The test data is unreliable and misleading because it doesn’t account for the fact that everyone’s fists are a different size and shape, meaning there is no control variable.
I propose that we find a volunteer who is willing to put their fist in everyone’s mouths. Alternatively, we could find an object with a fixed size and shape, such as a dildo or a volunteer’s left testicle, to use as our control variable.
This study is important, but as it stands our data doesn’t tell us whether any potential correlation is between mouth size, fist size, or both. We will only be able to speculate on the mechanism once we have trustworthy data, and as it stands we do not.
Thank you for your consideration, and I wish you luck in the peer review process.