Date:         Wed, 9 Apr 2003 19:13:40 +0000
Reply-To:     alejandro.jaramillo@ATT.NET
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         alejandro.jaramillo@ATT.NET
Subject:      Re: sampling and control groups (fwd)

The formula you are referring to is to do point estimates of a proportion. With
that formula you calculate the sample size to estimate the proportion of
something in a population.

To do a sample size calculation to compare a treatment group against a control
group you should define what you are comparing means, proportion or something
else. If you are testing a hypothesis between the test and control group, you
need to do a sample calculation that takes in consideration th significance and
power level of your test.

If you are interested in estimating the difference between the two groups then
doing a sample size estimation at a particular significance and margin of error
should be fine.

Refer to Sample size Methodology by M. M. Desu and D. Raghavarao, published by
Academic press. For the point estimates refer to Sampling Thechniques by
Cochran, published by Wiley.

Hope this helps,

Alejandro