| Simple domestic violence crosstabs [message #10856] |
Mon, 26 September 2016 05:50  |
will.dahlgreen
Messages: 1
Registered: September 2016
Location: London
|
Member |
|
|
Hi,
I'm a journalist looking into the disadvantages faced by indigenous women. I'm making simple crosstabs in R to show the variation in domestic violence by ethnicity, using the individual recode datasets for a number of countries.
I just want to check if I'm correct in using the individual recode datasets, and that all I need to do is weight the data by the domestic violence weight variable (D005 or V005 in the case of Peru). Or do I also need to filter the data by any other variables before I weight the data (eg marital status/whether selected for the domestic violence module)?
Many thanks,
Will
|
|
|
|
| Re: Simple domestic violence crosstabs [message #10873 is a reply to message #10856] |
Tue, 27 September 2016 11:00  |
Trevor-DHS
Messages: 803
Registered: January 2013
|
Senior Member |
|
|
Yes, the individual recode datasets are the correct datasets to use. Typically you would filter the cases by whether selected for the domestic violence module (see v044), although, women not selected would generally have a weight of 0 on d005. You may then need to sub-select to the group needed for a particular tabulation, e.g. for currently married women, say, depending on the tabulation.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
