Guidance on Using Weights for North-East India Analysis in NFHS-5 [message #29631] |
Fri, 12 July 2024 04:44 |
Saurav
Messages: 1 Registered: July 2024 Location: Shillong, India
|
Member |
|
|
Dear DHS Forum,
I am currently working on an analysis using the NFHS-5 dataset, specifically focusing on the North-East India region. I have the individual women's file, but it does not include regional-specific weights for North-East India. I would like to ensure that my analysis is representative of this specific region.
Could you please provide guidance on the following:
1. Are regional-specific weights available for the North-East India region in the NFHS-5 dataset, and if so, where can I find them?
2. If regional-specific weights are not available, is it recommended to use the national weights for regional-level analyses? Are there any adjustments or considerations I should keep in mind while doing this?
3. Are there any best practices or additional resources for conducting sub-national analyses using NFHS-5 data?
Thank you!
|
|
|
Re: Guidance on Using Weights for North-East India Analysis in NFHS-5 [message #29642 is a reply to message #29631] |
Fri, 12 July 2024 08:57 |
Bridgette-DHS
Messages: 3190 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
The state weights would be used for the analysis of a single state, but for a group of states, such as North-East India, you would use the national weights.
I suggest the following, although it is really not necessary. You could re-normalize the national weight, which has a mean of 1 million at the national level, so that it has a weight of 1 million for the region. You would do that by opening the IR file, reducing it to the North-East, and then entering these two lines (in Stata):
summarize v005
gen v005_region=1000000*v005/r(mean)
Then the mean weight of v005_region will be 1000000.
|
|
|