The DHS Program User Forum      
Discussions regarding The DHS Program data and results
Home » Data » Dataset use in Stata » Svy details Colombia 2010
Svy details Colombia 2010 [message #12783] Thu, 13 July 2017 05:16 Go to next message
mariano.salazar@ki.se is currently offline  mariano.salazar@ki.se
Messages: 1
Registered: July 2017
Member
Hi,

I am working with the aforementioned data set. One difficulty that I am experiencing is that it has been tricky to identify the variables used to define the survey structure in Stata. I am working with pregnant women who completed the domestic violence module.
According to the description in the survey report, the sample selection was done first by selecting departments, then municipalities (PSU), then then urban/rural blocks, then selection of "segments" and then people.

The variables department (SDEPTO) and municipalities (V021) as well as the sampling weights have been identified. But the other variables describing the strata (urban-rural blocks and "segments") are not easily identified.

In addition the variables containing the number of municipalities per departments, the number of blocks per municipality and the number of segments per block are difficult to identify.

I have read in detail the reports and the documentation that comes with the dataset and I have been unable to identify these variables. Can you kindly point them to me?

Kind regards,

Mariano Salazar

Re: Svy details Colombia 2010 [message #12813 is a reply to message #12783] Mon, 17 July 2017 13:27 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1073
Registered: February 2013
Senior Member
Following is a response from Chief of Data Processing, Guillermo Rojas:

Below is the definition for the geographical variables available for the survey. It is important to mention that because of confidentiality we don't include urban/rural blocks codes. If further details for the sample design are required, the user should contact Profamilia in Colombia (Rocio Murad) as we were not involved in that process.

• V001. Cluster number which are what is called "segments". Essentially these correspond to census enumeration areas
• SDEPTO : Correspond to country states (Departamento)
• SMUNIP: Municipality. This variable follows the country codification as defined by the Colombian statistical office (DANE) and it should be used in combination with SDEPTO. The Unique code for municipality is the concatenation of Departamento and Municipality (SSMMM where SS is SDEPTO and MMM is SMUNIP).
• SAREA/V025: Urban/rural classifications
• V021: Sample stratification. Essentially the aggregation of clusters (segments) where the number of households interviewed was low
• V022/V023: A combination of SDEPTO by urban/rural


Previous Topic: Variance Inflation Factor
Next Topic: Decomposition of concentration index
Goto Forum:
  


Current Time: Fri Jul 28 15:02:55 Eastern Daylight Time 2017