Data Analysis and Regression: A Second Course in Statistics

by Frederick Mosteller, John W. Tukey

Two mainstreams intermingle in this treatment of practical statistics: (a) a sequence of philosophical attitudes the student needs for effective data analysis, and (b) a flow of useful and adaptable techniques that make it possible to put these attitudes to work.

Paperback

First published January 1, 1977

Reviews

[Max Lybbert · Rating 4 out of 5]

Unlike Exploratory Data Analysis, this book covers much of what I'd expect a more modern statistics book to cover. There is some discussion of techniques that I think aren't widely used and wouldn't be in more modern books (for instance, regression techniques other than ordinary least squares), but I found those parts interesting. As implied by the subtitle ("a second course in statistics"), this book assumes the reader has some basic knowledge of statistics. Exploratory Data Analysis provides that basic knowledge, but other introductory textbooks would as well.

More modern books assume that the student will do all the arithmetic with software, and in many cases those books assume the student will use a particular program such as Excel or MATLAB. I read this book largely because I was interested in implementing software of that kind. Unfortunately, this book generally assumes that the student will do the relevant arithmetic with software and does not provide enough details to implement that software. Even so, it does present some formulas and some proofs, and if you don't plan to do the math "by hand," it should cover everything you'll need to know.

By far, my favorite part is the discussion of possible correlation between height and French fluency (clearly there shouldn't be any such correlation; but if "height" is correlated to age, and by extension "experience speaking French," there may well be, and that is something to consider before declaring that taller people speak French better).