Background and Objectives Self - report data is known to be unrel iable and susceptible to factors such as social desirability bias. Methods used for collecting self - report data has thus far been unsuccessful in ameliorating known obstacles to honest self - disclosure. Considering the current HIV/AIDS pandemic and relate d health crises, it is imperative that self - report data is an accurate depiction of reality, since it informs research requirements and designs as well as intervention designs and the evaluation of the efficacy of the interventions. Aim To evaluate and co mpare the efficacy of the Informal Confidential Voting Interview (ICVI) to the FTFI (Face - to - Face Interview) and the SAQ (Self - Administered Questionnaire) in enhancing self - disclosure and minimizing social desirability bias on sensitive topics of sexual ex perience and sexual activity. Study Design A sample of 110 undergraduate and post - graduate students at various tertiary education institutions in Pietermaritzburg were randomly allocated to the ICVI, the SAQ or the FTFI. The ICVI combined a face - to - face interview with a voting box method devised to enhance response anonymity. The FTFI and the SAQ were administered according to a standardized procedure to maximize confidentiality and self - disclosure. Results The self - disclosure scores were significant ly higher for the ICVI in comparison to the FTFI and the SAQ, with a p = 0.005. Post - hoc tests revealed that the ICVI performed significantly better in self - disclosure scores than the FTFI with p = 0.022 and the SAQ with p = 0.015. There was no significa nt difference in self - disclosure scores between the SAQ and the FTFI. Using the Marlowe - Crowne scale of social desirability bias, a significant difference in social desirability bias scores were achieved with p = 0.043. However, the post - hoc analysis ind icated no affirmative significant mean difference in social desirability score among any of the methods. Males displayed greater self - disclosure than females with p = 0.013, but for both sexes the ICVI group achieved the highest mean self - disclosure score s than the FTFI - and the SAQ group. Conclusion The results of this study concluded that the employment of ICVI fundamentally resulted in better quality data than the SAQ and the FTFI on topics of sensitivity and controversial behaviours. The findings ar e suggestive of the successful implementation of the ICVI method across potentially diverse research contexts that rely on self - report data, as the method is adaptable to the target population and its characteristics. Further research is warranted to buil d on its current design and facilitate the implementation of the ICVI across the wide disciplines of self - report data. / Thesis (M.Sc.) - University of KwaZulu-Natal, Pietermaritzburg, [2009]