Epistemic status: Not approved by the Dependent Institutional Review Board.
The study is, simply:
- We identify a cohort, say, 5-7 year old children in Mississippi.
- We obtain informed consent and register the participants.
- We announce that in ten years, there will be a Chess tournament between the participants, with a $25 million prize pool, including a $10 million prize for the champion chess player.
We then study the prize winners demographically.
We note that as we reject the concept of an independent Institutional Review Board, all IRB comments are inliine.
The study design review board suggests there be multiple separate competitions, with overlapping geographic and birth cohort domains, and for both chess and Go.
As an outline:
- Have $1 million chess competitions in North Alabama, Central Alabama, South Alabama, and North Dakota.
- Yes we know it's annoying that "North Alabama" is part of a state, and "North Dakota" is a state. That's not our department.
- Have $1 million go competitions in South Dakota, Mississippi (whites), and Mississippi (non-whites).
- Isn't that racist?
- If we want to study race, we must act in a way that supposes it exists.
- We consider whether anyone in Mississippi will have substantial problems in determining whether to register for either the "white" or "non-white" division.
- Possibly some racist white person will try to register their child in the non-white division, believing it will make it more likely that their child will win. We will simply tell them no. If they file a high-profile lawsuit, we will make an exception.
- On the other hand, it may not be clear if the quadroon should register as "white" or "non-white".
- Some of our higher math suggests there is no stable solution with only two options, but there are stable solutions with three options. "Solution" referring to the problem of assigning people to one of two categories based on "skin color" and "stated race identity". We may be able to get an approximation that works for "every 6 year old in Mississippi", however.
- As an alternative, we can have Mississippi (boys) and Mississippi (girls).
- Isn't that transphobic?
- If it were legal, we could probably establish a good betting pool on whether this would be more controversial than the race-based division.
- Mississippi is less populated than Alabama, and doesn't need 3 divisions. North and South isn't a great way to divide Mississippi because Jackson is in the middle of the state. And the question of a non-geographical divide is interesting, anyhow.
- We leave it as an exercise to the reader to figure out how people would whine about a height-based division.
- I expect that those people are going to hate and protest the study for other reasons having nothing to do with alleged transphobia. We do not prostrate ourselves for our enemies' feelings.
- As a second alternative, we can have Mississippi (last name A-K) and Mississippi (last name L-Z).
- Have a $5 million chess competition and a $5 million go competition across the four states.
Our efforts in the field of informed consent include:
- Having a publicly-available document listing the various risks that may result from the experiment.
- Registering participants well before the time of the competition.