Survivorship bias in science: is individual resilience the most important quality of a good scientist?

It is clear many scientists are leaving academe, but why they leave and who ends up remaining will have substantial implications for how we will train future scientists.