The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Weighting data » Weight about pooling 2003,2008,2013,2018 BR dataset in Nigeria (How to weight pooling dataset)
Re: Weight about pooling 2003,2008,2013,2018 BR dataset in Nigeria [message #19350 is a reply to message #18997] Wed, 03 June 2020 16:52 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3081
Registered: February 2013
Senior Member

Following is another response from DHS Research & Data Analysis Director, Tom Pullum:

The BR file has one record for every live birth. The mother is identified with caseid and the child with bidx. The unique ID for each record is caseid plus bidx. I do not see how or why you plan to convert the BR file to a file of women. If you want a file of women, I would recommend that you use the IR file. "Mothers" are women who have had children, i.e. who have v201>0.

There have been endless postings on pooling successive rounds from the same country. This makes sense and can be helpful if you want to describe changes from one round to another. For this purpose you do not need to change the weights at all.

If your goal is to synthesize the surveys, in the sense of calculating some kind of rate that ignores which round the case comes from, I would recommend against doing that. Analyzing successive cross-sections would be better than pooling the cross-sections. If you take the latter approach then you have to take account of the fact that the successive rounds have different sample sizes and possibly take account of the fact that the population size was changing. If you look through earlier postings you will find some relevant procedures.

 
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Sample weights and stratification - Nigeria 2008 and 2018
Next Topic: Weighted result's total gives different total value with unweighted one.
Goto Forum:
  


Current Time: Thu Jun 13 20:19:25 Coordinated Universal Time 2024