The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » General » Marriage History
Marriage History [message #2909] Tue, 09 September 2014 07:09 Go to next message
Lorretta is currently offline  Lorretta
Messages: 8
Registered: August 2013
Location: Nigeria
Member
I am conducting a study that aims to estimate the risk of divorce and separation in sub-Saharan Africa (Cameroon, Ghana, Nigeria, Kenya, Uganda, Tanzania). I have not seen any variable that captures time of marital dissolution in these countries. I need help on how to estimate the risk using survival analysis in the absence of complete marital history.


Lorretta
Re: Marriage History [message #2912 is a reply to message #2909] Tue, 09 September 2014 10:37 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
Unfortunately the time of dissolution of marriage is not collected in the DHS surveys. I can't think of a way of using survival analysis for what you want to do with the data available. Sorry.
Re: Marriage History [message #2913 is a reply to message #2909] Tue, 09 September 2014 10:38 Go to previous messageGo to next message
user-rhs is currently offline  user-rhs
Messages: 132
Registered: December 2013
Senior Member
Loretta,
This is a tough one. The short answer is: there is no good way for this using DHS. Information on marital status in the 5+ yrs preceding interview is captured in the reproductive calendar (VCAL) variable column 3 (ref: http://dhsprogram.com/pubs/pdf/DHSG4/Recode6_DHS_22March2013 _DHSG4.pdf, p.98), but the information is very limited and the time frame is quite short. I don't know much about marriage in W. / E. Africa, but I imagine divorce rates are lower than in the U.S. or Europe, so that may present additional challenges to you if the number of observations for whom marriage has been dissolved is low. Maybe if you can use a combination of age at first marriage, current age, VCAL_3, and marital status to model survival of marriage (the idea is that if the difference in age at first marriage is within 5 yrs of interview, you might have "complete" marital history for those people). Even this has its problems and you will have lots of right-censoring in your data.

The WB's living standards surveys may have info on marital hx. I think some of them might even be panel.

HTH,
RHS
Re: Marriage History [message #2914 is a reply to message #2913] Tue, 09 September 2014 10:58 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
While RHS is correct that some data can be found in vcal_3, they are limited to the last 5 years as he says, and, for the most part (with a couple of exceptions), these data have not been collected in the more recent surveys.
Re: Marriage History [message #2916 is a reply to message #2914] Tue, 09 September 2014 12:00 Go to previous message
Lorretta is currently offline  Lorretta
Messages: 8
Registered: August 2013
Location: Nigeria
Member
Thanks RHS for the useful information. Trevor is right, I have checked the datasets I have, there is no vcal_3 in the recent surveys. I will check WB.

Lorretta
Previous Topic: What does a ".a" response indicate?
Next Topic: Female Genital Cutting data - Nigeria 2008
Goto Forum:
  


Current Time: Fri Mar 29 07:31:03 Coordinated Universal Time 2024