Geographic Displacement Procedure and Georeferenced Data Release Policy for the Demographic and Health Surveys

Georeferencing population-based surveys such as the Demographic and Health Surveys (DHS) have many benefits. Most important researchers can analyze respondent locations spatially to identify geographical patterns associated with specific demographic and health outcomes and programs. Second the proximity of survey communities to geographic locations such as health centers roads and cities can serve as a proxy for access to services; and third data from sampled locations can be aggregated to form new units of analysis such as climatic zones or program intervention areas rather than being constrained to administrative units. However while it is important to make available to researchers analysts and policymakers the georeferenced data from population-based surveys it is also important to maintain the confidentiality of survey respondents. This report describes the geographic displacement procedures and georeferenced data release policy developed by the DHS project to protect the identity of survey respondents. The georeferenced data release policy applies specifically to the release of georeferenced data from DHS household surveys. It aims to balance the need to protect respondent confidentiality with the need to make available to the public analytically useful data. The policy incorporates two levels of protection: first data from the same enumeration area (EA) are aggregated to a single point coordinate; then the coordinate is geomasked through use of the Global Positioning System (GPS) coordinate displacement process. In DHS household surveys the GPS coordinate displacement process is carried out as follows: urban clusters are displaced a distance up to two kilometers (0-2 km) and rural clusters are displaced a distance up to five kilometers (0-5 km) with a further randomly-selected 1% (every 100th) of rural clusters displaced a distance up to 10 kilometers (0-10 km). (Excerpt)