The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Dataset use in Stata » Defining Strata Using Svyset: Many Datasets
Defining Strata Using Svyset: Many Datasets [message #492] Wed, 29 May 2013 16:08 Go to next message
adi.greif@yale.edu is currently offline  adi.greif@yale.edu
Messages: 5
Registered: May 2013
Member
I am looking at the following DHS datasets, and would like to use the svyset command in order to analyze subpopulations of the datasets. I understand that each survey is stratified slightly differently in the individual women's recode. So for some, the strata might be on v022, others v023 etc. Is there a list of the strata by dataset? Does anyone know which are appropriate for the following surveys? I assume that I should choose the strata by which psu's were chosen, rather than the highest level available strata, correct? (So for example, in Egypt the strata were done by urban/rural and by governorate at the highest level of stratification, but psu's were chosen based on a set of 33 lower-level areas nested within each of the urban/rural/governorate criteria).

Bangladesh2007
Benin2006"
Burkina-Faso2010
Cambodia2010
Cameroon2011
Chad2004
Comoros1996
Congo-Brazzaville2005
Cote-d-Ivoire1998
Egypt2008
Gabon2000
Ghana2008
Guinea2005
Guyana2009
Haiti2006
India1999
Jordan2007
Kenya2008
Lesotho2009
Madagascar2008
Malawi2010
Mali2006
Morocco2003
Niger2006
Nigeria2008
Pakistan2006
Senegal2010
Sierra-Leone2008
South-Africa1998
Swaziland2006
Togo1998
Uganda2011
Vietnam2002
Zambia2007
Zimbabwe2010"

[Updated on: Wed, 29 May 2013 16:50]

Report message to a moderator

Re: Defining Strata Using Svyset: Many Datasets [message #598 is a reply to message #492] Wed, 03 July 2013 18:33 Go to previous messageGo to next message
bsayer
Messages: 12
Registered: March 2013
Location: Silver Spring Maryland
Member
Just create a new stratum variable that is survey specific and then combine them. If you want extra insurance, do the same for the PSU variable. You do have to make sure the variable has the same name in all data sets.

Bryan Sayer
Statistician
Social & Scientific Systems, Inc.
Re: Defining Strata Using Svyset: Many Datasets [message #2061 is a reply to message #492] Wed, 23 April 2014 11:44 Go to previous messageGo to next message
mpkellytwo is currently offline  mpkellytwo
Messages: 2
Registered: April 2014
Location: North Carolina
Member
I am also working with Nigeria 2008 Woman's survey and was unclear if the strata is v023 or a combination of v022 and v023?

I'd appreciate it so much if you could let me know.
Re: Defining Strata Using Svyset: Many Datasets [message #2107 is a reply to message #2061] Tue, 29 April 2014 09:39 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member
Response to your post, from DHS Senior Sampling Specialist, Ruilin Ren:

For Nigeria DHS 2008 (NDHS 2008), the stratification is defined by crossing HV023 with HV025 at household level. At the individual level(women), by crossing V023 with V025 at women individual level. HV023/V023 define the states, the urban area and rural area each, is a stratum for every state.
Previous Topic: Multilevel modelling (ML)
Next Topic: Multilevel modelling (ML)
Goto Forum:
  


Current Time: Thu Mar 28 09:13:47 Coordinated Universal Time 2024