Student Number : 0411802R - MSc (Med) research report - Faculty of Health Sciences / Unintended pregnancy among the youth is a serious public health problem for it exposes the young women and the foetuses to higher risk of morbidity and mortality. Although recognised as a major public health problem, studies on it have been limited and localised in South Africa. Using the 1998 South African Demographic and Health Survey (SADHS) data set, this study examines the distribution of and factors associated with unintended pregnancy among South African youth. The 1998 SADHS was a nationally representative cross-sectional survey with a probability sample of twelve thousand (12 000) women between the ages of 15 and 49. The survey used a structured questionnaire to collect information on fertility issues in general from the respondents. Analysis was based on 1 395 observations which was arrived at after sub setting observations of women aged 15-24 that had pregnancy at the time of and/or three years preceding the survey. Logistic regression model was employed to estimate the effects of identified predictors on unintended pregnancy. The results show a high level of unintended pregnancy with only 29% of the pregnancies wanted. The level of unintended pregnancy varies by region and some socio-economic variables. Respondents from KwaZulu Natal had the highest percentage (81%) of unintended pregnancy while North West had the lowest at 56%. It was also found that the higher the educational level of the respondents, the higher the incidence of unwanted pregnancy. Among the respondents using modern method of contraception, 74% reported having unintended pregnancy while the probability of unintended pregnancy was found to decrease with increase in age at first intercourse. Overall, about 69% of respondents who reported unintended pregnancy had last sexual relation with their regular partners, 21% with marital partners and 10% with casual partners. Using stepwise logistic regression, five critical predictors of unintended pregnancy among South African youth were identified. These are: age group, region, marital status, education and relationship to the last sexual partner. The findings of this study have implications for reproductive health policies and programs in designing appropriate national programs for reducing the incidence of unintended pregnancy among South African youth. The need for further research into this area using triangulated methodology is recommended.