Discrepancy in population totals for postnatal checkup-India (NFHS 4) [message #21937] |
Fri, 08 January 2021 22:16 |
dgodha
Messages: 44 Registered: November 2016 Location: India
|
Member |
|
|
Hello DHS Experts,
While replicating tables 8.21 and 8.22, my population total estimate is 184,640.62 as opposed to 184,627 in the India DHS 2015-16 report. I have used the following commands.
*Use BR file
/*Generating weights and survey setting*/
gen wt=v005/1000000
svyset [pw=wt], psu(v021) strata(v025)
*PNC for those who delivered in a health facility
recode s460 (100/202 =1 "pnc within 2 days") (203/305= 2 "pnc 3-41 days") (306/400 = 3 "42 days or more") (900/1000 =9 "dk") (else=0 "no pnc") , gen(pnctime_hf)
*PNC for those who delivered at home or who had a check after they left the facility
recode m51 (100/202 =1 "pnc within 2 days") (203/305= 2 "pnc 3-41 days") (306/400 = 3 "42 days or more") (900/1000 =9 "dk") (else=0 "no pnc") , gen(pnctime_home)
*PNC for all women
gen pnc=pnctime_hf
replace pnc=pnctime_home if pnc==0
replace pnc=3 if pnc==0
lab de pnc 1"PNC within 2 days" 2"PNC 3-41 days" 3"No postnatal check" 9"Don't know"
lab val pnc pnc
*PNC by health provider
gen pncpr=s461
replace pncpr=m52 if pncpr==.
replace pncpr=0 if pnc==3
replace pncpr=99 if pnc==9
label de pr 0"No postnatal check" 11"Doctor" 12"ANM/Nurse/Mid-wife/LHV" 13"Other health personnel" 21"Dai(TBA)" 22"ASHA" 96"Other" 99"Don't know"
label val pncpr pr
tab pncpr
****
*Check with Table 8.21
svy:tab pnc if v208>0 & v208~=. & bidx==1,percent
*Check with Table 8.22
svy:tab pncpr if v208>0 & v208~=. & bidx==1,percent
I will request your advice on how to get the exact population total.
Thank you
Deepali
Deepali
|
|
|
|
|