The DHS Program User Forum: Weighting data » Weights used in the Pakistan 2017-18 Survey

Home » Data » Weighting data » Weights used in the Pakistan 2017-18 Survey

Show: Today's Messages :: Show Polls :: Message Navigator

Weights used in the Pakistan 2017-18 Survey [message #22865]

Wed, 26 May 2021 06:17

Saima
Messages: 10
Registered: October 2015
Location: Bowling Green, Ohio

Member

While going through PDHS data files for couples, I found this information on weighting.
"Tables in Pakistan were not produced for the full country (8 regions), and some regions (AJK and GB) are not included in the national totals. The standard weights are used to generate tabulations for 6 regions (HV005/HV028/V005/D005/MV005).
To generate any tabulations for the other 2 regions (GB or AJK with codes 5 and 7 respectively), use the CS weights ( SHV005/SHV028/SV005/SD005/SMV005).
The CS weights are 0 for regions different of 5 and 7. The standard weights are 0 for regions 5 and 7."

As for my analysis if I want to keep region 5 & 7, how can i weight my whole data set. For instance, I am using MV005 variable to weight my data but the values for this variable are zero for region 5 & 7. In this case, do I need to drop these two regions from my analysis or is there any way to weight the data overall?

Thanks,

Saima

Report message to a moderator

Re: Weights used in the Pakistan 2017-18 Survey [message #22866 is a reply to message #22865]

Wed, 26 May 2021 11:58

Bridgette-DHS
Messages: 3210
Registered: February 2013

Senior Member

Following is a response from DHS Lead Statistician, Ruilin Ren:

There is no way to estimate indicators for all Pakistan including provinces 5 and 7. This was the decision of the Pakistan government. Any attempt to produce indicators for all the 8 provinces together is discouraged.

Report message to a moderator

Re: Weights used in the Pakistan 2017-18 Survey [message #22869 is a reply to message #22866]

Thu, 27 May 2021 02:16

Saima
Messages: 10
Registered: October 2015
Location: Bowling Green, Ohio

Member

Thank you Sir for your response.

Saima

Saima

Report message to a moderator

Previous Topic:	Weights not normalized?
Next Topic:	Weighting for same cohort across multiple surveys

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Fri Dec 13 15:39:25 Coordinated Universal Time 2024