The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » General » Imputing missing DOBs for children
Imputing missing DOBs for children [message #3840] Sun, 22 February 2015 11:06 Go to next message
adev is currently offline  adev
Messages: 14
Registered: May 2013
Member
Hello -

I am working with a dataset that is adapted from DHS. However, the data providers have not done as thorough of a job of imputing dates of birth for children as the DHS Program. I have reviewed the methodology for the DHS DOB imputation. Does DHS share the code for imputing missing DOBs for children? I understand the process but was curious about the STATA code to actually impute dates.

I would appreciate any feedback on the most efficient way to do this, given that there are 4 panels in my data.

Thank you
Alka


Alka Dev
Graduate Fellow
CUNY Institute for Demographic Research
Re: Imputing missing DOBs for children [message #3843 is a reply to message #3840] Sun, 22 February 2015 19:23 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
We don't have Stata code for the imputation process. This process actually carries out a few hundred checks on the data and uses a lot of information to impute dates of a number of events, not just the dates of birth of children. The basic approach is described in http://dhsprogram.com/pubs/pdf/dhsg3/dhs_data_editing.pdf. If you are comfortable with the approach described here, I would suggest writing your own code. Sorry we don't have anything more useful.
Re: Imputing missing DOBs for children [message #8794 is a reply to message #3843] Fri, 18 December 2015 17:30 Go to previous messageGo to next message
adev is currently offline  adev
Messages: 14
Registered: May 2013
Member
Hi Trevor,

Does DHS provide information on the proportion of births that were missing DOB? I am comparing DHS to another HH survey and I think this information will be useful as I now what the proportions are for the other survey.

Thank you,
Alka


Alka Dev
Graduate Fellow
CUNY Institute for Demographic Research
Re: Imputing missing DOBs for children [message #8795 is a reply to message #8794] Fri, 18 December 2015 19:48 Go to previous message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
See variable B10. Cases with B10 = 1 have no missing information, others cases had missing or don't know for some part of the date of birth. You can also see the proportions in table C.3 in most reports.
Previous Topic: Are madrasas (and other religious schools) counted as formal education?
Next Topic: Missing 'language of inerview' variable in African surveys
Goto Forum:
  


Current Time: Thu Mar 28 10:02:46 Coordinated Universal Time 2024