The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Strata in surveys where v022-v025 does not seem to work
Strata in surveys where v022-v025 does not seem to work [message #30256] Wed, 23 October 2024 06:40 Go to next message
ewa.b is currently offline  ewa.b
Messages: 37
Registered: January 2014
Member
I am conducting a cross-country study and identifying the correct strata across many surveys. There are 4 surveys where I was not able to identify strata by looking in detail at (i) v022, (ii) v023, (iii) the combination of v024 and v025 (i.e., egen strata=group(v024 v025)), as well as the final report. Could you help me identifying strata in these surveys?:

1) South Africa (1998): report states there are 26 strata, but none of the (i), (ii), (iii) gives 26. Could you indicate how to identify the "health regions" that should be accounted for in the strata creation according to the report?

2) Cameroon (1991): report indicates 13 strata but none of the options (i)-(iii) give that.

3, 4) Senegal (1986) and Uganda (2000): neither v022 nor v023 indicates strata; in both surveys, combination of v024 and v025 gives 8 strata. Is that correct? (I was not able to identify information on this in the respective reports)

Thank you.
Re: Strata in surveys where v022-v025 does not seem to work [message #30257 is a reply to message #30256] Wed, 23 October 2024 07:17 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3199
Registered: February 2013
Senior Member
Following is a response from Senior DHS staff member, Tom Pullum:

The strata for surveys up to about 2020 are specified in the file "Survey_strata.do" (see attached). For more recent surveys the strata are given by v022 or v023 (they are the same). In most surveys, but not all, the strata are v024 x v025.

There have been other posts on this topic. This file has been posted several times, once very recently. Users are asked to search the forum before asking a question. There's always a good chance that your question has come up before.

Previous Topic: DHS weights and random effects in a GLMM
Next Topic: Weighting and pooling multicountry datasets
Goto Forum:
  


Current Time: Sat Nov 30 15:23:48 Coordinated Universal Time 2024