The DHS Program User Forum: Fertility

Home » Topics » Fertility » Birth interval (Estimation of the effect of birth interval on TFR)

Show: Today's Messages :: Show Polls :: Message Navigator

Birth interval [message #30122]

Sat, 28 September 2024 09:07

Tesfay
Messages: 10
Registered: April 2021

Member

I am trying to see the effect of birth interval on TFR using pooled Ethiopian DHS data. I want to use the tfr2 command for multivariate Poisson regression in Stata which is only programed on the IR file. So,
1. How can I merge birth interval variables (e.g. b11) to the IR file
2. How can I estimate the BI that is useful to explain the TFR?

Report message to a moderator

Re: Birth interval [message #30145 is a reply to message #30122]

Thu, 03 October 2024 14:15

Janet-DHS
Messages: 937
Registered: April 2022

Senior Member

Following is a response from DHS staff member, Tom Pullum:

This is an interesting question, but I suggest that you approach it differently, perhaps as "how does the probability of having another birth after m months change according to the value of m (and other characteristics, such as parity and age)?" You can look into this with the BR file. There has been considerable analysis of birth intervals.

The length of the birth interval is a characteristic of two successive births. It is only defined for a pair of births.

Your question goes beyond the scope of what DHS staff can answer on the forum. Perhaps other users will offer suggestions.

Report message to a moderator

Re: Birth interval [message #31608 is a reply to message #30145]

Mon, 24 November 2025 15:58

Tesfay
Messages: 10
Registered: April 2021

Member

Dear Professor Tom Pullum, peace & health to you!
I am bringing a previous question back to your attention. I was working on other parts of my dissertation and found it necessary to test whether the optimal birth interval (the rationale for FP-promotion in Ethiopia) will ultimately affect parity at n, particularly result in a reduction of reaching parity 5+. Here are my questions
1)Is my question logical and statistically testable? If so,
2)Can I use the b11 and v201 found in the BR file as they are (v201in BR!= v201 in IR? If not, can I merge average b11 from BR to IR?
3)If this approach is not reliable, can you give more tips on the issue that you suggested to me last time:
"I suggest that you approach it differently. 'How does the probability of having another birth after m months change according to the value of m (and other characteristics, such as parity and age)?"

Generally, I am trying to empirically???? test the contraceptive-->birth-interval-->parity path, using the EDHS data.

[Updated on: Mon, 24 November 2025 16:12]

Report message to a moderator

Previous Topic:

Using tfr2 to Calculate age specific marital fertility rate

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Nov 26 10:01:54 Coordinated Universal Time 2025