Re: st: mean difference test.

[Steve Samuels <sjsamuels@gmail.com>]

Compare Sample B to the rest of A. This is equivalent to comparing
B with all of A. There is an example in point 1 of:
http://www.stata.com/statalist/archive/2008-09/msg00869.html

Steve

On Sep 14, 2012, at 10:22 AM, Fernando Rios Avila wrote:

Dear Statalisters,
I wonder if anyone can point out what would be the appropriate test for
testing the mean differences of two samples (A and B), when Sample B is a
subsample of Sample A.
The context for this is to evaluate the severity of missing information in
sample B.
I ve been using a simple -ttest- , but it came to my attention that it
requires independent samples, which by construction is not the case in my
problem in hand.
thank you.
Fernando
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/