The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Child Health » Geospatial analysis
Geospatial analysis [message #25561] Wed, 09 November 2022 12:04 Go to next message
Endsofanias is currently offline  Endsofanias
Messages: 9
Registered: November 2022
Dear all, I want to see the geospatial distribution of The 3 nutrition indicators(hw70-hw72) in Ethiopia by using Ethiopian DHs data from 2016, given in the file ETKR71FL.So, I bothering about it how can link these variables with the geographical pattern of the data. What additional variables or covariates are needed to illustrate my analysis? Anyone who has such experience, please help me!!!!!!!!!!!!!
Say I want to see the regional variation of Undernutrition in Ethiopia by using EDHS 2016 of ETKR71FL of the country Ethiopia. But still, I am bothering about how to use these geographical variables, where Can I get the necessary covariates, or How can I link this outcome underweight to the geographical covariates. So, Guys, please help me!!

[Updated on: Sun, 13 November 2022 08:38]

Report message to a moderator

Re: Geospatial analysis [message #25610 is a reply to message #25561] Wed, 16 November 2022 11:08 Go to previous messageGo to next message
Janet-DHS is currently offline  Janet-DHS
Messages: 773
Registered: April 2022
Senior Member
Following is a response from DHS Senior Geospatial Data Scientist, Rose Donohue:

Users can link the recode file (ETKR71F) with the GPS dataset (ETGE71FL) and the geospatial covariate dataset (ETGC72FL) using the cluster number (hv001). To visualize the geographic distribution of the variables, you will need to calculate the percentage of stunted/wasted/underweight children at the cluster-level, and merge the GPS coordinates (from ETGE71FL) with the cluster-level values. Details on how to construct these nutrition indicators can be found here: https:// us.htm.

If you are interested in exploring the relationship between covariates and the nutrition indicators, the geospatial covariates dataset (ETGC72FL) provides a range of environmental, socioeconomic, and population variables. At The DHS Program, we prepare these geospatial covariates, extract values at the cluster locations, and make these datasets available for users. We have a report introducing users to the geospatial covariates prepared by The DHS Program that can be found here: We also have an accompanying manual describing the available covariates, which can be found here: https:// _Data_Description_2.pdf.
Re: Geospatial analysis [message #27240 is a reply to message #25561] Wed, 05 July 2023 13:13 Go to previous messageGo to next message
Hajra Adil Nawaz is currently offline  Hajra Adil Nawaz
Messages: 2
Registered: July 2023
respected sir, please help me regarding an issue. i want to assess stunting at a district level in pakistan . how may i make use of latitude, longitude to locate required clusters then identify and filter them using stata. what are going to be demand syntax to filter out those children in respective clusters identified. thankyou.
Re: Geospatial analysis [message #27291 is a reply to message #27240] Thu, 13 July 2023 12:42 Go to previous message
Janet-DHS is currently offline  Janet-DHS
Messages: 773
Registered: April 2022
Senior Member
Following is a response from DHS staff member, Tom Pullum:

There have been recent postings giving the Stata code for district-level estimates. Please do a search using keywords. For this purpose, using the 2017-18 survey, you would use the PR file and shdist (the variable for districts). (There is a variable for district in the previous Pakistan survey but NIPS has not provided the labels for the district codes.) You do not need to use the cluster identifier (hv001) or the GIS codes. Using hv001 and shdist you can tell which clusters are in each district. There are not enough children within clusters for statistically stable cluster-level estimates of stunting.
Previous Topic: merging HR and KR files and weighitng for analysis
Next Topic: Child Health and Nutrition
Goto Forum:

Current Time: Mon Jul 22 20:18:29 Coordinated Universal Time 2024