| Re: Calculation of prevalence using STATA command [message #318 is a reply to message #172] |
Mon, 15 April 2013 14:29   |
Bridgette-DHS
Messages: 3199
Registered: February 2013
|
Senior Member |
|
|
Here is a response from one of our DHS experts Kerry MacQuarrie, that should answer your question.
You are using the domestic violence weight correctly, with the difference between the n you estimated and that found in the final report being due to rounding only. The discrepancy between the 14.01% you arrived at and the 15.2% in the report is because Table 14.3 in the report is NOT restricted to sexual violence perpetuated by the husband/partner only (which also explains how the Table reports a figure for prevalence of sexual violence among never married women). Instead of relying on variable d108 (ever experienced any sexual violence (d105h-i) by husband/partner) alone, this table uses a combination of d108 and variables d124 and d125. D124 is ever forced to have sex by anyone other than husband/spouse in last 12 months and d125 is ever forced to perform unwanted sexual acts. I found that if I generated the following composite variable:
g anysv=.
replace anysv=0 if d108==0 | d124==0 | d125==0
replace anysv=1 if d108==1 | d124==1 | d125==1
that I was then able to replicate the figures in Table 14.3:
tab v501 anysv [iweight=dvwt], row
If you are interested in sexual violence perpetuated by husbands/partners and not by others, you are correctly using d108. Please note that there is another variable available in the dataset: d123, which captures whether women's first sexual initiation was forced and this variable, in conjunction with those that describe sexual initiation as occurring within (and only one) marriage is sometimes added to d108 to capture any sexual violence by husband/partner. This variable is not included in the composite variable producing the results for Table 14.3.
I hope this is helpful.
-
Attachment: capture.bmp
(Size: 666.34KB, Downloaded 3245 times)
[Updated on: Mon, 15 April 2013 14:30] Report message to a moderator |
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
