From listening to the various airlines presentations at job fairs I do recall them saying recency is big for non military and that they like people from "known quantities" aka military, regionals or large very well known flight departments.
Ironic.
Probably the most important attribute of a good pilot is his/her ability to exercise sound judgement in a fluid environment and make good decisions. Sooooooo, therefore, the airlines, allegedly filled with good pilots, abrogate their decision-making responsibility and abandon their judgement for the hassle-free (read fear-free) convenience of HR departments checking boxes as an arbitrary proxy for reality!?
I like that you put "known quantities" in quotes. Speaking only from my own experiences with many pilots from many backgrounds, I would definitely NOT make any assumptions about the consistency of product coming out of any particular aviation training or experience background. I've seen pilots with "perfect" resumés who hand flew like new students, others who made decisions like drunken despots. And I've seen pilots with "perfect" resumés who demonstrated capabilities close to perfection. It's the person, not the paper. It's the walk, not the talk. The FOI, and other reliable sources inform us that learning is a
demonstrated change in
behavior based on experience; It's not the experience itself, and it's not the paper stating that the experience allegedly took place.
So, all you HR Potentates: ... Don't Abrogate! Have 'em Demonstrate!*
Screening for hire is a lot like testing, or should be anyway, and could take a clue from testing science. The hiring process is a test seeking to measure something. So a good place to start would be to define -explicitly- what that
something is. Then design a tool to test for that something... a tool that is Comprehensive, Reliable, Valid, Discriminating (no, not
that kind), and Usable.
Nowhere in any exegeses of testing have I ever seen listed "eliminating fear" or "avoiding lawsuits" as desired testing characteristic goals.
And nowhere in any scientific research of any kind have I ever seen "reliance on vague assumptions" used as a proof of fact.
(Preemptive rebuttal to the $-based "usability" replies: Part of Usability is efficiency of grading... but NOT to the elimination or significant degradation of any of the other characteristics. I mean, if you look at Usability as ONLY efficiency, i.e dollars, you can then think of all the other characteristics as the products you are buying with your dollars. If you focus so greatly on usability (cost) that you end up with limited reliability, validity, and discrimination... then what the heck are you paying for?? It becomes kinda like my ol' Uncle Burt used to say, "Quality, Speed, or Price: Pick any Two!")
*In many cases, don't even bother with that; just give the job to the people capable of doing the job, and then fire yourselves for incompetence.