|
Re: regions AJK and GB in PDHS (Pakistan DHS) [message #24850 is a reply to message #24819] |
Wed, 20 July 2022 16:36  |
Janet-DHS
Messages: 495 Registered: April 2022
|
Senior Member |
|
|
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:
Those two regions can be analyzed separately (an analysis for AJK and a separate analysis for GB) but you cannot pool them and you cannot pool them with the rest of Pakistan. In the IR file, for example, the usual sample weight, v005, is 0 for those regions. The correct weight variable for these two regions is sv005, and it is normalized to have a mean of 1 (or 1000000) in each of them. To repeat, you cannot pool these regions with each other or with the rest of the country. The government of Pakistan has been very clear about not wanting users to do that.
|
|
|