The Elements of Graphing Data

Mr. Cleveland takes a topic which sounds very dry and makes it interesting with simple descriptions of methods for displaying quantitative data and illustrations of "do's" and "don'ts". It should be mandatory reading for anyone who is creating charts and graphs for widespread consumption.

Even if you've been graphing for decades or are a scientific or statistical sophisticate, this book is more valuable than you'd guess. You may know some stuff to help make your graphing better, but I bet there are many more principles, features, and techniques you simply never thought of. This book has these. But for even more incisive visualizations, you should get also Cleveland's "Visualizing Data". You'll need both books really. (There's not much overlap.) Even though making use of graphical perception principles increases the power of your graphs (the main topic of "Elements of Graphing Data"), there are even more incisive graph types you need to learn about; only a couple of these are in "Elements"; the others are in "Visualizing Data". After digesting Cleveland's two books, you will be a master data-behaviour elucidator. Once in a great while you may need the old statistical inference paradigm (test-statistics & p-values), but much more often you will be so glad you have the power of Cleveland's visualization paradigm to use instead. But again, you will need both "Elements of Graphing Data" and "Visualizing Data". Start with "Elements" though. The book reads easily, is interesting and has a bonus for those into perceptual psychology. A neato tidbit: the author's research results on graphical perception were given in part as graphs -- leading to the nifty "the medium is the message" thing. No matter what aspect of "Elements" you look at, it is simply marvelous -- all substance, and several points (not just a single point for a whole book like Tufte did in his book).

William Cleveland clearly describes how data can be presented to great effect. His description of visual perception spell out the "how to's" of graphing data. While many graphing programs are available in today's high technology environment, Cleveland's descriptions of how data can be presented into graphical format is enlightening. The book provides great examples of both superior and poor graphing presentation, focusing on how to encode graphs to allow for straightforward data analysis."The Elements of Graphiing Data" is a must for those who graph scientific data.

Tufte shows you why it's important to do graphs well. Cleveland shows you _how_.The last quarter of the book details experiments in human visual perception that rank how well we detect certain things: relative angles not on a common baseline (i.e. pie charts) justly come out at the bottom of the list.One of a only handful of books I've labelled "JXH ONLY". If I loan you my copy, know that you are special.