And nor should they Star. There's dozens and dozens even hundreds of potential variables you could measure, most of which aren't independant so in using them your in fact doubling up on your measurements. Barny is alluding to this I think. Of those left many won't have much effect when you calculate their likelihood ratios. With the greatest respect to Vortech, I strongly doubt there are 75 independant variables that make a measurable difference. I suspect Shaun is about right.

__________________
Pixie

"It's worth remembering that profit isn't profit until it's spent off the racecourse." -- Crash