Re: st: power calculation and sample size

[Ron�n Conroy <rconroy@rcsi.ie>]

On 6 Aib 2007, at 21:01, stata_user stata_user wrote:

> I am studying the occurrence of a  disease between two groups.
> I have an annual average event rate of 1.6 % and a follow up period  
> of 5 years, I want to have a power of 80%  with alpha  5% I am  
> looking to detect  a relative risk of 3.0 associated with  group1  
> versus group 2 .
Aside from help calculating the sample size, I can tell you, as a  
member of a research ethics committee, that an 80% power is not  
adequate.

There is a simple reason for this. An 80% power gives you one chance  
in five of coming to a false negative conclusion. Such false negative  
conclusions waste research money, as well as discouraging further  
research. And, of course, wasting money is an ethical issue, since  
money wasted on a project that has a high risk of failure is money  
that cannot be used to fund other research.

Ethics committees like to see power calculations for 90% and 90%  
power. This gives them an idea of the effect size that the study can  
reasonably be expected to detect. you will notice that Stata's  
default power is 90%.

=========
Ron�n Conroy
Royal College of Surgeons in Ireland
rconroy@rcsi.ie
+353 (0) 1 402 2431
+353 (0) 87 799 97 95
http://www.flickr.com/photos/ronanconroy




*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/