Weighted median and range with svy data

[Grace Strachan]

Hi all,

I'm hoping this is an easy response but I can't seem to find an answer to it. I'm working with survey data and trying to find what would be the equivalent of

sum age, detail
bysort district: sum age, detail

using the svy command. I want to find the weighted median and range of age for both the full data set and by district. The sum and by sort sum accept weights

sum mat_age [w=hh_womnumb], detail
bysort district: sum mat_age [w=hh_womnumb], detail

but doesn't account for clustering. I'm not sure if this matters. Any help is appreciated!

[David Radwin]

If you need survey-adjusted standard errors or confidence intervals, your best bet (and I think your only option) is epctile by Stas Kolenikov. In my experience it usually but not always reports standard errors.

. findit epctile