My issue of Scientific American also arrived yesterday, but I won't get to it for a while because I'm behind in my magazine reading. There's a stack of magazines about three inches high on my nightstand. I think I'm up to June.
An example. There's a large college whose student population breaks down into the various departments like this:
You're told Bob is a student at this college and asked what department he's most likely in. You'll guess Business Administration and have your best chance of being right.
But what if you're additionally told that Bob is very logical and organized and not very social. Now you'll guess Math and will most likely be wrong.
Or say you're instead told that Jill is a student at this college who is very active in women's rights and wants to run for Congress someday. Now you'll guess Woman's Studies and almost certainly be wrong.
The above demonstrates the power of soft facts (in this case, some personal information whose implications are ambiguous) to cause us to abandon hard ones (statistics).
Here's another example. There are two hospitals, one large, one small. On average 100 babies are born each day at the large hospital, only 10 at the small. The sex of the babies, on average, is 50/50 male/female. You're told that on one day at one of the hospitals that 60% of the babies were female and are asked whether it is more likely it was the large hospital, the small hospital, or could equally be either one.
The answer is that it is more likely the small hospital, since larger variations from the mean are more likely with smaller sample sizes.
The book was not written with politics in mind, and I'm trying to think how it might inform our current dilemma. I think it would be a lesson we're already familiar with, that constant chaos makes it difficult for people to focus on what is truly important.