4 ms·
> If you’ve got 300 applicants and 15 of them are Ivy League, you throw away the other 285 You'd be better off throwing away 285 at random. At least then you'r
by feoren 3y ago
> If you’ve got 300 applicants and 15 of them are Ivy League, you throw away the other 285
You'd be better off throwing away 285 at random. At least then you're not specifically selecting for the people who you are going to have the most competition for.
On your first pass, spend 30 seconds scanning each one and throw away any that just feel off. This is probably at least 50%, possibly a lot more. This is 2.5 hours.
On the 2nd pass, spend 2 minutes more thoroughly reading them, and pick your top 10 to 20. This is about an hour if you already threw away half.
In 3.5 hours you can get a much better pool of favorites than by just throwing away those from non top-tier schools.
- chongli 3y agoI think it’s the same reason people didn’t shop around in the 70’s/80’s, they just bought IBM. Or Microsoft in the 90’s/00’s. It’s not really about making the best possible decision, it’s the one that’s easiest to defend when the CEO comes around for a chat. “We hired the guy from Harvard” puts a smile on the face of the CEO, who went to Harvard himself. “We hired the guy from Podunk State” might not get you fired, but could be a bit awkward!