The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Dataset use in Stata » Women who gave birth in the last 2 years (Number of women who gave birth in the last 2 years)
Women who gave birth in the last 2 years [message #29323] Thu, 30 May 2024 07:40 Go to next message
vanganir@who.int is currently offline  vanganir@who.int
Messages: 2
Registered: May 2024
Member
The number of women who gave birth in the last 2 years is 13296 in the India Report (2019-21)-Table 13.15
However using the IR file the number of women who gave birth in the last 2 years comes out to be around 85,773
Please Help

using IR file
gen age = v008 - b3_01
gen xweight = v005/1000000
tabulate age if age < 24 [iw=xweight]
Re: Women who gave birth in the last 2 years [message #29328 is a reply to message #29323] Thu, 30 May 2024 11:55 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3203
Registered: February 2013
Senior Member

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

This puzzled me too, until I went to the text of Chapter 13 and saw that the chapter applies only to women in households that were selected for the state module. The variable that identifies such households is ssmod. It takes the value 1 for 108,785 women, only about 15% of the full sample. If you restrict to women who have ssmod=1 you should get a match.


Re: Women who gave birth in the last 2 years [message #29333 is a reply to message #29328] Fri, 31 May 2024 05:42 Go to previous messageGo to next message
vanganir@who.int is currently offline  vanganir@who.int
Messages: 2
Registered: May 2024
Member
Thank you very much. I had spent a lot of time trying to figure out the reason.

I wanted to further ask that if we are doing a state level analysis on certain behavioral indicators at a state level than we need to apply women's individual sample weight(v005) or state level weight(sweight).
Also will it be meaningful to do regression analysis as the sample size will reduce further.

Thanks and Regards
Re: Women who gave birth in the last 2 years [message #29337 is a reply to message #29333] Fri, 31 May 2024 09:32 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3203
Registered: February 2013
Senior Member

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

Selection for the state module (which also was the criterion for selection for the men's interview) was completely random, throughout the entire survey. So the general rule applies: use v005 for all analyses except state-specific analyses. For example, you would use sweight if you were ONLY looking at Punjab. Otherwise use v005.

Regression would be ok. For most purposes the sample size would be large enough, in terms of statistical power.
Previous Topic: Duration of Pregnancy
Next Topic: Strange Issues w/ Data Formatting from DHS
Goto Forum:
  


Current Time: Sat Dec 7 22:35:54 Coordinated Universal Time 2024