Assuming a Gaussian distribution, the critical difference for 80%
probability is 1.81.
This is derived by looking up the z value in a Gaussian distribution within
which 80% of the values lie (1.28 according to my table) and then
multiplying by the square root of two to account for the difference between
duplicate measurements.
Best wishes
Gordon Challand
----------
> From: Samuel Vasikaran <[log in to unmask]>
> To: [log in to unmask]
> Cc: [log in to unmask]
> Subject: Critical Difference
> Date: 05 January 1999 09:15
>
> A change in an analyte of 2.77 times the CV would be considered
> significant with 95% confidence. What change is needed for 80%
> confidence?
> Can anyone enlighten me please?
>
> Sam Vasikaran
> Royal Perth Hospital
> Western Australia
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|