Andrew,
This may be of help:
Young MJ et al. Sample size nomograms for interpreting negative clinical
studies. Ann Intern Med 1983;99:248-251.
Harold Goldberg, M.D.
Associate Professor of Medicine
University of Washington
On Fri, 11 Jun 1999, Andrew Jull wrote:
> Dear all
>
> I was musing on teaching critical appraisal after leading a workshop with
> some nurses yesterday, and it struck me that rather than have to use a
> computer porgram or go to a statistician to work out a study's power (for
> articles where it is not reported), it would probably be possible for the
> mathematically literate to create nomogram with logarithmic scales of sample
> size on one side, beta or power on the other side and p-value in the centre,
> much like the scale for likelihood ratios. That way one wouldn't have to
> rely on being in front of a computer, or scientific calculator or a
> statistician. Has anybody attempted this or something similar? Or am I (a
> member of the mathematically illiterate unless substantially assisted)
> musing a bit too creatively here?
>
> regards
> Andrew Jull
> Clinical Nurse Consultant
> Auckland Hospital
> NEW ZEALAND
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|