By David Salsburg
Another book as part of my statistics kick, this time much more relevant to what I wanted but not the style I was looking for. I guess you can’t get everything.
Salsburg is explicit that this book is made for those with little to no mathematical background, thus it has no equations and not even a good deal of explanation of the mathematics being developed. This was the major flaw in the book for me. I loved his detailed history of the various people developing statistics. He starts with Laplace and Bayes who laid the groundwork for this kind of thinking and then gets deep in Pearson, Fisher, Neyman, and many others who really developed the tools we use today in scientific experimentation. He talks about experiment design and hypothesis testing, the various lemmas and theorems as they were developed to deal with understanding agricultural and epidemiological and beer brewing success.
But he shies away from anything too mathematical, which was difficult for me because I didn’t know about many of the ideas these people were developing and thus gained no insight into the concepts except the high level explanations Salsburg thought were appropriate. I think I’d rather have read the book after having a more technical understanding of some of the material, such that the information in the book could have consolidated my learning instead of starting it.
Still, my main takeaway is that statistics led us to think more clearly about how to design and interpret experiments in fields where it is hard to isolate variables: agriculture, medicine, eventually molecular biology, and many others. A lot of this had to do with moving away from considering specific measurements and instead considering the distribution of the those measurement. It also led to a flourishing field of mathematics that dealt with many of the real world problems of experimentation.