The percentage in the gage represents the variance of cases as deviation of the average from the median

The median is calculated by:

A is array of double or whatever primitive you use for the metrics

Arrays.sort(a);

if (a.length == 0)

// no elements, just return 0

return 0;

else if (a.length % 2 == 1)

// odd-length, return middle element

return a[a.length/2];

else

// even-length, return average of middle two elements

// be sure to divide by 2.0 and not by 2!

return (a[a.length/2] + a[(a.length/2)-1])/2.0;

now suppose the median is med and average is avg

Return 100-(med/avg *100);

For the nodes popup present the distribution

X axis the PUI values

Y axis the count of PUIs falling in this value

Yaron Parasol