Since 1984, The DHS Program has provided technical assistance to more than 300 surveys in over 90 countries, advancing global understanding of health and population trends in developing countries. DHS has earned a worldwide reputation for collecting and disseminating accurate, nationally representative data on fertility, family planning, maternal and child health, gender, HIV/AIDS, malaria, and nutrition. The information is processed and presented in reports and data formats that describe the situation of the relevant country. Details on The DHS Program standard procedures, methodologies, and manuals can be found here.

  • Data Structure

    All data avialable is available in shapefile format for use in a GIS. Each DHS Program downloadable SDR file contains the following:

    • One file for use in a GIS. File choices are:
      • A shapefile for use in a GIS. Each shapefile is made up of 7 files containing the following extensions: .dbf, .prj, .sbn, .sbx, .shp, .shp.xml, .shx
      • A geodatabase file for use in ESRI ArcGIS products
    • An explanatory notes file containing citation and data use guidelines
    • A data availablility file containing a chart of data available for the indicators and surveys selected during the download process
    • A field names file that contains titles and indicator definitions
    • A XML metadata file containing general information about the shapefile as well as detailed information about each shapefile attribute
  • Subnational Boundaries

    GIS data are provided "as is" and boundary representations are not necessarily authoritative. The national and subnational geometry in the downloadable shapefiles comes from the following sources:


The DHS Program provides a standard set of spatially modeled map surfaces for recent population-based survey. Each modeled surfaces is produced using standardized geostatistical methods, publically available DHS data, and a standardized set of covariates across countries. Each map package contains a mean estimate surface, an uncertainty surface, and corresponding information on the model creation process and validation. Not all indicators are available for all countries, depending on availability of appropriate data inputs in any particular survey.

For more information on the modeling methods, refer toSpatial Analysis Report 11. For more information on the use of these surfaces in decision making, refer to Spatial Analysis Report 14.

  • Data Structure

    The download package is a ZIPPED folder which contains 5 components:

    • Mean estimate modeled surface (raster file in GeoTIFF format)
    • Uncertainty estimate modeled surface (raster file in GeoTIFF format)
    • Image of mean estimate modeled surface (PNG format)
    • Image of uncertainty estimate modeled surface (PNG format)
    • Indicator specific document on modeling procedures (PDF format)

    The modeled surface package naming conventions follow a standard derived from The DHS Program API (application program interface). The API provides standard names for country surveys and indicators, and allows users to find corresponding information about the country survey and associated information on the specific indicator. Each dataset has a standard naming convention that identifies the country survey year, the indicator, the type of data, and the version number. The fields are described below.

    These fields are combined for each component of the data packages. These are described in Table 3 with an example for the Ghana 2008 DHS Children under 5 stunted indicator.

    Folder NameSurveyID _SDRID_MS_v#GH2008DHS_CNNUTSCHA2_MS_v01
    DatasetsSurveyID _SDRID_MS_TYPE_v#
    Image FilesSurveyID _SDRID_MS_TYPE_v#
    DocumentationSurveyID _SDRID_MS_v#GH2008DHS_CNNUTSCHA2_MS_v01


The geospatial covariate datasets link survey cluster locations to ancillary data - known as covariates – that contain data on topics including population, climate, and environmental factors. This allows individuals with limited Geographic Information Systems (GIS) experience to conduct geospatial statistical analysis without having to manually source and link these covariates to cluster locations. These covariates are extracted from freely available global datasets. The covariates are packaged as a zip and include a .csv data file and a data description text file that defines each of the included covariates. Read more on our blog.

  • Data Structure

    The download is a zip file containing:

    • Covariates data related to the survey (CSV format)
    • DHS Covariates Extract Data Description file describing the covariate data and the extraction procedures used (PDF format)

    The covariate datasets are named using The DHS Program’s standard naming convention.

    TopicSpecific dataTimeframeUnitsSource
    AgricultureDrought episodes1980 - 2000Number of episodesSEDAC
    AgricultureGrowing season lengthSingle datasetMode of 16 categories representing a range of the number of days within the period of temperatures above 5 °C when moisture conditions are considered adequateIIASA/FAO
    EnvironmentAridity1960 - 1990Unitless index. Higher Aridity Index (AI) reflect more humid conditions, low AI reflect greater aridity.CGIAR
    EnvironmentEnhanced vegetation index1985, 1990, 1995, 2000, 2005, 2010, 2015Unitless index. Higher scores indicate higher vegetation vigor/photosynthetic activityAVHRR (1985-1995), MODIS (2000-2015)
    EnvironmentPotential evapotranspiration1950 - 2000Monthly estimate of PET (mm/month)CGIAR
    EnvironmentProximity to coast/ large lakesN/AMetersGSHHG
    EnvironmentProximity to Protected areaN/AMetersProtected Planet
    EnvironmentRainfall1985, 1990, 1995, 2000, 2005, 2010, 2015mm/yearCHIRPS
    EnvironmentTemperature (average monthly)January, February, March, April, May, June, July, August, September, October, November, December. Average for 1970-2000.Degrees CelsiusWorldClim
    HealthITN net coverage2000, 2005, 2010, 2015% of people who slept under an insecticide treated net (ITN) on any given nightMAP
    HealthMalaria2000, 2005, 2010, 2015Plasmodium falciparium malaria cases per person per year observedMAP
    InfrastructureNightlights2015 annual compositeAverage cloud-free radiance valuesVIIRS (DNB)
    InfrastructureProximity to national border2014MetersState Dept. LSIB
    InfrastructureTravel times2000Estimated travel time (minutes) to the nearest city of 50,000 or more people in year 2000FOROBS
    InfrastructureGlobal human footprint1995-2004Unitless index. Higher values indicate higher human influence.SEDAC
    InfrastructureGHS Built-up Grid1990, 2000, 2015Built-up presence index, range 0-1. Higher value is more confidence that it is built-upGHSL (Landsat) (EC JRC)
    Infrastructure GHS Settlement grid1990, 2000, 2015Mode of categories:
    • 1 = “rural cells” or base (BAS)
    • 2 = “urban clusters” or low density clusters (LDC)
    • 3 = “urban centres” or high density clusters (HDC)
    PopulationUN adjusted population count2000, 2005, 2010, 2015 (adjusted)Number of peopleGPW v4
    PopulationUN adjusted population density2000, 2005, 2010, 2015 (adjusted)Number of people/km2GPW v4
    PopulationPopulation Count2000, 2005, 2010, 2015Number of peopleWorldPop
    PopulationPopulation density2000, 2005, 2010, 2015Number of people/km2WorldPop


Over the past six decades, the U.S. Census Bureau has worked in over 100 countries to provide both short- and long-term technical assistance on a wide variety of topics related to censuses, surveys, and information systems. The U.S. Census Bureau has an innovative program to develop global, subnational population data and maps for use in international humanitarian relief and disaster planning. The data from the U.S. Census Bureau provides subnational 5-year age/sex group population estimates and projections for 2000-2015 for designated countries in the President's Emergency Plan for AIDS Relief (PEPFAR) program.