I'm questioning a methodology that uses self-reported numbers out of 10. Why not use empirical observations of subjects' reactions, as at least one previous tickling study has done?
It's still subjective, but subject's reactions (and the professional's judgement of it) are too.