The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » India » Cancer screening (Cervical cancer )
Cancer screening [message #25538] Sun, 06 November 2022 19:53 Go to next message
DHS Data explorer is currently offline  DHS Data explorer
Messages: 2
Registered: November 2022
Member
Hello,

I am using the India DHS 2019-21 to look at cervical cancer screening. I am trying to match the table in Stata by using the women's record (IAIR7DDT) file.
I am unable to match the numbers and percentages in "Table 12.7: Screening tests for cancer," in the second column, "Prevalence of cancer screening tests".
I created weight [wt=v005/1000000] and then used svyset [pw=wt], psu(v021) strata(v022). To tabulate the frequencies, I used tab sb79 [iw=gt].
I wondered if anyone would be able to advise me if I am using the right dataset, sample weights, and, most importantly, the right variable to determine cervical cancer screening.
Any suggestions would be much appreciated.

[Updated on: Sun, 06 November 2022 19:56]

Report message to a moderator

Re: Cancer screening [message #25606 is a reply to message #25538] Wed, 16 November 2022 09:43 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2635
Registered: February 2013
Senior Member
Following is a response from Senior DHS staff member, Tom Pullum:

The following code will reproduce the percentages in the "Cervix" column of table 12.7. In order to match the report, you must include "m" (for "missing" or "NA") in the tab commands. It could be argued that those cases should NOT be included in the denominators for the percentages. However, to get a match you have to do that.

I include the recodes for age and schooling; you can do the recode for religion of head = "other".

use "...IAPR7DFL.DTA", clear

keep if hv105>=15 & hv105<=49
gen age=1
replace age=2 if hv105>=20
replace age=3 if hv105>=25
replace age=4 if hv105>=35
label define age 1 "15-19" 2 "20-24" 3 "25-34" 4 "35-49"
label values age age

gen schooling=0
replace schooling=1 if hv108>=1
replace schooling=2 if hv108>=5
replace schooling=3 if hv108>=8
replace schooling=4 if hv108>=10
replace schooling=5 if hv108>=12
replace schooling=6 if hv108==98
label define schooling 0 "None" 1 "1-5" 2 "5-7" 3 "8-9" 4 "10-11" 5 "12" 6 "DK" 
label values schooling schooling

tab shb79 if hv104==2 & hv105>=15 & hv105<=49 [iweight=hv005/1000000], m


local lcovariates age hv025 schooling sh47 sh49 hv270
foreach lc of local lcovariates {
tab `lc' shb79 if hv104==2 & hv105>=15 & hv105<=49 [iweight=hv005/1000000], m row
}
Re: Cancer screening [message #25640 is a reply to message #25606] Mon, 21 November 2022 21:30 Go to previous message
DHS Data explorer is currently offline  DHS Data explorer
Messages: 2
Registered: November 2022
Member
Thank you.
Previous Topic: Caste data for 2019-21
Next Topic: Telangana State discrepancy in numbers of women with a live birth
Goto Forum:
  


Current Time: Sat Dec 3 19:44:26 Coordinated Universal Time 2022