I agree that the usability research should be transparent and data-driven. As a statistician, I'd rather have something I can actually critique instead of "we just didn't think it was user-friendly."