The DHS Program User Forum      
Discussions regarding The DHS Program data and results
Home » Topics » Mortality » Code for Infant mortality (Dead or Alive)
Code for Infant mortality (Dead or Alive) [message #11936] Thu, 09 March 2017 09:41 Go to next message
Nkechi Onwuka is currently offline  Nkechi Onwuka
Messages: 2
Registered: March 2017
Member

Please i'm having issues with locating/creating a code to analyse infant mortality (Dead or Alive) using Nigeria Demographic and Health Survey (2013) on SPSS. The only such code I can find on the data set is for Under-five mortality (Dead or Alive) on SPSS but I really need that of Infant mortality, ad it is the basis Of my research.
Re: Code for Infant mortality (Dead or Alive) [message #11968 is a reply to message #11936] Tue, 14 March 2017 07:05 Go to previous messageGo to next message
A.TIENDREBEOGO-BKF is currently offline  A.TIENDREBEOGO-BKF
Messages: 9
Registered: March 2017
Location: OUAGADOUGOU
Member
First locate your variables: B3 (date of birth), B5 (child is alive) , B7 (age at death in months), v008 (date of interview)
Then create your 5 years periods preceding survey like that:
compute period = trunc((v008-1 - b3)/60).
You will have for this variable 0 (5 years preceding), 1 (5-9 years preceding), 2 (10-14 years preceding), etc.
You can now determine your infant mortality. Let us take an example to be more expressive.
At this level, there are 2 considerations to take into account.
1. if you are only interested inside a period, for example, the past 5 years, all born in this period (period=0) are the denominator because all born in this period died before one year or survive above one year. The numerator (infant deaths) can be calculated like that:
Period=0 and B5=0 and B7<12.
2. if you are interested to all infant deaths occurred in a period, you have to add up those children of the preceding period, in this example period=1 (59 years preceding) who died during 5 years period before on year. You can determine them like that by creating before, the period of death:
if (B5=0) cmc_death=B3+B7.
compute perdeath=trunc((v008-1-cmc_death)/period).

Then determine the children in this case (born in preceding period and died in the ongoing period before on year)
Period=1 and perdeath=0 and B7<12.

We determine the deaths, let us calculate now the denominator for such children by calculating their number at the beginning of the ongoing period (here period=0).
*compute number of months all child lived:
compute months = 0.
if (B5 = 0) months = b7.
if (B5 = 1) months = (v008 - b3).

Let us determine the beginning of the ongoing period: start=v008-1-60.
Then the denominator is:
start-B3<12 and months>=start.

Thinking this will be helpul to you.
Re: Code for Infant mortality (Dead or Alive) [message #12020 is a reply to message #11968] Mon, 20 March 2017 15:42 Go to previous messageGo to next message
Nkechi Onwuka is currently offline  Nkechi Onwuka
Messages: 2
Registered: March 2017
Member

Thank you very much A.TIENDREBEOGO-BKF. This helps a lot
Re: Code for Infant mortality (Dead or Alive) [message #12745 is a reply to message #11968] Thu, 06 July 2017 18:01 Go to previous messageGo to next message
Davison is currently offline  Davison
Messages: 5
Registered: March 2017
Location: Germany
Member
Hello,

I don't understand what you mean with

compute perdeath=trunc((v008-1-cmc_death)/period).

You create the variable period and I understand that this are the periods where the child died but why should I divide with classes?

Best regards

Re: Code for Infant mortality (Dead or Alive) [message #12749 is a reply to message #12745] Fri, 07 July 2017 08:40 Go to previous message
A.TIENDREBEOGO-BKF is currently offline  A.TIENDREBEOGO-BKF
Messages: 9
Registered: March 2017
Location: OUAGADOUGOU
Member
You divide into period to have those born and dies and those born during the previous period and died for each five-year period considered. Thus, calculations can be made on each period or over a period of specific interest. Because the mortality during a specific period concerns children born and deceased during this period, plus children born during the previous period and deceased during the period considered before their 5th birthday.

Hoping to have responded to you and keeping me available for further information
Previous Topic: Notes on Child mortality calculation
Next Topic: GFR
Goto Forum:
  


Current Time: Tue Sep 19 09:31:11 Eastern Daylight Time 2017