No description available.

M.A. (Psychology) / Please refer to full text to view abstract

published_or_final_version / Clinical Psychology / Master / Master of Social Sciences

This study extended earlier research supporting discrepancy theory by including a multidimensional conceptualization of gender including attitudes, behaviors, and characteristics. Analyses revealed that gender discrepancy when assessed multidimensionally or unidimensionally (as in past research) was not significantly related to eating disordered symptomatology.

A wide body of research has investigated the possible pathogenic role of the family in the development of eating disorders. Within the context of the research which places family dynamics at the centre of psychopathology, little research attention has been given to the relationship between parent-adolescent attachment and eating disorders. There is currently no existing South African research in this area. This study aims to redress this balance by exploring the relationship between parental attachment (as measured by the Parental Attachment Questionnaire) and disordered eating (as measured by the Eating Disorders Inventory) among white female adolescents. The sample comprised 209 white female learners from a former 'model C' school in the Durban area. It was found that highly significant negative correlations existed between most of the subscales of the PAQ and EDI. Overall, canonical analysis revealed a significant relationship between parent-adolescent attachment and disordered eating. The relationship between the variables of attachment and disordered eating was very similar regardless of whether the two subscales of the PAQ (Affective Quality of Attachment and Parental Role in Providing Emotional Support) were combined or not. It was found that those adolescents who described their parental relationships as affectively positive and emotionally supportive and viewed their parents as supporting their independence, also described themselves as experiencing low levels of weight preoccupation, low levels of bulimic behaviour and interpersonal distrust, and high levels of personal effectiveness and interoceptive awareness. The above results are discussed in the light of the relevant available literature and research. The methodological and conceptual limitations of the study are explored and provide a basis for recommending possible future research. / Thesis (M.A.)-University of Natal, Pietermaritzburg, 1999.

The purpose of this dissertation was to undertake a conceptualization and empirical "mapping" of the ethic-of-care domain - often characterized as self-referential and lacking in rigor. The current focus is upon conventional forms of care, involving notions of moral "goodness" as self-silencing and -sacrificial. Employing a "super" circumplex as a prescriptive and descriptive tool, projected item analyses were undertaken as a theoretico-structural clarification of existing scale items, along with a provisional pool of additional items generated as part of Study 1. Based in part on proposed circular criteria, the Conventional Care Scales (CCS) were developed and submitted to a conjoint principal components analysis along with the battery of other care/dependency scale items. An examination of item circular distributions, factor loadings, alpha-contribution plots, and thematic content revealed several factors expressive of two "faces" of conventional care, submissive and ingenuous. These two forms were shown to have unique correlates with measures of adjustment, interpersonal competencies, other factors of the Five-Factor Model, false-self beliefs, and reported distress in narrated accounts of rejected care giving. Gender differences in the association between indices of adjustment and conventional care were also found. The range of conventional care was also extended through the development of scales reflecting other-directed and socially prescribed dimensions of conventional care. The factor structure for the CSS was also replicated in a second sample (Study 2). Participants for Studies 1 and 2 were composed of undergraduate students (N = 302 in both samples) who completed a battery of questionnaires in the first study and the CCS alone in the second study. Anticipating future clinical directions, secondary analyses using structural equation modelling were conducted on an existing data set (N = 92) which included measures of conventional care and perfectionism, along with indices of psychological adjustment and eating disordered attitudes. Results indicated that conventional care, for which there is little research, was more predictive of adjustment and eating disordered attitudes than perfectionism, for which there exists a large clinical literature. This research contributes to an understanding of ways in which an ethic of care can "go awry," as well as proposes a research platform upon which the clinical implications of morality and self-ideals can be investigated. These findings speak to both the constraints and prescriptions that can inform a philosophical ethic of care.

The purpose of this dissertation was to undertake a conceptualization and empirical "mapping" of the ethic-of-care domain - often characterized as self-referential and lacking in rigor. The current focus is upon conventional forms of care, involving notions of moral "goodness" as self-silencing and -sacrificial. Employing a "super" circumplex as a prescriptive and descriptive tool, projected item analyses were undertaken as a theoretico-structural clarification of existing scale items, along with a provisional pool of additional items generated as part of Study 1. Based in part on proposed circular criteria, the Conventional Care Scales (CCS) were developed and submitted to a conjoint principal components analysis along with the battery of other care/dependency scale items. An examination of item circular distributions, factor loadings, alpha-contribution plots, and thematic content revealed several factors expressive of two "faces" of conventional care, submissive and ingenuous. These two forms were shown to have unique correlates with measures of adjustment, interpersonal competencies, other factors of the Five-Factor Model, false-self beliefs, and reported distress in narrated accounts of rejected care giving. Gender differences in the association between indices of adjustment and conventional care were also found. The range of conventional care was also extended through the development of scales reflecting other-directed and socially prescribed dimensions of conventional care. The factor structure for the CSS was also replicated in a second sample (Study 2). Participants for Studies 1 and 2 were composed of undergraduate students (N = 302 in both samples) who completed a battery of questionnaires in the first study and the CCS alone in the second study. Anticipating future clinical directions, secondary analyses using structural equation modelling were conducted on an existing data set (N = 92) which included measures of conventional care and perfectionism, along with indices of psychological adjustment and eating disordered attitudes. Results indicated that conventional care, for which there is little research, was more predictive of adjustment and eating disordered attitudes than perfectionism, for which there exists a large clinical literature. This research contributes to an understanding of ways in which an ethic of care can "go awry," as well as proposes a research platform upon which the clinical implications of morality and self-ideals can be investigated. These findings speak to both the constraints and prescriptions that can inform a philosophical ethic of care. / Arts, Faculty of / Psychology, Department of / Graduate

The performances of 114 eating disordered females on measures of individual and family functioning were examined via t-tests, analyses of variance, and chi squares analyses. Subjects met DSM-III-R criteria for anorexia nervosa, anorexia nervosa with bulimia nervosa, bulimia nervosa, or eating disorder not otherwise specified. Measures included selected scales of the MMPI, Rotter's Internal-External Locus of Control Scale, the Bem Sex Role Inventory, the Eating Disorder Questionnaire, the Moos Family Environment Scale, and the Berren-Shisslak Family Dynamics Survey. First, performances on the measures were compared to normative samples and across diagnostic groups. As expected, the subjects differed from normative samples on many dependent measures, but the diagnostic groups differed little among themselves. Next, an examination of 45 clinical features suggested that subjects were not as symptomatically distinct as diagnosis might imply. Finally, diagnosis was set aside to examine the relationships of specific clinical features to the measures of individual and family functioning. These features included age, weight history, food binges, vomiting, laxative and diuretic use, food restriction, menstrual history, exercise, drug and alcohol abuse, symptom severity, inpatient treatment history, and additional diagnosis. Surprisingly, a history of anorectic weight was not related to any of the measures, but amenorrhea emerged as a clinical feature of some import, not just among the anorectic subjects but among subjects generally. Other findings suggested that patients who binge have difficulty with separation, that binges may be a metaphor for unsatisfied cravings for nurturance, and that vomiting and exercise may enhance a sense of separateness while laxative use may represent the private expulsion of anger. The need for family involvement in treatment was highlighted by many relationships between the clinical features and indices of family dysfunction.

In this study, the relationship between the risk factors of perfectionism and depression and eating disorder symptomatology was examined. A sample of female undergraduates completed the following three measures: the Beck Depression Inventory-II (BDI-II), the Multidimensional Perfectionism Scale (MPS), and the Questionnaire for Eating Disorder Diagnoses (Q-EDD). Results indicated that perfectionism is significantly positively correlated with depression. Higher levels of perfectionism were also associated with the presence of eating disorder symptoms, whereas lower levels of perfectionism were associated with the absence of eating disorder symptoms. Finally, an increase in perfectionism levels was predictive of eating disorder symptoms. Directions for future research and clinical implications are discussed. / Department of Counseling Psychology and Guidance Services

Research suggests that the eating disorders (anorexia nervosa and bulimia) represent a caricature of the sociocultural values placed on young women to achieve thinness and beauty ideals. Although eating disorders have long been thought to occur only in White, "Western" cultures, more recent research suggests that women from different cultural groups are presenting with unhealthy eating attitudes and behaviours. In South Africa's pluralistic cultural context, the effects ofcontinuous first-hand contact between cultures (acculturation) is an important area of research, especially in light ofthe hypothesised etiological role ofsociocultural factors in eating disorders. The present study aims to address the association between acculturation and disordered eating in a non-clinical sample of nursing students in Pietermaritzburg. Additionally, it aims to contribute to the development of a local acculturation instrument. The South African Acculturation Scale (SAAS) was developed based on the work of Berry (1976), Berry, Trimble and Olmedo (1986) and Berry (1997). The Individualism-Collectivism (INDCOL) scale (Hui, 1988) and the Eating Disorder Inventory (EDI, Garner & Olmsted, 1984) were included in the questionnaire profile A pilot study was undertaken on 28 students in the health arena, in order to assess the psychometric properties of the assessment instruments. The results of the pilot study yielded adequate reliability co-efficients for the SAAS, although the INDCOL scale yielded unexpectedly inconsistent results. The formal study adopted a cross-sectional design on a population of 155 nursing students. The sample consisted of37 Blacks, 33 Whites, 11 Indians and 7 Coloureds between 19 and 28 years of age. Additionally, the sample included 49 Blacks, 3 Whites, 11 Indians and 4 Coloureds greater than, or equal to 29 years ofage. The research findings suggest that both Black and White respondents display a propensity towards disordered eating. Black respondents scored higher on measures of the psychological correlates of eating disorders, and Whites scored higher on the attitudinal and behavioural measures of disordered eating. Partial support was obtained for the hypothesis that assimilation and individualist values are correlated to eating disorder pathology. The findings suggest that acculturating young women from diverse cultural and racial backgrounds present with a degree of risk for the development of eating disorders. / Thesis (M.Soc.Sc) - University of Natal, Pietermaritzburg, 2000.