The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » MULTILEVEL MODELING USING DHS SURVEYS: APPROXIMATE LEVEL-WEIGHTS for Kenya DHS Surveys
MULTILEVEL MODELING USING DHS SURVEYS: APPROXIMATE LEVEL-WEIGHTS for Kenya DHS Surveys [message #29116] Thu, 25 April 2024 09:10 Go to next message
geoK is currently offline  geoK
Messages: 44
Registered: May 2014
Member
Dear DHS Team,

I have been reviewing the DHS Methodological Reports 27 in order to create wt1 and wt2 weights for multilevel modeling using the latest two rounds of the Kenya DHS. I followed the example provided in the report, which used the Zimbabwe DHS, but I am now encountering difficulty identifying which tables and columns correspond exactly to Table A.2 and Table A.3. These are the tables where I need to extract figures for Stage 1 of the process. While it is clear for Zimbabwe, I am uncertain if I am selecting the correct tables and columns for the 2014 and 2022 KDHS Appendix A tables. Could you please confirm where I can locate the figures for Stage 1? Thank you!

[Updated on: Thu, 25 April 2024 09:12]

Report message to a moderator

Re: MULTILEVEL MODELING USING DHS SURVEYS: APPROXIMATE LEVEL-WEIGHTS for Kenya DHS Surveys [message #29124 is a reply to message #29116] Fri, 26 April 2024 07:02 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3049
Registered: February 2013
Senior Member

Following is a response from Senior DHS Staff member, Tom Pullum:

Unfortunately, we cannot provide much assistance with this topic. I recommend that you first identify the specific appendix tables in the Zimbabwe final report that were used in MR27. The Kenya 2022 survey was a DHS-8 survey and followed a slightly different tabulation plan, but I hope you can find the appendix tables in the Kenya report that correspond with the tables in the Zimbabwe report. I haven't checked, but the table titles should be very similar even if the table numbers have changed.

Ideally, DHS would routinely calculate the level weights when each survey is completed, and make them available in a small cluster-level file. We have recommended this, but without success. If other users have calculated the level-weights for other surveys, perhaps they could let us know and share them on the forum?
Previous Topic: Reclassifying admin areas
Next Topic: HR Files
Goto Forum:
  


Current Time: Sun May 5 12:25:41 Coordinated Universal Time 2024