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

st: RE: t-test for weighted data

From   "Lachenbruch, Peter" <[email protected]>
To   "'[email protected]'" <[email protected]>
Subject   st: RE: t-test for weighted data
Date   Wed, 27 Jan 2010 10:21:54 -0800

I suppose you could use regress q19 brand [weight=weight]


Peter A. Lachenbruch
Department of Public Health
Oregon State University
Corvallis, OR 97330
Phone: 541-737-3832
FAX: 541-737-4001

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Data Analytics Corp.
Sent: Wednesday, January 27, 2010 8:54 AM
To: [email protected]
Subject: st: t-test for weighted data


Suppose I have a data set with four variables from a market research 
survey: consumer ID, brand (say two brands, A and B), Q19 (responses to 
question 19 on a 1 - 7 scale), and a weight.  I need to compare the mean 
response on Q19 for the two brands.  With unweighted data, I would 
simply use a t test - no problem.  But with the weights I have an 
issue.  The t test does not accept weights.  If I do the test by 
constructing my own t-stat, the numerator would be just the difference 
in the weighted means - no problem there.  The issue is the standard 
error for the difference in weighted means.  Does anyone know a source 
on this standard error?  Does anyone know of a function (or have one) 
for doing a t-test for weighted data?




Walter R. Paczkowski, Ph.D.
Data Analytics Corp.
44 Hamilton Lane
Plainsboro, NJ 08536
(V) 609-936-8999
(F) 609-936-3733
[email protected]

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index