The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Mortality » Missing data for mm8 Date of sibling death
Missing data for mm8 Date of sibling death [message #15967] Thu, 11 October 2018 18:57 Go to next message
rkbroussard@prc.utexas.ed is currently offline  rkbroussard@prc.utexas.ed
Messages: 4
Registered: April 2018
Location: Austin, Texas
Member
Hello,

I recently discovered that I have missing data for variable mm8, which provides the CMC date of a sibling's death. I used the following code in Stata, which asks Stata to tell me the number of cases I have in which a woman says a sibling died, but then no date of death is provided for this sibling, and came up with 21% missing data.
*missing= 11750 (21 percent)
forvalues i = 1/20 {
count if mm2_`i'== 0 & mm8_`i' ==.
}

I read online in the 2014 Assessment of Maternal mortality Data (https://dhsprogram.com/pubs/pdf/MR13/MR13.pdf) that the variable mm8, which provides the CMC date of death of the sibling, is calculated and imputed by DHS analysts using mm6 (years ago that sibling died) and the interview date.

So if these values are imputed, why am I still finding missing data? Do I need to impute them myself?

Thank you for any assistance you can offer!
Kathleen


Kathleen Broussard
Re: Missing data for mm8 Date of sibling death [message #16037 is a reply to message #15967] Wed, 24 October 2018 17:13 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
Which dataset are you using? I can't tell from your post.
Re: Missing data for mm8 Date of sibling death [message #16045 is a reply to message #16037] Thu, 25 October 2018 19:53 Go to previous message
rkbroussard@prc.utexas.ed is currently offline  rkbroussard@prc.utexas.ed
Messages: 4
Registered: April 2018
Location: Austin, Texas
Member
Hi Trevor,

I am using DHS data from Peru for 2004-2012. (It's a compiled dataset). However, when I examine the data from 2012 exclusively, it shows that all values for mm8 are missing for siblings that died. I reshaped the data to long format. Looking at 2012 data, I used the code below to find the number of siblings that had a year of death reported, and this code returned a value of 0.

count if mm2 == 0 & mm8 !=.

Thank you for any assistance you can offer!
Kathleen



Kathleen Broussard
Previous Topic: infant mortality
Next Topic: Sibling mortality and household membership
Goto Forum:
  


Current Time: Thu Mar 28 12:52:31 Coordinated Universal Time 2024