Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Formula for weighted standard error

From   Steven Samuels <>
Subject   Re: st: Formula for weighted standard error
Date   Wed, 17 Nov 2010 15:40:48 -0500

However in an introductory sampling book, the word "linearized" might not appear: look for the formula for the estimated SE of the mean with clustered, stratified, and weighted sampling.

The formula is also given in any sampling book and you can find it in online class notes as well. By the way, proper spelling of our program is "Stata", not "STATA", as shown in the FAQ Section 9


On Nov 17, 2010, at 2:42 PM, Stas Kolenikov wrote:

On Wed, Nov 17, 2010 at 1:08 PM, Emre Ozaltin <> wrote:
I am running the command
"svy: mean"

using pweights

"svyset [pweight=X], psu(Y) strata(Z) singleunit(certainty)"

Can anyone tell me the formula that STATA uses in calculating the
standard error in this case (reported as "Linearized Std. Err.")

Two abbreviations come to mind: OMG and RTFM. The general formula is
in the manual, and so is the discussion of the singleton PSU options
(and how the variances are computed for those options), although you
might have to spend a minute or two locating it if it is not right
there in the entry for -svyset-.

Stas Kolenikov, also found at
Small print: I use this email account for mailing lists only.
*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index