So you want to be a political scientist?

Much of this applies to economics, other than the part about not getting a job.

It can also be "read" as a critique of those who would make strength of identification the sole metric by which to judge empirical work.