| Re: Discrepancies in Observations (Children's File 2014 Ghana DHS) [message #9990 is a reply to message #9932] |
Tue, 14 June 2016 10:29   |
Bridgette-DHS
Messages: 3230
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Tom Pullum:
The KR file (and the BR file) has a record for each child born in the five years before the interview, including children who have died. If you do "tab b5,m" you will see that of the 5884 children in the KR file, 289 died before the survey. If you do "tab b8,m" you will see that age in years is given for all 5595 children who are still alive; it is not given, i.e. is missing, for the 289 children who have died. I think that will explain the discrepancies you are finding. Those children are not missing, but current age is not defined for them.
Note also that all numbers in DHS reports, unless stated explicitly to the contrary, are weighted. If you do "tab b5 [iweight=v005/1000000]" you will see that the 5884 UNWEIGHTED children correspond with 5695 WEIGHTED children.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
