Date: Fri, 15 May 1998 20:21:52 GMT
Reply-To: Richard F Ulrich <wpilib+@PITT.EDU>
Sender: "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From: Richard F Ulrich <wpilib+@PITT.EDU>
Organization: University of Pittsburgh
Subject: Re: Calculation of sample size - power
Christopher W. Hall (c_hall@PA.AUSOM.NET.AU) wrote:
: A nurse who is undertaking postgraduate palliative care nursing research
: has been asked to explain to her ethics committee how her sample size was
: determined. Her supervisor, currently overseas, calculated the sample size
: "using a computer program" and fled without explaining the process or
: rationale used. Can someone please provide a description of how sample size
: is calculated or provide a reference that might be an appropriate starting
: point. Thanks in advance for your assistance.
"How it is calculated" depends on the *test*, the comparison.
Jacob Cohen's textbook includes more discussion than most.
There are some Q&A about power in the stats FAQ which you can
find from my homepage.
Online, you can find a source of programs in my stats FAQ.
The program GPOWER is one of the nice ones, and it comes with
a 40-page document, which convinces me that the authors of
this package do know what they are doing. (That is not always
a sure thing.)
Rich Ulrich, biostatistician email@example.com
http://www.pitt.edu/~wpilib/index.html Univ. of Pittsburgh