child mortality ,mother autonomy and husband/partner related variables [message #25251] |
Fri, 23 September 2022 02:05 |
titas
Messages: 5 Registered: September 2022
|
Member |
|
|
I want to see the decision making with respect to healthcare, household purchases[v743 variables] and whether beating justified variables by husband[v744 variables] of mothers of children born in last 5 years of survey and link that with child mortality variables[for India NFHS4 and NFHS5] .I am using the children recode file of India NFHS4 and NFHS5 but with respect to full sample size ,there are a lot of missing values[same thing for husband/partner's age, education, occupation].When I want to link the mother's autonomy and husband/partner's related variables for mothers with children born in the last five years to child mortality for children born in last five years do I need to merge children file with any other files to get increased sample size or should I stick to the children file only??
|
|
|
Re: child mortality ,mother autonomy and husband/partner related variables [message #25294 is a reply to message #25251] |
Wed, 28 September 2022 14:52 |
Janet-DHS
Messages: 852 Registered: April 2022
|
Senior Member |
|
|
Following is a response from DHS staff member Tom Pullum:
Many of the variables you are working with are only included in a subsample. The men's and DV data were only collected for 1/6 of households. The subsample was constructed as alternate households (every second household) in 1/3 of the clusters. However, because the India surveys are so large, even 1/6 will give you enough cases. I suggest that you just reduce the data files to the cases (women, children, and men) in the selected households and then do your merges. Note that v024 (or mv024 or hv024) must be included in the full id for the cases. These merges have been described in other postings.
|
|
|