This study focuses on the development, implementation and evaluation of a research-based legally defensible forensic interview protocol for social workers in South Africa in order to improve the social worker's knowledge, understanding and skills regarding forensic assessment interviews in the field of child sexual abuse. The motivation for the study has its origin in the fact that current legislation in South Africa requires that all cases of alleged sexual abuse have to be reported to either social workers or the South African Police Service. Social workers often find themselves having to deal with allegations of sexual abuse, and are faced with the challenging task of conducting assessment interviews, where the result of such interview will be a determining factor in the final outcome of the legal proceedings. The research problem is that there is currently no national research-based forensic interview protocol for social workers in South Africa. A quantitative research approach was followed. Applied research was used to address immediate problems that are encountered by professionals in practice. Intervention research was used as research methodology. The quasi-experimental design was applied where a comparison of two groups were done. The hypothesis formulated for this study is: If this interview protocol will be applied in cases of alleged sexual abuse against children of the middle childhood, it would facilitate the disclosure in a more legally acceptable and defensible manner. An interview protocol with seven definite phases was developed. A self-developed checklist consisting of 119 fundamentals compiled from comprehensive literature study, consultations with experts and extensive experience of the researcher, was used as a measuring instrument. The researcher applied the newly developed interview protocol with ten girls (experimental group) in the middle childhood who have allegedly been sexually abused. The interviews were audio-recorded and evaluated by means of the self-developed checklist. An independent social worker also interviewed ten allegedly sexually abused children (comparison group) of the same age and gender as the experimental group. This social worker, however, made use of her own interview protocol. She represents the social workers in South Africa. The interviews were also audio-recorded and evaluated by means of the self-developed checklist. An independent professional person coded a sample of 50% of all interviews. All data were submitted to the Department Statistics of University of Pretoria who has done the statistical analysis. In chapter two the phenomenon of child sexual abuse is discussed and aspects that interviewers need to take in consideration were highlighted. In chapter three all aspects of child development are discussed. Developmental issues which need to be accommodated during forensic interviews are explored. Chapter four focuses on interview techniques in the forensic context, and international guidelines on forensic interviews were explored. This information was used to develop the proposed forensic interview protocol. In chapter five the proposed seven-phase forensic interview protocol was discussed. In chapter six all the data that was collected was quantified, analysed and interpreted with the assistance of the Department of Statistics of the University of Pretoria. The statistical analysis showed that in five of the seven phases a statistically significant difference was found between the experimental and comparison groups. The results suggest that the seven-phase forensic interview protocol was successfully implemented, and could be considered a new development and thus a contribution to the social work profession. However further research with a larger sample of children is needed. Conclusions and recommendations (chapter 7) of this study are put forward in accordance with the process that was followed in developing and implementing the interview protocol which would assist social workers when dealing with alleged victims of child sexual abuse. / Thesis (DPhil)--University of Pretoria, 2007. / Social Work and Criminology / unrestricted