The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » M_h data for Myanmar, M_h & M data for Pakistan
M_h data for Myanmar, M_h & M data for Pakistan [message #24417] Thu, 12 May 2022 04:35 Go to next message
w@dhs
Messages: 5
Registered: December 2019
Member
Hello
May I ask a couple of questions? I'm trying to use weighting as guided in the MR-27 using Myanmar and Pakistan data. However, I do not have some Infos in their reports.

1. For Myanmar, I do not see the average number of households per cluster (M_h) by strata. (For the total number of census clusters (A_h) is found in Table A3).
2. For Pakistan, Table A1. gives me A_h, however, the average number of households per cluster (M_h) was not found in the report. Which figures should I take for M_h?
3. Also, for Pakistan, the total number of households in the country was not found. Where should I get it?

Thank you.

[Updated on: Thu, 12 May 2022 04:41]

Report message to a moderator

Re: M_h data for Myanmar, M_h & M data for Pakistan [message #24436 is a reply to message #24417] Mon, 16 May 2022 14:34 Go to previous message
Janet-DHS is currently offline  Janet-DHS
Messages: 199
Registered: April 2022
Senior Member
Following is a response from DHS Senior Sampling Specialist, Mahmoud Elkasabi and Ruilin Ren:

We do not have the required information for these two surveys because we never had access to the entire sampling frame. For the Myanmar survey, the sample was selected from a master sample. DHS was only given the master sample allocation, which is given in table A.3 in the final report. For the Pakistan 2017-18 survey, DHS was not given the number of households in the selected clusters in the sample file.

Our guidelines for estimating multi-level weights do not extend to this kind of situation with such incomplete information about the sampling frame. You may be able to find some other approach, but unfortunately we cannot help.
Previous Topic: Adjusting weights on binary variables
Next Topic: create a weighted population variable
Goto Forum:
  


Current Time: Thu Sep 29 00:52:37 Coordinated Universal Time 2022