4 ms·
Sure, you will miss people, and you will miss good people. But since there is (according to you) some correlation, it might act as a heuristic, employed when it
by claudius 14y ago
Sure, you will miss people, and you will miss good people. But since there is (according to you) some correlation, it might act as a heuristic, employed when it is simply not feasible to implement better tests.
- pessimizer 14y agoBut what if it's a better heuristic for finding young men than engineers?
- claudius 14y agoI would argue it is also a very good heuristic for finding Star Wars fans. Your point being?
- pessimizer 14y agoMy point being that if 80% of programmers are young men, and knowledge of Star Wars jokes (or whatever) can find young men at an 80% accuracy from a general list, but good programmers at only a 64% accuracy from a list of programmers, you should just cut out the rationalization and just hire based on sex and race. You haven't discovered a heuristic for hiring good programmers, you've discovered a heuristic for confirming your own preconceptions, and for reenforcing stereotypes.
- claudius 14y agoTake false positives into account – you will find more young men who are not engineers than Star Wars fans who are not engineers (for suitable values of engineers and personal experience). Since such false positives are extremely annoying[0] when looking for interesting people in a party, you want to minimise them first, not the false negatives. [0] A false positive takes you n minutes to discover and abandon, whereas a false negative ideally doesn’t concern you at all, assuming there are enough interesting people.