The DHS Program User Forum: Dataset use in Stata

Home » Data » Dataset use in Stata » SVY command

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

SVY command [message #10844]

Fri, 23 September 2016 11:03

Lorretta
Messages: 8
Registered: August 2013
Location: Nigeria

Member

Each time I used svy command for 2013 Nigeria DHS in a regression model, only the odds ratio or coefficients will appear, other columns will be empty with the following "Note: missing standard errors because of stratum with single sampling unit". What could be the problem? I define the svyset as svyset [pw=wt], psu(v021) strata (v022)

Lorretta

Report message to a moderator

Re: SVY command [message #10857 is a reply to message #10844]

Mon, 26 September 2016 08:11

Bridgette-DHS
Messages: 3199
Registered: February 2013

Senior Member

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

Hi Lorretta--This used to be a very annoying problem with Stata, but beginning with Stata 13, as I recall, a "singleunit" option was added to svyset. After "strata (v022)", with a space but not a comma, add any one of the following three specifications:

singleunit(centered)
singleunit(scaled)
singleunit(certainty)

I have tried all of them and there are usually differences, but very small differences, between the three options. I usually use the first one but don't have very good reasons for doing that. Maybe someone else can give a reason for picking one over the others.

Thanks!

Report message to a moderator