The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Nutrition and Anthropometry » Linking household food security, maternal and child nutritional status in Nepal DHS 2016
Linking household food security, maternal and child nutritional status in Nepal DHS 2016 [message #14408] Wed, 04 April 2018 05:37 Go to next message
cmaitr
Messages: 1
Registered: March 2018
Member
I'm a new user of DHS. I'm using Nepal DHS 2016.
I am having a problem with estimation in stata which needs to be addressed quite urgently.
I would like to estimate the impact of household food security status on child nutritional status (stunting, wasting, birth weight) controlling for maternal characteristics (such as maternal education, age and bmi)and household characteristics such as wealth, access to toilet, drinking water, electricity etc.
Which files should I link? I'm using stata15.

Is it possible to get all information from PR file? Or should link HR, KR and IR ?

When I merged IR and PR (PR as master , using IR) and ran a regression on the above variables (tried stunting and hc70 both)I'm getting a 'no observation' message in stata?

I tried the same regression for stunting using PR file alone, but I'm getting the same message . If I drop the women level variables (such as ha40 in PR file ) and keep the household level variables only, regression is going fine. I must be missing something basic but can not identify what.


Would someone please advise.

[Updated on: Wed, 04 April 2018 05:41]

Report message to a moderator

Re: Linking household food security, maternal and child nutritional status in Nepal DHS 2016 [message #14452 is a reply to message #14408] Fri, 06 April 2018 18:54 Go to previous message
Reduced-For(u)m
Messages: 292
Registered: March 2013
Senior Member

I think the easiest way is to use the children's recode (KR). It will have one row (observation) for each child, and include all their height/weight info. You should not have to do any merging at all, unless you need something really weird from one of the other files.
Previous Topic: Standard deviations for Stunting + Co
Next Topic: how to recode scores
Goto Forum:
  


Current Time: Mon May 20 21:48:38 Coordinated Universal Time 2024