Merging children to their mothers [message #13991] |
Thu, 01 February 2018 05:36 |
tschoudhary
Messages: 7 Registered: January 2018
|
Member |
|
|
I am trying to analyse if maternal height is associated with stunting among under five children in the NFHS 4 data set (india using stata). The PR files given me the exact numbers of weighted cases to calculate stunting as per the national report. But it doesn't contain data on the height of children's mothers. this is available in the IR file but i am not able to merge the data for child's haz scores (hc70) with the height of their mothers.
i have tried the following
for the PR file
gen v001 = hv001
gen v002 = hv002
gen v003 = hvidx
sort v001 v002 v003
save as "temp.dta"
use the ir file
sort v001 v002 v003
merge v001 v002 v003 using "temp.dta"
i get additional observations but maternal height isn't linked to the child haz scores (hc70)
tab hc70 if ha3 !=. shows 0 observations.
kindly guide.
|
|
|
Re: Merging children to their mothers [message #13995 is a reply to message #13991] |
Thu, 01 February 2018 10:58 |
Bridgette-DHS
Messages: 3208 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
You do not need to merge any files to link the child's and mother's anthropometric information. Go to the KR file (IAKR71FL.dta). In this file, the units are children under five and most of the mother's information is attached to the child's record. The child's z scores are hw70, hw71, and hw72. They are exactly the same as hc70, hc71, and hc72, respectively, in the PR file. The mother's anthropometric variables are v437 to v447. For the mother, we would usually select BMI (v445) as the best indicator.
|
|
|
|