The generation of spatial population distributions from census centroid data

Census data are commonly used in geographical analysis and to inform planning purposes, though at the disaggregate level the basis of enumeration poses difficulties. In this paper an approach to surface generation is described that offers the prospect of revealing an underlying population distribution from centroid-based data which is independent of zonal geography. It is suggested that this can serve a wide variety of analytical, cartographic, and policy purposes, including the creation of spatial indicators of economic and social conditions and enhancing the value of census data. The approach is illustrated by reference to an analysis of part of the valleys of South Wales, in the United Kingdom.