| Place of Delivery [message #23599] |
Tue, 19 October 2021 05:42  |
Niranjana
Messages: 13
Registered: October 2019
|
Member |
|
|
Hello,
I am using the India IR file to look at the role of certain variables on place of delivery (m15_1 - m15_6) using STATA. As I understand, place of delivery question is asked for each birth in the last five years to each woman who has given birth in the last 5 years. With the outcome variable being place of delivery, I want to know how to recode m15_1 - m15_6 to create a single variable for the mother. I have created a count variable as follows:
label list `: value label m15_1'
forval j = 1/4 {
gen placedel`j' = 0
}
forval i = 1/6 {
replace placedel1 = placedel1 + inrange(m15_`i', 20, 27)
replace placedel2 = placedel2 + inrange(m15_`i', 30, 32)
replace placedel3 = placedel3 + inrange(m15_`i', 10, 13)
replace placedel4 = placedel4 + inlist(m15_`i', 33, 96)
}
This creates a count variable which gives the number of times a respondent has given birth in a specific place. Counting instances opens the door to summary variables of the form never, (count 0), ever (count 1 to 6) and always (count 6). I was hoping if this captures all the information on place of delivery by respondent. If there is another way it could be recoded particularly to reflect the BR file, do let me know. My numbers however, do not match with Table 8.13 in https://dhsprogram.com/pubs/pdf/FR339/FR339.pdf. Thank you.
|
|
|
|
| Re: Place of Delivery [message #23605 is a reply to message #23599] |
Tue, 19 October 2021 12:43  |
Bridgette-DHS
Messages: 2801
Registered: February 2013
|
Senior Member |
|
|
Following is a response from DHS Research & Data Analysis Director, Tom Pullum:
In Table 8.13 the units of analysis are births, rather than women. You need to shift to the KR file and m15 (no subscripts). The characteristics of the mother are already on the child's record in the KR file.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
