| South Africa 2016 [message #28112] |
Tue, 14 November 2023 04:36  |
Carolina
Messages: 1
Registered: November 2023
|
Member |
|
|
Hello,
We merged the individual record (women only) whith the HIV results and found that only 2'726 matched observations, while there were 8'514 elligible women that were interviewed.
We just want to make sure that we completed the merge correctly. Is this number accurate ? Why isn't there a 50% split between men and women who were tested ?
Thank you for your response.
Carolina
|
|
|
|
| Re: South Africa 2016 [message #28136 is a reply to message #28112] |
Thu, 16 November 2023 14:39  |
Janet-DHS
Messages: 888
Registered: April 2022
|
Senior Member |
|
|
Following is a response from DHS staff member, Tom Pullum:
The Stata lines below show how to merge the AR file with both the women and the men. Of the 4862 cases in the AR file, 2136 are men and 2726 are women. These numbers are fairly close, and the difference is mainly due to higher refusal rates (for HIV testing) for men than for women. The number of cases in the IR file (8514) does not tell you anything about how many men are in the AR file.
* Specify workspace
cd e:\DHS\DHS_data\scratch
* Prepare AR file
use "...ZAAR71FL.DTA", clear
gen cluster=hivclust
gen hh=hivnumb
gen line=hivline
gen in_AR=1
sort cluster hh line
save ARtemp.dta, replace
* Prepare IR file
use "...ZAIR71FL.DTA", clear
gen cluster=v001
gen hh=v002
gen line=v003
gen sex=2
keep cluster hh line sex
save IRtemp.dta, replace
* Prepare MR file
use "...ZAMR71FL.DTA", clear
gen cluster=mv001
gen hh=mv002
gen line=mv003
gen sex=1
keep cluster hh line sex
save MRtemp.dta, replace
* Append IR and MR and merge with AR
append using IRtemp.dta
sort cluster hh line
merge cluster hh line using ARtemp.dta
keep if _merge==3
tab sex if in_AR==1
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
