Following is a response from Senior DHS staff member, Tom Pullum:
Here is the Stata code for the first column of table 16.1.2 in the Kenya 2014 report. The code for women (table 16.1.1) is the same except the mv variables for men are v variables for women. I had to go back to the original CSPro code. For some covariates, there is one category for which my estimate is a little lower than the final report. There may still be something missing but I cannot put more time into this. You can modify this to get the other columns of the table.
* Kenya 2014, Domestic Violence Module
* MEN: Table 16.1.2
use "C:\Users\26216\ICF\Analysis - Shared Resources\Data\DHSdata\KEMR72FL.DTA", clear
* Construct indicator for first column, ever violence
gen dv_phy=0 if mv044==1
foreach ll in a b c d e f {
replace dv_phy=1 if md105`ll'<. & md105`ll'>0
}
replace dv_phy=1 if md115y==0
replace dv_phy=1 if md118y==0
replace dv_phy=1 if md130a<. & md130a>0
* Some recodes of covariates
gen age5=mv013
replace age5=4 if mv013>4
replace age5=5 if mv013>5
label define age5 1 "15-19" 2 "20-24" 3 "25-29" 4 "30-39" 5 "40-49"
label values age5 age5
gen marstat=mv501
replace marstat=1 if mv501==2
replace marstat=2 if mv501>2
label define marstat 0 "Never married" 1 "Married" 2 "Divorced, etc."
label values marstat marstat
* Distributions for first column (not complete)
tab age5 dv_phy if mv013<=7 [iweight=md005/1000000], row
tab mv130 dv_phy if mv013<=7 [iweight=md005/1000000], row
tab mv025 dv_phy if mv013<=7 [iweight=md005/1000000], row
tab mv024 dv_phy if mv013<=7 [iweight=md005/1000000], row
tab marstat dv_phy if mv013<=7 [iweight=md005/1000000], row
tab mv149 dv_phy if mv013<=7 [iweight=md005/1000000], row
tab mv190 dv_phy if mv013<=7 [iweight=md005/1000000], row
summarize dv_phy if mv013==8 [iweight=md005/1000000]
summarize dv_phy [iweight=md005/1000000]
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
