Putting People in Their Place: An Overview

This 3 year project is developing methods that situate public use census microdata in space at finer resolutions than are currently available in the public domain. The methods impute a set of sample weights (probabilities) that allocate microdata records to census tracts, and with the addition of ancillary data, even to sub-tract levels. Imputed sample weights respect publicly available census tract statistics, as well as correlations between census data and ancillary data.  We refine estimates to sub-tract levels using conventional dasymetric modeling as well as using more advanced statistical modeling.

The methodology allows for quantification of uncertainty and simulations of small area populations. Our project will deliver scripts to run the statistical models, sample datasets and usage guidelines to make the tools we build accessible to anyone who has access to R, Java, and other public domain software environments. These tools should be useful to urban geographers, land use planners, transportation specialists, demographers and educators.