HOW TO USE DESIGNROC.for
************************
DESIGNROC.for is an interactive FORTRAN program for estimating sample size and/
or power in ROC studies involving the partial area under the curve (AUC) or
sensitivity at a fixed false-positive rate (FPR). The program also computes
sample size/power for the area under the full curve, but because it requires
estimates of parameters A and B (see below), it may not be as user-friendly as
rocpower.sas (unless estimates of A and B are readily available). This program
is for situations involving the comparison of two curves (paired or unpaired
design).

You will first need to copy the program DESIGNROC.for into your home directory.
Then you will need to issue the following statements in your command tool after
the % prompt:

source /vol/vni/ipt/bin/iptsetup.csh
f77 DESIGNROC.for -o designroc
designroc

The program will then prompt you for information which you will enter directly
into the command tool window. You will need to have estimates of parameters A
and B for each curve (under the alternative hypothesis that there is a
difference) and for both curves under the null hypothesis (i.e., AUC1=AUC2).

A is the standardized difference between the diseased and nondiseased
populations for a single curve. In other words, 

A = (meanD - meanN)/S(diseased)
where
'meanD' is the mean rating (or continuous measurement) among diseased patients
'meanN' is the mean rating (or continuous measurement) among normal patients
'S(diseased)' is the standard deviation of the rating (or conts measurement)
    among diseased patients

B is the ratio of the standard errors of the ratings (or continuous
measurements) of normal and diseased patients; specifically,

B= S(normal)/S(diseased)