[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Re: statalist-digest V4 #1153
Quoting statalist-digest <firstname.lastname@example.org>:
> statalist-digest Sunday, February 9 2003 Volume 04 : Number
> Paul O'Brien wrote:
> Can I calculate the sample size for an equivalence trial with binary
And Philip Ryan replied with a useful program,
ending with the warning
> Remember this simple program uses normal approximations, so when dealing
> with proportions less than about 0.2 you are on shaky ground.
to which Paul replied
> Unfortunately I am dealing with proportions of the order of 0.01.
In fact, Philip's warning may be over-cautious.
For rare events, both the Normal approximation and the
power depend more on the number of events than on the
proportion. You are approximating more & more closely
to a Binomial distribution. For the numbers of events
necessary for an equivalence study, Normal approximation is
not going to be a problem.
Using sampsi with 90% power:
p1 p2 n1 e1
.1 .2 286 28.6
.01 .02 3300 33.0
.001 .002 33444 33.4
.0001 .0002 334875 33.5
Paul T Seed (Please note change of e-mail address)
Lecturer in Medical Statistics
Maternal & Fetal Research Unit,
Department of Obstetrics & Gynaecology,
GKT School of Med., KCL.
10th Floor, North Wing,
St Thomas' Hospital, Lambeth Palace Road,
London SE1 7EH
tel (+44) (0) 20 7960 5889 (except Thursdays)
fax (+44) (0) 20 7620 1227
* For searches and help try: