| Double burden of malnutrition [message #18159] |
Tue, 01 October 2019 09:04  |
rwlchoi
Messages: 2
Registered: September 2019
|
Member |
|
|
Hi,
I'm researching double burden of malnutrition (child stunting and maternal obesity at the same time in the same household.) in Pakistan using PDHS 2017-18.
The question is that how to make a pair of stunting child and overweight mother in the same household. I figured out that I have to use v001 v002 v003 to identify the household number, but I still don't know how to merge children and mother in the same household.
Also, if one mother gives birth to several children, how should I handle it for the research?
As mentioned above, I'm using STATA 14 SE.
Thank you in advance.
|
|
|
|
| Re: Double burden of malnutrition [message #18175 is a reply to message #18159] |
Fri, 04 October 2019 11:56  |
Liz-DHS
Messages: 1516
Registered: February 2013
|
Senior Member |
|
|
Dear User, a response from Dr. Tom Pullum:
"Except for a very small number of women who were in the household survey and were eligible for the survey of women (hv117=1) but were not interviewed, all pairs of mothers and children in the same household can be identified in the IR and KR files. There are many ways to construct the file you want, for example by merging the IR and KR files. However, the easiest way is as follows, using ONLY the IR file and "reshape long":
use v0* v1* v4* hw70* hw71* hw72* using "... PKIR71FL.DTA" , clear
reshape long hw70_ hw71_ hw72_, i(v001 v002 v003) j(hidx)
rename hw7*_ hw7*
drop if hw70==.
You have to modify the path to the data file. You can add more variables for the mother or for the child.
The reshaped file gives you one record per child (age 0-4), with the mother's data (including her anthropometry variables, such as v437 and v444) attached to the child's data. You can then construct the stunting (etc.) variables for the mother and the child and then construct a variable to describe the joint status of the mother and the child."
To get help with STATA click here:
[Updated on: Mon, 07 October 2019 10:04] Report message to a moderator |
|
|
|
| Re: Double burden of malnutrition [message #18194 is a reply to message #18175] |
Wed, 09 October 2019 16:32 |
Liz-DHS
Messages: 1516
Registered: February 2013
|
Senior Member |
|
|
Dear Wonil, a more detailed response from Dr. Tom Pullum based on some input from you:
Quote:Dear Wonil,
Your "use" line has dropped the hw70-hw72 variables. You need to put them back in. You don't need hidx* and hidxa*. That is, change the "use" line to this:
use v0* v1* v4* seduc v714 h32l* h32a* m18* b4* hw1* h11* m39a* m39* h80a* h80b* h80c* h15j* h15e* hw70* hw71* hw72* using "E:\졸업 논문\교수님 2차 (by 19.07.15)\파키스탄 데이터\PK_2017-18_DHS_07252019_2215_133252\PKIR71DT\PKIR7 1FL.DTA " , clear
Your "use" line includes "hw1*. That brings in hw1 and hw10-hw19 (but there is no hw14). In the "reshape" line you will need to list these separately. "m39*" brings in both m39* and m39a*, so you can drop "m39a*".
The "b" variables, including "b4", come from the birth history and have indices 01 through 20. The variables for children under 5 have indices 1-6. You need to add the following lines to change b4:
gen b4_1=b4_01
gen b4_2=b4_02
gen b4_3=b4_03
gen b4_4=b4_04
gen b4_5=b4_05
gen b4_6=b4_06
drop b4_0* b4_10-b4_20
reshape long h32l_ h32a_ m18_ b4_ hw1_ h10_ h11_ h12_ h13_ h15_ h16_ h17_ h18_ h19_ m39a_ m39_ h80a_ h80b_ h80c_ h15j_ h15e_ hw70_ hw71_ hw72_, i(v001 v002 v003) j(hidx)
Therefore. the following lines will work (after you have changed the path):
use v0* v1* v4* h32l* h32a* m18* b4* hw1* m39* h80a* h80b* h80c* h15j* h15e* hw70* hw71* hw72* using ...PKIR71FL.DTA", clear
gen b4_1=b4_01
gen b4_2=b4_02
gen b4_3=b4_03
gen b4_4=b4_04
gen b4_5=b4_05
gen b4_6=b4_06
drop b4_01-b4_20
reshape long h32l_ h32a_ m18_ b4_ hw1_ hw10_ hw11_ hw12_ hw13_ hw15_ hw16_ hw17_ hw18_ hw19_ m39a_ m39_ h80a_ h80b_ h80c_ h15j_ h15e_ hw70_ hw71_ hw72_, i(v001 v002 v003) j(hidx)
Good luck--Tom
|
|
|
|
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
