I’m reading Daniel Kahneman’s Thinking Fast and Slow. It begins amusingly with his telling us how he and Amos Tversky discovered – during a seminar of Kahneman’s at which Tversky was a guest speaker, their first collaboration – that even statisticians are bad at intuitive statistics.
He tells us about the resemblance heuristic, and starts with a question.
As you consider the next question, please assume that Steve was selected at random from a representative sample:
An individual has been described by a neighbor as follows: “Steve is very shy and withdrawn, invariably helpful but with little interest in people or the world of reality. A meek and tidy soul, he has a need for order and structure, and a passion for detail.” Is Steve more likely to be a librarian or a farmer?
First, of course I knew the obvious answer was the wrong one and I could see that “librarian” was the obvious and therefore wrong one – but I think I would not have chosen librarian even if I hadn’t known the obvious answer was wrong. I can tell you why.
It’s because I frequent libraries a good deal, and I think about things like “what would it be like to be a librarian/farmer/acrobat?” I already know that being a librarian would not be a good fit for someone who is very shy and withdrawn, because librarians spend much of their time interacting with strangers, and besides, colleagues. I also know that farming can be very solitary and even that some people choose it for that very reason.
That’s not actually why librarian is the wrong answer; it’s because there are twenty farmers for every one librarian, and I wouldn’t have considered that at all, so I would still have been wrong, but I would have gotten the right answer for the wrong reason.
I’m a terrible intuitive statistician. I’m confident of that.