Over at herdingcats, Glen is unhappy with most people’s understanding of statistics. http://herdingcats.typepad.com/my_weblog/2010/12/q-2.html
If he means that Taleb has oversimplified what it means to make a estimate or a prediction, he’s right. It is simple statistical process control. Estimates have confidence intervals and predictions have prediction intervals. I see the lack of understanding when I teach PSP. I see this all the time when I teach PSP. The Student’s T is pretty basic, but most people seem to see it for the first time when I teach it.
Maybe the problem starts in high school, but I think it’s deeper. The press seldom shows any understanding at all. The constant reports, usually misrepresenting, the latest study of some hazard need some context. I usually don’t see this context. Even worse, professionals who should know better use the statistics naively.
Maybe the problem starts in high school, but it doesn’t end there. The lack of understanding of probability and statistics in the press is shocking.
You cannot be an informed person today without at least a basic understanding of probability and statistics. If you want a gentle introduction, or have a high school kid, I read the following two books from Darrell Huff in high school and I highly recommend them.