SPATIAL UNCERTAINTY IN SMALL AREA ANAYLSIS FROM SURVEY AND ADMINISTRATIVE DATA ABSTRACT Small area data are ubiquitous in public health including incident or prevalent disease counts, population sizes, socioeconomic covariates, and environmental exposures. Geographic information systems (GISs) link and merge such data but do not account for multiple sources of uncertainty. Small area estimation and disease mapping techniques provide tools for maintaining geographic precision of smaller areas and statistical precision of larger sample sizes via “borrowing information” from neighboring areas. To date, these methods do not readily incorporate design-based uncertainty from the components. Our proposed research project involves the development of statistical methodology and associated software tools as part of an evolving reproducible workflow to incorporate local population and covariate uncertainty into spatial models of disease risk.