Anonymisation of cluster locations
At the heart of any reputable international household survey programme is respondent protection. The MICS Programme is applying the present global gold standard method of anonymisation of cluster locations, as developed by the DHS Program. This is achieved through a custom-built software plugin working on the open-source Geographic Information System (GIS) software QGIS. The plugin will carry MICS branding and terminology but can be used for any household survey.
Each cluster is displaced on shapefiles with boundaries confirmed by governments as representing the individual survey’s sampling frame. Displacement is at a random angle and at random distance within a set of standard parameters, including a condition to remain within the original subnational boundary. The software plugin automates, error-checks, and visualises the geospatial anonymisation process. The output is a set of anonymised cluster locations.
The original, un-anonymised data are never shared outside the Global MICS Team without the explicit permission of the individual National Statistics Office (NSO). The anonymisation process for the traditional non-spatial data collected at household and individual level remains the same as always, ensuring that no respondent is identifiable in publicly available data.
While raw location data are very useful in themselves, the covariates are what transform these data into a powerful instrument for analyses. The covariates effectively link the traditional MICS data with ancillary data containing data on different topics such as population estimations and climatic factors. Opportunities for such linked analysis are endless and constrained only by availability of external geospatial data. Users will no longer need to rely on advanced technical GIS expertise to work with data - usage of the covariates makes it a lot easier to perform geospatial statistical analysis.
As part of the QGIS software plugin, covariates are extracted from existing open-source databases such as WorldPop, European Commission's Global Human Settlement Layer (GHSL), FAO, and others. The covariates are calculated for buffer zones of clusters or for the centre of clusters. The buffer zone is a maximum area in which one cluster can be displaced.
Spatial Data Repository
The MICS GIS Initiative is intensively working on making boundary shapefiles of all MICS surveys publicly available. These shapefiles will be for survey reporting domains and will be shared upon confirmation by NSOs as correct representations of the sample frames and the survey universe. As the boundary data can be highly sensitive, the MICS GIS documentation will clearly state that boundaries do not necessarily carry endorsements of the United Nations or national governments but are used solely for the purpose of visualising survey data on maps.