Hacker News Reader

Comment by Rebelgecko | Hacker News Reader

That sample hiring quiz (https://kaisoapbox.com/projects/faa_biographical_assessment) is batshit insane, to the point I can hardly believe it's real (but it seems like it is?)

View on HN

I read through the list. It seems to be standard Meyers-Briggs type questions. I don't think that this is that useful a metric, but I have been given similar questions by private sector companies and was also given a weirder one by the patent office in 2006.

If you look at how they're scored, it's worse than Meyers-Briggs. Wanting a specific personality type for the job makes a lot more sense than many of these. Most don't give any points, many have labels that increase in number but the points aren't logically related to them (unemployed for 3-4 months before this job? No points for you! 1-2 or 5-6? Have a lot of points!). Even ignoring the rest of the story, it's a very flawed test.

Yeah the way the points are arbitrary makes it very clear that the intention is for only people who know the answers ahead of time to pass.

It was an underhanded way to do favoritism (by giving out the answers to the associations whose back you want to scratch) not a complex way to do equity.

I don't understand why

* applicants are heavily penalized if they took 1 music class in college (whereas 0 or 2 classes is treated as a good thing)

* Applicants are penalized if science.wasn't their worst course in high school. However they're also penalized if science was their worst course in college.

* It seems wrong that the way you heard about the job is directly used to determine whether or not you'll get it

Questions about school, personality and experience? That's... just about normal to fill out for corp jobs. It's questionable if it's useful, but it's very common, far from batshit insane.

The insanity is in the scoring for it. You should try out the quiz and see how many questions are graded in a way that you expect. Although I did pass, purely due to being bad at history in college.

I've seen worse tests. It's stupid, but it's weird how aggressive people seem about it now compared to this topic getting lots of attention in the industry over decades now. Even this year WSJ mentions astrology, numerology, graphology in business https://archive.is/ZvJKX

So why is this specific case something that seems to get people out with pitchforks? There are thousands of other cases and they should've been all laughed at until nobody proposes a personality test again. It's all bad and the attention on this one case makes me doubt people genuinely care about profiling and broken hiring in general. (Rather than joining the dei-bad bandwagon)

Because, unlike most professions, ATC is immediately, personally responsible for making decisions for which a slight mistake could instantly claim the lives of hundreds of people.

We've not heard of anyone not qualified getting a job, only not hiring enough people. This has been an issue for many many years now. On the other hand there's lots of other professions with their own profiled hiring issues that claim lives over extended periods of time. So yeah, it still smells like an artificially popular topic. Especially since many changes have already been made this year but we're still seeing the same gripes reposted - but https://www.transportation.gov/briefing-room/us-transportati... hasn't been posted even once.

They hired more people who then failed the academy and the FAA only has budget for a fixed number of seats. This resulted in a shortage of trainees making it to towers.

And the test doesn't test for that well. I passed and honestly shouldn't. Few of the questions about if I'd be a good ATC mattered much.

Because the arbitrary scoring makes it clear that it was just corruption, designed for the answers to be given to some people ahead of time, so that you could hire them instead of other people, backed by (artificial) data.

People don't like corruption. Which is good, because corruption is bad.

I had similar for KPMG!

Reply on news.ycombinator.com