Recent years have seen widespread use of small-area maps based on census data enriched by relationships estimated from household surveys that predict variables, such as income, not covered by the census. The purpose is to obtain putatively precise estimates of poverty and inequality for small areas for which no or few observations are available in the survey. We argue that to usefully match survey and census data in this way requires a degree of spatial homogeneity for which the method provides no basis and which is unlikely to be satisfied in practice. We document the potential empirical relevance of such concerns using data from the 2000 census of Mexico.
This content is only available as a PDF.
Copyright by the President and Fellows of Harvard College and the Massachusetts Institute of Technology