29 August 2008

Recent Reading 1

Yesterday while the net was down at home, I did some real reading -- G. H. Hardy's A Mathematician's Apology. It's a good entry to how a serious creative mathematician looks at his subject, with some interesting general philosophy. I don't agree with everything he says; but he says it well, so I have some work to do to crystallize exactly how and why we disagree. Fun!

Not a recent reading as a classic that I've recently mentioned elsewhere. It's a very good book to read if you read ... pretty much anything. How to Lie With Statistics by Darrell Huff, illustrations by Irving Geis. The recent mention was prompted by someone who wanted to illustrate that there's no trend or not much -- by showing a graph where the vertical axis was far larger than the data range. This is a technique for lying straight out of the book. Once you've read it (it is old, 1954, so some current methods may not be mentioned, you should be able to recognize other methods of distortion.

2 comments:

John Mashey said...

Huff's book is an indispensable part of the "detecting BS" kit; from recent long Deltoid post, here are a few more:

[BES2001] Joel Best, Damned Lies and Statistics - Untangling numbers from the media, politicians, and activists, 2001 (A).

[CAP1987] Nicholas Capaldi, The Art of Deception, 1987 (C).

[HUF1954] Darrell Huff, How to Lie with Statistics, 1954 (A). Classic, indispensable, cheap.

[JON1995] Gerald Everett Jones, How to Lie with Charts, 1995 (A).

[MON1991] Mark Monmonier, How to Lie with Maps, 1991 (A).

[TUF1983] Edward Tufte, The Visual Display of Quantitative Information, 1983 (A). Most about doing it right, but Chapter 2 is about doing it wrong, and recognizing such. A truly wonderful and beautiful book, as are Tufte's later three, all of which are worth having for anyone who wants inspiration for good presentation of data.

Robert Grumbine said...

Thanks for the additional books.

I'll encourage people here to go over to John's full post at Deltoid and see both his fuller thoughts, and the rest of his reading list. http://scienceblogs.com/deltoid/2008/08/john_mashey_on_how_to_learn_ab.php

The ones I've read are good. One that I particularly like that's not there is an older book -- Martin Gardner's Fads and Fallacies in the Name of Science. Like the Science: Good, Bad, and Bogus by the same author which is on the list, it looks at things which claim to be science and then how they fail to be so. My preference for it is that it goes into the ideas, and their failure to be science, in greater detail. 1957 publication. My copy was an inexpensive edition from Dover.