Not nearly as enjoyable as Innumeracy. Paulos starts off describing the relationship between stories and statistics. Anecdotes and examples abound showing the danger of applying one without the other, and he explains how statistics grew out of narrative. As the book continues, discussions that link numbers and narrative become increasingly abstract, and I found I lost interest. In addition, I found the arguments to be fairly short, with insufficient linkage – there was too much jumping around.
The straw that broke the camel’s back, however, was the appalling copyediting. There were a number of glaring errors – the most innocuous, simple typos, but on more than one occasion, a word was mis-spelled or -written so as to invert the meaning of a sentence. For example, there are a kind of logical or probability tree used in statistic that allow one to plug in probabilities and arrive at an overall probability of a certain situation. Software packages exist to help one manage these trees. Of course, if garbage inputs are provided, the result is worse than useless. In making this point, though, the book claims that the software packages that work with these trees can do no more than ensure that the trees are internally inconsistent!
Sheesh.