Causation can sometimes not be proven with all the data points (full population). Sometimes statistics are not enough to prove causation.

Anyways, I like your analysis!