Twins and mortality [message #14455] |
Sat, 07 April 2018 08:12 |
behappy.ina77@gmail.com
Messages: 3 Registered: April 2018 Location: Indonesia
|
Member |
|
|
Hi there,
I want to investigate between twin (multiple births) with mortality (neonatal, post neonatal and infant). Can you explain how to get those variable? I use the data with code IDKR63FL.SAV. Can I recode those variables directly from B0 = child is twin? How can we know that the twin died one or both of them?
Need your respond and thank you before :)
Regards,
Bintang
|
|
|
Re: Twins and mortality [message #14470 is a reply to message #14455] |
Tue, 10 April 2018 14:30 |
Liz-DHS
Messages: 1516 Registered: February 2013
|
Senior Member |
|
|
A response from Dr. Tom Pullum:
Quote:
You need to put both children in the pair onto a single record. I will paste below the lines to do this in Stata. I do not use SPSS. Perhaps you can figure out the Stata lines or find someone to translate them.
I use the BR file rather than the KR file, just because there is no need to restrict to births in the past five years. The procedure involves scratch files (intermediate work files that don't need to be saved). You will have to change the paths.
Quote:
set more off
cd e:\DHS\DHS_data\scratch
use "C:\Users\26216\ICF\Analysis - Shared Resources\Data\DHSdata\IDBR63FL.DTA"
* Construct a data file that has the b variables for both children of a twin birth on one record
* Must allow for the possibility that a mother may have more than one set of twins
numlabel,add
tab b0
keep if b0==1 | b0==2
keep v0* b*
drop b9-b16
rename b* b*_
egen sequence=seq(),by(v001 v002 v003)
reshape wide b*_, i(v001 v002 v003) j(sequence)
tab1 b0*
* 11 women had two pairs of twins; must move them down
save ID63_twins1234.dta, replace
keep v* b*3 b*4
rename b*3 b*1
rename b*4 b*2
save ID63_twins34.dta, replace
use ID63_twins1234.dta, replace
keep v* b*1 b*2
append using ID63_twins34.dta
sort v002 v002 v003 bord_1
save ID63_twins12.dta, replace
Thank you!
[Updated on: Tue, 10 April 2018 14:32] Report message to a moderator
|
|
|
|
|
Re: Twins and mortality [message #15011 is a reply to message #14455] |
Fri, 25 May 2018 02:21 |
behappy.ina77@gmail.com
Messages: 3 Registered: April 2018 Location: Indonesia
|
Member |
|
|
Dear all,
I want to make sure about my project,
I am analyzing multiple birth (twin) and neonatal mortality. I am using IDKR because my population is all live births from 2007-2012. I choose my independent variable (multiple births) from B0 which code 0= single birth and 1 (1st of multiple) and 2 (2nd of multiple)as twins. I exclude code 3,4, and 5 because I only analyze for twins not triple or so son. Is it correct what I am doing?
Is it correct that 0 is single birth and 1 and so on as birth order for the twins?
Need you guidance,
Thank you
Regards,
Bintang
|
|
|