The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Sampling » Stratification
Re: Stratification [message #1472 is a reply to message #1467] Mon, 03 March 2014 12:12 Go to previous messageGo to previous message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 795
Registered: January 2013
Senior Member
The implicit stratification is an ordering that has been made prior to selection of the clusters, and it is that ordering that permits the implicit stratification. However, I would suggest using the explicit strata defined at the time of the sample design - that is what DHS does now. The strata are generally the urban and rural areas in each region, but not always. Read appendix A of the final reports to confirm what is used for each survey.

I would not say that the sampling errors are incorrect or inadvisable. With using the implicit stratification, the assumption is that there is some improvement over simply randomly selecting the sample in a strata, thus the sampling errors tend to be slightly smaller and the confidence intervals slightly narrower. However, DHS has decided to be more conservative in its estimates of sampling error and confidence intervals and using the original design strata, you will get slightly wider confidence intervals. Of course, the difference is typically tiny and frequently negligible.
 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Specific description of the data
Next Topic: Two or Three Sampling Stages
Goto Forum:
  


Current Time: Tue May 14 09:40:48 Coordinated Universal Time 2024