The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » General » ever married sample
ever married sample [message #15428] Wed, 18 July 2018 05:55 Go to next message
lynn.alghazzi@gmail.com is currently offline  lynn.alghazzi@gmail.com
Messages: 8
Registered: July 2018
Member
Hello, if i am working with a dataset like India for example that includes all women, how do i limited to just the ever married sample so that it matches datasets of countries that include only ever married women? do i drop never married in STATA?
Re: ever married sample [message #15431 is a reply to message #15428] Wed, 18 July 2018 10:36 Go to previous messageGo to next message
HanR is currently offline  HanR
Messages: 19
Registered: March 2013
Location: Netherlands
Member
Hello,

Yes that would normally the case, ever married would then be currently living in union/living with a man and formerly in union/ living with a man. Please see variables V501 and/or V502. Please note that India is a special case where never married includes those who are married but where gauna has not yet been performed.
If you plan on doing fertility analysis then there may be some extra complications in this case, because you would then need so called all women factors, which are not ready available in all women samples.
Re: ever married sample [message #15434 is a reply to message #15431] Wed, 18 July 2018 11:27 Go to previous messageGo to next message
lynn.alghazzi@gmail.com is currently offline  lynn.alghazzi@gmail.com
Messages: 8
Registered: July 2018
Member
i am actually only studying the determinants of seeking help in many of the countries that include the domestic violence section so i think that dropping out the non married women is enough so that its matched with Jordan and Egypt that include only ever married women. i am not sure also whether to apply (keep v044==1) so that only women asked about domestic violence are included before i run my regression or if there is no need.
One last thing, my Stata knowledge is very little so which command can i use to create a seeking help variable that includes only whether help is sought or not regardless of who it is sought from?
Re: ever married sample [message #15467 is a reply to message #15434] Wed, 25 July 2018 09:16 Go to previous message
Liz-DHS
Messages: 1516
Registered: February 2013
Senior Member
From data processing specialist Han Raggers:
Quote:

"Yes one could select only those cases where V044 = 1 and then weight your observations by D005/ 1 000 000. In general the Domestic Violence Module is only asked to ever married women."


For Stata Tutorial: http://www.cpc.unc.edu/research/tools/data_analysis/statatutorial
Previous Topic: generating a new variable
Next Topic: SHDIST Variable for Ghana
Goto Forum:
  


Current Time: Thu Dec 2 10:24:21 Coordinated Universal Time 2021