The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Nutrition and Anthropometry » Honduras 2012 data child <5
Honduras 2012 data child <5 [message #14240] Fri, 09 March 2018 10:03
sc.sanchez is currently offline  sc.sanchez
Messages: 5
Registered: January 2017
Dear experts

We are using Honduras survey 2012 in order to calculate the National Child Poverty Index. We are using PR data file and we have IR, MR and KR to have all possible information of every household member. We have a problem with information of children under 5 years-old, becouse we found a lot of missing values, and appear registered in his mother, not in a child. We think that there is a problem whit the merge, becouse where we merge KR to PR the distribution of variable changes. Please, help us to decorate where is the mistake.

We using this coding in stata:

use HNHR62FL, replace
save hogar, replace

use HNPR62FL, replace
tab hvidx, m
clonevar v003=hvidx
egen id_hogar = concat(hv001 hv002)
destring id_hogar, replace
save personas, replace

use HNIR62FL, replace
clonevar hvidx=v003
rename v001 hv001
rename v002 hv002
save mujeres, replace

use HNMR62FL, replace
clonevar hvidx=mv003
rename mv001 hv001
rename mv002 hv002
save hombres, replace

use HNKR62FL, replace
tab b16, m
drop if b16==. // delate dead childs
drop if b16==0 // delate childs that not live in household
clonevar hvidx=b16
rename v001 hv001
rename v002 hv002
save niños, replace

use personas, replace
merge 1:1 hv001 hv002 hvidx using mujeres, generate(mujeres)
tab mujeres
merge 1:1 hv001 hv002 hvidx using hombres, generate(hombres)
tab hombres
merge 1:1 hv001 hv002 hvidx using niños, generate(child)
tab child
merge m:1 hv001 hv002 using hogar, generate(household)
tab household
Read Message
Previous Topic: anthropocentric data of egypt men
Next Topic: Maternal anemia in Egypt
Goto Forum:

Current Time: Mon Jun 17 07:32:00 Coordinated Universal Time 2024