The restoration of the self. New York: International Universities Press. Lasch, C. (1978). The culture of narcissism. New York: W.W. Norton. Livingstone, R. (1985). Reappraising a British structuralist. The Times Higher Education Supplement, No.685, 15. Lowen, A. (1983). Narcissism: Denial of the true self. New York: Macmillan. McCutcheon, G. (1981). On the interpretation of classroom observations. Educational Researcher, 10(5), 5-10. McFarlin, D.B., Baumeister, R.F., & Blascovich, J. (1984). On knowing when to quit: Task failure, self-esteem, advice, and nonproductive persistence. Journal of Personality, 52(2), 138-155. 128 Mead, G. (1934). Mind, self and society. Chicago: University of Chicago Press. Millon, T. (1982). Millon Clinical Multiaxial Inventory Manual. Minneapolis: National Computer Systems. Mischel, W. (1968). Personality and assessment. New York: Wiley. Ornstein, P.H. (1974). A discussion of the paper by Otto F. Kernberg on "Further contributions to the treatment of narcissistic personalities." International Journal of Psychoanalysis, 55, 241-247. Parkerson, J., Lomax, R., Schiller, D., & Walberg, H. (1984). Exploring causal models of educational achievement. Journal of Educational Psychology, 76, 638-646. Prifitera, A. & Ryan, J.J. (1984). Validity of the Narcissistic Personality Inventory (NPI) in a psychiatric sample. Journal of Clinical Psychology, 40(1), 140-142. Raskin, R.N. related? Raskin, R.N. (1980). Narcissism and creativity: Psychological Reports, 46, 55-60. Are they & Hall, C.S. (1979) personality inventory. Psychological A narcissistic 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(2), 159-162. Scheirer, M.A. achievement Educational & Kraut, R.E. (1979). Increasing educational via self-concept change. Review of Research, 49(1), 131-149. Shrauger, J.S. & Sorman, P.B. (1977). initial success and failure, and determinants of persistence. Journal of Clinical Psychology, 45, 784-795. Self-evaluation, improvement as Consulting and Smith, L. (1978). An evolving logic of participant observation, educational ethnography, and other case studies. Review of Educational Research, 6, 316-377. Solomon, R.S. personality 50, 463-466. (1982). Validity of the MMPI narcissistic disorder scale. Psychological Reports, The New York Academy of Sciences. (1987). The Sciences, 27(1), 24. 129 Thomas, J. (1980). Agency and achievement. Review of Educational Research, 50, 213-240. Thompson, W. (1985). Environmental effects and educational performance. Alberta Journal of Educational Research, 31(1), 11-25. Travis, L.D. (1975). Political economy, social learning, and activism: Toward a theory of educational turmoil. Unpublished doctoral dissertation. University of Alberta. Travis, L.D. & Mah, T. (1986). Narcissism, education, and debasement through elevation. Proceedings of the Pacific Region Association For Higher Education. Seattle,WA:PRAHE Travis, L.D. & Violato, C. (1982). A possible path past some personality study problems. Alberta Journal of Educational Research, 28, 19-30. Travis, L.D., Violato, C, & White, W. (1982). The transsituational consistency of behavioral persistence in children:. A study of person-systems variability. Unpublished manuscript. University of British Columbia, Vancouver. Urist, J. (1977). The Rorschach test and the assessment of object relations. Journal of Personality Assessment, 41, 3-9. Vigilante, F.W. (1983). Students' narcissism and academic performance. Social Casework: The Journal of Contemporary Social Work, 602-608. Violato, C. & Travis, L.D. (1985). Sex as a moderator variable in the academic achievement of elementary school children: A structural analysis. B.C. Journal of Special Education, 9(3), 215-230. Violato, C, White, W., & Travis, L.D. (1984). Some concurrent, criterion-related data on validity for the Quick Test based on three Canadian samples. Psychological Reports, 54, 775-782. Walberg, H. (1984). Synthesis of research on teaching. In M.C. Whittrock (Ed.), Third handbook of research on teaching. Washington, DC: American Educational Research Association. Watson, P.J., Grisham, S.O., Trotter, M.V., & Biderman, M.D. (1984). Narcissism and empathy: Validity evidence for the narcissistic personality inventory. Journal of Personality Assessment, 48(3), 301-305. 130 Weiss, H. & Sherman, J. (1973). Internal-external control as a predictor of task effort and satisfaction subsequent to failure. Journal of Applied Psychology, 57, 132-136. Wittgenstein, L. (1953). Philosophical investigations. New York: Macmillan. Young, M.F. (1959). An investigation of narcissism and correlates of narcissism in schizophrenics, neurotics, and normals. Dissertation Abstracts International, 20, 3394. 131 APPENDIX A PROJECTIVE DEVICE The individual administration of a series of stimulus cards each containing a graphic representation of an adult-child relationship provides a projective measure of narcissism for all subjects. (The projective device is expected to tap any narcissistic configurations within the subject via the emergence of an idealizing or mirroring transference.) Use of the device is predicated on the assumption that each of the stimulus cards carries equal weight. At this point, the psychometric properties of the instrument are not fully known, and pursuit of same will form the basis of a future study. Hence for present purposes, each card is treated as equivalent to any other in the series. The projective device consists of 14 stimulus cards which depict 8 prototypical situations. All but two of these are the subject of 2 cards so that the gender of the child can be alternated for each scene as follows: card 1 - (dental office). A male dentist with a boy patient. The patient is reclining on the dentist's chair. The dentist is standing between the patient and a table of dental instruments. card 8 - (dental office). A male dentist with a girl patient. As above. card 2 - (bedroom). A mother with her son. The boy and his mother are standing face to face. The mother is pointing upwards while looking down at the floor, where objects are scattered between the two figures. Behind the boy is a bed with creases in the sheets. card 9 - (bedroom). A mother with her daughter. As above. card 3 - (tennis lesson). A female recreation leader with a boy student. The student and the recreation leader are standing on opposite sides of the net/ court. The recreation leader is holding a ball in one hand and a racket in the other. The student is watching with his racket held down. card 10 - (tennis lesson). A male recreation leader with a girl student. As above. card 4 - (house). An adult with a child (both of ambiguous sex. The child is standing on the ground beside a house and is holding up a ladder. The adult is standing on the roof of the house. 132 card 11 - (open sea). A man and a woman with a child (sex ambiguous). The child is sitting in a boat in the middle of the ocean. Nearby, the man is standing on top of an overturned sailboat. The woman is hanging onto the end of the boat. Signs of rough weather are evident in the picture. card 5 - (nurse's office). A female nurse with a boy patient. The patient is sitting on the edge of a cot. The nurse is standing between the patient and a table of medical instruments. card 12 - (nurse's office). A female nurse with a girl patient. As above. card 6 - (classroom). A male teacher with a boy student. The student is sitting in his desk with an open book before him. The teacher is standing beside the student and is pointing at the open book. card 13 - (classroom). A male teacher and a girl student. As above. card 7 - (camp). A boy scout leader with a boy scout. The boy scout and the scout leader are standing face to face. The scout leader is pointing. Signs of night are evident in the background. card 14 - (camp). A girl guide leader with a girl guide. As above. ** NOTE: The following arrangement of the 14 stimulus cards is consistent with the order of administration. 136 137 139 140 141 1 143 145 146 147 APPENDIX B RECORD FORM FOR PROJECTIVE DEVICE Background Information Name _ Code Birth Order/No. of Siblings Sex Parents' Occupations Age Date Tested GradClinical Judgment Indicate clinical judgment before the actual scoring of responses. Overall Judgment: Notable signs of narcissism No apparent signs of narcissism Judgment of Degree: High ( ) Medium ( ) Low Judgment of Type: Narcissistic-obsessional Non-obsessional Notes: 148 Scoring Matrix Directions: For each card, score "1" for each indicator that appears in a subject's response. Score "0" for each indicator that is not present in the response. Score "NP" for no projection; and score "NR" for no response. Record scores for both representational (R) and presentational (P) indicators. (Illustrative indicators for each of the manifestations is provided in the scoring guide. Manifest evidence of narcissism should bear at least a "family resemblance'* to the example provided for a particular indicator.) Sub-total the scores for each card- and for each indicator. Add the sub-totals of both the cards and. the indicators to derive a total score for the subject. (Check procedure: the sum of the columns should equal the sum of the rows.) Stimulus Cards Indicators of Narcissism 1 2 3 4 5 6 7 8 9 10 11 12 13 14 T 0 T A L envy R P devaluate R P depreciate R P aggrandize R P omnipotence R P attention R P self-esteem R P entitlement R P exploit R P idealize R P empathy R P TOTAL 149 PROFILE ANALYSIS Clinicians who wish to draw a profile should first transfer the subject's sub-totals of the scores for each card to the row of boxes below. Then mark an "x" on the dot corresponding to the sub-totaled score for each card, and draw a line connecting the "x" 's. Stimulus Cards 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Sub-totaled Score 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 150 SCORINGGUIDE Indicators of Narcissism and Examples Representation (signed meaning) a. narcissistic envy (e.g., "The girl would give anything to be like her guide leader."). b. devaluation of help or support (e.g., "They are putting the camp in order. The boy is doing most of the work, and the scout leader is not doing his share. The leader isn't doing much; he's just giving orders."). c. depreciatory attacks on the helper (e.g., "She hates her teacher; he's useless."). d. aggrandizing of the self (overestimation of one's abilities and achievements) (e.g., "The girl is the best athlete in her class, so she can play her own way."). e. sense of omnipotence (preoccupation with fantasies of unlimited success, power, brilliance, beauty, or ideal love) (e.g., "The water is full of sharks, but he will swim across and save the people from drowning."). f. exhibitionistic need for constant attention and admiration (e.g., "The dentist keeps staring at her because her teeth are so perfect."). g. fragility of self-esteem (cool indifference or feelings or expressions of rage, revenge, hostility, inferiority, shame, humiliation, or emptiness in response to criticism, indifference of others, or defeat) (e.g., "She feels hurt because her teacher told her to do the assignment over again because it wasn't good enough. He says she can do better, but she doesn't care; it's just a junkie piece to pass the time."). h. sense of entitlement (e.g., "The boy is surprised and annoyed that his mother did not clean up his room for him."). i. exploitativeness in interpersonal relationships (e.g., "She is trying to be really nice to her teacher so that she can get into his good books."). j. overidealization and devaluation in interpersonal relationships (e.g., "The nurse is his favorite person; she's a dream."). (e.g., "She thinks she's so great! She's always telling him what to eat, how much sleep he needs, when to wash. She makes him sick!"). 151 k. lack of capacity for empathy (e.g., "The girl is annoyed with her father because he took so long to fix the roof that he's too tired to take her to the beach, and now she has to help him down too."). Presentation (expression) 1. inflection (harshness of voice). 2. amplitude (loudness of voice). 3. pitch (subdued voice or increasing height of pitch indicating rising emotion). 4. cadence (rapidity of expression indicating increased arousal). 5. lability of emotion (i.e., depression; excitation). 6. tone (meaning implied by tone) (i.e., envy; -regret; sarcasm; shame). 7. sense (soul or vision; organizing or shaping principle) (i.e., Aristotle's six aspects of poetry: melos (melody); lexis (diction); spectacle; mythos (plot); ethos (characters and setting); dianoia (thought). 152 APPENDIX C NPI Naie_; Date Sex Age Education Occupation Instructions: For each question, circle A or B according to how you feel about yourself. Do not skip any questions. Example: A I love licorice. B I hate licorice. If you like licorice, circle A. If you don't, circle B. 1. A I am a fairly sensitive person. B I am more sensitive than most other people. 2. A I have a natural talent for influencing people. B I am not good at influencing people. 3. A Modesty doesn't become me. B I am essentially a modest person. 4. A Superiority is something that you acquire with experience. B Superiority is something you are born with. 5. A I would do almost anything on a dare. B I tend to be a fairly cautious person. 6. A I would be willing to describe myself as a strong personality. B I would be reluctant to describe myself as a strong personality. 7. A When people compliment me I sometimes get embarrassed. B I know that I am good because everybody keeps telling me s o. 8. A The thought of ruling the world frightens the hell out of me. B If I ruled the world it would be a much better place. 9. A People just naturally gravitate towards me. B Some people like me. 153 10. A I can usually talk my way out of anything. B I try to accept the consequences of my behavior. 11. A When I play a game I don't mind losing once in a whi1e. B When I play a game I hate to lose. 12. A I prefer to blend in with the crowd. B I like to be the center of attention. 13. A I will be a success. B I'm not too concerned about success. 14. A I am no better or no worse than most people. B I think I am a special person. 15. A I am not sure if I would make a good leader. B I see myself as a good leader. 16. A I am assertive. B I wish I were more assertive. 17. .A I like having authority over other people. B I don't mind following orders. 18. A There is a lot that I can learn from other people. B People can learn a great deal from me. 19. A I find it easy to manipulate people. B I don't like it when I find myself manipulating people. 20. A I insist upon getting the respect that is due me. B I usually get the respect that I deserve. 21. A I don't particularly like to show off my body. B I like to display my body. 22. A I can read people like a book. B People are sometimes hard to understand. 23. A If I feel competent I am willing to take responsibility for making decisions. B I like to take the responsibility for making decisions. 24. A I am at my best when the situation is at its worst. B Sometimes I don't handle difficult situations too well. 25. A I just want to be reasonably happy. B I want to amount to something in the eyes of the wor1d. 154 26. A My body is nothing special. B I like to look at my body. 27. A Beauty is in the eyes of the beholder. B I have good taste when it comes to beauty. 28. A I try not to be a show off. B I am apt to show off if I get the chance. » 29. A I always know what I am doing. B Sometimes I'm not sure of what I am doing. 30. A I sometimes depend on people to get things done. B I rarely depend on anyone else to get things done. 31. A I'm always in perfect health. B Sometimes I get sick. 32. A Sometimes I tell good stories. B Everybody likes to hear my stories. 33. A I usually dominate any conversation. B At times I am capable of dominating a conversation. 34. A I expect a great deal from other people. B I like to do things for other people. 35. A I will never be satisfied until I get all that I deserve. B I take my satisfactions as they come. 36. A Compliments embarrass me. B I like to be complimented. 37. A My basic responsibility is to be aware of the needs of others. B My basic responsibility is to be aware of my own needs. 38. A I have a strong will to power. B Power for its own sake doesn't interest me. 39. A I don't very much care about new fads and fashions. B I like to start new fads and fashions. 40. A I am envious of other people's good fortune. B I enjoy seeing other people have good fortune. 41. A I am loved because I am lovable. B I am loved because I give love. 42. A I like to look at myself in the mirror. B I am not particularly interested in looking at myself in the mirror. 155 43. A I am not especially witty or clever. B I am witty and clever. 44. A I really like to be the center of attention. B It makes me uncomfortable to be the center of attent ion. 45. A I can live my life in any way I want to. B People can't always live their lives in terms of what they want. 46. A Being an authority doesn't mean that much to me. B People always seem to recognize my authority. 47. A I would prefer to be a leader. B It makes little difference to me whether I am a leader or not. 48. A Iamgoingtobeagreatperson. B I hope I am going to be successful. 49. A People sometimes believe what I tell them. B I can make anybody believe anything I want them to. 50. A I am a born leader. B Leadership is a quality that takes a long time to develop. 51. A I wish someone would someday write my biography. B I don't like people to pry into my life for any reason. 52. A I get upset when people don't notice how I look when I go out in public. B I don't mind blending into the crowd when I go out in public. 53. A I am more capable than other people. B There is a lot that I can learn from other people. 54. A I am much like everybody else. B I am an extraordinary person. 

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 22 12
Republic of Korea 20 2
United States 7 5
France 3 0
Japan 3 0
City Views Downloads
Unknown 25 9
Beijing 19 0
Ashburn 3 0
Shenzhen 3 12
Tokyo 3 0
Redmond 1 0
Phoenix 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097751/manifest

Comment

Related Items