The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Strata and PSUs in Egypt KR 2014 for sampling weight
Strata and PSUs in Egypt KR 2014 for sampling weight [message #22387] Fri, 05 March 2021 06:03 Go to next message
mkoshida is currently offline  mkoshida
Messages: 2
Registered: February 2021
Member
According to the final report of Egypt DHS 2014 (AppendixB), 51 sampling strata and 884 PSUs were used in sampling process.

(From Final Report
"Prior to the first stage of selection, each of the 27 governorates was first stratified into urban and rural areas, yielding a total of 51 sampling strata since Cairo, Suez and Port Said do not have rural areas"
The first sampling stage involved the selection of 926 shiakhas/villages, as the Primary
Sampling Units (PSUs)" "Because of security issues,.... the 42 shiakhas/villages selected in those governorates for the 2014 EDHS sample were not included in the survey. This reduced the total number of sampling PSUs in the 2014 EDHS from 926 to 884 PSUs. ")

However, when I tried to weight data, number of strata and PSUs shown in STATA do not match with the number reported in the final report.
my code and results are following:

svyset[pw=wgt], psu(v021) strata (v023)

pweight: wgt
VCE: linearized
Single unit: missing
Strata 1: v023
SU 1: v021
FPC 1: <zero>

svy: tab v106
(running tabulate on estimation sample)

Number of strata = 47 Number of obs = 15,848
Number of PSUs = 876 Population size = 15,667.944
Design df = 829
------------

I tried with v022 and (v024×v025) for Strata, neither did match with the final report.
If anyone could help me figure out with this, I would sincerely appreciate it.

Thank you in advance.

Re: Strata and PSUs in Egypt KR 2014 for sampling weight [message #22404 is a reply to message #22387] Mon, 08 March 2021 10:06 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3035
Registered: February 2013
Senior Member

Following is a response from DHS Senior Sampling Specialist, Mahmoud Elkasabi:

When I run exactly the same Stata code on the IR file, I get the correct answer. You may have filtered out some cases. Please go back to the original IR file.
Previous Topic: Weight in Indonesia DHS 2012 for un insurance coverage (None)
Next Topic: Appending files
Goto Forum:
  


Current Time: Fri Apr 19 17:09:16 Coordinated Universal Time 2024