The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Dataset use (other programs) » Uttar Pradesh districtwise Neonatal, infant, postnatal and under five Mortality data (Stata for required processing Uttar Pradesh districtwise Neonatal, infant, postnatal and under five Mortality data From NFHS 5 Datasets)
Re: Uttar Pradesh districtwise Neonatal, infant, postnatal and under five Mortality data [message #26999 is a reply to message #26996] Thu, 08 June 2023 07:43 Go to previous messageGo to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3199
Registered: February 2013
Senior Member
Following is a response from Senior DHS staff member, Tom Pullum:

Your request goes far beyond the scope of the forum. Also what you propose to do would not be useful, because of the statistical uncertainty of district-level estimates.

However, I have prepared something that may be useful. I wrote a Stata program to calculate, within the districts of Uttar Pradesh, the proportion of children born in the past five years who had died by the date of the survey, along with 95% confidence intervals adjusted for the survey design. This program could be of interest to other users so I will paste it below (as a text file). Since you wanted an Excel file, I am attaching the Excel file that is produced by this program.

You will see that the confidence intervals are wide. The confidence intervals for the standard rates would be much wider. The proportion calculated here is probably the best you can do to identify variation in child survival across districts.


* Program to calculate the proportion of children in the NFHS5 KR file
*  who survived to the date of the survey, in districts in Uttar Pradesh

* Specify a workspace
cd e:\DHS\DHS_data\scratch

* Open the KR file; children born in the past 5 years
use "C:\Users\26216\ICF\Analysis - Shared Resources\Data\DHSdata\IAKR7EFL.DTA", clear 

describe v024

* The label for v024 is V024; list it
label list V024

* Uttar Pradesh is v024=9 
keep if v024==9

* How many districts are in Uttar Pradesh?
codebook sdist

* 75 districts

svyset v001 [pweight=v005], strata(v023) singleunit(centered)
svy: proportion b5, over(sdist)

* the lines with the output "0" give the proportion of children who died, with a confidence interval

* Save the results and copy into Excel

matrix S=r(table)
matrix T=S'

* construct a file with the district names and n's, weighted and unweighted
gen unwtdn=1
gen wtdn=sweight/1000000

collapse (sum) *wtdn,by(sdist)
rename sdist district
label values district SDIST
label variable unwtdn "N (unweighted)"
label variable wtdn "N (weighted)"
gen line=_n
sort line
save stub.dta, replace

clear
svmat T
matrix list T
rename T1 b
rename T2 se
rename T3 t
rename T4 pvalue
rename T5 ll
rename T6 ul
rename T7 df
rename T8 crit
rename T9 eform
gen line=_n
keep if line<=75
sort line
merge line using stub.dta
tab _merge

rename b P
rename ll L
rename ul U
label variable P "Proportion dead"
label variable L "Low end, 95% c.i."
label variable U "Upper end, 95% c.i."
keep  district P L U *wtdn
order district P L U *wtdn

list, table clean
export excel using child_survival_by_district.xlsx, replace firstrow(var)
 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: NFHS5 mortality tables
Next Topic: Disability prevalence question
Goto Forum:
  


Current Time: Wed Nov 27 04:08:03 Coordinated Universal Time 2024