No description available.

The purpose of this investigation was to empirically test an ecological model of child abuse. The orientation of this model is to address the contexts in which individuals function. The three contexts pivotal to the current investigation are the ontogenic system, representing personality traits and the quality of care received by the mother in her childhood the microsystem, representing the dyadic relationships between the mother and her child, and the mother and the rest of her family; the microsystem, representing the relationship between the mother and her social network as well as the impact of life stress. A prospective method of investigation was used. One hundred seven women were selected from a larger pool of women previously identified as high risk. Half of these women were clearly abusing their children while the other half were providing adequate care. Discriminant function analyses were employed to determine the rates of prediction into abusing and nonabu-sing groups, first for individual systems, then for all three systems simultaneously. The hypotheses predicted a higher percentage of correct classification when all systems were considered together, rather than individually. These hypotheses were supported. An 86% rate of correct classification was obtained when ontogenic, microsystem and mesosystem variables were entered together. This is in contrast to a 76% rate of correct classification for ontogenic, 69% for microsystem and 76% for the mesosystem. The most powerful predictors were the quality of care the mother received in her own childhood, family continuity and life stress. Results from additional descriptive analyses suggest that women who abuse their children are not more socially isolated than those who do not abuse, nor are they more impulsive or hostile. Stress appeared to be an important variable only for those women who had, themselves, been victims of abuse. The results of this investigation suggest that multivariate methods are a fruitful direction for future inquiry into abuse etiology. / Graduate and Postdoctoral Studies / Graduate

The Rosebush Visualization Technique (RVT) (Allan & Crandall,1986) a projective art technique, and Q sort Methodology were combined in an experiment to test their usefulness in assessing sexual, physical and emotional abuse. Forty two children between the ages of 8 and 13 inclusive were selected from four groups; the sexually abused with a mean age of 9.5, physically abused with a mean age of 10.75, emotionally neglected with a mean age of 9.0; and children who had suffered neither abuse nor neglect with a mean age of 9.7. In accordance with the RVT, the subjects were instructed to take a minute or two to imagine themselves as a rosebush. Each child then inspected 24 drawings of rosebushes selected from a previous study using the RVT and representing a range of trauma from no known trauma to severe trauma, from amongst the four groups. Each child then selected and ordered the 24 drawings into those rosebushes that were most and least like their own visualizations. Each child subsequently developed a Q-sort of the 24 drawings that was analyzed by both Q-sort methodology and analysis of variance techniques. The results indicated that the Q-sort analyses did not support the hypotheses that traumatized children would select drawings completed by similarly traumatized children. However, a follow up repeated measures analysis of variance indicated a significant joint group x picture effect at the .05 level of significance. A study of these joint effects indicated that the two different groups, traumatized and nontraumatized, did in fact differentially identify some of the pictures, but not on the basis expected. This finding provides support for the hypothesis that children's preference for a set of pictures reflects their own background of abuse or non abuse. The implications of this finding for clinical practice and research are discussed. / Education, Faculty of / Educational and Counselling Psychology, and Special Education (ECPS), Department of / Graduate

M.A.(Psychological Research) / Child maltreatment has occurred over the ages, but has been recognised as such only during the present century. The concept of maltreatment is culturally bound and therefore dependent on value judgements within a community. What is acceptable in one society may be rejected in another. Medical professionals discovered skeletal injuries in young children that was eventually traced to harsh treatment by parents. Physical abuse was thus identified, resulting in public and professional awareness and involvement. Later neglect and sexual abuse was identified as variations of maltreatment and legislation was instituted for the protection of children. Maltreated children manifest symptoms of behavioural, emotional and scholastic problems, and antisocial behaviour such as crime is being ascribed to child abuse. Research shows a tendency of placing emotional factors at the centre of maltreatment and the concept of psychological maltreatment was thus born. The present study was an attempt at indicating the relationship between maltreatment by parents and their children's ability at forming personal attachments. A questionnaire was developed in an attempt to operationalize the concept of psychological maltreatment and make it quantifiable. A comparison between this questionnaire and the PHSF relationship questionnaire indicated that a damaged capacity to form personal attachments may be traced back to psychological maltreatment. The implications of these findings were briefly discussed.

It was recommended by the American Psychological Association and the American Bar Association in October 1999, that the time had come for psychologists to do more than act as expert witnesses, and that their influence should be felt " ... at the first level of intervention" (Foxhall, 2000, p. 1). The first level of intervention is considered, in this study, to be the decision-making process. This research examines the decision-making processes used by four of the disciplines integral to the child abuse multi-disciplinary team. Social workers, child protection officers of the South African Police Services, psychologists and prosecutors currently involved in the child abuse arena in South Africa were given case vignettes to anchor their decision-making on simulated examples. On the basis of these in-depth interviews, using a Grounded Theory methodology, three specific questions are addressed, namely: how decisions are made and substantiated; how they differ across disciplines; and how they compare to recommended protocol. Results indicate interesting differences across these four disciplines. The differences between the decision-making processes in the disciplines seem to be rooted in their philosophical tenets, which give rise to specific aims in addressing child abuse. Differences between the decision-making processes and recommended protocol were also noted in cases where guidelines are available. In addition differences in expected outcomes of the vignettes, use ofassumptions and a level of mistrust between professionals was shown. If this research can help to sensitise professionals to begin to understand their own decision-making processes and those of their fellow decision-makers, then perhaps key role players and policy makers may be one step closer to responding to the challenge of child sexual abuse in South Africa. / Thesis (M.A.)-University of Natal, Pietermaritzburg, 2002.