Robert P. Abelson (1928–2005)

This book was a life changing experience. I have no memory of how I found it, but I had it at about the time I went back to college, circa 1977. I probably just picked it up in the bookstore. It's beginnings of what became cognitive science, and the approach to artificial intelligence via social psychology and linguistics was, at the time, revolutionary. This book was where I met SAM, which appears in other Schank books as well.

Most of the concepts in this book will seem old hat to anyone show more who has been in the field for a while, or is newly arrived, but in 1977 it kept me up nights. It's still an interesting basis from which to approach the field. show less

So you understand how to manipulate data and can make fancy charts. Great. Those statistics are still just numbers on a page -- that most people skim -- unless you can turn them into useful arguments. Abelson uses the acronym MAGIC (Magnitude, Articulation, Generality, Interesting, Credibility) to describe the five components of a persuasive statistical argument. I especially like that he uses examples to illustrate his points.

While this book is light on formulas and math, you still need a show more basic understanding of statistics to get the most out of it. This book is probably most often used by grad students in the social sciences, but every journalist, salesperson, politician, and activist should own a copy, too. show less

This book wasn't quite what I thought it would be. The title led me to expect an analysis of statistical methodology in comparison to other forms of "principled argument", but it's really just a how-to-do-good-research guide. The examples are heavily tilted towards the author's own speciality, social psychology. It seems like a good supplement to theoretical textbooks for researchers who write papers containing statistical data.

"It will be of true interest and lasting value to beginning graduate students and seasoned researchers alike."

I shoulda read the blurb more carefully. I am neither of those audiences. And I've been reading about the use and abuse of statistics in more recent science books, so ok.

Nov. 2021