I don’t have a great answer, but I just wanted to mention that this is something we (Rethink Priotities) have thought about and will probably come back to in analysing the many years of data from the EA survey. Because people are not experimentally aside an age to learn about this, it will be hard to make strong causal inference is. Still I suspect that data will lead to more insight than the absence of data. Also, everyone is playing a very long game, note that orgs/projects like Charity Elections are introducing EA ideas in high schools and could possibly do some exogenous assignment/randomisation.
I don’t have a great answer, but I just wanted to mention that this is something we (Rethink Priotities) have thought about and will probably come back to in analysing the many years of data from the EA survey. Because people are not experimentally aside an age to learn about this, it will be hard to make strong causal inference is. Still I suspect that data will lead to more insight than the absence of data. Also, everyone is playing a very long game, note that orgs/projects like Charity Elections are introducing EA ideas in high schools and could possibly do some exogenous assignment/randomisation.