Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

st: Puzzling sampsi calculations when power is very low


From   Sam Richardson <samr@stanfordalumni.org>
To   statalist@hsphsun2.harvard.edu
Subject   st: Puzzling sampsi calculations when power is very low
Date   Wed, 15 May 2013 13:26:46 -0500

I'm using -sampsi- in Stata 12.1 to calculate power for a two-sided
comparison of proportions when I know the power is extremely low, and
I'm getting very strange results. Can anyone please explain the
following?

sampsi .05 .045, n1(300) n2(300)

This gives power of 0.0312, which is less than alpha. How is this
possible? (Is Stata only calculating the power to reject the null
hypothesis on the "correct" side where we conclude p2 is significantly
less than p1?)

The second puzzle is this:

sampsi .05 .049, n1(300) n2(300)

This gives power of 0.0338, which is higher than the previous power,
despite the effect size being smaller. How is this possible?

I'm using:
Stata/IC 12.1 for Mac (64-bit Intel)
Revision 20 Mar 2013

Thank you!
~Sam
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index