Definitely more context, but I don't think that paper really backs up the assertion. He asked finance professionals or students, not social scientists. And he asked them at a seminar where I'm guessing they didn't have access to a calculator or computer, so it looks like a lot of them just used a rule of thumb.

The question they asked reads like a trick question to me. And coming up with the right answer would have required working backwards through statistical formula It's the type of thing you do an "Introduction to Statistics" class and then don't think about too often because it doesn't have a ton of practical applications. It seems a lot easier to just ballpark it.