Density Equalizing Map Projections: Techniques and Applications
DEANE W. MERRILL, S. Selvin, Möhr · eScholarship (California Digital Library) · 1992
In the statistical investigation of geographic disease clusters, one needs to correct for inevitable variations in population density.Density equalizing map projections (DEMP) have been proposed as an alternative to conventional methods.In a DEMP, a geographic map is transformed so that population density is everywhere equal.Disease cases plotted on the transformed map should have a uniform distribution if risk is everywhere equal.Because one does not need to defme arbitrary geographic subareas, the full geographic detail of the original data is preserved.Unlike previous cartogram algorithms, the DEMP algorithm described here is a continuous transformation which provides correct area magnification over the entire map domain.This fact permits analysis of geographic disease distributions by simple statistical methods which rely upon the assumption of uniform population density.Here we describe recent progress, including new objective and constraint functions, diagnostic tests and graphic displays, and estimates of computing requirements.The use and interpretation of the DEMP technique in epidemiologic applications is discussed.