(Academy of Science of South Africa (ASSAf), 2016-07) Mokhele, T.; Mutanga, O.; Ahmed, F.
The use of a single geographical unit to both collect and disseminate census data is common in many
countries across the world, especially in developing countries. In South Africa this approach poses some
challenges, as the design of small geographical units called enumeration areas to facilitate data collection
differs considerably from the design of units that aid data analysis and interpretation. We aimed to create
optimised census output areas using the Automated Zone-design Tool (AZTool) program, using the 2001
census enumeration areas as building blocks at various spatial levels, for both rural and urban settings in two
South African provinces. The results were consistent and stable. The primary criterion of the confidentiality
limit of 500 people was respected at all geographical levels or regions, in both urban and rural settings,
for newly created optimised output areas. For the second criterion, lower intra-area correlation values at
lower geographical levels for both rural and urban areas showed that higher geographical levels produced
more homogeneous output areas than did lower geographical levels or regions. Our obtained intra-area
correlation of 0.62 for the two provinces combined indicated that the selected homogeneity variables were
good indicators of social homogeneity for creating optimised output areas in South Africa. We conclude
that the AZTool software can be used to effectively and objectively create optimised output areas for South
African data. Further research on the comparison of the newly created output areas with existing output areas
in South Africa should be explored.