The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Geographic Data » Correct approach for regional analysis (representativeness)
Correct approach for regional analysis (representativeness) [message #17531] Mon, 08 April 2019 11:49
SusanneK is currently offline  SusanneK
Messages: 1
Registered: April 2019

As I am not experienced in analysing survey data, I have a question regarding the best practice for sub-national data analysis. According to what I have investigated so far, the DHS surveys are representative at a national and sometimes regional level.

Would it be (statistically correct) possible to analysis different surveys of the same country (maybe also including non-DHS surveys measuring the same variable) on a sub-national spatial level (e.g. district)?

Example: The aim is to compare, for example, child health at the district (or livelihood zone) level using cross-sectional or/and longitudinal (non) DHS data from different rounds.
What would be necessary steps to ensure the representativeness of the data? Which approaches could be used for this analysis?

It would be great if someone could help me out in this matter.

Kind regards, Susanne

[Updated on: Mon, 08 April 2019 11:53]

Report message to a moderator

Previous Topic: Linking survey data to GPS data
Goto Forum:

Current Time: Tue Apr 23 02:46:14 Eastern Daylight Time 2019