The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Domestic Violence » RWANDA 2014-2015 Domestic Violence Tables
RWANDA 2014-2015 Domestic Violence Tables [message #15439] Thu, 19 July 2018 12:05 Go to previous message
j.cml is currently offline  j.cml
Messages: 2
Registered: July 2018
Location: Kigali

I am currently analyzing the module on domestic violence in DHS 2015-2015 for Rwanda (using the Women's recode: RWIR70DT.DTA)
I am using the weight for domestic violence but I have trouble replicating the exact figures of DHS table 17.1.1

For Table 17.1.1, the total Percentage who have ever experienced physical violence since age 15, displayed in the report is 34.5 , but I systematically obtain 34.07.
I suspect the 34.5 includes unweighted cases but I was wondering how to reach this number with STATA. I also obtain small discrepancies in each of the crosstab statistics

My code is as follows:

gen wgt= v005/1000000
gen wvwgt = d005/1000000

gen psu=v021

gen strata=v023

svyset psu [pweight=wvwgt], strata(strata)

g anyvlc2=.
replace anyvlc2=0 if d106==0 | d107==0 | d115y!=0 & d115y<. | d130a==0
replace anyvlc2=1 if d106==1 | d107==1 | d115y==0 | d130a>0 & d130a<4
lab var anyvlc2 "Percentage who have ever experienced physical violence since age 15"
lab def yesno 0 "No" 1 "Yes"
lab values anyvlc2 yesno
svy: tab anyvlc2, percent ci

I have also tried

gen anyvlc = 0 if v044 == 1
ds d105a-d105f d105j d130a
foreach var in `r(varlist)'{
replace anyvlc = 1 if `var' > 0 & `var' < 5
replace anyvlc = 1 if d115y==0
lab var anyvlc "Percentage who have ever experienced physical violence since age 15 "
lab values anyvlc yesno
svy: tab anyvlc, percent

I would appreciate any guidance on how to replicate the exact figures displayed in the table using STATA.

Read Message
Read Message
Previous Topic: domestic violence
Next Topic: Multilevel Weights
Goto Forum:

Current Time: Sat Feb 29 07:31:36 Eastern Standard Time 2020