The DHS Program User Forum: Reproductive Health

Home » Topics » Reproductive Health » Mothers age at birth

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

Mothers age at birth [message #1459]

Fri, 28 February 2014 02:52

pkaburi
Messages: 9
Registered: February 2014
Location: Nairobi

Member

Hi,

I am using DHS data set and I am unable to locate the variable mother age at birth. How can i generate the variable?

regards

Patrick

Report message to a moderator

Re: Mothers age at birth [message #1461 is a reply to message #1459]

Fri, 28 February 2014 09:44

Liz-DHS
Messages: 1516
Registered: February 2013

Senior Member

Dear User,
Maternal age at birth is computed as follows: (B3-V011)/12.
You may also want to refer to The Guide to DHS Statistics. http://www.measuredhs.com/pubs/pdf/DHSG1/Guide_to_DHS_Statis tics_29Oct2012_DHSG1.pdf The document is bookmarked. Under the bookmark: III. Guide to Indivdual Statistics, the 4th bookmark under that section, you will find "Age at First Birth". You may also find the Standard Recode Manual http://www.measuredhs.com/pubs/pdf/DHSG4/Recode6_DHS_22March 2013_DHSG4.pdf useful as you look at the recode variables for your data.
Thank you.

Report message to a moderator

Re: Mothers age at birth [message #10836 is a reply to message #1461]

Thu, 22 September 2016 21:21

mnisha
Messages: 5
Registered: January 2016
Location: Sydney, Australia

Member

Hi,
I am working on BDHS 2014 data. I want to calculate 'mother's age at delivery' for the stillbirths. Could you please suggest me which dataset I need to use to calculate mother's age for the stillbirths- BR or IR? Does BR dataset contain any information for the stillbirths?
Regards,
Nisha

Report message to a moderator

Re: Mothers age at birth [message #10840 is a reply to message #10836]

Fri, 23 September 2016 09:39

Trevor-DHS
Messages: 803
Registered: January 2013

Senior Member

Stillbirths are not included in the birth history - only live births. If you want to calculate mother's age at delivery for live births, you can use the BR (or KR) dataset and the following code:

gen mage_deliv = int((b3 - v011)/12)

The only information on stillbirths is collected in the calendar, and presented in table 8.5 of the report. This is estimated based on any non-live births in the last five years reported in the calendar. The data for the calendar (vcal_1) can be found in the IR file.

Report message to a moderator

Re: Mothers age at birth [message #10842 is a reply to message #1459]

Fri, 23 September 2016 10:23

mnisha
Messages: 5
Registered: January 2016
Location: Sydney, Australia

Member

Hi,
I am trying to calculate inter-pregnancy interval (in months) for stillbirths from BDHS 2014 data. Could you please help me to find the right way to calculate that variable?

Regards,
Nisha

[Updated on: Mon, 05 December 2016 01:15]

Report message to a moderator

Re: Mothers age at birth [message #26108 is a reply to message #1459]

Wed, 08 February 2023 12:23

vkyni
Messages: 4
Registered: February 2023

Member

Hi,

I'm working on marriage to first birth interval (months) (v221 variable). I'm using PH model, for that I need event as first live birth, could you please help me to get that?

[Updated on: Wed, 08 February 2023 12:25]

Report message to a moderator

Re: Mothers age at birth [message #26143 is a reply to message #26108]

Fri, 10 February 2023 12:06

Janet-DHS
Messages: 888
Registered: April 2022

Senior Member

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

Does "PH model" refer to the Philippines surveys, or does it mean something else?

The interval from first union or marriage to first (live) birth is given in months by v221, as you say. It can be negative, unlike all subsequent birth intervals. If it is negative, the code is 996. What else do you need? What do you mean by "event as first live birth"?

Report message to a moderator

Re: Mothers age at birth [message #26150 is a reply to message #26143]

Sat, 11 February 2023 07:22

vkyni
Messages: 4
Registered: February 2023

Member

Thank you for getting back to me.
The PH model refers to the Cox Proportional Hazard Model. In that model, my time variable is v221, and we also need to specify the event, so my event is the first live birth. So, how can we get a variable in binary form that refers to the number of first live births a woman had?

Report message to a moderator

Re: Mothers age at birth [message #26194 is a reply to message #26150]

Wed, 15 February 2023 11:31

Janet-DHS
Messages: 888
Registered: April 2022

Senior Member

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

The term "birth" is synonymous with "live birth". A woman has had a first birth if she has had any births. The total number of births a woman has had is given by v201. There are no missing values for this variable. Women with v201=0 have had no births at all. Women with v201>0 have had a first birth. You can construct a binary variable that is 0 if v201=0 and 1 if v201>0.

Report message to a moderator

Re: Mothers age at birth [message #26953 is a reply to message #1459]

Thu, 01 June 2023 12:07

Koray
Messages: 11
Registered: June 2023

Member

Please how do I calculate the age of mother at birth in DHS VII dataset

Report message to a moderator

Re: Mothers age at birth [message #26954 is a reply to message #26953]

Thu, 01 June 2023 12:12

Trevor-DHS
Messages: 803
Registered: January 2013

Senior Member

The age of the mother at birth is already given earlier in the thread:
age of mother at birth is (b3-v011)/12, then take the integer part using the truncate function (trunc in Stata and SPSS).

Report message to a moderator

Re: Mothers age at birth [message #26955 is a reply to message #26954]

Thu, 01 June 2023 20:19

Koray
Messages: 11
Registered: June 2023

Member

How have used what is already shown in the thread but it isn't working. Kindly, post the do-file used for Liberia DHS 19-20 table 8.4.
My variables of interest are the mother's age at birth and how it was weighted.

Attachment: Screenshot 2023-06-02 081805.png
(Size: 40.73KB, Downloaded 140 times)

Report message to a moderator

Re: Mothers age at birth [message #26957 is a reply to message #26955]

Fri, 02 June 2023 10:52

Janet-DHS
Messages: 888
Registered: April 2022

Senior Member

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

It appears that you tried to calculate mother's age at the birth using the IR (women's) file. In that file, b3 occurs many times, as b3_01, b3_02, etc., for the different births the woman had. You want to use the KR file, in which there is one record per child and b3 gives the cmc of birth of that specific child. Let us know if you still have a problem.

Report message to a moderator

Re: Mothers age at birth [message #26959 is a reply to message #26955]

Fri, 02 June 2023 12:27

Trevor-DHS
Messages: 803
Registered: January 2013

Senior Member

I see now that you are interested in the perinatal mortality data, so you would need the mother's age at birth for both the live births and the stillbirths. We did not use Stata for the tabulations in the final report, so we don't have Liberia specific state code to share - the tables are produced in CSPro for the report. However, we do have generic code available in Stata from our code repository on GitHub. Go to Chap08_CM and then see the code in PMR.do that is used to produce table 8.4. This provides all of the details for how to calculate the perinatal mortality. You don't need to worry about using the KR file as Tom has suggested - that would apply if you are just looking at live births, but for perinatal mortality it is better to use the IR file.

Report message to a moderator

Re: Mothers age at birth [message #26971 is a reply to message #26957]

Sat, 03 June 2023 10:59

Koray
Messages: 11
Registered: June 2023

Member

The response has been so helpful. Thank you

Report message to a moderator

Re: Mothers age at birth [message #26972 is a reply to message #26959]

Sat, 03 June 2023 11:00

Koray
Messages: 11
Registered: June 2023

Member

This response has been useful. However, in using the weights is it possible to run chi2 test?

Report message to a moderator

Re: Mothers age at birth [message #26997 is a reply to message #26972]

Wed, 07 June 2023 19:13

Trevor-DHS
Messages: 803
Registered: January 2013

Senior Member

It is possible to use chi-squared test with weights.

Report message to a moderator

Re: Mothers age at birth [message #27003 is a reply to message #26972]

Thu, 08 June 2023 13:33

Janet-DHS
Messages: 888
Registered: April 2022

Senior Member

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

Your question does not really belong in this thread, but we will try to answer it if you can be more specific about where and why you want to calculate a chi2 test.

When doing a statistical test with DHS data, you need to take into account the design effects of clustering and stratification as well as weighting. Tests should be done with individual-level data, not with the aggregated data that is given in a typical two-way table of frequencies. You can test whether two categorical variables A and B are independent using svyset and svy: mlogit A i.B, which will produce a test statistic.

Report message to a moderator

Re: Mothers age at birth [message #27008 is a reply to message #26997]

Fri, 09 June 2023 05:07

Koray
Messages: 11
Registered: June 2023

Member

This is what I had when I tried running a chi2 test on perinatal death (weighted) and pregnancy interval. Similar response is give when I use [aw=v005/1000000]

tabulate perina_death preg_interval [iw=v005/1000000], chi2
option chi2 not allowed
r(198);

Report message to a moderator

Re: Mothers age at birth [message #27142 is a reply to message #27008]

Tue, 20 June 2023 11:22

Janet-DHS
Messages: 888
Registered: April 2022

Senior Member

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

When doing a statistical test with DHS data, you need to take into account the design effects of clustering and stratification as well as weights. You can test whether two categorical variables A and B are independent using svyset and svy: mlogit A i.B . This will produce a test statistic for the model, which is equivalent to a test of the null hypothesis that A and B are independent.

Report message to a moderator

Previous Topic:	Total number of pregnancies (abortion+stillbirths+live births)
Next Topic:	Place of Delivery

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Mon Nov 25 10:29:22 Coordinated Universal Time 2024