The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Nepal » Births in the two years preceding the survey
Births in the two years preceding the survey [message #28202] Tue, 28 November 2023 23:34 Go to next message
bshrestha is currently offline  bshrestha
Messages: 2
Registered: December 2018
Member
In 2021 NDHS, how do we calculate 'Births in the two years preceding the survey'?

[Updated on: Tue, 28 November 2023 23:35]

Report message to a moderator

Re: Births in the two years preceding the survey [message #28267 is a reply to message #28202] Wed, 06 December 2023 16:55 Go to previous message
Janet-DHS is currently offline  Janet-DHS
Messages: 893
Registered: April 2022
Senior Member
Following is a response from DHS staff member, Tom Pullum:

The Stata code pasted below will calculate the number of births in the past 1, 2, 3, 4, or 5 years using the IR file, in surveys that include b19. The IR file already includes variables for the past 3 and 5 years, and this approach matches those numbers. However, the construction of v209 in the IR file, for the past year, is inconsistent, because it is calculated as the number of births 0-12 months ago. It should not include month 12, but it does include month 12. In the definition of the past 3 years or the past 5 years, DHS uses completed months 0-35 or 0-59, respectively. I would recommend that you follow that approach, as I have done here.

The inclusion of month 12 in v209 goes back to the time before DHS surveys included b19, and made some sense then, but I would classify it as a mistake in surveys with b19.

Your subject line says "two years" but your question asks about "three years". I am giving the code for 1, 2, 3, 4, or 5 in order to cover more options. It's easy to generalize further.


* Births in the past X years in surveys that include b19, such as Nepal 2022

use "C:\Users\26216\ICF\Analysis - Shared Resources\Data\DHSdata\NPIR82FL.DTA", clear

rename *_0* *_*

gen B1=0
gen B1alt=0
gen B2=0
gen B3=0
gen B4=0
gen B5=0

forvalues li=1/20 {
replace B1=B1+1 if b19_`li'<12
replace B1alt=B1alt+1 if b19_`li'<=12
replace B2=B2+1 if b19_`li'<24
replace B3=B3+1 if b19_`li'<36
replace B4=B4+1 if b19_`li'<48
replace B5=B5+1 if b19_`li'<60
}

label variable B1 "Births in the past year"
label variable B1alt "Births in the past 13 months"
label variable B2 "Births in the past 2 years"
label variable B3 "Births in the past 3 years"
label variable B4 "Births in the past 4 years"
label variable B5 "Births in the past 5 years"

* Some of these are already in the standard recode file
* Note that v209 has an inconsistent construction
* v209 births in past year (calculated as months 0-12)
* v238 births in last three years (calculated as months 0-35)
* v208 births in last five years (calculated as months 0-59)

correlate v209 B1
correlate v209 B1alt
correlate v238 B3
correlate v208 B5
Previous Topic: Intimate partner violence against women
Next Topic: Impac guideline missing in recode
Goto Forum:
  


Current Time: Thu Nov 28 16:55:45 Coordinated Universal Time 2024