calculations

[simon]

Luke Scicluna wrote:

>This I would
> like to do by weighting the resulting value from the subtraction of
> the two means by the standard deviation (or error of the two means).
> Is such a calculation possible.
>         Is this clear !!!!!! I hope so statistics is not my field.
>         Thanks for any possible ideas.


I _think_ that the maths is:

If you have A +/- SD(A) and B +/- SD(A) then the standard deviation 
(SD) of A-B is:

SD(A-B) = SQRT(SD(A)^2 + SD(B)^2)

However, a second opinion would be nice....

Simon

_______________________________________________________________
Dr Simon B Easterbrook-Smith    Phone: (+) 612 351 3905
Department of Biochemistry      FAX: (+) 612 351 4726
University of Sydney            Email: sbe at biochem.usyd.edu.au
Sydney NSW 2006
AUSTRALIA
_______________________________________________________________
Every complex question has a simple, straight-forward,
easy to understand, wrong, answer.
_______________________________________________________________