| replicating child health data--does under 5 ever include children older than 60 months? [message #12368] |
Fri, 05 May 2017 16:49  |
AnnM
Messages: 3
Registered: March 2013
Location: Boston
|
Member |
|
|
I'm working with a dataset from Madagascar, and am trying to replicate a result (vaccinations). The only way I can replicate it is to take as the denominator (base data set) for under 5s as follows (stata commands):
generate lastbirth5year=1 if cdoi - cdob <60 & col_sec3==1 (in other words, if the child is <60 months of age and the last born child). And then using the command :
replace lastbirth5year=1 if lastbirth5year==. & qw305 ~=.. Where qw305 is the question of whether or not someone had a prenatal consult with the pregnancy.
When I put these together, I get the same numbers that are in the report. Is this something standard in DHS? Maybe to include the children whose ages/birth months might be unknown? Or is it more usually the case that you just rely on the CMC age being less than 60 months, and this represents an error in the report? Because some of these children who have been included in the calculations have "cage"s up to 65 months. Thus my confusion. Many thanks!
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
