The DHS Program User Forum: Dataset use in Stata » Nigeria 2008 Strata for Svyset

Home » Data » Dataset use in Stata » Nigeria 2008 Strata for Svyset

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

Nigeria 2008 Strata for Svyset [message #2062]

Wed, 23 April 2014 11:47

mpkellytwo
Messages: 2
Registered: April 2014
Location: North Carolina

Member

Hi all,

I am a young researcher working with the woman's questionnaire for DHS 2008 Nigeria. I originally used v023 to specify the strata in my svyset command, but now I'm second guessing myself. The sampling documentation indicates that EAs were stratified by urban/rural. Do I need to combine v022 and v023 to create the strata variable?

This is what I currently have:
svyset [pweight=hh_wgt], psu(v021) strata(v023)

Thanks!!!

Report message to a moderator

Re: Nigeria 2008 Strata for Svyset [message #2075 is a reply to message #2062]

Thu, 24 April 2014 15:14

Liz-DHS
Messages: 1516
Registered: February 2013

Senior Member

Here is a response from one of our sampling experts, Dr. Ruilin Ren:
Unfortunately, neither of the two variables alone, HV022 and HV023, describes correctly the sampling stratification of the Nigeria DHS 2008 (NDHS 2008). The NDHS 2008 used HV023 crossing HV025 as stratification, means, each state was stratified into two strata, an urban stratum and a rural stratum. We have some problems in the recode data files for the stratification variables HV022/V022/MV022 for surveys earlier than 2009. The content we put in the variable was often not consistent with the label of the variable. This situation has been improved in later surveys. For the case of NDHS 2008, please use Hv023xHV025 as stratification.

Thank you!

Report message to a moderator

Re: Nigeria 2008 Strata for Svyset [message #22563 is a reply to message #2075]

Thu, 01 April 2021 15:38

lbdrew
Messages: 7
Registered: September 2019

Member

Hello,
I am also using Nigeria 2008 data, but I do not see hv023 and hv025. I only see v023 and v025.
Would the correct code in stata be:

gen wgt=v005/1000000
svyset[pw=wgt], psu(v021) strata(v023xv025)

Report message to a moderator

Re: Nigeria 2008 Strata for Svyset [message #22584 is a reply to message #22563]

Wed, 07 April 2021 11:32

Bridgette-DHS
Messages: 3214
Registered: February 2013

Senior Member

The variables hv023 and hv025 will be in the Household Recode dataset - HR (not the Individual Recode dataset - IR).

Report message to a moderator

Previous Topic:	Calculating ICC
Next Topic:	Data name and data do not match

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Dec 18 23:18:36 Coordinated Universal Time 2024