The DHS Program User Forum: Other countries » Honduras svyset variables

Home » Countries » Other countries » Honduras svyset variables

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

Honduras svyset variables [message #2661]

Sun, 03 August 2014 08:49

ABLR
Messages: 10
Registered: August 2014

Member

Greetings,
I'm working with a number of DHS datasets including Honduras 2011, and I'm having a hard time figuring out which variables to use for svyset command in Stata (other than the weight) when working with the individual data for women and men. My Spanish is not that great, and Google Translate is not helping as much as I'd like.

Is this correct?
cluster: v021 (identical to v001)
strata: v023? or do I need to group v024 and v025?
weight: v005/1000000

Thank you in advance!

Report message to a moderator

Re: Honduras svyset variables [message #2669 is a reply to message #2661]

Mon, 04 August 2014 09:51

Bridgette-DHS
Messages: 3199
Registered: February 2013

Senior Member

Following is a response from Senior DHS Specialist, Tom Pullum:

For the weights, you can use either hv005 or hv005/1000000. If you try both of them, you will see that the results will be the same no matter which you use. Stata automatically normalizes the weights to have a mean of 1.

For the clusters, yes, you can use hv001 or hv021. They are exactly the same.

For the strata, use something like "egen strata=group(hv023 hv025)". You would get virtually the same thing with "egen strata=group(hv024 hv025)". hv023 and hv024 differ only in that hv023 pulls out two specific urbanized areas within two provinces that are combined with the other urban parts of those two provinces in hv024.

Report message to a moderator