The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Child Health » Pooled Analysis (Question regarding pooled analysis )
Pooled Analysis [message #20268] Sat, 17 October 2020 08:11 Go to next message
abpromiti is currently offline  abpromiti
Messages: 20
Registered: December 2017
Location: Dhaka
Member
Hello

I want to conduct pooled analysis using data from Bangladesh standard DHS 2014, India Standard DHS 2015-16, Pakistan Standard DHS 2017-18.
All will be KR file.
I am unable to understand how these datasets can be merged.
Can anyone give me any outline on this issue?
Re: Pooled Analysis [message #20276 is a reply to message #20268] Mon, 19 October 2020 09:33 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2025
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

You want to append the files, not merge. If you pool the surveys the results will be dominated by the India survey, because that survey was so large. If you re-scale the weights (v005) so they will be proportional to population size, the results will be dominated by India because its population is so large. There have been many posts on this sort of thing. DHS staff cannot provide further assistance but I hope other users can help.
Previous Topic: Help with number generation
Next Topic: SPSS to STATA commands
Goto Forum:
  


Current Time: Fri Nov 27 17:57:10 Coordinated Universal Time 2020