Sibling mortality and household membership [message #16041] |
Thu, 25 October 2018 07:47 |
MatthiasArnold
Messages: 1 Registered: October 2018
|
Member |
|
|
Hi,
I am working on the Malawian DHS 2015/16. I am interested in getting a better understanding about determinants of adult mortality. I have calculated adult mortality as described in the post by Trevor.
I have estimated the sisterhood mortality in specific socio-economic subgroups. However, I do find odd results especially when linking household wealth and residence to mortality rates. Basically the richer households are, the higher is the probability of sisterhood mortality.
I was wondering if there is additional information in the respective questionnaire that allows identifying were the sisters/brothers actually lived, if they were in the same household or in others. Unfortunately it appears that most of the information that would help here was not collected in Malawi. Is there a way to connect the sibling information to the household roster information?
|
|
|
Re: Sibling mortality and household membership [message #16088 is a reply to message #16041] |
Fri, 02 November 2018 11:59 |
Liz-DHS
Messages: 1516 Registered: February 2013
|
Senior Member |
|
|
Dear User, a response from technical expert, Mr. Trevor Croft:
Quote:
We don't have data on where the siblings live(d), and they could be in very different wealth quintiles.
Disaggregating by wealth quintile is highly problematic as there would be very few deaths in each quintile and the confidence intervals would be enormous. The surveys are not designed to produce estimate of maternal mortality by quintile and I would not trust any estimates produced by quintile.
A response from senior data processing expert, Mr. Albert Themme:
Quote:
no info to link "sisters" to the "household roster", since we do not collect line numbers to link them. We also do not have a way to find out which sisters live in the household and which do not.
Thank you!
|
|
|