Jittering Impacts Raster- and Distance-based Geostatistical Analyses of DHS Data
share
Fine-scale covariate rasters are routinely used in geostatistical models for mapping demographic and health indicators based on household surveys from the Demographic and Health Surveys (DHS) program. However, the observations in these surveys have GPS coordinates that are jittered for privacy purposes. We demonstrate the need to account for this jittering when analysing DHS data and propose a computationally efficient approach. We analyse the prevalence of completion of secondary education among 20-39 year old women in Nigeria in 2018 based on the 2018 DHS survey in Nigera, and demonstrate substantial changes in the estimates of range and fixed effects compared to ignoring jittering. Then based on a simulation study that mimics the dataset, we demonstrate that accounting for jittering reduces attenuation in the estimated coefficients and improves predictions.
READ FULL TEXT