Holding a population and housing census is both a momentous undertaking and a costly
challenge for any government. It requires vast planning and organising, but the results are
vital for constructive planning. The aim of a census is to collect, process and disseminate
detailed statistics on population size, composition and distribution at small-area level. As
an undercount could affect the trust and use of census data, a major challenge in any
census is dealing with the effects of an undercount. This article investigates some of the
causes of the undercount in the South African National census of 2011, and how these
causes vary across different geographic areas. The aim of the investigation is to determine
the relationship between settlement type and the undercount in the 2011-census with the
intention of laying the groundwork for lower undercounts in future censuses and survey
implementation strategies. The key objectives of this article are to determine whether
geographic location affects the census undercount and to understand how results based
on geographic location are distributed spatially across the country.