The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » using wieght, cluster, strata
using wieght, cluster, strata [message #26224] Mon, 20 February 2023 03:41 Go to next message
Alanood is currently offline  Alanood
Messages: 18
Registered: November 2022
Member
I am a phd stundent, currently I am doing my 1st paper and I am using DHS dataset for Egypt (2008) and (2014).
The title of my paper is The impact of maternal employment in child nutrition by looking at HAZ, WAZ, and WHZ.

I found one of the litertaure that used the DHS dataset, Rashad et al (2019),
"To avoid any bias in the coefficients from the over-sampled population and adjust the standard
errors for the survey clustering effect (Deaton, 1997), all the regression analyses have corrected for
the survey design, i.e. the sampling weight, the cluster, and the strata were all taken into account."




The authors tested the same variables that I am interested in, and I am wondering why? and if I should do the same for my dataset?
if yes, please provide me with the STATA comand.



Thank you





[Updated on: Mon, 20 February 2023 05:33]

Report message to a moderator

Re: using wieght, cluster, strata [message #26238 is a reply to message #26224] Tue, 21 February 2023 08:27 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member

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

I assume you are using the KR files. These files contain the outcomes for children whose mothers are included in the survey, and virtually all of the mother's information is on the same records. For the Egypt 2014 survey, this file is EGKR61FL.dta. For the 2008 survey, replace "61" with "5A".

All regressions and other estimation commands should include svyset and svy, not just those with anthropometry variables. There have been many posting on the forum about svyset and svy. You can also read about them if you enter "help svyset" in Stata.

The svyset command for your data would be "svyset v021 [pweight=v005], strata(v023) singleunit(centered)". For most surveys, v001 and v021 are exactly the same, but for the Egypt surveys they are different and you should use v021, which is the PSU ID code. The sampling weights are v005 and the strata are v023. Then the estimation commands will begin with "svy: ".

There have been many DHS reports related to anthropometry. You can find them under the Publications tab and "Analytical" on the website. Also you can search for related posts on the forum.

Good luck!
Previous Topic: Melogit and Weights
Next Topic: Weighting at regional level
Goto Forum:
  


Current Time: Thu Mar 28 07:38:17 Coordinated Universal Time 2024