The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Ethiopia » Calculating Age structure explanatory variables on Household Wealth
Re: Calculating Age structure explanatory variables on Household Wealth [message #11209 is a reply to message #11205] Thu, 17 November 2016 20:34 Go to previous messageGo to previous message
xrl1g11 is currently offline  xrl1g11
Messages: 4
Registered: November 2016
Location: England
Member
Dear Tom,

Thank you very much! Everything worked out fine.

I have one last question, I would like to include a variable in my regression analysis that is a measure of household health or household member health from the PR file. I created child stunting as a health variable (from a coding that was listed on this forum), which worked, but it has reduced the sample size from 77,727 to only 10,000 (as it reflects only children under 5). Also, at the household level, the variable cannot be created as the head of household cannot be a child under 5. One variable from the PR file that I feel may suffice, would be the BMI variable (I believe HA40 is BMI for women and HB40 for men).

Is it possible to create a separate BMI variable (one for male, one for female, and possibly one for child) which is categorized into under 18.5 BMI (indicator of chronic malnutrition) and normal BMI range? If so, what would be the coding for it?

I'm looking for a good general indicator for health and aim to include it in the regression analysis (along with the other variables which have all been successfully created); I would like to examine how household health or household member's health is related to their wealth level. I'd imagine if BMI was the variable, the odds of those with a BMI less than 18.5 are more likely to be from poorer households as opposed to richer households (something along those lines).

If I could get help with creating this variable or another potential health indicator that is commonly used by the DHS, that would be great!

P.S. I have also merged IR into PR to obtain some health variables from the woman's recode but ideally i'd like to just work with the PR file.

Any guidance would be greatly appreciated! Thanks for your time.

Regards,
Xavier


[Updated on: Thu, 17 November 2016 20:39]

Report message to a moderator

 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: 2016 dataset
Next Topic: Breastfeeding Initiation in Ethiopian population
Goto Forum:
  


Current Time: Wed Feb 5 20:49:05 Coordinated Universal Time 2025