Indonesia neonatal mortality 2002/2007 [message #3828] |
Fri, 20 February 2015 02:20 |
adev
Messages: 14 Registered: May 2013
|
Member |
|
|
Hello -
I am using the following stata code to calculate the number of neonatal deaths:
gen neodeaths= 0
forvalues i = 1/20 {
replace neodeaths = neodeaths+1 if b3_`i' != . & b3_`i' >= v008-59 & b5_`i' == 0 & b7_`i' == 0
}
The code produces the same number as published neonatal deaths for IDHS2007 (308 deaths) but not for IDHS2002 (My 296 deaths vs. published 290 deaths). However, using 296 produces the published NMR (20/1000) and using 290 does not (19/1000). Is it ok to assume that my 2002 calculations are correct? Not sure why this discrepancy would exit for one survey but not another.
Thank you,
Alka
Alka Dev
Graduate Fellow
CUNY Institute for Demographic Research
|
|
|