This doesn't seem particularly useful, as the data set is small, and biased toward those who are willing to report their salaries on twitter. Aside from the pos

This doesn't seem particularly useful, as the data set is small, and biased toward those who are willing to report their salaries on twitter. Aside from the possibility that at least some of these reports misreport the actual salary, and likely over-state, it also seems likely that those who report salaries on twitter make above average relative to their peers.