I agree with Petro: time digging a standard size hole. It seems subjective, but with a little repetition I think you could easily differentiate the capabilities of different models. Small, thin, uncomfortable models that would show little resistance to penetrating the soil would be penalized because they are a PITA to use and can't remove very much material.

Dig in different materiel: sand, gravel,... and try to find some soil with consistent roots. Roots are the hardest to test for but the most common issue while in service. Inconsistency in root structure could throw off results.

Best of luck. I eagerly anticipate your results (not quite as eagerly as I have anticipated the results of digging a hole in the back country a few time though)