The DHS Program User Forum: Dataset use in Stata » Extracting Zimbabwe households with at least 1 birth in last 2 years

Home » Data » Dataset use in Stata » Extracting Zimbabwe households with at least 1 birth in last 2 years

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

Extracting Zimbabwe households with at least 1 birth in last 2 years [message #2718]

Thu, 14 August 2014 19:10

swdalb
Messages: 5
Registered: August 2014

Member

Hi everyone,

I am working with the ZWBR62DT.dta data - so the Zim 2011 DHS STATA data - and I want to extract data on contraceptive use from any woman in a household where at least one woman in that household has had a birth in the 2 years prior to the survey. The only available data for Zim 2011 is BR, KR, GE and HR, so I'm not sure which one I should use to get all of this data. I don't think that the BR dataset contains any information on women in households that haven't had births because... this is a birth level dataset, but I'm also not sure that any of the other datasets have the variables I'm interested in (contraceptive use / modern contraceptive use). Could someone please advise me on (1) which dataset to use and (2) how to extract this information? Thanks so much.

-Sean

Report message to a moderator

Re: Extracting Zimbabwe households with at least 1 birth in last 2 years [message #2722 is a reply to message #2718]

Fri, 15 August 2014 01:49

Trevor-DHS
Messages: 803
Registered: January 2013

Senior Member

Hi Sean,
1) First you need to decide on what your unit of analysis will be - that usually determines the dataset to use. It sounds like you will be analyzing women, so you should use a file that has women as the unit of analysis, and you are right that the BR dataset is the wrong file - you should use the IR file.
2) Second you need to select which women to include in your analysis. If you are interested in analyzing women who gave birth in the last two years, then you need to look at the date of the last birth compared to the date of interview. v008 is the century month code (cmc) of the date of interview and b3_01 is the cmc of the date of the last birth, thus you should women with the following if clause:

if v008 - b3_01 < 24

Report message to a moderator

Re: Extracting Zimbabwe households with at least 1 birth in last 2 years [message #2726 is a reply to message #2722]

Fri, 15 August 2014 11:34

swdalb
Messages: 5
Registered: August 2014

Member

Hi Trevor,

Thanks for the response. Good to know about the CMC data too - I had been conditioning on day / month / year of interview and birth day, so it was needlessly complicated it seems. Thanks again!

-Sean

Report message to a moderator

Previous Topic:	Replicating vaccination rate
Next Topic:	survey set preparation for Pakistan DHS 2012-13

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Mon Nov 25 00:59:42 Coordinated Universal Time 2024