Any other ideas? If you are at the HR of Lufthansa and get a thousand resumes that you have to bring down to 20 you want to invite, how do you start?
I would put all the candidates in a Excel spreadsheet and pick 100 candidates at random. Then you have them sit an exam with questions about things the company think are important.
A better approach would be to select the number you actually want, using a random draw.
Let’s face it, who wants to employ unlucky people?
Post-Germanwings, you might look at whether the candidate makes eye contact during the interview 
I would put all the candidates in a Excel spreadsheet and pick 100 candidates at random.
But then you would also get some of those candidates, who really only could achieve 76% marks in their third or fourth attempt at the exam. They will fail the first or second question of your company test. It really would be a waste of time and money to invite them. Somewhere, a line must be drawn. And if ever one or two of these guys (they exist, I’ve met a few of them – “sub-monkey standard” really!) should ever make it into the cockpit of an airliner I will never fly as passenger again.
But then you would also get some of those candidates, who really only could achieve 76% marks in their third or fourth attempt at the exam.
Yes and that’s where we disagree (and that’s ok). I don’t think these marks are in any way correlated to the qualities that make someone a good pilot. If someone is working two jobs to pay his ATPL while another is sleeping at his parents and paying with daddy’s second mortage, then guess who will have all day to monkey train the database to reach 99.9% …
Still, showing a higher result shows you invested some effort to achieve it. Of course it is bad to just know the answers but this is a result you can present immediately, showing some dedication.
By the way, on my ATPL exams I had about 3-5 questions (out of 50-70) per subject which were not from the available databases. Especially in navigation there was one question with 5 points which was not from the database and was not close to any one I had seen. So not knowing the material meant you couldn’t get it. And I remember calculating that if I didn’t get that one right, I had to get almost all other correct to pass the 75% because 5 points was a huge amount of points on that subject.
I don’t think these marks are in any way correlated to the qualities that make someone a good pilot.
“Not fully correlated” I would rather say. There are certainly some good pilots out there who have less than perfect theoretical understanding or not the best memory or who just can’t be bothered to study air law and human factors. But do those belong in the cockpit of an airliner? As long as the airlines own those airliners, it’s their decision and we know what they base that decision on.
Kerwin wrote:
If someone is working two jobs to pay his ATPL …
He too knows that the chances of getting an airline job will depend to some degree on the marks of his exam. If he is really dedicated then he will find a way to free himself from his duties in the 6 or 8 weeks before the exam and study hard to achieve what’s expected from him. Others have done it and this is what the employers expect to see. Otherwise, his two jobs will be wasted lifetime.
Kerwin wrote:
…while another is sleeping at his parents and paying with daddy’s second mortage,…
But these boys and girls have an advantage everywhere in life, be it law school or medical school or airlines. Not much we can do about that I think.
I think the danger is in selecting the “above 90% non-monkey” types based solely on that criteria.
Most serious airlines conduct psychological tests (debatable value really), simulator tests, some form of theoretical tests, interviews (multiple) and medical tests as well.
These all form the final decision. Sure, there needs to be some form of initial selection and while it would be easy to use the “90% non-monkey” mark to do that, I sincerely believe that you lose out on some rather important qualities in people.
Let’s say you use the 90% limit. You want 20 candidates out of 800. So, with many of these doing better than 90% perhaps the bar is raised closer to 99%. These are the most super intelligent guys and gals, obviously – they achieved the highest marks, right? But, then when you get to the interviews you may find that these “nerds” are completely impossible to work with, because they are “the best” and lack any humanistic capability whatsoever. You’ll find two out of twenty that actually have some people skills as well – these should be at NASA on their way to Mars, not flying for RYR or some other shitty company – really.
The nerdy types probably don’t belong in a cockpit monitoring an autopilot for 6 hours – they should be employed as scientists, develop the world and take us to new frontiers.
Seriously, some of the “greatest” pilots in history did not belong in the “above 90% non-monkey” class initially. They got there eventually. But, they were damn good pilots.
Peter wrote:
I suppose 90% does mean the student is going to be intelligent
90% is hardly an “A” at the university, typically a good “B”. In that respect a 90% represents about the 10-30% best students by following the ECTS scale. If you have 100 applications for one job, it seems far from unreasonable to start with the 10-30 best ones for an interview.
some of the “greatest” pilots in history did not belong in the “above 90% non-monkey” class
But the airlines do not require the “greatest” pilots – too expensive, at least. Possibly too difficult personalities, on top.
The airlines need the best pilots from the cheapest. That’s the situation now and in the future even more.
