South Africa has a well documented history of male labour migration. Thus, living in a family with at least one parent away (particularly the father) for a long period of time is a common phenomena of childhood for many children in that country. Since the effect of migration on health and mortality may be mixed, it is important to explore this relationship closely. Studies of this kind present the opportunity to identify intervention programs appropriate in different populations in our attempt to reach the WHO Millennium Development Goal on infant and child mortality. Therefore, the goal of this paper was to assess whether or not father s migration improved child mortality using secondary data from Agincourt Health and Demographic Surveillance Site in South Africa collected in 2002. Using logistic regression the authors compared child mortality between children born in a family where the father had left the family to work in another place with those whose fathers were non-migrants controlling for other demographic and socio-economic variables.