Accounting for survey design in Bayesian disaggregation of survey-based areal estimates of proportions: An application to the American Community Survey

Understanding the effects of social determinants health on outcomes requires data characteristics neighborhoods in which subjects live. However, estimates these are often aggregated over space and time a fashion that diminishes their utility. Take, for example, from American Community Survey (ACS), multiyear nationwide survey administered by U.S. Census Bureau: small municipal areas 5-year periods, whereas 1-year only available with populations >65,000. Researchers may wish to use ACS studies population characterize neighborhood-level exposures. not properly temporal changes or align temporally other study, while coarse spatial resolution utility characterizing neighborhood exposure. To circumvent this issue, paper we propose modeling framework disaggregate proportions derived sampling surveys, explicitly accounts design effect. We illustrate our model applying it data, generating poverty state Michigan at fine spatiotemporal resolution.