Oh great! It seems we have some discrepency in what equations or some other issues that the various calculators are using.

Using http://www.geistware.com/rcmodeling/cg_super_calc.htm with your numbers I get 30.1 inches back from the front wing's leading edge for a 10% stability margin. Now that is certainly close enough to what you got from the Canard calculator. But it's a pretty big jump away for both of those from the 26.85 you got from the other one.

I'd suggest to your buddy that he makes a little 12 inch or so scale test glider from simple 1/16 sheet wings with a 1/4 inch square stick fuselage. One of the wings shoudl be adjustable for incidence angle. Then play with the CG location. What you're looking for is how far back he can get the CG and then trim the adjustable wing's incidence to get a stable glide where it doesn't try to diverge from stable flight. This means that a gentle push should glide in a flat stable glide. A hearty push should show some signs of raising the nose up to a stall. Even if it doesn't recover before it hits the ground the key is that the model MUST nose up even with the CG as far back as you can get it. When it gets to where the model does not nose up with a fast push or even arcs down then he went too far and the CG needs to move back forward a bit and the wing re-trimmed for a stable glide. This final setting is the most rearward position he should put the CG. But don't think that putting it WAY forward is good either. Like the baby bear's porridge you want to get it "just right".

It'll be interesting to see what the test glider says about all three numbers.